Holt Software WinOOT, MacOOT, DOS OOT, Unix X OOT, TOOT Programming Language Environment Reference Manual

OBJECT ORIENTED TURING
REFERENCE MANUAL
Seventh Edition
Version 1.0
R. C. Holt
with
Tom West
Holt Software Associates Inc.
203 College Street, Suite 305
Toronto, Ontario M5T 1P9
www.holtsoft.com
(416) 978-8363
ISBN 0-921598-35-1
All right reserved. No part of this book may be
reproduced in any way, or by any means, without
permission of the author.
© 1999 by the authors
ISBN 0-921598-35-1
Printed in Canada
For more information about Turing such as
textbooks, resources and additional materials, check
Holt Software's web site located at:
http://www.holtsoft.com/turing
Table of Contents
CHAPTER ONE : INTRODUCTION
1
CHAPTER TWO : OOT FOR WINDOWS
9
Language Evolution
Other Material about Turing
Two Main Implementations: Turing and OOT
Using Turing and OOT for Teaching
Design of the Language
Dirt and Danger
Implementing Turing and OOT
Basic Concepts in Turing
Compile-Time Expressions
Documentation of Language Features
OOT Predefined Units
Changes to OOT
Installing OOT for Windows
Minimum System Requirements
Installing Under Windows 3.1
De-installing Under Windows 3.1
Enabling Virtual Memory Under Windows 3.1
Changing Monitor Resolution Under Win 3.1
Installing Under Windows 95
MS Windows Drivers for WinOOT
Using OOT for Windows
Starting WinOOT
WinOOT Environment
OOT Windows - Control Panel
OOT Windows - Directory Viewer
OOT Windows - Editor Window
OOT Windows - Error Viewer
OOT Windows - Run Window
OOT Windows - Debugger Window
Editing Text - Keyboard Editing Commands
Editing Text - Selection, Cutting and Pasting
Editing Text - Searching and Replacing Text
Running Programs - Input and Output Redirection
Running Programs - Programs with Multiple Parts
Menu Commands - Control Panel
Menu Commands - Directory Viewer
Menu Commands - Editor Window
Debugging - Stepping and Tracing
Debugging - Showing Variables
Tutorial
1
1
2
2
3
3
4
4
6
6
7
8
9
9
9
12
12
12
14
15
16
16
17
18
20
21
22
23
24
26
27
27
29
31
32
34
36
40
42
45
Table of Contents
i
CHAPTER THREE : OOT FOR MACINTOSH
49
CHAPTER FOUR : OOT FOR DOS
81
Installing OOT for Macintosh
Minimum System Requirements
Installing Under Macintosh
De-installing OOT for Macintosh
Using OOT for Macintosh
Starting MacOOT
MacOOT Environment
MacOOT Windows - Editor Window
MacOOT Windows - Run Window
MacOOT Windows - Debugger Window
MacOOT Windows - Debugging Controller
MacOOT Windows - Error Viewer
Editing Text - Keyboard Editing Commands
Editing Text - Selection, Cutting and Pasting
Editing Text - Searching and Replacing Text
Running Programs Input and Output Redirection
Running Programs Programs with Multiple Parts
Menu Commands
Debugging - Stepping and Tracing
Showing Variables
Tutorial
Installing OOT for DOS
Minimum System Requirements
Installing Under DOS
De-installing Under DOS
Using OOT for DOS
Starting DOS OOT
Selecting a Menu Item in DOS OOT
Loading a Program in DOS OOT
Editing a Program in DOS OOT
Saving a Program in DOS OOT
Running a Program in DOS OOT
To Exit DOS OOT
OOT for DOS Editor Reference
The Menu Commands:
The Window Commands
Selecting Text
Mouse Movement
Dialog Boxes
The Pick File
Start Up Options
Compilation to Executable File
Input and Output Redirection
ii Object Oriented Turing Reference Manual
49
49
49
52
53
53
53
55
56
57
59
60
61
62
62
64
64
66
66
67
72
74
77
81
81
81
85
86
86
87
88
90
92
93
93
94
94
99
99
100
101
114
115
119
119
Summary
120
CHAPTER FIVE : THE GUI MODULE
123
CHAPTER SIX : IMPORTING GRAPHICS INTO TURING
165
Introduction
Terms
Example Programs
General Principles of the GUI Library
Active and Passive Widgets
Keyboard Shortcuts
Background Color
Widget Sizes
Positioning Text Labels (Aligning Labels with Widgets)
Canvases
Multiple Windows
The GUI Library Internals
GUI Module Routines Summary
Widgets - Common Routines
Widgets - Buttons
Widgets - Check Boxes
Widgets - Radio Buttons
Widgets - Picture Buttons
Widgets - Picture Radio Buttons
Widgets - Sliders
Widgets - Scroll Bars
Widgets - Canvases
Widgets - Text Fields
Widgets - Text Boxes
Widgets - Lines
Widgets - Frames
Widgets - Labelled Frames
Widgets - Labels
Widgets - Pictures
Widgets - Menus
Widgets - General Routines
Introduction
Importing BMP Images in WinOOT
Pic.ScreenLoad
Pic.ScreenSave
Importing PICT Images in MacOOT
Pic.ScreenLoad
Pic.ScreenSave
Importing PCX Images in DOS OOT
PCX2TM2
Pic.ScreenLoad
Pic.ScreenSave
123
125
125
125
127
130
130
131
132
133
135
136
138
141
142
143
145
146
147
148
150
152
154
155
156
157
158
159
160
161
162
165
165
166
167
167
167
168
168
168
170
171
Table of Contents
iii
CHAPTER SEVEN : LANGUAGE FEATURES
abs absolute value function
addr address of a variable
addressint type
all all members of a set
and operator
anyclass the ancestor of all classes
arctan arctangent function (radians)
arctand arctangent function (degrees)
array type
assert statement
assignability of expression to variable
assignment statement
begin statement
bind declaration
bits extraction
body declaration
boolean true-false type
break debugger pause statement
buttonmoved has a mouse event occurred
buttonwait get a mouse event procedure
case selection statement
catenation (+) joining together strings
ceil real-to-integer function
char type
char(n) type
cheat type cheating
checked compiler directive
chr integer-to-character function
class declaration
clock millisecs used procedure
close file statement
cls clear screen graphics procedure
collection declaration
color text color graphics procedure
colour text colour graphics procedure
colorback background color procedure
colourback background colour procedure
comment remark statement
comparisonOperator
Concurrency
Concurrency.empty
Concurrency.getpriority
Concurrency.setpriority
iv Object Oriented Turing Reference Manual
173
159
160
160
161
162
163
164
164
165
168
169
170
171
172
172
174
175
176
178
180
181
182
183
184
185
187
189
189
190
195
195
196
197
198
198
199
199
200
200
201
202
202
203
Concurrency.simutime
condition declaration
Config
Config.Display
Config.Lang
Config.Machine
const constant declaration
constantReference use of a constant
cos cosine function (radians)
cosd cosine function (degrees)
date procedure
declaration create a variable
deferred subprogram declaration
delay procedure
Dir
Dir.Change
Dir.Close
Dir.Create
Dir.Current
Dir.Delete
Dir.Get
Dir.GetLong
Dir.Open
div integer truncating division operator
Draw
Draw.Arc
Draw.Box
Draw.Cls
Draw.Dot
Draw.Fill
Draw.FillArc
Draw.FillBox
Draw.FillMapleLeaf
Draw.FillOval
Draw.FillPolygon
Draw.FillStar
Draw.Line
Draw.MapleLeaf
Draw.Oval
Draw.Polygon
Draw.Star
Draw.Text
empty condition function
enum enumerated type
203
204
206
206
207
208
209
210
210
211
211
212
213
214
215
216
216
217
218
218
219
220
222
223
223
224
225
226
226
227
228
229
230
231
232
233
234
234
235
236
237
238
240
240
Table of Contents
v
enumeratedValue enumerated value
eof end-of-file function
equivalence of types
erealstr real-to-string function
Error
Error.Last
Error.LastMsg
Error.LastStr
Error.Msg
Error.Str
Error.Trip
ErrorNum
Exceptions
exit statement
exp exponentiation function
explicitCharConstant character literal
explicitConstant literal
explicitIntegerConstant integer literal
explicitRealConstant real literal
explicitStringConstant string literal
explicitTrueFalseConstant boolean literal
expn expression
export list
external declaration
false boolean value (not true)
fetcharg fetch argument function
File
File.Copy
File.Delete
File.DiskFree
File.Exists
File.Rename
File.Status
flexible array initialization
Font
Font.Draw
Font.Free
Font.GetName
Font.GetSize
Font.GetStyle
Font.Name
Font.New
Font.Sizes
Font.StartName
vi Object Oriented Turing Reference Manual
241
242
243
244
245
246
246
247
248
249
249
250
250
250
251
252
252
253
254
255
256
257
258
260
261
261
262
263
264
264
265
266
267
268
270
271
272
273
273
274
275
275
278
279
Font.StartSize
Font.Width
for statement
fork statement
forward subprogram declaration
frealstr real-to-string function
free statement
function declaration
functionCall
get file statement
getch get character procedure
getchar get character function
getenv get environment function
getpid get process id function
getpriority function
GUI
GUI.AddLine
hasch has character function
id (identifier) name of an item in a program
if statement
#if used for conditional compilation
implement clause
implement by clause
import list
in member of a set
include source files
index find pattern in string function
indexType
indirection operator (@)
infix operator
inherit inheritance clause
init array initialization
Input
Input.getch
Input.getchar
Input.hasch
Input.Pause
int integer type
intn n-byte integer type
intreal integer-to-real function
intstr integer-to-string function
invariant assertion
Joystick
Joystick.GetInfo
279
280
281
283
284
285
286
287
288
289
293
294
295
295
296
296
296
393
394
394
396
397
399
399
401
402
403
404
404
405
407
409
410
411
412
413
413
414
415
416
416
417
418
418
Table of Contents
vii
Limits
ln natural logarithm function
locate procedure
locatexy graphics procedure
loop statement
lower bound
Math
max maximum function
maxcol maximum column function
maxcolor graphics function
maxint maximum integer function
maxnat maximum natural number function
maxrow maximum row function
maxx graphics function
maxy graphics function
min minimum function
minint minimum integer function
minnat minimum natural number function
mod modulo operator
module declaration
monitor declaration
Mouse
Mouse.ButtonChoose
Mouse.ButtonMoved
Mouse.ButtonWait
Mouse.Hide
Mouse.Show
Mouse.Where
Music
Music.Play
Music.PlayFile
Music.Sound
Music.SoundOff
named type
nargs number of arguments
nat natural number type
natn n-byte natural number type
natreal natural number to real function
natstr natural-number-to-string function
420
421
421
422
423
424
424
425
426
426
427
427
428
428
429
430
430
431
432
433
436
439
439
440
442
444
445
446
447
447
448
449
450
451
451
452
453
454
454
455
Net
All subprograms in the Net unit are exported qualified (and thus must be prefaced
with "Net.").
455
nil pointer to a collection
468
viii Object Oriented Turing Reference Manual
not true/false (boolean) operator
objectclass of a pointer
opaque type
open file statement
or operator
ord character-to-integer function
palette graphics procedure
parallelget parallel port function
parallelput parallel port procedure
paramDeclaration parameter declaration
pause statement
PC
PC.InPortWord
PC.Interrupt
PC.InterruptSegs
PC.OutPort
PC.OutPortWord
PC.ParallelGet
PC.ParallelPut
PC.SerialGet
PC.SerialHasch
PC.SerialPut
pervasive declaration modifier
Pic
Pic.FileNew
Pic.Free
Pic.New
Pic.Save
Pic.ScreenLoad
Pic.ScreenSave
play procedure
playdone function
pointer type
post assertion
pre assertion
precedence of operators
pred predecessor function
prefix operator
procedure declaration
procedureCall statement
process declaration
program a (main) program
put statement
quit fail statement
469
469
470
470
473
474
474
475
476
477
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
495
495
496
498
499
500
501
501
504
504
505
507
507
508
510
511
512
513
515
Table of Contents
ix
Rand
rand random real number procedure
Rand.Int
Rand.Reset
Rand.Seed
Rand.Set
randomize procedure
read file statement
.c.read file statement
real the real number type
realn n-byte real number type
realstr real-to-string function
record type
register use machine register
rem remainder operator
repeat make copies of string function
result statement
return statement
RGB
RGB.AddColor
RGB.GetColor
RGB.maxcolor
RGB.SetColor
scalar type
seek (file) statement
self pointer to current object
separator between tokens in a program
serialget serial port function
serialput serial port procedure
set type
setConstructor
setpriority procedure
setscreen graphics procedure
shl shift left operator
shr shift right operator
sign function
signal wake up a process statement
simutime simulated time function
sin sine function (radians)
sind sine function (degrees)
sizeof size of a type
sizepic graphics function
skip used in get statement
skip used in put statement
x Object Oriented Turing Reference Manual
516
517
518
520
521
521
522
523
523
524
525
526
527
528
528
529
529
530
531
531
532
533
534
535
536
537
537
538
539
540
541
541
542
545
545
546
546
547
548
548
549
549
550
551
sound statement
551
Sprite
552
All subprograms in the Sprite unit are exported qualified (and thus must be
prefaced with "Sprite.").
552
standardType
560
statement
560
statementsAndDeclarations
562
Str
562
Stream
563
Stream.Flush
564
Stream.FlushAll
565
string comparison
565
string type
566
strint string-to-integer function
567
strintok string-to-integer function
568
strnat string to natural number function
568
strnatok string to natural number function
569
strreal string-to-real function
569
strrealok string-to-real function
570
subprogramHeader
570
subprogramType
571
subrangeType
572
substring of another string
573
succ successor function
574
Sys
575
Sys.FetchArg
576
Sys.GetEnv
577
Sys.GetPid
577
Sys.Nargs
578
sysclock millisecs used procedure
579
system statement
579
tag statement
581
takepic graphics procedure
582
tell file statement
583
Text
584
Text.Cls
585
Text.Color
585
Text.Colour
585
Text.ColorBack
586
Text.ColourBack
586
Text.Locate
587
Text.LocateXY
588
Text.maxcol
589
Text.maxrow
589
Table of Contents
xi
Text.WhatChar
Text.WhatCol
Text.WhatColor
Text.WhatColour
Text.WhatColorBack
Text.WhatColourBack
Text.WhatRow
Time.DateSec
Time.Delay
Time.Elapsed
Time.ElapsedCPU
Time.PartsSec
Time.Sec
Time.SecDate
Time.SecParts
time time of day as a string procedure
token in input
true boolean value (not false)
TypeConv
type declaration
typeSpec type specification
unchecked compiler directive
union type
unit file containing module, monitor, or class
unqualified export
upper bound
var declaration
variableReference use of a variable
View
View.ClipAdd
View.ClipOff
View.ClipSet
View.GetActivePage
View.GetVisiblePage
View.maxcolor
View.maxcolour
View.maxx
View.maxy
View.Set
View.SetActivePage
View.SetVisiblePage
View.WhatDotColor
Text.WhatDotColour
wallclock seconds since 1/1/1970 procedure
xii Object Oriented Turing Reference Manual
590
591
591
591
592
592
593
595
596
596
597
598
598
599
600
600
601
602
602
603
604
605
606
607
609
609
610
611
612
613
614
614
615
616
617
617
618
618
619
622
623
624
624
626
whatcol cursor position function
whatcolor text color graphics function
whatcolorback color of background function
whatdotcolor graphics function
whatpalette graphics function
whatrow cursor position function
whattextchar graphics function
whattextcolor graphics function
whattextcolorback graphics function
Window
Window.Close
Window.GetActive
Window.GetPosition
Window.GetSelect
Window.Hide
Window.Open
Window.Select
Window.Set
Window.SetActive
Window.SetPosition
Window.Show
write file statement
xor exclusive "or" operator
APPENDICES
Appendix A
Appendix B
Appendix C
Appendix D
Appendix E
Appendix F
Appendix G
Appendix H
626
627
627
628
629
629
630
631
631
632
633
634
634
635
636
636
637
638
639
640
641
641
642
Predefined Subprograms and Keywords
OOT Predefined Subprograms
Turing Predefined Subprograms
Operators
File Statements
Control Constructs
IBM PC Keyboard Codes
Turing (IBM PC) Character Set
693
696
750
775
778
779
780
782
Table of Contents
xiii
Acknowledgments
T.L. West has added extensive materials to this manual since the original edition.
C. Stephenson has greatly contributed in her role as editor of the manual. Brenda Kosky
created the cover design.
J.R.Cordy, the co-designer of the Turing Language, is acknowledged as the source of
much of the material in this Manual. S.G. Perelgut contributed the Introductory section of
this manual.
Technical Questions
Any technical problems or questions should be directed to our technical support line at
Holt Software Associates. The number for this support line is (416) 978-8363 or
1-800-361-8324. Holt Software's web site, containing a wealth of Turing information
including free downloadable resources, is located at:
http://www.holtsoft.com/turing
xiv
Object Oriented Turing Reference Manual
Legend
On the upper right of each predefined subprogram is a little symbol that describes
whether the subprogram is available under a version of Object Oriented Turing. The
versions of OOT referred to here are:
Win
DOS
Mac
X
TOOT
Term
Object Oriented Turing for Windows version 3.0
Object Oriented Turing for DOS version 2.5
Object Oriented Turing for Macintosh version 1.5
Object Oriented Turing for X Windows version
Command line version Object Oriented Turing version
This version of Object Oriented Turing does not exist at the time of this
writing. It is included for completeness and refers to a version written for
ASCII terminals under UNIX.
The symbols are as follows:
This feature is currently implemented.
This feature is not implemented at the time of writing. Check the release
notes provided with your software to see if this feature has been
implemented.
This feature is implemented, but there are restrictions to its use. Consult
the subprogram description for more details.
This feature will not be implemented. This usually applies to predefined
subprograms that don’t make sense for a particular environment (such as
the Window subprograms in a non-window environment such as DOS
OOT, or a pixel graphics subprogram under a command line environment
such as TOOT).
Implemented
in WinOOT
Implemented
in DOS OOT
To Be
Implemented
in a Future
Version of
MacOOT
Win DOS Mac
X TOOT Term
Implemented
in X OOT with
Restrictions
(See Text)
Will Not Be
Implemented
in TOOT
Will Not Be
Implemented
in Terminal
OOT
Acknowledgements
xv
Chapter 1
Introduction
Object Oriented Turing is a high-level programming language that is easy to learn. This
Manual has been written to provide easily accessible answers to technical questions about
the language. It includes a list of the language features arranged in alphabetic order.
This Manual is directed to readers who have already been introduced to Turing or
know a related language, such as Pascal, Basic or C and have some familiarity with
operating systems such as UNIX.
While there are two languages currently available, Turing and Object Oriented Turing,
this manual will, in most cases, abbreviate Object Oriented Turing to Turing. When
referring to the original language Turing, it will use "Original Turing".
Language Evolution
Original Turing, as originally designed in 1982, is a Pascal-like language. It has been
used extensively for teaching from grade 8 through graduate school. Original Turing as
extended with graphics and binary file input/output is currently supported on IBM PC
compatibles using MS-DOS and Apple Macintoshes. Object Oriented Turing (often
abbreviated OOT or just Turing), was developed in the late 80's. It runs on IBM PC
compatibles running Microsoft Windows or MS-DOS, Apple Macintoshes and UNIX/Linux
under X Windows. Object Oriented Turing extends Original Turing with two sets of
features. The first includes concurrency as well as systems programming features that allow
the language to be used in place of C. The second is the object oriented features, which
include classes, inheritance and polymorphism.
Other Material about Turing
Holt Software Associates Inc. also publishes J.N.P. Hume's Turing Tutorial Guide and
Problem Solving and Programming in Turing which provide a beginner level introduction to
the language using many examples. Advanced books on the language include An
Introduction to Computer Science Using the Turing Programming Language, Programming:
Concepts and Paradigms, and Data Structures: An Object-Oriented Approach. For more
information on books published by Holt Software, check Holt Software's web site
http://www.holtsoft.com/turing.
For the computer science expert, the book The Turing Programming Language: Design and
Definition [Prentice-Hall publishers] provides detailed information about how and why
Original Turing was designed. That book also contains both the Turing Report, which is the
official defining document for the Original Turing language, and the formal (mathematical)
definition of the language. The technical article The Turing Programming Language,
appearing the Communications of the Association for Computing Machinery in December
1988, provides the best overview of Original Turing and its extension to Turing Plus.
Two Main Implementations: Turing and OOT
There are two main implementations of Turing, one called Original Turing that runs on
IBM PC compatibles and Apple Macintoshes, and the other called Object Oriented Turing
(OOT) that runs on UNIX under X Windows, and IBM PC compatibles running Microsoft
Windows or DOS and the Apple Macintosh. Original Turing proper supports graphics
features and binary input/output (read, write, etc.). OOT supports all the features of
Original Turing as well as concurrency and systems programming features and object
oriented features (classes, inheritance and polymorphism).
There are sets of features that are available in some versions of Original Turing but not
in others, as annotated by x's in the following table. With the descriptions in this Manual,
there are annotations such as [Pixel graphics only] that give the feature set to which the
description belongs. This table lists an additional version: TOOT (Terminal OOT, a
command line form of OOT available only in UNIX).
FEATURE SET
VERSION
WinOOT
MacOOT
DOS OOT
Unix X OOT
TOOT
Char
Graphics
x
Pixel
Graphics
x
x
x
x
Sound
UNIX
OOT
x
x
x
x
x
x
x
x
x
x
Using Turing and OOT for Teaching
Original Turing and OOT are ideal for teaching introductory Computer Science
concepts and good programming habits. OOT is also ideal for teaching higher level
concepts such as object oriented programming, programming in the large and concurrency.
The concurrency and systems programming features of OOT are ideal for teaching systems
programming (including operating systems and compilers). These features support the
functionality and performance of C but eliminate the need for dealing with C’s
uncontrolled crashes.
2
Object Oriented Turing Reference Manual
Design of the Language
The Turing language has been designed to be a general purpose language, meaning
that it is potentially the language of choice for a wide class of applications. Its convenience
and expressive power make it particularly attractive for learning and teaching. Because of
its clean syntax, Turing programs are relatively easy to read and write.
Turing helps programmers write more reliable programs by eliminating or constraining
error-prone language features. For example, Turing eliminates the goto statement. It also
provides many compile-time and run-time checks to catch bugs before they lead to disaster.
These checks guarantee that a Turing program behaves according to specification. If it does
not, a warning message is given.
Turing has been designed to eliminate the security problems of languages such as
Pascal. For example, Turing's variant records (unions) have an explicit tag field that
determines the active variant of the record. Run-time checks guarantee that the program
can never access fields that have been assigned in a different variant. In principle, a Turing
compiler prevents function side effects (changing values outside of the function) and aliasing
(having more than one way to modify a given value). However, existing implementations
do not enforce these restrictions.
Turing has modules, which serve as information hiding units. These modules are objects
in the sense of "object-oriented programming". They allow the programmer to divide the
program into units that have precisely controlled interfaces. OOT extends Turing's modules
so they can have multiple instances. This extension, called classes, supports inheritance
(called expansion) and polymorphism (overriding subprograms in expansions).
SmallTalk, one of the best known object oriented languages, considers that primitive
items, such as integers, are "objects". This requires the beginner to unlearn classical
mathematical concepts. OOT avoids this puritanical approach by considering that the
integers and other well understood concepts are exactly those of mathematics. C++, another
well known object oriented language, grafts classes into a structure that requires the
beginning programmer to deal with source inclusion, conditional compilation, header files
and a linkage editor. OOT avoids this complication by using classes and modules as the
language's compilation units. The result, in OOT, is that programming in the large becomes
object oriented and easy for the beginner (as well as the expert). OOT does not support
automatic recovery of object space (garbage collection) or operator overloading.
Dirt and Danger
OOT contains a full set of systems programming features. This includes features that
provide full access to the underlying run-time representation. While this access is needed
(only occasionally!) by the systems programmer, it comes at a great cost in terms of
portability to other machine platforms, reliability and productivity. As a result, the design
of these features has isolated them so their use can be controlled.
Dirt. Language features that have implementation-dependent results are called dirty.
For example, taking the address of a variable or inspecting the internal representation of a
real number are both dirty actions. Features such as these are marked as Dirty in the
Chapter 1 : Introduction
3
Manual. In the absence of dirty actions, a program can be expected to run with the same
results on different platforms, for example on both IBM PC compatibles and Macintoshes.
Danger. Language features that allow changes to arbitrary machine memory, or that
can cause an uncontrolled crash are called dangerous. For example, storing a value at a
specified machine address or where a pointer points without checking the validity of the
pointer are dangerous actions. Features such as these are marked as Dangerous. A
programmer should resort to dangerous actions only when this is obviously necessary and
the inherent loss in security can be justified.
In low level languages such as C, it is not possible to master even strings and arrays
without dealing with the dangerous concepts of machine addresses. In C there is no
syntactic differentiation of dirty or dangerous features. As a result, it is difficult to control
program quality.
Implementing Turing and OOT
Original Turing was designed so it could be supported by either compilers or
interpreters. At present (1999) the Original Turing system used on IBM PC compatibles and
Apple Macintoshes is called an interpreter, although technically speaking it is a compiler
that generates pseudo-code.
The OOT system is based on the Original Turing interpreter. Although a compiler is
under consideration for OOT, at the present time, OOT is supported by a pseudo-code
system. The OOT system makes extensive use of the GUI (graphical user interface) facilities
the platforms it runs under to provide the user with an extremely rapid edit-and-test cycle,
even when the program consists of a large number of classes and modules. In the spirit of
object oriented programming, the OOT system supports browsing of reusable classes and
rapid prototyping.
The Original Turing and OOT compilers and interpreters are all written in Turing Plus,
a systems programming language developed from Turing.
Basic Concepts in Turing
Like most programming languages, Turing has variables whose values are changed by
assignment statements. Example:
var i : int
% This declaration creates variable i
i := 10
% This assignment sets the value of i to 10
This example uses comments that begin with a percent sign (%) and which end with the
end of the line. The names of items, such as i, are identifiers.
4
Object Oriented Turing Reference Manual
A named constant is a named item whose value cannot change. In the following, c and x
are named constants:
const c := 25
const x := sin ( y ) ** 2
In this example, c's value is known at compile-time, so it is a compile-time (or manifest)
value. In the Pascal language, all named constants are compile-time values, but in Turing
these values may be computed at run-time, as is the case for x in this example.
An explicit constant (or literal constant) is a constant that denotes its own value. For
example, 27 is an explicit constant and so is "Hello".
In a Turing program, each named item such as a variable has a lifetime, which is called
its scope. The lifetime of the item begins with its declaration and lasts to the end of the
construct (such as a procedure or a loop) in which it is declared. More detail can be found
under declaration in the main text. Turing's scope rules tell you where an item can be used.
These rules are very similar to rules in Pascal.
Turing, like Pascal, has two kinds of subprograms, called procedures and functions.
Procedures can be thought of as named sequences of statements, and functions can be
thought of as operators that map values to values.
Turing allows you to declare an array whose size is not known until run time; these are
called dynamic arrays. This is in contrast to languages such as Pascal in which the sizes of
arrays must be known at compile time.
Turing can be thought of either as an algorithmic language, in which you write
algorithms that will not necessarily be run on a computer, or as a programming language
whose purpose is to be used on a computer. In the former case, the language is called Ideal
Turing and is a mathematical notation, much as algebra and arithmetic are mathematical
notations. In Ideal Turing, numbers have perfect accuracy (this is the nature of pure
mathematics) and there is no limit on the size of a program or its data. By contrast,
Implemented Turing is a language that you can use on actual computers.
Implemented Turing (which we usually call simply Turing) approximates Ideal Turing
as closely as is practical. However, its integers have a fixed maximum range (31 bits of
precision) and its real numbers have limited precision (about 16 decimal digits) and limited
exponent size. Whenever Implemented Turing cannot carry out your program in the same
way as Ideal Turing, it gives you a warning. It does not, however, warn you about the
limited precision of real numbers. For example, if your program tries to create an array
with a million elements in it, but there is not enough space, you will be warned.
Implemented Turing checks to make sure that your running program is meaningful.
For example, it checks that variables are initialized before being used, that array subscripts
are in bounds, and that pointers locate actual values. Checking guarantees faithful execution,
which means that your program behaves as it would in Ideal Turing. If it does not, Turing
gives a warning message. In production versions of Turing, checking can be turned off, for
maximal efficiency of production programs.
Chapter 1 : Introduction
5
Compile-Time Expressions
In certain situations, the Turing language requires you to use values that are known at
compile time. Such values are called compile-time values. For example, the maximum length
of a string type, such as the n in string(n), must be known at compile time. You are not
allowed to read a value from the input and use it as the maximum length of a string. Here
is a list of the places where compile-time values are required:
(a) The maximum size n of a string, as in string(n).
(b) The value of a label in a case statement or in a union type.
(c) Each value used in init to initialize an array, record or union.
(d) The lower bound L and upper bound U of a subrange, as in L .. U. There is,
however, one exception to this rule. The exception, called a dynamic array, is an array
declared using var or const that is not part of a record, union, or another array. In a
dynamic array, the upper bound U (but not the lower bound L) is allowed to be read or
computed at run time.
Case (d) implies that the size of set types is known at compile time. The technical
definition of a compile-time value is any expression that consists of only:
(1) explicit integer, real, boolean and string constants, such as 25, false, and
"Charles DeGaulle", as well as enumerated values such as color.red.
(2) Set constructors containing only compile-time values or all.
(3) Named constants that name compile-time values.
(4) Results of the integer operators +, -, *, div and mod.
(5) Results of the chr and ord functions.
(6) Results of the catenation operator +.
(7) Parenthesized versions of any of these expressions.
(8) OOT adds explicit character constants such as 'z' and 'Hello there'.
You are not allowed to use run-time values (any values not covered by items 1-7) in the
places listed above (a-d).
Documentation of Language Features
The following syntax notation is used in this Manual:
{ item }
[ item ]
means any number of repetitions of the item (or none at all)
means the item is optional
In the sections of the Manual that give an alphabetic list of language features,
keywords, operators and delimiters, such as get, procedure, := and ) are written in
6
Object Oriented Turing Reference Manual
boldface. Comments in example Turing programs are written in italics in a smaller font
than the rest of the program. Identifiers are written in italics. Explicit constants such as 27 or
"Hello" are written normally.
In these sections, features are presented by giving their syntax, a general description of
their meaning, examples of their use, and then details about the feature. The description is
intended to satisfy the reader who mainly wants to know basic information about the item,
for example, that the string type represents character strings. The examples illustrate
important patterns of usage of the item, and should in many cases answer the question in
mind. The detailed information that follows gives a full technical description of the item as
well as references to related items. Virtually all the information in the Turing Report
appears in this Manual, although not in its original form. This Manual takes liberties with
the original syntax of the language to make explanations easier to understand. For example,
the Manual describes declarations as a single item, and then explains restrictions that
disallow certain declarations from appearing in particular contexts. By contrast, the Report
uses the form of the context free syntax to imply, in a less obvious way, these same
restrictions.
OOT Predefined Units
In 1994, changes were made in the way that OOT organizes predefined subprograms.
Instead of a monolithic group of predefined subprograms, the subprograms are now
organized into a number of different units. By default, all the predefined units are
automatically made available to the user's program, but it is planned in future versions of
OOT to allow users to select which units they wish to use in their program.
This Reference Manual lists the OOT predefined units in Chapter Four: Language
Features, but it lists the subprograms in these units in Chapter Five: OOT Predefined Units.
Subprograms in the predefined units are exported either qualified or unqualified.
Those subprograms that are exported unqualified do not require the unit name in the
call (an example of this is the round function which is found in the TypeConv unit). These
subprograms are also listed in the Language Features chapter of the Reference Manual. The
subprograms exported qualified require the unit name in the call (an example of this is the
Create subprogram in the Dir unit which must be called Dir.Create). Lastly, there are a
number of subprograms that because implementation reasons are not actually placed in
any predefined unit (an example of this is the abs function, that returns one of two types
depending on the parameter type). However, each of these subprograms conceptually
belongs to one of the units and is listed in the summary of that unit.
There is a separate unit called Student that contains all the predefined subprograms
that are available in Original Turing (and are not exported by any other unit) and exports
them all unqualified. This means that all Turing programs will still run perfectly under
OOT. All the subprograms in the Student unit have counterparts in the rest of the OOT
predefined units, which means that a program does not need the Student unit at all.
Chapter 1 : Introduction
7
The predefined naming convention has been changed to take into account this new
organization. The new convention is as follows:
Turing predefined subprograms (defined in Turing proper) are in lower case. Example:
drawbox, clock.
OOT predefined subprograms that are exported unqualified are all in lower case.
Example: maxint, simutime.
OOT predefined units have each the first letter of each word in the name capitalized.
Example: Draw, TypeConv.
OOT predefined subprograms that are exported qualified have the first letter of each
word in the name capitalized. Example: Create, GetLong.
OOT predefined constants that are exported qualified have the first letter in each word
after the first word in the name capitalized. Example: eBadStream,
cdScreenWidth.
Changes to OOT
Turing is still an evolving language. For the most up-to-date information on Turing,
check Holt Software's web site located at http://www.holtsoft.com/turing.
8
Object Oriented Turing Reference Manual
Chapter 2
OOT for Windows
Installing OOT for Windows
Minimum System Requirements
Object Oriented Turing for Windows requires either:
(a)
Microsoft Windows 95 or Windows NT
5 Mb of disk space
(b)
Microsoft Windows 3.1
386 or better processor
5 Mb of disk space
4 Mb of RAM and virtual memory turned on, or
8 Mb of RAM.
If you are uncertain as to whether your system is a 386, you can determine this by
selecting About Program Manager from the Help menu in the Program Manager window.
The last few lines of the dialog box that appears should read 386 Enhanced Mode if your
system is equipped with a 386 or better.
If you are uncertain as to the amount of memory on your system and whether you need
to activate virtual memory, select About Program Manager from the Help menu in the
Program Manager window. The second last line indicates the amount of memory available.
If the line indicates that less than 4,000 KB of memory are free, you will need to enable
virtual memory (detailed below).
Installing Under Windows 3.1
1) Start Microsoft Windows.
If at the DOS prompt, type WIN to start Windows.
2) Place the OOT for Windows Disk into drive A or B.
3) Select Run from the File menu in the Program Manager.
Run dialog box
4) A dialog box labelled Run appears. In it, there is an edit box labelled Command Line:.
Enter into the box:
A:\SETUP
or
B:\SETUP
if the disk is in the B drive.
5) A dialog box appears labelled Registration Information. Please fill it out with your
name and (if applicable) institution and press the OK button.
6) Another dialog box appears labelled Registration Number. Please fill in the edit box
with your registration number.
Important: For the commercial version, your registration number is on a sticker
labelled OReg# followed by 8 digits. The 8 digits are your registration number. The
sticker will be in one of three places:
1) In the lower left hand corner on the first page of the Turing Reference Manual.
2) On the actual disk itself (if purchased without the Turing Reference Manual).
3) On the envelope that contained the software (if purchased with a textbook
containing the software).
For the academic version, the registration number is found on the loose leaf release
notes that accompanied the diskette. The academic registration number is 8
characters, an ‘A’ followed by 7 digits.
10 Object Oriented Turing Reference Manual
OOT for Windows Installation dialog box
7) Another dialog box appears labelled OOT for Windows Installation. The edit box
indicates the directory into which the software will be installed. By default, this
directory is c:\winoot which is suitable for most users. Press the OK button to accept
the directory name.
8) The installation begins. A dialog box labelled OOT for Windows Installation appears.
It contains a progress bar that indicates the amount of software already installed.
Progress Indicator window
9) If you are running on a system without a Math coprocessor (a 386 or a 486SX), when
the installation is finished, the system will also install a driver in the SYSTEM.INI file
and display a dialog box to that effect. This driver is used to perform math coprocessor
emulation on systems without a math coprocessor. The added line reads:
device=c:\winoot\wemu387.386
Chapter 2 : OOT for Windows
11
De-installing Under Windows 3.1
1) Delete the icons - Select the WinOOT icon in the OOT for Windows program group and
select Delete from the File menu. Click on the Yes button to delete the icon. Then select
the OOT for Windows program group and once again select Delete from the File menu
and click Yes to delete the program group.
2) Delete the driver reference - If you are running on a 386 or 486SX, the installer placed a
driver in the SYSTEM.INI file. Using Notepad or any other text editor, open
c:\windows\system.ini, edit out the line:
device=c:\winoot\wemu387.386
and save the result.
3) Delete the WinOOT files - Start the File Manager found in the Main program group.
Select the winoot directory on the left- hand side. Select Delete from the File menu. A
Delete dialog box comes up indicating the directory to delete. Make certain this is the
directory into which you installed the OOT files. Normally this should read
“c:\winoot”. Select OK. Another dialog box labelled Confirm Directory Delete
appears. Check the directory name and then select Yes to All. A dialog box labelled
Confirm File Delete appears. Click on the Yes to All button. All the WinOOT files are
now deleted.
Enabling Virtual Memory Under Windows 3.1
1) Open the Main program group and double click on 386 Enhanced icon.
2) A dialog box labelled 386 Enhanced appears. Click on the button labelled Virtual
Memory found on the right-hand side.
3) A dialog box labelled Virtual Memory appears. Click on the button labelled Change>>
on the right-hand side.
4) The box expands. In the edit box labelled New Size found at the bottom, enter a figure
of at least 4000. MS Windows will already have put a figure there depending on the
amount of available disk space. If it is larger than 4000, feel free to use the amount
indicated (although you should probably enter a figure no more than 8000). Press OK
once a figure is entered.
5) A dialog box appears asking you to restart your machine. You should do so at this time.
Changing Monitor Resolution Under Win 3.1
Having installed WinOOT, you should now (if possible) set the resolution of the
monitor to at least 800x600. It is not required, but it will certainly make the experience of
using WinOOT somewhat easier as the windows won't overlap as much. Unfortunately,
there isn't a single way of changing the resolution using MS Windows. We will describe the
most common method, but you should consult your MS Windows and display card
manuals before changing the monitor resolution.
12 Object Oriented Turing Reference Manual
The most common method of changing the monitor resolution is to change the
resolution from DOS. Enter the WINDOWS directory by typing:
cd \windows
Then run the MS Windows SETUP utility by typing:
setup
This displays a list labelled System Information.
Windows Setup Screen
In this list there is a line titled Display. This line will normally list the display card and
the resolution. Using the arrow keys, move the cursor up to this line and press the ENTER
key. At that point a list of all the possible resolutions for the display card and monitor
appears.
Windows Setup Monitor Selection Screen
Chapter 2 : OOT for Windows
13
Use the arrow keys to select a resolution (WinOOT functions best with at least
800x600x16 colors, although higher resolutions are fine) and press the ENTER key. This
returns you to the System Information screen. Move the cursor to the bottom line, which
says Accept the configuration shown above and press the ENTER key.
You then get another screen asking for confirmation of your choice.
Windows Setup Confirmation Screen
Pressing ENTER gives you one last screen of explanatory messages and you continue
by pressing the ENTER key one last time. Finally you will be returned to DOS and are
ready to start Windows.
Installing Under Windows 95
1) Place the OOT for Windows Disk into drive A or B.
2) Press the Start button and select Run.
3) A dialog box labelled Run appears. There is an edit box labelled Command Line:.
Enter into the box:
A:\SETUP
or B:\SETUP if the disk is in the B drive.
4) A dialog box appears labelled Registration Information. Please fill it out with your
name and (if applicable) institution and press the OK button.
5) Another dialog box appears labelled Registration Number. Please fill in the edit box
with your registration number.
Important: For the commercial version, your registration number is on a sticker
labelled OReg# followed by 8 digits. The 8 digits are your registration number. The
sticker will be in one of three places:
1) In the lower left hand corner on the first page of the Turing Reference Manual.
14 Object Oriented Turing Reference Manual
2) On the actual disk itself (if purchased without the Turing Reference Manual).
3) On the envelope that contained the software (if purchased with a textbook
containing the software).
For the academic version, the registration number is found on the loose leaf release
notes that accompanied the diskette. The academic registration number is 8
characters, an ‘A’ followed by 7 digits.
6) Another dialog box appears labelled OOT for Windows Installation. The edit box
indicates the directory into which the software will be installed. By default, this
directory is c:\winoot which is suitable for most users. Press the OK button to accept
the directory name.
7) The installation begins. A dialog box labelled OOT for Windows Installation appears.
It contains a progress bar indicating the directory into which the software will be
installed. When it is completed, another dialog box appears stating that the installation
is complete.
8) If you are running on a system without a Math coprocessor (a 386 or a 486SX), the
system will also install a driver in the system.ini file. This is used to perform math
coprocessor emulation on systems without a math coprocessor. The line reads
device=c:\winoot\wemu387.386
If you wish to remove WinOOT from your hard drive, you will need to eliminate this
line from the system.ini file before deleting the directory containing the OOT files.
MS Windows Drivers for WinOOT
A note about the installed driver. In order to run, WinOOT installs a line calling a
driver in the system.ini file in the \windows directory. The driver itself is placed in the
directory containing the WinOOT files. If the WinOOT directory is ever deleted, it will be
necessary to manually edit the system.ini file and remove the line calling the driver.
Specifically, you must remove the line:
device=c:\winoot\wemu387.386
The directory specified in this line contains the WinOOT files. WinOOT uses this driver
to emulate a 387 math coprocessor on systems that do not have one (the 386 and the 486SX).
As well, if WinOOT is ever accidentally installed twice, it is possible to end up with two
copies of the device= line in your system.ini file. In this case, it will be necessary to edit out
the second occurrence of the line. You will know if this occurs because you will not be able
to enter MS Windows and will get a message indicating that there are two identical device=
lines in the system.ini file.
Chapter 2 : OOT for Windows
15
Using OOT for Windows
Starting WinOOT
Once OOT is installed, you can launch the OOT for Windows application.
Windows 3.1
If the OOT for Windows program group is not already open, look for the program
group icon labelled OOT for Windows and double click it. It opens and you see the
contents of program group which contains a single icon also called OOT for Windows.
Double click on this icon to startup WinOOT.
Windows 95
To start WinOOT in Windows 95, press the Start button and select Programs from the
pop-up menu. From the Programs submenu, select OOT for Windows. From the OOT for
Windows submenu, select the OOT for Windows program. Release the mouse to startup
WinOOT.
Both
Once WinOOT is launched, you are presented with the initial startup dialog box which
displays the version of OOT and the name of the person for whom the software is licensed.
Please note that it is not legal to give the software accompanying this book to another
person. The software is licensed for use by the purchaser of the book alone.
Pressing any key or clicking the mouse will advance to the next dialog box. This dialog
box allows the user to start up the Turing environment or view OOT demonstration
programs. You may wish to use these demonstration programs to familiarize yourself with
the language.
16 Object Oriented Turing Reference Manual
OOT Introduction Window
Besides Start up OOT and Demonstrations of OOT, you can select an overview of
OOT, the current release notes for OOT, and other information, all of which will bring up
text files with appropriate information relating to the button pressed.
Click on the Start up OOT button to start the OOT programming environment.
WinOOT Environment
When the WinOOT environment starts up, it creates three windows. In the upper-left
corner, there is the Control Panel, in the upper-right corner of the screen there is the
Directory Viewer, and in the lower-left corner, there is an Editor window. There are other
types of windows that are used by WinOOT as well. These are the Error Viewer, Debugger
windows and Run windows.
Chapter 2 : OOT for Windows
17
Screen Shot of WinOOT Running
OOT Windows - Control Panel
Control Panel
The Control Panel is a unique window (like the Directory Viewer and Error Viewer).
The OOT Control Panel contains the menus for commands that, in general, do not involve
changing the contents of a individual file. The commands found in the Control Panel are
for controlling the environment as a whole, determining how programs are executed, and
getting information about WinOOT and user programs. The Control Panel also contains a
status message bar that is very useful. If you find yourself confused, look at the status
18 Object Oriented Turing Reference Manual
message bar of the Control Panel. It will tell you if you are currently executing a program,
if there were errors encountered, and so on.
The Control Panel contains six parts. There is the window title bar, which also displays
the version of OOT running. Note that you may be running a newer version than seen here
as the software included with the book will be updated regularly. Beneath the title bar is
the menu bar. The contents of the menu bar are documented in the section titled Menus Control Panel. Beneath the menu bar is the status message bar, where all messages are
displayed. Look at this area when you are unsure as to what OOT is doing. To the right of
the status message bar, the name of the current main program is displayed. See Running
Programs - Programs with Multiple Parts for more information about the main program.
Beneath the status message bar are a series of buttons. The Run Main button compiles
and executes the main program (displayed above and to the right of the status message bar).
When a program is executing, this button changes to Stop and stops program execution
when pressed. The next button is the Pause button. This button pauses a program that is
executing. When a program is paused, you can see the execution line or view variables (see
Stepping and Tracing and Showing Variables for more information). When a program is
paused, the Pause button changes to Resume. Pressing this button causes execution to
continue from where it left off.
The Step button causes OOT to execute the next statement of a paused program
(usually highlighted) and then pause, highlighting the line next line to be executed. If the
statement to be executed is a call, an importation of a module, or is creating an object, OOT
will step into the subprogram (or module or class initialization). This button is inactive
when the program is executing and is not paused.
The Step Over button acts like Step except that if the current line is a call such as to a
subprogram, pressing Step Over will cause the call to be made and execution to be paused
at the line following the call.
The Step Return button operates like Step or Step Over, except that execution
continues until the subprogram, module, or class initialization completes. The line
highlighted will be the subprogram call. Pressing Step advances execution to the following
line.
The Tracing button is a special button in that pressing it will toggle tracing on and off,
as noted by the check box in the button. If tracing is on, whenever a program is executing
the next line to be executed will be highlighted.
The speed of execution is controlled by the Trace Speed slider situated to the right of
the buttons. This slider controls how fast the program executes while tracing is on. The
more to the left the slider is placed, the longer OOT will pause on each line before executing
the next line. When slid to the far right, OOT will execute the program (while displaying
the next line to be executed) without pausing. See Stepping and Tracing for more details.
Chapter 2 : OOT for Windows
19
OOT Windows - Directory Viewer
Directory Viewer
The Directory Viewer is used to view the current directory and handle file operations.
The top of the Directory viewer is the title bar. Beneath the title bar are the menus which
are described in Menus - Directory Viewer. Just beneath the menus, the Directory viewer
shows the name of the directory that is currently being displayed. Beneath the name, the
amount of space free on the current disk is displayed. Below that is the main window
which shows the files and directories in the current directory. You can use the menu
commands to delete, rename, and copy these files, as well as to change the current directory
The main window displays the parent directory as [..]. Double clicking on it moves you
up one directory level. The root (or top) directory is displayed as [\]. Double clicking on
this symbol moves you to the root directory. Following the root directory, all subdirectories
of this directory are listed in square brackets ([ ]). Double clicking on any of them moves
you to the directory. Finally, all files in the directory are listed following the subdirectories.
You can open any file into an Editor window by double clicking on the filename.
By default, the Directory Viewer displays all files. You can, however, specify the types
of files you want displayed by selecting a different choice from the pop-up menu at the
bottom left.
To change the directory being viewed, select the desired drive from the pop-up list of
all available drives in the lower-right hand side.
By default, when you run a program, the directory displayed in the Directory Viewer
is the execution directory. Files that are opened or created will be created relative to this
directory. You can change this directory before you run your program, causing OOT to use
a different directory to read and write data files. Note that file names specified using an
20 Object Oriented Turing Reference Manual
absolute path name do not use the execution directory when determining where the file is
to be opened.
OOT Windows - Editor Window
An Editor window is used to enter and edit programs or to display text files.
The title bar of an Editor window contains the name of the file displayed in the
window. If the file has not yet been named, it will appear as unnamed. Beneath the title bar
is the menu bar. The menus contain the commands that allow you to edit the program and
make changes to it. The contents of the menu bar are discussed in Menus - Editor Window.
Below the menu bar are five buttons used to control aspects of program execution as
well as allowing the user to quickly close the file window. The Run button causes the
program in the window to start execution. While the program is executing, the button
changes to Stop. Pressing the Stop button immediately terminates execution of the
program and causes the button to revert to Run.
The Pause button is used to temporarily halt execution of (pause) an executing
program. While a program is paused, you can use the Step, Step Over, and Step Return
buttons on the Control Panel or the Step button in the Editor window to step through
execution of the program one line at a time. When there is no program executing, the Pause
button is disabled. When a program is paused, the Pause button changes to Resume and
pressing it resumes execution from where the program was halted.
Editor Window
Chapter 2 : OOT for Windows
21
The Step button is identical to the Step button in the Control Panel. While a program is
stopped or paused, pressing the Step button causes OOT to execute one line of the program
and highlight the next line to be executed.
The Trace is identical to the Tracing button in the Control Panel. This button selects
whether tracing should occur while a program is being executed. When tracing is turned
on, the program highlights each line as the line is executed. This slows down execution
speed tremendously, but allows you to see how the program is being executed. Pressing the
Trace button starts tracing either immediately (if the program is currently executing) or the
next time the program is run. When tracing is turned on, the check box in the Trace button
has an “X” in it. When the Trace button is pressed a second time, tracing is turned off and
the “X” in the check box in the button is removed. You can control the speed at which the
program execution is traced by using the Trace Speed slider in the Control Panel.
The Show Vars button is similar to the Trace button in that it toggles showing variables
on and off. When selected, this button causes the currently executing program to display
the list of units (if more than one), the global variables of the main unit, and the stack of the
primary process. See Showing Variables for more information. Note that even when the
Show Vars button is pressed again in order to turn this feature off, any still visible
Debugger windows will be updated, slowing execution down. In order to stop the
debugger from continuing to display variables, you must have turned Show Vars off and
closed all open Debugger windows.
Beneath the buttons is the area displaying the text of the file. Text can go off the right
and left edges of the screen. To move beyond the edges, select a line that extends beyond
the edge of the screen and press the End key. To move off the left side of the screen
(assuming that left-hand side of the screen is not column 1), press the Home key. This will
move the text cursor to the beginning of the line and force the window to display column 1
on the left-hand side.
Beneath the text area is the status message bar. This displays any status messages and
is identical to the status bar in the Control Panel. To the right of the status message bar is
the cursor location information. It displays the line and column number of the text cursor
and the total number of lines in the file.
OOT Windows - Error Viewer
This window lists any error messages that occur while compiling or executing the
program. Note that there is no menu bar associated with the Error Viewer.
The body of the window displays all the error messages concerning the program. Each
message starts with the file and line number that the error occurred on, followed by the text
of the error. If the error extends beyond the right-hand side of the screen, a scroll bar will
appear along the bottom of the error message area allowing you to view the entire error
message.
Beneath the error message area are five buttons. The Show line button displays the line
and (if possible) the token on the line that the error took place. If there are multiple errors in
the program, the error currently highlighted in the Error Viewer is used to display the line.
22 Object Oriented Turing Reference Manual
Error Viewer
The Next button highlights the next error (wrapping around to the top error if
necessary). It also displays the line the error occurred on and the token causing the error (if
applicable). The Prev button highlights the previous error.
The Hide warnings button causes the Error Viewer not to display any warnings, only
errors. This can be useful when certain constructs in the program are generating warnings
that can be safely ignored. However, it is very hazardous to hide warnings as a matter of
course. Warnings will often indicate a problem in the program that is not strictly a
compilation error but may result in incorrect execution. By hiding the warning, the user
does not get a chance to notice the problem and correct it before running the program.
The Close button closes the Error Viewer. The Error Viewer will reappear the next
time an error occurs.
OOT Windows - Run Window
Run Window Displaying Output from %oot/examples/widget/widget.dem
Chapter 2 : OOT for Windows
23
All output from an OOT program appears in the Run window. In the example above,
there is a large amount of graphical output appearing in the window. (Note that what
appears as a menu bar at the top of the window is produced by the output of the OOT
program being executed. It is not part of the Run window itself.)
By default Run windows are 25 columns by 80 rows, although you can change this
using the OOT Preferences. All output appears in the Run window except output being
sent to files. It is also possible to have multiple Run windows open with output being sent
to each one. The Window.Open in OOT can open another window.
To print the contents of a Run window, press Control+P when the window is selected.
At present, you cannot save the contents of a Run window. Note that if you want to save
the output of a run, you will need to redirect your output to a file. This is discussed in
detail in Running Programs - Input and Output Redirection.
OOT Windows - Debugger Window
Debugger Window - Unit List
Debugger Window Showing Variable Contents
24 Object Oriented Turing Reference Manual
Debugger Window Showing Array Contents
Whenever the user selects Show Vars, OOT displays the output in Debugger
windows. There are three types of debugger windows. There is the Unit List Debugger
window, which displays all the units found in the currently executing program. There are
Variable Display Debugger windows, such as the second debugger window displayed.
Lastly, there are Array debugger windows, such as the third window displayed. These
display the contents of arrays. There are no menus associated with Debugger windows.
For more information about viewing variables see Showing Variables.
Chapter 2 : OOT for Windows
25
Editing Text - Keyboard Editing Commands
There are a variety of keyboard commands that can be used in an Editor window. By
learning these commands, you can edit your file more quickly and efficiently.
Moving the cursor:
Up Arrow Key
Move cursor up a line.
Down Arrow Key
Move cursor down a line.
Left Arrow Key
Move cursor left a character.
Right Arrow Key
Move cursor right a character.
Home key
Move cursor to the beginning of the current line.
End key
Move cursor to the end of the current line.
Page Up key (PgUp)
Move the cursor up a page.
Page Down key (PgDn)
Move the cursor down a page.
Ctrl+Left Arrow
Move the cursor left a word.
Ctrl+Right Arrow
Move the cursor right a word.
Ctrl+Home
Move the cursor to the top of the program.
Ctrl+End
Move the cursor to the bottom of the program.
Ctrl+Page Up, Ctrl+U
Move the cursor up half a page.
Ctrl+Page Dn, Ctrl+D
Move the cursor down half a page.
Ctrl+G
Go to specified line in the file (type the line number in
the status bar).
Editing the text
Del
Delete character to right of cursor.
Backspace
Delete character to left of cursor.
Ctrl+J
Join the current line with following line.
Ctrl+E
Delete from the cursor to the end of the current line.
Ctrl+K
Delete the current line, place the cursor at beginning of
following line.
Ctrl+Y
Insert a blank line before the current line, place the
cursor at the beginning of the blank line.
Ctrl+L
Show search/replace dialog.
Ctrl+Shift %
Comment a selected range of lines.
Ctrl+5
Uncomment a selected range of lines.
Ctrl+Enter
Structure completion.
Ctrl+M
Structure match.
Ctrl+I
Paragraph program.
26 Object Oriented Turing Reference Manual
Editing Text - Selection, Cutting and Pasting
To move or duplicate large sections of text, select text, move it to the paste buffer (also
called the clipboard) and then paste it into the text. Text can also be selected in order to
perform other operations, such as commenting out or uncommenting a section of program,
deleting a section, and so on.
Selection can be accomplished using the mouse or the keyboard. To select text using the
mouse, click at the text position where the selection is to start. Holding the mouse button
down, drag the mouse to the location in the text where the selection is to end. The selection
grows from the starting position (called the anchor point) to the current mouse position. You
can see the highlighted text easily as the selection is displayed in blue.
To select text using the keyboard, use the cursor keys to move the text cursor to where
the selection is to begin. This location becomes the selection’s anchor point. Holding the
shift key down, use the arrow keys to move the text cursor around. The text between the
anchor point and the current text cursor position is highlighted as the selection.
Once the text is selected, you can perform various operations on the selection. The
simple operations are Cut, Copy, and Paste. Cut copies the selected text to the paste buffer
and then deletes the text in the document. Use this when you want to move text to a
different location in the document. Copy copies the selected text to the paste buffer but
does not affect the location. Paste places the contents of the paste buffer at the text cursor
location.
Note that using Paste does not delete the contents of the paste buffer. It is possible to
paste the contents of the paste buffer several times without cutting or copying each time.
Editing Text - Searching and Replacing Text
When editing documents, it is often helpful to be able to either find or replace
occurrences of one string for another in a file. This capability is called search and replace.
Search/Replace Dialog Box
To display the Search/Replace dialog, select Show search/replace from the Edit menu
in an Editor window (or press Control+L). The bottom of the Editor window displays the
Search/Replace dialog. To make the dialog box disappear, select Hide search/replace from
the Edit menu (the Show search/replace menu item becomes Hide search/replace when
the Search/Replace dialog is visible), press Control+L or press the Cancel button in the
Search/Replace dialog.
Chapter 2 : OOT for Windows
27
There are several sections to the Search/Replace dialog. The most important parts are
the two text boxes in the middle called the search field and the replace field. The user is
expected to enter the text that is to be found in the search field and enter the text that is to
replace the found text in the replace field. The contents of the search field is called the search
text. The contents of the replace field is called the replace text. When a search is done, OOT
scans the program text looking for occurrences of the search text. When the replace
command is given, the found text is replaced with the replace text.
In the top right of the Search/Replace dialog there are a pair of radio buttons labelled
Forward and Reverse. When Forward is selected, OOT searches forward (downward)
through the program searching for occurrences of the search text. When Reverse is selected,
OOT searches backwards (upward) through the program searching for occurrences of the
search text. To the right of the radio buttons is a check box labelled Wrap-around. When this
box is checked, whenever a forward search reaches the bottom of a file, the search is
continued from the top (although it stops if it searches the entire file and does not find
anything). Likewise, if Wrap-around is selected, any backwards search that hits the top of
the file continues searching from the very bottom.
The check box to the right of Wrap-around is Case sensitive. When Case sensitive is
selected, searches only match text in the program that is identical to the search text. When
Case sensitive is not selected, searches will match text in the program that matches the
search text, disregarding the case of both.
There are five buttons in the Search/Replace dialog. The Search For button
immediately searches either forward or backward for the next occurrence of the search text
in the program, scrolls to the location in the Editor window, and highlights the matching
selection. The Replace By button replaces the selection found using the Search For button
with the replace text.
The Replace and Search button replaces the currently found text with the contents of
the replace text and then immediately searches for the next (or previous) occurrence of the
search text.
The Replace All button goes through the file in the Editor window from top to bottom
and replaces all occurrences of the search text with the replace text.
Lastly, the Cancel button causes the Search/Replace dialog to disappear.
A search and replace is usually done by entering text in both the Search field and the
Replace field, doing a Search For the text and then pressing either the Replace and Search
button to replace a found occurrence of the contents of the Search field or alternatively
pressing Search For in order to skip over an occurrence. In both cases, the next occurrence
of the contents of the Search field is highlighted. This continues until the entire file has been
checked.
28 Object Oriented Turing Reference Manual
Running Programs - Input and Output Redirection
To run a program, press the Run button in the Control Panel or Editor window, or
select the Run menu item from the Run menu in either type of window. You can also press
F1 when an Editor window or the Control Panel is selected. The Run command compiles
and executes the selected window (if the currently selected window is an Editor
window),the last selected Editor window (if the currently selected window is the Control
Panel), or a fixed selected program. See Running Programs - Programs with Multiple Parts for
more details on how to select the file to be run.
Once the program is running, a Run window will be created and the output of the
program displayed in the Run window. You cannot save the contents of a Run window,
but you can print it by pressing Ctrl+P on the keyboard while the Run window is active.
You can send all the text output of a program to a printer by redirecting the output of
the program to a file using the Run with Arguments command, opening the text file
produced in an Editor window, and printing the file by selecting Print from the File menu.
Arguments and Redirection Dialog Box
You can also specify that input should come from a file instead of the keyboard. Both
these operations are performed by selecting Run with Args from the Run menu. When you
do so, the Arguments and Redirection dialog box appears.
This dialog box has several parts. The first part is the Arguments field. You can enter
any text in the Arguments field and the text will be passed as command line arguments to
the program when it executes. The program can read the command line arguments using
the OOT predefined subprograms nargs and fetcharg.
In the Input redirection field, you can either type in a file name or press the Browse
button to the right to bring up a file dialog which allows you to select the file that should be
used. If this field is filled in, when the program executes, any input that would normally
Chapter 2 : OOT for Windows
29
have been read from the keyboard (usually using get) will be read from the specified file
instead. Note that the input read from a file with not be echoed in the Run window.
In the Output redirection field, as in the Input redirection field, you can either enter a
file name into the field or press the Browse button beside the field to select a file name to be
used for output redirection. When this field is filled in, any text output (usually using put)
will be sent to the specified file instead. The file produced will be a text file and can be
opened as an Editor window. Note that if no output is to be sent to the screen at all, OOT
will not display a Run window. Note also that graphical output will still be sent to the Run
window and not sent to the file.
The Output redirection field shows the maximum number of errors that you wish to
have returned at once from the compiler.
Beneath the Maximum Errors field is a check box that sets whether all subsequent runs
should run with the arguments set in this dialog box. If set, the Run button in the Editor
windows and Control Panel change to say Run w/ Args to remind you that the input and
output may be redirected.
Beneath the Maximum Errors field are four buttons. The Run button runs the program
with the information that you have entered here. The Clear button wipes out all the
information in the dialog box. The dialog box always stores the information about the last
Run with Arguments. If you want to change all the values quickly, use the Clear button.
The OK button is used to set the Arguments and Redirection without actually running the
program. The Cancel button aborts the run without changing the contents of the dialog
box.
Besides the Run and Run with Arguments menu items, the other useful menu items in
the Run menu are the Set preprocessor symbols, to set preprocessor symbols for #if
statements in your program, and the Compiler reset menu item which resets the compiler,
forcing a recompilation of all the units. This is useful if the environment shows odd
behavior.
In future versions of the software, there may be a Compile menu item in the Run
menu. This will be used for producing standalone versions of compiled programs
(essentially EXE files that you can run without having OOT present). When this becomes
available, an item in will appear in the Help menu explaining the use of this feature.
Running Programs - Programs with Multiple Parts
In OOT, editing and executing a program consisting of a single file (no import or
include statements) is straightforward. When the Run or Run with Arguments command
is given, OOT compiles and executes the contents of the active Editor window, or if no
Editor window is active, the last active Editor window. The program that is compiled and
executed when the run command is given is called the main program and its name is
displayed in the Control Panel.
In cases where a user is editing and executing programs consisting of several units
(usually a main program and several modules or classes in separate files), it is not always
obvious which file should be run. Running the currently active Editor window will not
work if the user is editing an Editor window containing one of the classes the program uses
instead of the main program itself.
30 Object Oriented Turing Reference Manual
Rather than always having to activate the window containing the main program before
selecting Run, OOT provides an alternative method of specifying which program to
compile when the run command is selected.
This is done by setting the main program to be either fixed or floating. When the main
program is floating (the default method), the main program is either the currently active
Editor window, or if no Editor window is active, the last active Editor window. If the main
program is fixed then the main program is a specific file that the user previously chose.
For example, a file called database.t uses two units readdata.tu and writedata.tu. The user
would specify database.t as the main program. Then regardless of which file is being edited,
when the run command is given, OOT compiles and executes database.t.
If the user later wants to run a different program, she or he sets the main program to be
floating. OOT then compiles and executes the active Editor window.
The following commands relate to selecting the main program. The Run button in the
Control Panel and in Editor windows compiles and runs the main program. The current
main program along with whether it is fixed or floating is found to the right of the status
message bar in the Control Panel. See the section labelled Control Panel for a picture.
The Show main program menu item in the File menu of the Control Panel activates
the main program, opening it into an Editor window, if it is not already open.
The Run and Run with Arguments commands in the Control Panel and the Editor
windows compile and execute the main program.
The Fix main program menu item in the Change/set dir menu in the Directory viewer
fixes the main program to be the file currently selected in the Directory viewer. Using this
command means that the main program need not be opened to execute it, a convenience
when only the units of a program need modification.
The Fix main program menu item in the File menu in the Editor window fixes the
main program to be the contents of active Editor window.
The Float main program in both the Directory Viewer and the Editor windows causes
the main program to float once again.
Menu Commands - Control Panel
File
New: Open an empty untitled Editor window.
Open: Bring up the Open dialog box to allow the user to open a file into an Editor
window.
Show main program: Display the main program window. If hidden, bring it to the
front and make it active.
Show directory viewer: Display the Directory Viewer. If hidden, bring it to the front
and make it active.
Chapter 2 : OOT for Windows
31
OOT Preferences: Bring up the OOT Preference dialog box to allow the user to change
a variety of features in OOT such as the default size of Editor windows and Run
windows, the cursor shape, and so on.
Quit OOT: Quit the OOT environment. Close all OOT windows. Ask the user whether
she or he wishes to save the contents of Editor windows that have been modified.
Run
Run main: Run the main program. The main program is specified in the Control Panel
to the right of the status message bar.
Run with arguments:
Running Programs.
Bring up the Arguments and Redirection dialog box. See
Single Step: Execute a single line of the program and highlight the next line of the
program to be executed. See Stepping and Tracing.
Pause Execution: Pause execution of the program. While the program is paused, the
user can step through execution and view variables. This menu item becomes
Resume Execution when the program is paused.
Stop Execution: Terminate execution of the program.
Stop/Close Run Window: Terminate execution of the program and close all the Run
windows.
Set Preprocessor Symbols: Bring up a dialog box allowing the user to specify all the
preprocessor symbols for use with the #if statements in an OOT program.
Compiler Reset: Reset the compiler. Throw away all compiled units.
Debug
Show unit list: Display the list of all units (main unit, modules, and classes) used by
the currently executing program. See Show Variables.
Show processes: Display the list of all processes used by the currently executing
program. See Show Variables.
Show queues: Display the list of all processes used by the currently executing
program by the queue that they are in (the queues can be executing, waiting for
event, paused, and so on).
Trace on/off: Turn tracing on or off. Same as pressing the Trace button in the Control
Panel. See Stepping and Tracing.
Show vars on/off: Turn Show Vars on or off. Same as pressing the Show Vars button
in the Control Panel. See Show Variables.
Show allocated objects: Show some objects allocated by the currently executed OOT
program. These objects will include file stream numbers, directory stream numbers,
Pics, and other elements from the predefined modules in later versions of the OOT
software.
32 Object Oriented Turing Reference Manual
Show run errors: Display the Run Error Viewer. When this box is visible, any non-fatal
run time errors (such as file opening failure, illegal font numbers, etc.) are
displayed here. Use this when predefined subprograms do not perform as
expected. Note that it is better to check and handle possible failures in the program
itself.
Show memory usage: Display a dialog box containing the total amount of free
memory, the largest contiguous block of memory and the percentage of Windows
resources in use. If OOT is running out of memory, select this item before running
the program. You should have a minimum of 1 megabyte of free memory before
the program executes.
Show execution line: Highlight the line that is currently being executed. Useful for
finding out what section of the program is currently being run.
Show definition: Display the definition of the highlighted variable. This can only be
used when a Debugger window was previously selected. The source declaration of
the variable being highlighted in the Debugger window will be highlighted.
Help
OOT Reference: A submenu containing help files about the help system, the OOT
environment, the editor, the debugger, and other parts of the environment.
About: Information about the current version of OOT for Windows and how to contact
Holt Software.
Authors: The list of those who have contributed to the OOT for Windows software.
Versions: Versions of different parts of the OOT environment.
Information: Display dynamic information about the OOT environment in the status
message bar. Information items include the name of the main program, the current
run time arguments, the current preprocessor symbols, the current input file, the
current output file, the execution directory, and the current directory.
Windows
This menu contains the list of all open windows in OOT. Non–unique windows are
found in submenus labelled Editor Windows, Run Windows and Debug Windows.
Selecting a menu item from this menu causes the corresponding window to come to the
front and to become active.
Menu Commands - Directory Viewer
File
New: Open an empty untitled Editor window.
Open: Bring up the Open dialog box to allow the user to open a file into an Editor
window.
Chapter 2 : OOT for Windows
33
Refresh: Refresh the Directory Viewer, displaying any new files and removing any
older files.
Run current main: Run the current main file.
Stop Execution: Terminate execution of the program.
Pause/Resume execution: Pause execution of the program. While the program is
paused, the user can step through execution and view variables. If the program is
already paused, then resume execution.
Select all files:
Viewer.
Select all files in the directory being displayed by the Directory
Quit OOT: Quit the OOT environment. Close all OOT windows. Ask the user whether
she or he wishes to save the contents of Editor windows that have been modified.
Close Viewer: Close the directory viewer. Selecting Directory Viewer menu item from
the Window menu causes the Directory Viewer to reappear.
Change/Set dir
OOT [%oot]: Move to the OOT directory. By default, this directory is the directory
where the OOT for Windows executable is located.
Job [%job]: Move to the job directory. The job directory is selected by user using the
Set job directory menu item in this menu. It is assumed that the user would set the
job directory to the directory containing most of the files for the project that he or
she is working on at the moment.
Home [%home]: Move to the user’s home directory. By default, this is the directory
that OOT starts up in. The startup directory is specified in the Working Directory
field of the Program Item Properties dialog box for Object Oriented Turing for
Windows. This dialog box is modified from within the Windows Program Manager,
not within OOT.
Help [%help]: Move to the OOT Help directory. Various help text files are stored here.
Temp [%tmp]: Move to the directory specified by the TMP environment variable in
DOS. This directory is also used by Microsoft Windows for various temporary files.
Other...: Move to the directory specified in the dialog box that appears.
Fix main program: Set the selected file to be the main program. A single file must be
selected in the Directory Viewer. See Running Programs - Programs with Multiple
Parts.
Float main program: Set the main file floating again. See Running Programs - Programs
with Multiple Parts.
Fix exec. directory: Set the current directory to be the execution directory. If this is set,
then all execution takes place in this directory rather than the current directory.
Float exec directory: Set the execution directory floating again.
Set OOT [%oot] directory: Set the current directory to be the OOT directory. Note that
many example programs use the %oot directory as the way of locating system files.
34 Object Oriented Turing Reference Manual
Set home [%home] directory: Set the current directory to be the home directory.
Set job [%job] directory: Set the current directory to be the job directory.
Utilities
Delete files [del/rm]: Delete any selected files.
Rename file/dir [ren]: Rename the selected file or directory. A dialog box comes up
allowing the user to specify the new name.
Move files [move]: Move a set of files to a new directory by allowing the user to
specify the location of the directory to move the files to in a dialog box.
Copy files [copy/cp]: Copy a set of files to a new directory by allowing the user to
specify the location of the directory to copy the files to in a dialog box.
Backup files: Make a copy of any selected files with the .BAK extension.
Delete directory [del+rmdir]: Delete the selected directory and its contents.
Create directory [mkdir]: Create an empty directory with the name specified in the
dialog box that appears.
File pattern search [grep]: Search the contents of a set of files for a specified pattern.
Add to OOT arguments: Add the currently selected files to the OOT run time
arguments list. See Running Programs for more details.
Set as OOT input file: Select the currently selected file to be the OOT input redirection
file. See Running Programs for more details.
Set as OOT output file: Select the currently selected file to be the OOT output
redirection file. See Running Programs for more details.
Menu Commands - Editor Window
File
New: Open an empty untitled Editor window.
Open: Bring up the Open dialog box to allow the user to open a file into an Editor
window.
Close: Close this Editor window. If the window has been changed since it was last
saved, there will be an opportunity to save it.
Save: Save the contents of the Editor window in the file associated with the window. If
the window is unnamed, bring up a Save As dialog box to allow the user to select
the name and directory of the file to be saved.
Save As: Bring up the Save As dialog box. Save the contents of the Editor window in
the file selected. Note that this changes the file associated with the window to be
the new file.
Revert: Discard any changes since the program was last saved.
Chapter 2 : OOT for Windows
35
Print: Print the program on the printer selected in Microsoft Windows.
Navigation: A submenu of commands to change the text cursor position. Mostly used
to remind users of the keyboard shortcuts.
Fix main program: Make this the main program. Whenever a the user selects Run,
either from menu, button or keyboard shortcut, this will be the program that is run
until another file is made the main program or Float main program is selected. See
Running Programs - Programs with Multiple Parts.
Float main program: Cause the main program to float again. When the main program
is floating, the active or last active Editor window will be the one that is run when
the run command is given. See Running Programs - Programs with Multiple Parts.
Redraw: Redraw the contents of the Editor window. Select this command if for some
reason the appearance of the contents of the Editor window gets scrambled.
OOT Preferences: Bring up the OOT Preference dialog box which allows the user to
change a variety of features in OOT (such as the default size of Editor windows
and Run windows, the cursor shape, and so on).
Quit OOT: Quit the OOT environment. Close all OOT windows. Ask the user whether
she or he wishes to save the contents of Editor windows that have been modified.
Edit
Undo: Undo the last operation, restoring the contents of the window to what they were
before the operation.
Cut: Copy the selection to the paste buffer and then delete the selection. See Selection,
Cutting and Pasting for more details.
Copy: Copy the selection to the paste buffer. See Selection, Cutting and Pasting for more
details.
Paste: Paste the contents of the paste buffer at the text cursor. If there is a selection, the
paste buffer replaces the selection. See Selection, Cutting and Pasting for more
details.
Join Lines: Join the line that the text cursor is on with the following line, adding a
space between the two lines.
Delete to EOL: Delete from the text cursor to the end of the line that the text cursor is
on.
Delete line: Delete the entire line that the text cursor is on.
Open new line: Insert a new empty line on the line before the line that the text cursor
is on and place the text cursor at the beginning of this new line.
Show search/replace: Show the Search/Replace dialog box at the end of the window.
If the dialog box is already visible, this menu item becomes Hide Search/Replace
and selecting it hides the Search/Replace dialog box. See Searching and Replacing
Text for more details.
36 Object Oriented Turing Reference Manual
Comment range: “Comment out” all lines in the selection. It places "%–" at the
beginning of all lines in the selection. Any line starting with a "%" is treated as a
comment by the compiler and ignored. This is useful for temporarily disabling
sections of the program while testing it.
Uncomment range: “Uncomment” all lines in the selection. Removes the "%–" at the
beginning of any lines in the selection. This command is used in conjunction with
the Comment range command to reenable sections of the program that were
disabled using the Comment range command.
Structure completion: If the text cursor is at an OOT program construct (for example
loop, case, and so on.), this command places the end-construct after the construct,
adds a blank line between the two, and moves the text cursor there, indenting the
line by four spaces. For example, if you have:
loop^
where ^ represents the text cursor, selecting Structure completion produces the
following:
loop
^
end loop
Structure match: If the text cursor is at an OOT program construct (for example loop,
case, etc.), or at an end-construct (for example end loop, end case, etc.), this
command moves the text cursor to the end of that particular construct. If at the end
of the construct, Structure match moves to the beginning of the construct. For
example, if you have:
for i : 1..20^
loop
loop
put "A"
end loop
end loop
end for
where ^ represents the text cursor, selecting Structure match moves the cursor to
the beginning of the end for:
for i : 1..20
loop
loop
put "A"
end loop
end loop
^end for
Chapter 2 : OOT for Windows
37
selecting Structure match a second time moves the cursor to the beginning of the
first line.
Paragraph file: Indent the file, splitting lines that are too long. This command formats
the file, indenting constructs properly and splitting lines that are longer than 78
characters or have more than one statement on them (if possible). For example, if
you have:
for i : 1..20
loop
if x = 3 then put "A" end if
end loop
end for
selecting Paragraph file produces the following:
for i : 1..20
loop
if x = 3 then
put "A"
end if
end loop
end for
Run
Run: Run the main program. If the main program is floating, run this Editor window.
You can always determine the current main program by looking to the right of the
status message bar in the Control Panel.
Run with Args: Bring up the Arguments and Redirection dialog box. See Running
Programs - Input and Output Redirection.
Pause execution: Pause execution of the program. While the program is paused, the
user can step through execution and view variables. This menu item becomes
Resume Execution when the program is paused.
Single Step: Execute a single line of code and highlight the next line of code to be
executed. See Stepping and Tracing.
Stop execution: Terminate execution of the program.
Set preprocessor symbols: Bring up a dialog box allowing the user to specify all the
preprocessor symbols for use with the #if statements in an OOT program.
Compiler reset: Reset the compiler. Throw away all compiled units.
Debug
Next error: Highlight the next error in the list of errors. The window where the error
lies will be brought to the front and activated. The line where the error occurred
38 Object Oriented Turing Reference Manual
will be highlighted and, if possible, the token causing the error will be marked. If at
the end of the list, wrap around and highlight the first error.
Previous error: Highlight the previous error in the list of errors. If at the beginning of
the list, wrap around and highlight the last error.
Show unit list: Display the list of all units (main unit, modules, and classes) used by
the currently executing program. See Show Variables.
Show processes: Display the list of all processes used by the currently executing
program. See Show Variables.
Show queues: Display the list of all processes used by the currently executing
program by the queue that they are in (the queues can be executing, waiting for
event, paused, and so on.).
Show allocated objects: Show some objects allocated by the currently executed OOT
program. These objects will include file stream numbers, directory stream numbers,
Pics, and other elements from the predefined modules in later versions of the OOT
software.
Show run errors: Display the Run Error Viewer. When this box is visible, any non-fatal
run time errors (such as file opening failure, illegal font numbers, etc.) are
displayed here. Use this when predefined subprograms do not perform as
expected. Note that it is better to check and handle possible failures in the program
itself.
Show execution line: Highlight the line that is currently being executed. Useful for
finding out what section of the program is currently being run.
Trace on/off: Turn tracing on or off. Same as pressing the Trace button in the Control
Panel. See Stepping and Tracing.
Show vars on/off: Turn Show Vars on or off. Same as pressing the Show Vars button
in the Control Panel. See Show Variables.
Help
Help with Editor: Display a text file containing help on the editor. In later versions of
OOT, if there are commands in the editor that are not documented here, check the
Editor help file using this command for instructions on use.
Keyword usage: If the text cursor is at an OOT predefined subprogram or part of the
OOT language, this command displays a one line summary of the OOT Reference
Manual entry in the status bar.
Keyword reference: If the text cursor is at an OOT predefined subprogram or part of
the OOT language, this command displays a window containing the OOT
Reference Manual entry on the item.
OOT Reference: A submenu containing help files about the help system, the OOT
environment, the editor, the debugger, and other parts of the environment.
About: Information about the current version of OOT for Windows and how to contact
Holt Software.
Chapter 2 : OOT for Windows
39
Authors: The list of those who have contributed to the OOT for Windows software.
Versions: Versions of different parts of the OOT environment.
Information: Display dynamic information about the OOT environment in the status
message bar. Information items include the name of the main program and
information from the arguments and redirection dialog box.
Windows
This menu contains the list of all open windows in OOT. Non–unique windows are
found in submenus labelled Editor Windows, Run Windows and Debug Windows.
Selecting a menu item from this menu causes the corresponding window to come to the
front and to become active.
Debugging - Stepping and Tracing
It is very useful to be able to follow the execution of a program visually. OOT lets you
see each line of a program as it is executed. By doing so, you can see exactly how the
program is executing, compare it with how it should be executing, and discover bugs.
As well, viewing the execution of the program can be an instructional aid to learning
how various language constructs operate.
Stepping and tracing provide two ways of seeing a program execute. Stepping is when
OOT executes a small chunk (usually one line) of a program and then waits for another
command. Tracing is when a line being executed is highlighted for a set amount of time,
then the next line, and so forth. In Tracing, execution continues until the user gives a
command to stop execution (or the program terminates). In stepping, the user must give a
command to continue after each line (or chunk) or code.
Tracing
To start OOT tracing a program, press the Trace button in an Editor window (causing
the check box in the button to be filled). If a program is already executing, OOT starts
highlighting the line being currently executed and continues to execute the program.
Otherwise, OOT starts tracing execution the next time a program is run. To turn off tracing,
press the Trace button again. (Note that the check box becomes empty.)
You can control the speed at which tracing occurs by adjusting the slider labelled Trace
Speed on the right-hand side of the Control Panel. To have each line executed more
quickly, move the slider to the right. To slow execution down, move it to the left. Anytime
you want execution to resume normally, click the Trace button again. You can go back and
forth turning tracing on and off as you deem appropriate.
Stepping
You can also examine execution one line at a time (i.e. stepping). While the program is
running, click on the Pause button in the Editor window (note that it changes to read
Resume). Once the program is paused, the three buttons in the Control Panel labelled
40 Object Oriented Turing Reference Manual
Step, Step Into, and Step Return are enabled. Clicking the Step button causes OOT to
execute one line of code (the next line will be highlighted). If you are at a call to a
subprogram, the Step button enters the subprogram and highlights the first line of it. If you
wish to trace execution over the subprogram (in other words, execute the subprogram all at
once and pause at the statement after the call), click on the Step Over button. As well, if
you are in a subprogram and wish to quickly skip the rest of a subprogram, click the Step
Return button, which executes the rest of the procedure and pauses execution at the
statement following the call to the subprogram.
Note that pressing Step Return in the main program is the same as pressing the Step
button.
Note that if you are stepping over multiple lines (by doing a Step Over or a Step
Return) while tracing is turned on, OOT will not highlight the individual lines being
executed.
When stepping through a program, you can always continue full speed execution by
making certain that tracing is turned off and pressing the Resume button.
Breakpoints
OOT does not have the ability to dynamically set breakpoints (points where execution
of the program stops and allows the user to start stepping or tracing). However, the user
can set a breakpoint in the program. This is done by placing the break statement anywhere
in the program. When the program reaches the break statement, execution is paused just as
if the user pressed the Pause button at that point. The user can then issue any of the
stepping, tracing or view variable commands, or press the Resume button to continue
execution. The user can have as many break statements in the program as he or she wishes.
Execution will pause at each break statement.
Debugging - Showing Variables
OOT is capable of showing all the variables in a program. To understand how OOT
displays the variables, you must understand the way the variables are grouped. First there
are unit variables, grouped one window per unit, and then there are stack variables,
grouped one window per process. A program consisting of one unit (no import statements)
and one process (no concurrency) would have one window of unit variables and one
window of stack variables. When observed in OOT, the two debugging windows would be
considered the "global" variables and the "local" variables.
A unit is either the main program or it is a module or class in a separate file with the
keyword unit at the top of the file. OOT programs use units using the import statement.
Unit variables are those in a given unit that are declared outside of any subprograms. These
variables are common to all subprograms in a unit. In the main program unit, they are
known as global variables.
A process is one thread of execution. Each process has its own local variables. The stack
variables for a particular process consist of all the local variables in the subprograms that
the process is currently in. In simple terms, stack variables are called local variables.
Chapter 2 : OOT for Windows
41
Displaying Variables
To display all the variables in a simple program, press the Show Vars button in the
Editor window. This causes two windows to appear. The first window is called Unit View:
MAIN which contains all the variables global to the main program. The second window,
called Stack Trace: Main Program, contains the list of variables on the stack (the “local”
variables). When the button is pressed, the check box in the Show Vars button is set. While
the check box is set, whenever a program starts running, the two variable windows will be
displayed.
Stop Displaying Variables
To stop displaying variables, close the Unit View: MAIN and the Stack Trace: Main
Program windows. Pressing the Show Vars button will not hide the windows or stop them
from updating. It is only used to force the two windows to appear.
When debugging a larger program, if you want execution to resume at full speed, you
must close all the debug windows to stop OOT from updating them. As well, Show Vars
must be turned off, otherwise the Unit View: MAIN and the Stack Trace: Main Program
windows will be opened when execution begins again. You can locate all open debug
windows by using the Window menu and checking the Debug Windows menu item to list
all open debug windows.
Variables Window
The variable windows show the values of the variables. Individual records will have
the word record by them where the value would appear. Following that, indented, will be
the field names of the variable and their values.
Arrays have the word array instead of a value. Double clicking the line causes another
window to appear that shows the value of the individual fields of the array. If it is a twodimensional array, clicking on the horizontal slider at the bottom of the window switches
between the different values for the first index.
The variable windows are updated every time a line is executed, so the user can see the
values change as the program executes. As always, the user can pause the program and
move step by step or start the program tracing.
The two variable windows are associated with each other, clicking on one window
makes both windows visible.
Stack Window
The stack window also contains the calls to subprograms. The format is: the name of
the function flush with the left side of the window, variables local to that subprogram
indented. At the end of the subprogram's variables is a line of dashes. As the subprogram
calls are made deeper and deeper, the list of variables grows downwards. As the program
returns from a call, the contents of the stack window shrinks.
Note that each process has its own stack window.
42 Object Oriented Turing Reference Manual
Viewing Variables in Large Programs
Pressing the Show Vars button in a program with multiple units brings up the Unit
View: Main window containing all the variables global to the main program. If the
program uses multiple units, the Unit List window also appears. This window shows a list
of all the non-predefined units used by the program. Double clicking any of the units in the
list causes the Unit View window for that unit to appear. Thus the user can select specific
units to see the variables of and leave all the other units unopened.
Likewise, if the program has multiple processes and the Show Vars button is pressed,
the Current Processes window appears listing all the currently executing processes. Double
clicking on any process in the list causes the Stack Trace window for that process to appear.
Thus you can select which processes you wish to inspect and leave the other processes
unopened.
Show Unit List
This command lists all the units created by the user (predefined units do not appear)
used by the program. Double clicking on a unit in the unit list brings up the Unit View
window for that particular unit.
Show Processes
This command shows all the processes and their status. Double clicking a process
brings up the Stack Trace window for that particular process.
You can display all the variables in the entire program using the Show Unit List and
Show Processes commands.
Chapter 2 : OOT for Windows
43
Tutorial
The following material is a very brief tutorial in creating, editing, running, printing,
and saving a file. These are the initial activities required to start using OOT to write
programs. As soon as you have mastered the basic activities, you are encouraged to read
the other sections of this manual to find out the other commands at your disposal.
How to Create a New Window
If you have just started OOT for Windows, there may already be an Editor window
labelled unnamed. You can start entering text in this window.
If you need to create another new window, select New from the File menu in the either
the Control Panel or any other Editor window.
Note that when you save the program with a specified file name, the window title will
reflect the new name.
How to Open an Existing File
In the Directory Viewer, locate the file that you want to open. If it is on a different
drive, select the appropriate drive from the pop-up menu of drives found in the lower-right
corner. To go into a directory (directories are listed first and enclosed in square brackets),
double click on the directory name. To go up to the directory above, double click on the [..]
in the list of files. To quickly move to the top level directory in the drive, double click on the
[\]. Once you have found the file that you want to open, double click on it to open the file
in an Editor window.
When you select Save from the File menu, the contents of the window will be saved
into the file, regardless of the current directory.
How to Edit a Program
In the simplest case, you enter text by just typing. Press the Enter key to start the next
line. You can erase text by pressing the Backspace key, which deletes the character directly
behind the cursor (the small upward pointing arrow).
You can move the cursor by clicking the location in the Editor window that you wish to
move the cursor to. At this point, you can insert text by typing, or delete text using the
Backspace key.
Note that you can add blank lines at any point in the program by moving the cursor to
the desired location and pressing the Enter key as often as required.
44 Object Oriented Turing Reference Manual
As the program expands beyond a single screen, use the scroll bars on the right-hand
side to scroll the program up and down.
You can select a section of text using the mouse. Click down at the start of the text that
you wish to select, and with the mouse button held down, drag the mouse to the end of the
text that you wish selected and release the mouse button.
Once you have a selection, you can delete it by hitting the Backspace key or place a copy
of it in the paste buffer by selecting Copy from the Edit menu. Once text is in the paste buffer,
you can place a copy of the text at the location of the cursor by selecting Paste from the File
menu.
When you have entered some portion of your program, you can Paragraph (indent) the
program by pressing the F2 function key or selecting Paragraph file from the Edit menu.
Note that this is just the beginning of the editor. Check the Editor Menu section and the
Editor Keystroke commands for more commands to make editing programs easier.
How to Print a Program
To print a copy of your program, select Print from the File menu. Note that you must
have already have a printer selected.
How to Save a Program
To save a copy of your program, select Save from the File menu. If you loaded the
program from the disk, this saves the contents of the window in the file that you opened. If
the window is unnamed, a dialog box labelled Save As appears. You can now enter the file
name for this program.
If you opened this window from a file and wish to save the contents of the window in a
different file, select Save As from the File menu. This opens the Save As dialog box which
gives you the opportunity to specify a different file name. The original file will not be
changed.
How to Run a Program
To run a program, press the F1 function key or select Run from the Run menu. OOT
then attempts to compile the program. Status messages appear at the bottom of the Editor
window (and in the Control Panel) as the program is compiled.
If errors are found in the program then compilation is unsuccessful. The Error window
appears with the list of errors in the program and the location in the program where the
first error occurred is highlighted. Correct the errors in the program and then run the
program again.
If the program finished compiling successfully (several seconds the first time, a second
or two subsequent times), execution starts. A Run window appears in the lower-right
corner and all output from the program appears there. You also enter all input to the
Chapter 2 : OOT for Windows
45
program here. Note that the Run window must be active in order for keyboard input to be
sent to the OOT program.
You can halt execution of the program at any time by pressing the Stop button in the
Editor window. Once the program has finished execution, the status message in the Editor
window states “Execution finished”. At this point, you can print or close the Run window.
You can close the Run window by double clicking on the Control menu (the box in the
upper-left corner of the window with the horizontal bar running through it).
How to Print a Program’s Output
To print the output of the Run window, make certain that the Run window is selected
and the press Control-P (that is the “Control” key and the “P” key pressed simultaneously).
If you need to print out more than one screen’s worth of output, see the section labelled
Input Redirection.
How to Quit OOT
To finish using OOT, select Quit OOT from the File menu of either any Editor window
or the Control Panel.
46 Object Oriented Turing Reference Manual
Chapter 2 : OOT for Windows
47
Chapter 3
OOT for Macintosh
Installing OOT for Macintosh
Minimum System Requirements
Object Oriented Turing for Macintosh requires:
System Software:
System 7.0 or greater
Hardware:
68020/68030/68040/PowerPC Macintosh
3MB of free RAM
5 Mb of disk space
If you are uncertain as to which version of the system software you are running, switch
to the Finder and select About This Macintosh under the Apple menu. In the upper right
of the dialog box that appears, the version of the system software you are currently running
is displayed. This must be version 7.0 or greater.
The only Macintoshes that don’t have a processor capable of running OOT for
Macintosh (MacOOT) are the Macintosh 128, 512, Plus, SE, Classic, original Portable and
PowerBook 100.
If you are uncertain as to the amount of memory on your system and whether you need
to activate virtual memory, switch to the Finder and select About This Macintosh under
the Apple menu. The figure titled Largest Unused Block is the amount of free memory
available. If the line indicates that less than 3,000 KB of memory are free, you will need to
enable virtual memory (detailed below). Note that quitting applications also frees up
memory.
Installing Under Macintosh
1) Place the original Object Oriented Turing for Macintosh disk into the disk drive.
2) Double click on the icon labelled MacOOT 1.2.2 Installer. Note that the version
number will probably be different.
3) A window shows up showing the OOT Folder on the left hand side and all the disks
onto which OOT can be installed on the right.
4) Drag the OOT folder onto the disk where you want OOT for Macintosh to be installed.
5) An information box comes up telling you which file is being installed.
50 Object Oriented Turing Reference Manual
6) When all the files are installed, a dialog box comes up allowing you to do another
installation or continue. Select the Quit button (or press Return).
7) Another dialog box appears telling you which version of MacOOT is being installed.
8) If this is the first time you have installed the software, a reminder to send in your
registration card appears. Please send in your registration card!
9) Another dialog box appears. This one has spaces to enter your name, address and
registration number. (Note the Academic version has a space only for the institution
and registration number.) Please fill it in. If this is not the first time the software has
been installed, all fields but the registration number will already be filled in and can’t
be changed. The user must then only fill in the registration number.
Chapter 3 : OOT for Macintosh
51
Important: For the commercial version, your registration number is on a sticker
labelled OReg# followed by 8 digits. The 8 digits are your registration number. The
sticker will be in one of two places:
1) In the lower left hand corner on the first page of the Turing Reference Manual.
2) On the actual disk itself (if purchased without the Turing Reference Manual).
For the academic version, the registration number is found on the loose leaf release
notes that accompanied the diskette. The academic registration number is 8
characters, an ‘A’ followed by 7 digits.
10) Once you have filled in all the appropriate fields and clicked OK, the dialog box will
disappear and registration is now complete. You can now run OOT for Macintosh by
double clicking on the OOT for Macintosh icon found in the OOT Folder on your hard
disk. Please keep the original diskette safe in case you need to reinstall the software.
De-installing OOT for Macintosh
1) Drag the OOT Folder from your hard disk into the trash.
2) Open up the Preferences folder found in the System Folder and drag the MacOOT
Preferences file found there into the trash.
52 Object Oriented Turing Reference Manual
Using OOT for Macintosh
Starting MacOOT
Once OOT is installed, you can launch the OOT for Macintosh application.
On the hard drive where you installed OOT for Macintosh (MacOOT), you will see a
folder called OOT Folder.
Open up the folder by double clicking on it. The folder contains a number of files and
folders and an icon labelled MacOOT.
To start MacOOT, double click on the icon labelled MacOOT. You can also launch
MacOOT by double clicking on any MacOOT file. MacOOT files look like this.
MacOOT Environment
Once MacOOT is launched, you are presented with the initial alert dialog box which
displays the version of OOT and the name of the person for whom the software is licensed.
MacOOT also compiles the built in modules at this time. The alert box disappears after a
few seconds.
If MacOOT is double clicked to start it up, it displays an untitled Editor window in the
upper-left corner, otherwise it opens up the MacOOT file that the user clicked on into an
Editor Window. Along with Editor windows, which contain user programs, there are
several other kinds of windows in MacOOT: Run windows which are the windows
Chapter 3 : OOT for Macintosh
53
produced when your program runs, an Error Window which contains a list of all the errors
found in your program, a Status window which contains the messages that tell you what is
being compiled or executed, the Debugging Controller window, which is used to step,
trace and view variables in a program, and Debug windows which are windows
containing debugging information.
What follows is a description of the windows that appear in MacOOT.
54 Object Oriented Turing Reference Manual
MacOOT Windows - Editor Window
An Editor window is used to enter and edit programs or to display text files.
The title bar of an Editor window contains the name of the file displayed in the
window. If the file has not yet been named, it will appear as Untitled.
Editor Window
The Editor window is where all programs are displayed and modified. If the program
extends of the bottom of the window, the right hand scroll bar is activated and you can use
it to scroll through the program. If any line in the program is longer than the width of the
window, then the bottom scroll bar is activated. You can change the default size of an
Editor window in the Preferences dialog box.
MacOOT Windows - Run Window
Run Window Displaying Output from %oot/examples/widget/widget.dem
All output from an OOT program appears in the Run window. In the example above,
there is a large amount of graphical output appearing in the window. (Note that what
appears as a menu bar at the top of the window is produced by the output of the OOT
program being executed. It is not part of the Run window itself.)
By default Run windows are 25 columns by 80 rows, although you can change this
using the OOT Preferences. All output appears in the Run window except output being
sent to files. It is not yet possible to have multiple Run windows open with output being
sent to each window, although this will be implemented in a future version of MacOOT. At
that point, a call to Window.Open will open another Run window.
To print the contents of a Run window, select the Print menu item from the File menu
or press P when the window is active (in front).
To save the contents of a Run window, select Save As from the File menu when the
Run window is active. If the Run window is in graphics mode, the contents are saved as a
PICT format graphics file. If the Run window was in text mode, then the contents of the
window and any text that scrolled off the top of the window are saved in a text file. You
Chapter 3 : OOT for Macintosh
55
can also save the output of a run by redirecting your output to a file. This is discussed in
detail in Running Programs - Input and Output Redirection.
MacOOT Windows - Debugger Window
Debugger Window - Unit List
Debugger Window Showing Variable Contents
Debugger Window Showing Array Contents
Whenever the user turns on Show Vars using the Debugger Controller, OOT displays
the output in Debugger windows. There are three types of debugger windows. There is the
Unit List Debugger window, which displays all the units found in the currently executing
program. There are Variable Display Debugger windows, such as the second debugger
window displayed. Clicking on objects or arrays in these windows will cause another
Debugger window to open displaying the contents of the object or array. Lastly, there are
Array debugger windows, such as the third window displayed. These display the contents
of arrays. For more information about viewing variables see Showing Variables.
Records are displayed with the fields of the record indented in order to show their
place in the record. Like a regular variable, a field of a record containing an object or an
array can be double clicked on to display the object or array in a separate window.
MacOOT does not currently have the ability to display two dimensional arrays. In
order to view variables in such an array, the array must be declared as an “array of arrays”
(e.g. array 1 .. 10 of array 1 .. 5 of int).
56 Object Oriented Turing Reference Manual
MacOOT Windows - Debugging Controller
The Debugging Controller is a unique window in MacOOT (i.e. you can only have one
of Debugging Controller visible at a time). The Debugging Controller is also a floating
window, which means that it always appears on top of all other windows, even when it is
not active.
The Run button compiles and executes the main program (either the active window or a
specified window selected by the Fix Main Program menu item in the File menu). When a
program is executing, this button changes to Stop and stops program execution when
pressed. The next button is the Pause button. This button pauses a program that is
executing. When a program is paused, you can see the execution line, view variables or set
breakpoints (see Stepping and Tracing and Showing Variables for more information). When a
program is paused, the Pause button changes to Resume. Pressing this button causes
execution to continue from where it left off.
The Step button causes OOT to execute the next statement of a paused program
(usually highlighted) and then pause, highlighting the line next line to be executed. If the
statement to be executed is a call, an importation of a module, or is creating an object, OOT
will step into the subprogram (or module or class initialization). This button is inactive
when the program is executing and is not paused.
The Step Over button acts like Step except that if the current line is a call such as to a
subprogram, pressing Step Over will cause the call to be made and execution to be paused
at the line following the call.
The Step Return button operates like Step or Step Over, except that execution
continues until the subprogram, module, or class initialization completes. The line
highlighted will be the subprogram call. Pressing Step advances execution to the following
line.
The Tracing check box will toggle tracing on and off, as noted by the check box in the
button. If tracing is on, whenever a program is executing the next line to be executed will be
highlighted.
The speed of execution is controlled by the Trace Speed slider situated to the right of
the Tracing check box. This slider controls how fast the program executes while tracing is
on. The more to the left the slider is placed, the longer OOT will pause on each line before
executing the next line. When slid to the far right, OOT will execute the program (while
displaying the next line to be executed) without pausing. See Stepping and Tracing for more
details.
The Show Vars check box is similar to the Tracing check box in that it toggles showing
variables on and off. When selected, this check box causes the currently executing program
to display the list of units (if more than one), the global variables of the main unit, and the
Chapter 3 : OOT for Macintosh
57
stack of the primary process. See Showing Variables for more information. Note that even
when the Show Vars check box is pressed again in order to turn this feature off, any still
visible Debugger windows will be updated, slowing execution down. In order to stop the
debugger from continuing to display variables, you must have turned Show Vars off and
closed all open Debugger windows. (You can close all the Debugger windows and hide the
Debugging Controller) by selecting the Hide Debugger menu item from the Debug menu.
Beneath the buttons is the status message. This displays the current execution status
allowing the user to see at a glance whether the program is executing, paused or has been
halted.
MacOOT Windows - Error Viewer
This window lists any error messages that occur while compiling or executing the
program.
Error Viewer
The body of the window displays all the error messages concerning the program. Each
line contains the text of the error followed by the line number and filename that the error
occurred on. Normally the line and file information is not visible as it is off the right hand
side of the window. You can scroll the Error Messages window to the right to get the line
and file information.
To go to the location of an error in the program, double click on the line in the Error
Viewer containing the error message. The line will be highlighted and the file containing
the error will be made active and the line containing the error will be highlighted.
58 Object Oriented Turing Reference Manual
Editor window with error highlighted
Editing Text - Keyboard Editing Commands
There are a variety of keyboard commands that can be used in an Editor window. By
learning these commands, you can edit your file more quickly and efficiently.
Moving the cursor:
Up Arrow Key
Down Arrow Key
Left Arrow Key
Right Arrow Key
Option+Left Arrow
Option+Right Arrow
Move cursor up a line.
Move cursor down a line.
Move cursor left a character.
Move cursor right a character.
Move the cursor left a word.
Move the cursor right a word.
Editing the text
Delete
F1
F2, I
F8
F9
Delete character to left of cursor.
Run the main program.
Paragraph program.
Get one line reference on selected keyword.
Get Turing Reference Manual entry on selected
keyword.
Chapter 3 : OOT for Macintosh
59
Editing Text - Selection, Cutting and Pasting
To move or duplicate large sections of text, select text, move it to the paste buffer (also
called the clipboard) and then paste it into the text. Text can also be selected in order to
perform other operations, such as commenting out or uncommenting a section of program,
deleting a section, and so on.
Selection can be accomplished using the mouse or the keyboard. To select text using the
mouse, click at the text position where the selection is to start. Holding the mouse button
down, drag the mouse to the location in the text where the selection is to end. The selection
grows from the starting position (called the anchor point) to the current mouse position. You
can see the highlighted text easily as the selection is displayed in the system highlight color
(specified in the General control panel).
To select text using the keyboard, use the cursor keys to move the text cursor to where
the selection is to begin. This location becomes the selection’s anchor point. Holding the
shift key down, use the arrow keys to move the text cursor around. The text between the
anchor point and the current text cursor position is highlighted as the selection.
Once the text is selected, you can perform various operations on the selection. The
simple operations are Cut, Copy, and Paste. Cut copies the selected text to the paste buffer
and then deletes the text in the document. Use this when you want to move text to a
different location in the document. Copy copies the selected text to the paste buffer but
does not affect the location. Paste places the contents of the paste buffer at the text cursor
location.
Note that using Paste does not delete the contents of the paste buffer. It is possible to
paste the contents of the paste buffer several times without cutting or copying each time.
Editing Text - Searching and Replacing Text
When editing documents, it is often helpful to be able to either find or replace
occurrences of one string for another in a file. This capability is called search and replace. To
find a particular line in the text, select Find from the Search menu. The Find dialog box
appears giving you the opportunity to specify the text to search for.
Find Dialog Box
Enter the text to be searched for in the text box labelled Find What?. The text in the text
box is called the search text.
60 Object Oriented Turing Reference Manual
Beneath the search text are two radio buttons labelled Literal and Pattern Match. You
can either search for a literal match (find the text in the Editor window that exactly matches
the search text) or using pattern match, in which case you can enter any regular expression
in the text box and OOT will match anything matching the regular expression.
There are two check boxes to the right of the radio buttons. The Search Backwards text
box causes OOT to look for the search text backwards starting from the beginning of the
current selection, rather than forwards from the end of the current selection.
When the Wrap-Around check box is checked, whenever a forward search reaches the
bottom of a file, the search is continued from the top (although it stops if it searches the
entire file and does not find anything). Likewise, if Wrap-Around is selected, any
backwards search that hits the top of the file continues searching from the very bottom.
Pressing the Find button (or pressing Return) causes the Find to take place. If the search
text is found, then the Editor window is scrolled so that the matching text is visible and the
matching text itself is highlighted. Once you have found the text, you can search for the
next occurrence of the search text by typing G or selecting Find Next from the Search
menu. No dialog box is displayed for Find Next.
To search for another occurrence of text that appears in the Editor window, highlight
the text in the Editor window to search for using the mouse and then press H or select the
Find Selection menu item from the Search menu. If you press the shift key with either G
or H, the direction of the search is reversed. (Handy for “backing up” when you are
searching for a particular occurrence of the text and you skip over the desired text by
accident.)
To replace text, you use the Change command (select Change from the Search menu or
press L. This command brings up the Change dialog box. The Change dialog box has
space to enter the text to search for (the search text) and the text to replace the search text
with (the replace text).
Change Dialog Box
The radio buttons and check boxes are identical to the Find dialog box. The buttons at
the button are different. For one, the Change dialog box doesn’t disappear when a button is
pressed. When the user selects Find Next, OOT searches for the next occurrence of the
search text. When it finds it, it highlights the matching text in the Editor window and
enables the Change, then Find and Change buttons, making Change, then Find the default
button.
The user can then either change the matching text to the replace text and immediately
search for the next occurrence of the search text (by selecting Change, then Find), change
the matching text to the replace text and observe the results (by selecting Change), change
every occurrence of the search text in the Editor window to the replace text (by selecting
Chapter 3 : OOT for Macintosh
61
Change All) or skipping the current match and finding the next match (by selecting Find
Next).
If there is no text specified for the replace text, then a doing a change is the same as
deleting the matching text.
You can also use F for Find Next. To hide the Change dialog box, just click the close
button on the window.
Running Programs Input and Output Redirection
To run a program, select the Run menu item from the Run menu or press R. You can
also press F1 when an Editor window is selected. The Run command compiles and
executes the selected window or a fixed selected program. See Running Programs - Programs
with Multiple Parts for more details on how to select the file to be run.
Once the program is running, a Run window will be created and the output of the
program displayed in the Run window. You can print and save the output in the Run
window using the Print and Save As commands.
You can also specify that input should come from a file instead of the keyboard. Both
these operations are performed by selecting Run with Arguments from the Run menu.
When you do so, the Run with Arguments dialog box appears.
Arguments and Redirection Dialog Box
This dialog box has several parts. The first part is the Arguments field. You can enter
any text in the Arguments field and the text will be passed as command line arguments to
the program when it executes. The program can read the command line arguments using
the OOT predefined subprograms nargs and fetcharg.
In the Input From section you choose one of the three radio buttons. If you select either
File or File with Echo, then an Open File dialog box comes up asking you to select the file
that you want to use as input (i.e. with get statements) to the program. If you choose File,
any input read from the file will not be echoed to the screen. If you choose File with Echo,
then anytime a line of input is read, it is also echoed to the output window.
In the Output From section, you choose one of three radio buttons. If you select either
File or Screen and File, then a Save File dialog box comes up allowing you to specify the
name of the file the output should be sent to. Note that only text output (i.e. put statements)
will be redirected to a file. No graphics will be redirected. If the user selected File, then the
output will be sent to the file and no text output will be visible by the user of the program.
62 Object Oriented Turing Reference Manual
If the user selected Screen and File then the text output will appear on the screen as well as
being sent to a file.
In future versions of the software, there may be a Compile menu item in the Run
menu. This will be used for producing standalone versions of compiled programs
(essentially applications that you can run without having OOT present). When this becomes
available, an item in will appear in the Help menu explaining the use of this feature.
Running Programs Programs with Multiple Parts
In OOT, editing and executing a program consisting of a single file (no import or
include statements) is straightforward. When the Run or Run with Arguments command
is given, OOT compiles and executes the contents of the active Editor window, or if no
Editor window is active, the last active Editor window. The program that is compiled and
executed when the run command is given is called the main program.
In cases where a user is editing and executing programs consisting of several units
(usually a main program and several modules or classes in separate files), it is not always
obvious which file should be run. Running the currently active Editor window will not
work if the active window is an Editor window containing one of the classes the program
uses instead of the main program itself.
Rather than always having to activate the window containing the main program before
selecting Run, OOT provides an alternative method of specifying which program to
compile when the run command is selected.
This is done by setting the main program to be either fixed or floating. When the main
program is floating (the default method), the main program is either the currently active
Editor window, or if no Editor window is active, the last active Editor window. If the main
program is fixed then the main program is a specific file that the user previously chose.
For example, a file called database.t uses two units readdata.tu and writedata.tu. The user
would specify database.t as the main program. Then regardless of which file is being edited,
when the run command is given, OOT compiles and executes database.t.
If the user later wants to run a different program, she or he sets the main program to be
floating. OOT then compiles and executes the active Editor window.
The following commands relate to selecting the main program. The Run command
compiles and runs the main program. The name of the current main program can be found in
the Run menu where the Run menu item is either labelled Run if the main program is
floating or Run “filename” if the main program has been fixed to be filename.
The Run and Run with Arguments commands compile and execute the main program.
The Fix Main Program menu item in the File menu fixes the main program to be the
contents of active Editor window. The Float Main Program causes the main program to float
once again.
Chapter 3 : OOT for Macintosh
63
Menu Commands
Apple ()
About MacOOT: Display the current version and creator of OOT and MacOOT.
File
New: Open an empty untitled Editor window.
Open: Bring up the Open dialog box to allow the user to open a file into an Editor
window.
Close: Close this Editor window. If the window has been changed since it was last
saved, there will be an opportunity to save it.
Save: Save the contents of the Editor window in the file associated with the window. If
the window is unnamed, bring up a Save As dialog box to allow the user to select
the name and directory of the file to be saved.
Save As: Bring up the Save As dialog box. Save the contents of the Editor window in
the file selected. Note that this changes the file associated with the window to be
the new file.
Revert to Saved: Discard any changes since the program was last saved.
Page Setup: Configure the printer.
Print: Print the program on the printer selected by Chooser.
Fix Main Program: Fix the active Editor window to be the main program. Whenever a
the user selects Run, either from menu or keyboard shortcut, this will be the
program that is run until another file is made the main program or Float main
program is selected. See Running Programs - Programs with Multiple Parts.
Float Main Program: Cause the main program to float again. When the main program
is floating, the active or last active Editor window will be the one that is run when
the run command is given. See Running Programs - Programs with Multiple Parts.
Preferences: Display a series of preferences. There are three debugging preference
dialogs which can be toggled between using the Next and Previous buttons. These
are the Editor, Printing and Running preferences. Using the Set Default button
sets all the preferences for all three dialog boxes to their default settings. Pressing
Cancel means that the preferences are not changed. Pressing OK sets the
preferences to whatever the user selected.
Memory Usage: Displays a dialog box containing the amount of free memory available
under MacOOT. Use this to check if you are running out of memory (in which case,
increase the memory partition of MacOOT using the Finder).
64 Object Oriented Turing Reference Manual
Quit : Quit the OOT environment. Close all OOT windows. Ask the user whether she
or he wishes to save the contents of Editor windows that have been modified.
Edit
Undo: Undo the last operation, restoring the contents of the window to what they were
before the operation.
Cut: Copy the selection to the paste buffer and then delete the selection. See Selection,
Cutting and Pasting for more details.
Copy: Copy the selection to the paste buffer. See Selection, Cutting and Pasting for more
details.
Paste: Paste the contents of the paste buffer at the text cursor. If there is a selection, the
paste buffer replaces the selection. See Selection, Cutting and Pasting for more
details.
Clear: Delete the selected text. Don’t place it in the paste buffer.
Select All: Selects all the text in the active Editor window.
Comment: “Comment out” all lines in the selection. It places "%" at the beginning of all
lines in the selection. This is useful for temporarily disabling sections of the
program while testing it.
Uncomment: “Uncomment” all lines in the selection. Removes the "%" at the
beginning of any lines in the selection. This command is used in conjunction with
the Comment range command to reenable sections of the program that were
disabled using the Comment range command.
Structure completion: This feature is not yet implemented. Future versions may
implement it, so it is documented here. If the text cursor is at an OOT program
construct (for example loop, case, and so on.), this command places the endconstruct after the construct, adds a blank line between the two, and moves the text
cursor there, indenting the line by four spaces. For example, if you have:
loop^
where ^ represents the text cursor, selecting Structure completion produces the
following:
loop
^
end loop
Structure match: This feature is not yet implemented. Future versions may implement
it, so it is documented here. If the text cursor is at an OOT program construct (for
example loop, case, etc.), or at an end-construct (for example end loop, end case,
etc.), this command moves the text cursor to the end of that particular construct. If
at the end of the construct, Structure match moves to the beginning of the
construct. For example, if you have:
Chapter 3 : OOT for Macintosh
65
for i : 1..20^
loop
loop
put "A"
end loop
end loop
end for
where ^ represents the text cursor, selecting Structure match moves the cursor to
the beginning of the end for:
for i : 1..20
loop
loop
put "A"
end loop
end loop
^end for
selecting Structure match a second time moves the cursor to the beginning of the
first line.
Paragraph: Indent the file, splitting lines that are too long. This command formats the
file, indenting constructs properly and splitting lines that are longer than 78
characters or have more than one statement on them (if possible). For example, if
you have:
for i : 1..20
loop
if x = 3 then put "A" end if
end loop
end for
selecting Paragraph produces the following:
for i : 1..20
loop
if x = 3 then
put "A"
end if
end loop
end for
Search
Find: Find the next occurrence of a specified string. Brings up the Find dialog box. See
Searching and Replacing Text for more information.
66 Object Oriented Turing Reference Manual
Find Next: Find the next occurrence of a previously specified string. Note that
Shift+G does a search in the opposite direction from previously specified in the
Find Box. (This is a handy way of going back if you press G once to often.)
Find Selection : This will find the next occurrence of the currently selected text.
Dimmed if there is no selection or if the selection is too long. Shift+H will search
backwards for the same string.
Find Next Error: Highlight location of the next error in an Editor window.
Change: Replace specified text with another specified text. You can also change all
occurrences.
Go To Line: Jumps to a specified line in the window. Useful if you have a listing with
line numbers on it.
Mark
This window contains a list of all the procedure, functions, modules, monitors,
processes and classes of the current Editor window. It is updated every time the
window is opened or paragraphed. Selecting on an item in the window causes the
declaration of the item to be displayed. (i.e. the Editor window scrolls to the beginning
of the item selected.)
Windows
This menu contains the list of all open windows in OOT. Non–unique windows are
found in submenus labelled Editor Window, Run Window, Help Window and Debug
Window. Selecting a menu item from this menu causes the corresponding window to come
to the front and to become active.
Tile Windows: Resize all the Editor windows so that they cover the entire screen.
Stack Windows: This stacks the windows on top of each other, offsetting each one
down and to the right of the one beneath it.
Editor Window: A sub-menu containing all the open Editor windows.
Run Window: A sub-menu containing all the open Run windows.
Help Window: A sub-menu containing all the open Help windows.
Debug Window: A sub-menu containing all the open Debug windows.
Change Dialog: Bring the change dialog box to the front. If the dialog box isn't visible,
make it visible.
Debugger Controller: If the debugger controller isn't visible, make it visible.
Error Window: Brings the error window to the front. If the error window isn't visible,
make it visible.
Status Window: Brings the status window to the front. If the status window isn't
visible, make it visible.
Chapter 3 : OOT for Macintosh
67
Debug
Show/Hide Debugger: This either makes the Debugger Controller visible or hides the
Debugger Controller and all the Debugger windows.
Show/Hide Breakpoints: This will cause all program windows and the unit window to
have a column on the left hand side displayed (or hidden). Clicking by a line in the
left hand column causes a breakpoint to be displayed (a red diamond). Clicking
again causes the breakpoint to be removed. You can place or remove breakpoints
while the program is executing. See Show Setting Breakpoints.
Clear Breakpoints: This erases all breakpoints.
Show unit list: Display the list of all units (main unit, modules, and classes) used by
the currently executing program. See Show Variables.
Show processes: Display the list of all processes used by the currently executing
program. See Show Variables.
Show queues: Display the list of all processes used by the currently executing
program by the queue that they are in (the queues can be executing, waiting for
event, paused, and so on.).
Show allocated objects: Show some objects allocated by the currently executed OOT
program. These objects will include file stream numbers, directory stream numbers,
Pics, and other elements from the predefined modules in later versions of the OOT
software.
Show Memory Usage: This is not currently implemented. When implemented, it will
bring up a debugger window displaying all allocated memory (anything using the
new keyword) including the location in the source.
Show Execution Line: This command highlights the line in the source that the selected
process is currently executing. Useful for seeing what section of code is being
executed without pausing the program.
Show Definition: This command highlights the line in the source that the selected
item was defined on.
Run
Run: Run the main program. If the main program is floating, run the active Editor
window. You can always determine the current main program by the fact that the
name is in quotes in the Run menu (as in Run “filename”).
Run with Args: Run a program (as above) but bring up the Run with Arguments
dialog box. See Running Programs - Input and Output Redirection.
Stop execution: Terminate execution of compilation of the program immediately.
Set preprocessor symbols: This feature is not yet implemented. It is documented for
later use. Bring up a dialog box allowing the user to specify all the preprocessor
symbols for use with the #if statements in an OOT program.
Compiler reset: Reset the compiler. Throw away all compiled units.
68 Object Oriented Turing Reference Manual
Help
Help with Editor: Display a text file containing help on the editor. In later versions of
OOT, if there are commands in the editor that are not documented here, check the
Editor help file using this command for instructions on use.
Help with Debugger: Display a text file containing help for the debugger.
Keyword usage: If the text cursor is at an OOT predefined subprogram or part of the
OOT language, display a one line summary of the Turing Reference Manual entry in
an alert box. (Shortcut F8.)
Keyword reference: If the text cursor is at an OOT predefined subprogram or part of
the OOT language, this display a window containing the Turing Reference Manual
entry on the item. (Shortcut F9.)
OOT Reference: A sub-menu containing help files about the release notes, the OOT
language in comparison to Turing Plus and converting Turing Plus programs to
OOT.
Information: This feature is not yet implemented.
Debugging - Stepping and Tracing
It is very useful to be able to follow the execution of a program visually. OOT lets you
see each line of a program as it is executed. By doing so, you can see exactly how the
program is executing, compare it with how it should be executing, and discover bugs.
As well, viewing the execution of the program can be an instructional aid to learning
how various language constructs operate.
Stepping and tracing provide two ways of seeing a program execute. Stepping is when
OOT executes a small chunk (usually one line) of a program and then waits for another
command. Tracing is when a line being executed is highlighted for a set amount of time,
then the next line, and so forth. In Tracing, execution continues until the user gives a
command to stop execution (or the program terminates). In stepping, the user must give a
command to continue after each line (or chunk) or code.
In MacOOT, to do any stepping or tracing, you need to bring up the Debugger
Controller. This is done by selecting Show Debugger from the Debug menu.
Tracing
To start OOT tracing a program, press the Trace button in an Debugger Controller. If a
program is already executing, OOT starts highlighting the line being currently executed
and continues to execute the program. Otherwise, OOT starts tracing execution the next
time a program is run. To turn off tracing, press the Trace button again. (Note that the
check box becomes empty.)
You can control the speed at which tracing occurs by adjusting the slider labelled Trace
Speed on the Debugger Controller. To have each line executed more quickly, move the
slider to the right. To slow execution down, move it to the left. Anytime you want execution
Chapter 3 : OOT for Macintosh
69
to resume normally, click the Trace button again. You can go back and forth turning tracing
on and off as you deem appropriate.
Stepping
You can also examine execution one line at a time (i.e. stepping). While the program is
running, click on the Pause button in the Debugger Controller (note that it changes to read
Resume). Once the program is paused, the three buttons in the Debugger Controller
labelled Step, Step Into, and Step Return are enabled. Clicking the Step button causes
OOT to execute one line of code (the next line will be highlighted). If you are at a call to a
subprogram, the Step button enters the subprogram and highlights the first line of it. If you
wish to trace execution over the subprogram (in other words, execute the subprogram all at
once and pause at the statement after the call), click on the Step Over button. As well, if
you are in a subprogram and wish to quickly skip the rest of a subprogram, click the Step
Return button, which executes the rest of the procedure and pauses execution at the
statement following the call to the subprogram.
Note that pressing Step Return in the main program is the same as pressing the Step
button.
Note that if you are stepping over multiple lines (by doing a Step Over or a Step
Return) while tracing is turned on, OOT will not highlight the individual lines being
executed.
When stepping through a program, you can always continue full speed execution by
making certain that tracing is turned off and pressing the Resume button.
Breakpoints
MacOOT has the ability to dynamically set breakpoints (points where execution of the
program stops and allows the user to start stepping or tracing). To set a break point, the
user selects Show Breakpoints in the Debug menu. All Editor Windows and the Unit List
Debugger Window display a column on the left hand side. Clicking in that column to the
left of a line causes a red diamond to be displayed adjacent to the line. When execution
reaches that line, then the program pauses just as if the use has pressed the Pause button.
To remove a breakpoint, click on the red diamond and the breakpoint is removed.
Setting a breakpoint on a unit (from within the Unit List Debugger Window) will cause
execution to pause whenever any line of that unit is executed.
To clear all the breakpoints from a program, select Clear Breakpoints from the Debug
menu.
Editor window with two breakpoints set
70 Object Oriented Turing Reference Manual
The user can set a static breakpoint in the program. This is done by placing the break
statement anywhere in the program. When the program reaches the break statement,
execution is paused just as if the user pressed the Pause button at that point. The user can
then issue any of the stepping, tracing or view variable commands, or press the Resume
button to continue execution. The user can have as many break statements in the program
as he or she wishes. Execution will pause at each break statement.
Showing Variables
OOT is capable of showing all the variables in a program. To understand how OOT
displays the variables, you must understand the way the variables are grouped. First there
are unit variables, grouped one window per unit, and then there are stack variables,
grouped one window per process. A program consisting of one unit (no import statements)
and one process (no concurrency) would have one window of unit variables and one
window of stack variables. When observed in OOT, the two debugging windows would be
considered the "global" variables and the "local" variables.
A unit is either the main program or it is a module or class in a separate file with the
keyword unit at the top of the file. OOT programs use units using the import statement.
Unit variables are those in a given unit that are declared outside of any subprograms. These
variables are common to all subprograms in a unit. In the main program unit, they are
known as global variables.
A process is one thread of execution. Each process has its own local variables. The stack
variables for a particular process consist of all the local variables in the subprograms that
the process is currently in. In simple terms, stack variables are called local variables.
Displaying Variables
To display all the variables in a simple program, press the Show Vars button in the
Debugger Controller. This causes two windows to appear. The first window is called Unit
View: MAIN which contains all the variables global to the main program. The second
window, called Stack Trace: Main Program, contains the list of variables on the stack (the
“local” variables). When the button is pressed, the check box in the Show Vars button is
set. While the check box is set, whenever a program starts running, the two variable
windows will be displayed.
Stop Displaying Variables
To stop displaying variables, close the Unit View: MAIN and the Stack Trace: Main
Program windows. Pressing the Show Vars button will not hide the windows or stop them
from updating. It is only used to force the two windows to appear.
When debugging a larger program, if you want execution to resume at full speed, you
must close all the debug windows to stop OOT from updating them (select Hide Debugger
from the Debug menu). As well, Show Vars must be turned off, otherwise the Unit View:
MAIN and the Stack Trace: Main Program windows will be opened when execution
begins again.
Chapter 3 : OOT for Macintosh
71
Variables Window
The variable windows show the values of the variables. Individual records will have
the word record by them where the value would appear. Following that, indented, will be
the field names of the variable and their values.
Arrays have the word array instead of a value. Double clicking the line causes another
window to appear that shows the value of the individual fields of the array. If it is a twodimensional array, clicking on the horizontal slider at the bottom of the window switches
between the different values for the first index.
The variable windows are updated every time a line is executed, so the user can see the
values change as the program executes. As always, the user can pause the program and
move step by step or start the program tracing.
The two variable windows are associated with each other, clicking on one window
makes both windows visible.
Stack Window
The stack window also contains the calls to subprograms. The format is: the name of
the function flush with the left side of the window, variables local to that subprogram
indented. At the end of the subprogram's variables is a line of dashes. As the subprogram
calls are made deeper and deeper, the list of variables grows downwards. As the program
returns from a call, the contents of the stack window shrinks.
Note that each process has its own stack window.
Viewing Variables in Large Programs
Pressing the Show Vars button in a program with multiple units brings up the Unit
View: Main window containing all the variables global to the main program. If the
program uses multiple units, the Unit List window also appears. This window shows a list
of all the non-predefined units used by the program. Double clicking any of the units in the
list causes the Unit View window for that unit to appear. Thus the user can select specific
units to see the variables of and leave all the other units unopened.
Likewise, if the program has multiple processes and the Show Vars button is pressed,
the Current Processes window appears listing all the currently executing processes. Double
clicking on any process in the list causes the Stack Trace window for that process to appear.
Thus you can select which processes you wish to inspect and leave the other processes
unopened.
Show Unit List
This command lists all the units created by the user (predefined units do not appear)
used by the program. Double clicking on a unit in the unit list brings up the Unit View
window for that particular unit.
72 Object Oriented Turing Reference Manual
Show Processes
This command shows all the processes and their status. Double clicking a process
brings up the Stack Trace window for that particular process.
You can display all the variables in the entire program using the Show Unit List and
Show Processes commands.
Tutorial
The following material is a very brief tutorial in creating, editing, running, printing,
and saving a file. These are the initial activities required to start using OOT to write
programs. As soon as you have mastered the basic activities, you are encouraged to read
the other sections of this manual to find out the other commands at your disposal.
How to Create a New Window
If you have just started OOT for Macintosh, there may already be an Editor window
labelled untitled. You can start entering text in this window.
If you need to create another new window, select New from the File menu.
Note that when you save the program with a specified file name, the window title will
reflect the new name.
How to Open an Existing File
Select Open from the File menu. This brings up the standard Macintosh file selector
dialog box. You can select a file by either double clicking on the file you want to open or
clicking once to select the file and then pressing the Open button.
When you select Save from the File menu, the contents of the window will be saved
into the file you just opened, regardless of the current directory.
How to Edit a Program
In the simplest case, you enter text by just typing. Press the Enter key to start the next
line. You can erase text by pressing the Backspace key, which deletes the character directly
behind the cursor (the small upward pointing arrow).
You can move the cursor by clicking the location in the Editor window that you wish to
move the cursor to. At this point, you can insert text by typing, or delete text using the
Backspace key.
Chapter 3 : OOT for Macintosh
73
Note that you can add blank lines at any point in the program by moving the cursor to
the desired location and pressing the Enter key as often as required.
As the program expands beyond a single screen, use the scroll bars on the right-hand
side to scroll the program up and down.
You can select a section of text using the mouse. Click down at the start of the text that
you wish to select, and with the mouse button held down, drag the mouse to the end of the
text that you wish selected and release the mouse button.
Once you have a selection, you can delete it by hitting the Backspace key or place a copy
of it in the paste buffer by selecting Copy from the Edit menu. Once text is in the paste buffer,
you can place a copy of the text at the location of the cursor by selecting Paste from the File
menu.
When you have entered some portion of your program, you can Paragraph (indent) the
program by pressing the F2 function key or selecting Paragraph file from the Edit menu.
How to Print a Program
To print a copy of your program, select Print from the File menu. Note that you must
have already have a printer selected.
How to Save a Program
To save a copy of your program, select Save from the File menu. If you loaded the
program from the disk, this saves the contents of the window in the file that you opened. If
the window is unnamed, a dialog box labelled Save As appears. You can now enter the file
name for this program.
If you opened this window from a file and wish to save the contents of the window in a
different file, select Save As from the File menu. This opens the Save As dialog box which
gives you the opportunity to specify a different file name. The original file will not be
changed.
How to Run a Program
To run a program, press the F1 function key or select Run from the Run menu. OOT
then attempts to compile the program. The Status window appears letting you know that
OOT is compiling your program.
If errors are found in the program then compilation is unsuccessful. The Error window
appears with the list of errors in the program and the location in the program where the
first error occurred is highlighted. Correct the errors in the program and then run the
program again.
If the program finished compiling successfully (several seconds the first time, a second
or two subsequent times), execution starts. A Run window appears in the lower-right
corner and all output from the program appears there. You also enter all input to the
74 Object Oriented Turing Reference Manual
program here. Note that the Run window must be active in order for keyboard input to be
sent to the OOT program.
You can halt execution of the program by Selecting Stop Execution from the Run
menu. At this point, you can print or close the Run window.
How to Print a Program’s Output
To print the output of the Run window, make certain that the Run window is active
and the select Print from the File menu. If you need to print out more than one screen’s
worth of output, see the section labelled Input Redirection.
How to Quit OOT
To finish using OOT, select Quit from the File menu.
Chapter 3 : OOT for Macintosh
75
76 Object Oriented Turing Reference Manual
Chapter 4
OOT for DOS
Installing OOT for DOS
Minimum System Requirements
DOS OOT requires an IBM PC compatible with a 386 processor and a minimum of 4
megabytes of memory. It also requires DOS 3.0 or better.
Installing Under DOS
To run DOS OOT, you require a minimum of a 386 system (386SX is completely
acceptable) and 4 Mb of Random Access Memory (RAM). The more memory you have, the
more DOS OOT can use, so extra memory is always helpful. You must also have 5 Mb of
disk space available on your hard disk.
DOS OOT comes in two varieties. The commercial version requires the user to enter a
registration number that is provided on the title page of this book (if purchased with
software). The academic version requires a registration number that is provided separately.
While the installation of the two is very similar, differences will be noted with the tag
Commercial or Academic where a section applies only to one or the other.
Installation of DOS OOT is fairly straight forward. Place the installation diskette in
drive A or B. Then type:
A:\INSTALL or B:\INSTALL
This starts the DOS OOT installer, which displays a window telling you what it intends
to install.
Installer message telling you what it's going to install
Commercial: If this is the first time you've installed software, you'll get a notice
reminding you to mail your registration card back.
Registration Card Reminder for Commercial Version
Commercial: Once these screens have passed, you'll continue on to the registration
screen. The registration information screen asks you to enter some information about
yourself and to enter your registration number. After you've installed the software
once, you'll only be required to enter the registration number to install it again.
Registration Information Screen (Commercial)
Registration Information Screen Filled In (Commercial)
Commercial: The registration number must be filled in before you can continue the
installation. You'll find the registration number on the lower-right corner of the title
page of the Turing Reference Manual.
Academic: The first time you install the software, you'll be asked to enter the name of
the organization (school or university). After the first time, this screen will not appear.
78 Object Oriented Turing Reference Manual
Registration Information Screen (Academic)
You can move between the different fields using the up and down arrow keys. When
you are finished, move to the OK button and press ENTER. You will be given a chance to
correct the information. If you decide to cancel the installation, move to the Cancel button
using the up and down arrow keys and press ENTER.
Once you have entered the registration information, you will be asked to specify the
installation directory (the directory in which DOS OOT is to be installed). Once you have
entered the directory name, then move to the OK button and press ENTER. Selecting the
CANCEL button will cancel the installation.
Installation Directory Screen
If the disk you have chosen does not have enough space, a message will appear and
allow you to select another disk. You can try to complete the installation even if there is
theoretically not enough disk space. You might want to do this if you are installing over an
old copy of DOS OOT
Once you have chosen the directory, the installation starts. You will get the following
sequence of messages: Installing DOS OOT, Verifying Software and Installing Support
Files. Each of these steps could take up to several minutes, so allow some time for the
operation.
If there is a previous version of the software already on the disk, the installer gives you
the option of whether you wish to install over top of the old version (and thus eliminate it)
or create an additional one. Note that WinOOT and DOS OOT share the same OOT support
files. Consequently if you have already installed WinOOT, there is no need to re-install the
OOT Support Files again.
After the installation is complete, you will see the final message box indicating that the
installation was successful.
Chapter 4 : OOT for DOS
79
Installation Complete Message
If there are any problems with the installation, please contact Technical Support for
Holt Software Associates Inc. at (416) 978-8363.
De-installing Under DOS
To remove DOS OOT from the system, delete the directory that DOS OOT was installed
in. Under DOS, this can be done using the deltree or del /s commands, depending on the
version of DOS installed.
80 Object Oriented Turing Reference Manual
Using OOT for DOS
Starting DOS OOT
You can start using DOS OOT by either typing DOSOOT alone or with a filename (for
example, DOSOOT PROGRAM.T). In the first instance, you get a blank window and no
name is associated with your work until you try to save the file to the disk. The second
form associates a filename with the work you will be doing and, if the file already exists,
will initialize your OOT Environment to the contents of that file. There are other command
line options for starting DOS OOT in specialized circumstances, which will be discussed
later.
When you start up DOS OOT, you will see the menu bar at the top of the screen, the
status bar at the bottom and an alert box in the center of the screen. The alert box gives the
version number and the address of the publisher. Clicking the mouse or pressing any key
will get rid of the alert box.
The Initial Screen With Start Up Message
Hitting any key or pressing the mouse button will get rid of the start up box.
The Initial Screen With No File Loaded
The bar at the top is the menu bar. Commands in the editor are usually given via the
menus. The menu bar at the top of the program lists the menus. When a menu is selected, a
menu bar is revealed showing the possible commands in that menu. There is the File menu,
which mostly covers those commands that affect the entire file, the Edit menu which is
used to make changes in the file itself, the Search menu which is used to both find and
replace text in the file, and the Run menu which has commands pertaining to running the
program.
Chapter 4 : OOT for DOS
81
At the bottom is the status bar. It always shows the line and column that the cursor is
currently on and the name of the file being edited. The time of day appears in the upper left
hand corner. The display of the time can be turned off by using the noclock startup option.
Selecting a Menu Item in DOS OOT
There are several ways to select menu items from a menu.
•
•
•
•
Using the keyboard, press the Alt key and the first letter of the corresponding menu
(i.e. Alt-F for the File menu, Alt-E for the Edit menu, etc.). The menu now appears and
the top menu item in the menu is highlighted. Using the arrow keys, you move the
highlight up and down. When the menu item you wish to select is highlighted, press
the Enter key. The menu item is selected.
As before, press the Alt key and the first letter of the corresponding menu (i.e. Alt-F for
the File menu, Alt-E for the Edit menu, etc.). The menu now appears with the top
menu item in the menu highlighted. You will note that most of the menu items have a
letter highlighted in them. (For example the N in New, the O in Open and the i in Pick
are all highlighted in the File menu.) By pressing the key that corresponds to the letter,
the menu item is selected. Thus, by pressing Alt-F followed by O, the Open menu item
would be selected from the File menu.
Many of the menu items have a short cut listed beside the menu item name. For
example, the New menu item has Ctrl+N listed beside it and the Open menu item has
Ctrl+O and F3 listed beside it. You can select a menu item without even selecting a
menu by using the short cut. Thus pressing Ctrl+O in the editor would cause the Open
menu item to be selected without actually displaying the menu.
You can also use the mouse to select menu items. Press and hold the mouse down on
the menu name. The menu now appears. Slide the mouse down with the button still
pressed. As the mouse cursor goes over a menu item, the menu item is highlighted.
Letting go of the mouse button will select the highlighted menu item.
If you want to select a different menu from the one that is current showing, you can:
•
•
use the left and right arrow keys to display the next or previous menu, in which case
the previous menu disappears and the new menu appears, or
bring the mouse up to another menu name, in which case the previous menu
disappears and the new menu appears.
If you decide not to select any menu at all, you can:
•
•
press the Esc key, in which case all the menus disappear and no menu is selected, or
move the mouse cursor off any of the menus and then release the mouse button, in
which case all the menus disappear and no menu is selected.
82 Object Oriented Turing Reference Manual
Loading a Program in DOS OOT
To edit a program, you must select the Open menu item from the File menu. Using the
keyboard, you can press Alt-F which displays the File menu:
The New menu item is highlighted. To select the Open menu item, you must use the
arrow keys and press the down arrow until the Open menu item is highlighted. At that
point, press Enter. This opens the Open File dialog box. As described above, this is only
one way of selecting the Open menu item. You could also press O once the File menu was
displayed, or you could have pressed Ctrl+O in the editor. You could also use the mouse to
select the Open menu item.
The Open File dialog box has five sections.
•
•
•
•
The top section contains the current directory name (C:\TURING\TEST\*.*). In the
current example, it is listing all the files in the directory as is indicated by the *.* at the
end of the path name. You can specify it to list just the OOT files by pressing F8 (i.e. the
files that end in ".T"). If you press F8 again, it goes back to listing all the files.
The second section contains the buttons that you can choose. In this case, the Tab key
will allow you to select another drive. The F8 key will cause the dialog box to display
only directories and files ending in ".T". The F9 key sends you immediately back to the
directory in which you started up DOS OOT. Clicking the mouse on any of the buttons
has the same affect as pressing the key for that button.
The third section is called the File Name section and allows you to directly type the
name of a file or directory that you wish to load (or enter). If you type a file name and
press Enter, that file will be opened. If you type a drive or directory name and press
Enter, then the Open File dialog box is changed to display the contents of the new
directory.
This fourth section is called the File section and contains the names of the files and
directories in the directory specified at the top of the dialog box. You maneuver
through this section using the arrow keys, or using the mouse key to click on a file or
directory name directly. When a file is highlighted, pressing Enter opens that file.
Chapter 4 : OOT for DOS
83
•
Double clicking on a file name also opens the file. If you select a directory instead of a
file, the Open File dialog box is changed to display the contents of the new directory.
The last section is called the Drive section and contains the drives that can be selected.
Like the file section, you can use the arrow keys to select different drives. When the
drive you want is highlighted, press Enter and the Open File dialog box will change to
reflect the new drive and directory. Once again, double clicking on a drive will select
the drive. To use the keyboard, once you are in the drive section (having pressed Tab to
get there), you press Tab once again to get back to the File section.
You can always press Esc to cancel the Open File dialog box without opening a file.
Once you have selected a file, the file is opened and its contents appear in the Editor
window. The status bar changes to reflect the name of the file just loaded. You also receive
a message in the message bar (the section below the status bar) indicating how many lines
were read in. At this point you are ready to start editing.
Editing a Program in DOS OOT
This is a sample screen with the program EXAMPLE.T just read in. Notice that the first
and last lines on the screen extend beyond the right side of the screen. The right arrow on
the right hand side is a reminder that the line goes beyond the right hand side of the screen.
If you move the cursor off the right hand side of the screen, the screen scrolls horizontally.
At that point, there will be left arrows reminding you that there is text to the left of the
screen, as seen on the next page.
84 Object Oriented Turing Reference Manual
File
Headers for open,
close, read, write
close, read, write
Text File
Device
Bodies for open,
close, read, write
close, read, write
Header for ioCtl
Tape
Bodies for open, close,
read, write, ioCtl
close, read, write
Disk
Bodies for open, close,
read, write, ioCtl
close, read, write
You can use the arrow keys to maneuver around the text, and type anything you like.
By default, you start in insert mode, which means that anything you type is inserted in
front of the character that the cursor is highlighting, instead of overwriting it. You can
switch between insert and overstrike mode by pressing the Insert key on the keypad. Note
that in insert mode, the cursor is a block, while in overstrike mode the cursor is an
underline.
Saving a Program in DOS OOT
To save the program, you need only go into the File menu and select the Save menu
item. To save the file with another name, select the Save As... menu item instead. The
Save As dialog box appears (it will also appear if you select the Save menu item with a file
that has not yet been named). The Save As dialog box looks like this.
The Save As dialog box is similar to the Open File dialog box, except that it has an
extra section. The File Name section contains a place for you to enter the name of the file.
The cursor will be blinking in the first character of the File Name section. This is where
you type in the name. If you wish to save the file in a different directory, you can specify
the path name in the File Name section, or you can press Tab to get to the File section and
select directories. Pressing Tab again moves you to the Drive section. A further Tab cycles
around to the File Name section again. Pressing Tab repeatedly cycles between the File
Name section, the File section and the Drive section.
Chapter 4 : OOT for DOS
85
Running a Program in DOS OOT
Select the Run menu by pressing Alt-R or using the mouse.
The Run menu item should already be highlighted. Press Enter to run the program.
Once again note that you could have run the program by pressing F1 or Ctrl+R without
selecting a menu.
The program executes, and eventually you get a message telling you to press any key to
continue. Upon doing so, you are returned to the program window. If there was an
execution error, or if you terminated execution, the cursor is located on the line where the
error or termination occurred. If it was an error, the line is also highlighted. Either way a
message is displayed in the message area at the bottom of the screen.
To Exit DOS OOT
To exit DOS OOT, go to the File menu (Alt-F) and select the Exit menu item. If the
program has been modified and has not been saved, you get a chance to save it before
quitting.
86 Object Oriented Turing Reference Manual
OOT for DOS Editor Reference
The Menu Commands:
The File Menu
New: This erases the file currently in memory and starts fresh. If the file has been
changed and not saved, you get a chance to save it.
Open... This displays the Open File dialog box. The user can select the file to be
opened from the File section.
Pick... This displays the Pick File dialog box which is a list of the last nine files opened.
The user can select one of the files and it will be opened.
Save: Save the program in the window under the current file name. If there is no file
name specified, treat the command as a Save As.
Save As... This displays the Save As dialog box. The user must enter a name for the
file with the option of changing directory and drive.
Save Selection... This command is only active when there is a selection. If text has
been selected, the user may save the selection using this command. This displays
the Save As dialog box. Only the lines selected are saved to the file. This also
Chapter 4 : OOT for DOS
87
causes the selection to occupy complete lines (you can't save partial lines to the
file).
Revert: This reloads the file from the disk, eliminating the changes made since the last
save. The user is asked to confirm the change.
Print... This displays the Print dialog box. The user can select whether or not to print a
header on the file and to print with line numbers.
Print Selection... This command is only active when there is a selection. If text has
been selected, the user may print the selection using this command. This displays
the Print dialog box. This also causes the selection to occupy complete lines (you
can't print partial lines).
Preferences... This command is not available at this time and is always disabled.
File Command: This command is not available at this time and is always disabled.
Memory Info: This command displays a dialog box showing the current amount of
memory available in the editor and on the PC.
Shell: This command allows the user temporary access to DOS. Type "exit" at the DOS
prompt to return to DOS OOT. If the noshell startup option is selected, this menu
item is always disabled.
Exit: This quits DOS OOT.
The Edit Menu
Undo: This command is not available at this time and is always disabled.
Cut: This command is only active when there is a selection. This copies the selection
into the paste buffer and deletes the selection.
Copy: This command is only active when there is a selection. This copies the selection
into the paste buffer. It does not change the selection.
Paste: This command places the contents of the paste buffer into the file. If there is no
selection, it places it in front of the cursor. If there is a selection, it replaces the
selection.
Clear: This command is only active when there is a selection. This deletes the selection.
Select All: This command selects the entire document.
Comment: This command is only active when there is a selection. This command places
"%" in front of each line in the selection, making the lines into comments. This also
causes the selection to occupy complete lines (you can't make a partial line a
comment).
Uncomment: This command is only active when there is a selection. This command
removes the "%" from the front of each line in the selection. You get an error
88 Object Oriented Turing Reference Manual
message if you try to uncomment a line that is not a comment. This also causes the
selection to occupy complete lines.
The Search Menu
Find... This displays the Find dialog box. It allows the user to set the text to be searched
for and to specify the direction and whether the find should "wrap" (i.e. continue
looking from the beginning once it reaches the end of file).
Find Again: This command is only active when there is text that has been specified by
the Find command. This command immediately looks for the next occurrence of
the string specified by the Find command.
Find Error: This command is only active when there are errors in a Turing program.
This jumps to the next error in the program. The error messages for that line will be
displayed at the bottom of the program.
Change: This displays the Change dialog box. It allows you to do substitutions in the
text. You can cycle through each occurrence of some text in the file, changing or
leaving each occurrence. There is also an option to change every occurrence of
some specified text in a file.
Goto Line: This displays a dialog box that allows you to enter a line number. Once the
line number is entered, the program immediately jumps to that line.
The Mark Menu
The Mark menu is contains the names of the first twenty procedures, functions,
modules or classes in the currently opened file. Selecting a menu item causes the editor to
scroll to the beginning of the construct selected. This feature can be used to help navigate
through large files.
New procedures, functions, modules or classes are loaded into the Mark menu
whenever the program is either loaded or paragraphed.
The Run Menu
Run: This command runs the program. If there are any compile-time errors, it
immediately returns, jumping to the first line with an error in it. All errors are
displayed just below the status bar. Any lines with errors are also highlighted.
Chapter 4 : OOT for DOS
89
Run with Args... This displays the Run with Arguments dialog box. This allows the
user to specify the files that DOS OOT should read input from and write output to.
Also allows the user to specify any run-time arguments to Turing.
Paragraph: This command paragraphs (indents) the program.
Fix Main Program: This brings up a dialog box allowing the user to specify the main
program to be run whenever the user specified the Run command. This allows the
user to edit a file containing a unit or an include file used by the main program and
run the main program without having top load it into the editor each time.
Float Main Program: This command causes OOT to run the program in the editor
when the Run command is given. This is used after the Fix Main Program
command is given in order to allow the user to run different programs.
Fix Run Directory: This displays the Set Directory dialog box. This sets the directory
from which the program will be run. If the program opens files for reading or
writing, it will look for them here unless a path is specified.
Float Run Directory: This sets the execution directory to be the directory that the main
program is saved in. If the main program has not been saved, then the program is
run in the current directory (the directory you see when you select the Open
command).
Reset Compiler: This resets the compiler. When the compiler is reset, the next time the
program is run all the built in modules will be recompiled as well. This command
can be used if the compiler is acting oddly in any way.
Compile to .EXE... This command is used to produce stand alone executables that
allow the program to be run without having a copy of DOS OOT present. This
command produces a “.EXE” file that combined with the DOS4GW.EXE that is
provided with DOS OOT allows the program to be run on any 386 or better
machine.
When the command is given, the Save Executable dialog box appears. This is
similar to a Save File dialog box, except the default file name ends with .EXE. If the
user wishes to rename the file, the file should always end in .EXE. After that, the
Compile Options dialog box appears. This allows the user to select among options
available for compiled programs. Once the options are selected, then the program
is compiled. In order to execute the program, the user will have to leave DOS OOT.
The Help Menu
Note that for each of these commands, the cursor must be in or at the end of a keyword
listed in the Turing Reference Manual.
90 Object Oriented Turing Reference Manual
Keyword Usage: This produces a one line summary at the bottom of the screen of the
keyword selected.
Keyword Reference: This produces the entire Turing Reference Manual entry for the
keyword selected. It appears in the Editor window and you can scroll up and
down the entry using the arrow and page up/down keys. All other commands are
disabled except the Esc key which returns the user to their program. The status bar
changes from cyan to green when displaying a reference manual entry to further
indicate that the user is not in the editor.
The Window Commands
There are a number of keypad commands that you can use to maneuver around the
window, make selections and delete text.
Up/Down/Left/Right Arrows
Move up/down/left/right a character.
Home/End
Move to the beginning/end of the line.
Page Up/Down
Move up/down page (16 lines).
Ctrl+Left/Right Arrow
Move left/right a word.
Ctrl+Page Up/Page Down
Move to the top/bottom of the file.
Ctrl+Home/End
Move to the top/bottom of the screen.
Insert
This changes the current mode from Insert to
Overstrike or Overstrike to Insert. The cursor is a block
when in insert mode and an underscore in overstrike
mode.
Ctrl+Y
Delete the current line and place it in the paste buffer.
Selecting Text
In order to be able to cut and paste blocks of text in the file, you need to be able to select
that text. When text is selected, it is indicated by changing the background color of the text.
By default, DOS OOT gives selected text a background color of blue.
Once the text is selected, it is possible to cut (remove) it, paste (place a copy of the text)
it, print it, save it or even turn it into a set of comments.
To select text with the keyboard, you hold down the Shift key and use the cursor
movement commands. Thus Shift+Right Arrow selects one character. Shift+Down Arrow
selects one line of text. You can extend the selection by continuing to hold the shift key
down while using the cursor movement commands. Note that you can select a page of a
program at a time or even from the cursor to the beginning or end of a program by using
combinations of the Shift, Ctrl, Home, End, Page Up and Page Down keys.
To select text using the mouse, move the mouse cursor to the point where you wish the
selection to begin and press the mouse button. Holding the mouse button down, move the
Chapter 4 : OOT for DOS
91
mouse cursor to where you would like the selection to end and then release the mouse
button. While holding the mouse button down, you can cause the screen to scroll by
moving the mouse off the top, bottom, left or right of the screen.
Note that if there is a selection active and you do not hold the shift key down when
using the cursor keys, or if you click the mouse anywhere else, the selection will become
inactive (become unhighlighted).
Once the text is selected, you can operate on the text in a variety of ways. Commands
that have an * work only on complete lines. When one of these commands is executed, the
selection is extended to make complete lines.
Save Selection *
Saves the current selection.
Print Selection *
Prints the current selection.
Comment *
Makes the current selection a comment. This is a quick way of
commenting out a section of code.
Uncomment *
Removes the comment symbol from the lines in the current
selection. It will give an error if you try to uncomment a selection
with lines that aren't comments.
Cut
Copies the selection to the paste buffer and deletes it from the file.
Copy
Copies the selection to the paste buffer but leaves the selection on
the screen untouched.
Paste
This replaces the current selection with whatever is in the paste
buffer.
Clear
This deletes the selection from the file.
Any keystroke
Any typing automatically replaces the selection with the keystroke
typed. Pressing the backspace or the delete key deletes the current
selection.
Mouse Movement
•
•
•
•
•
The mouse can be used in a variety of ways in the file window. First, clicking in the file
window moves the cursor to that point in the window.
Double-clicking on a word selects the word.
Triple-clicking on a line selects the line.
Click-dragging the mouse will select text from the point the mouse was clicked on to
the point that the button was released. When dragging the mouse, moving the mouse
beyond the top of the screen causes the screen to scroll upward, while moving the
mouse below the status bar causes it to scroll downwards.
Clicking on the right-most column of the screen moves the cursor five spaces to the
right. Holding the button down causes the cursor to continuously move to the right.
92 Object Oriented Turing Reference Manual
•
Clicking on the left-most column of the screen when the screen has been scrolled to the
right (i.e. when the left margin has left arrows in it) moves the cursor five spaces to the
left. Holding the button down causes the cursor to continuously move to the left.
Dialog Boxes
The following applies to all dialog boxes:
The current default button has its brackets (the < >) in bright white.
In each button "<>", radio button "()" or checkbox"[]" there is often a character
highlighted in bright white. Pressing either the letter (in dialog boxes where there is no text
field such as the Print dialog box) or Ctrl+letter (in dialog boxes where there is a text field
such as the Find dialog box) will cause a button to activate, a radio button to select that
button, or a check box to be checked (or unchecked if it was already checked).
Pressing Enter activates the default button.
Pressing Esc cancels the dialog box without doing anything.
Pressing Tab cycles between the various objects in the dialog box.
You use the mouse to activate a button, select a radio button, or toggle a checkbox by
clicking once on the item.
Some dialog boxes list all the files and directories in a particular directory. An
individual file or directory can be selected either by double clicking on it with the mouse or
by pressing Enter when the file or directory is highlighted. You can change which file or
directory is highlighted by using the arrow keys. The ".." directory moves the directory
being displayed by the File dialog box to the parent directory.
Often buttons or files may be grayed-out in a dialog box. This means that the buttons or
files are disabled and can not be selected.
Some dialog boxes list all the valid drives. The File dialog box can be switched to a
particular drive by either double-clicking on the drive with the mouse, or by pressing Enter
when the drive is highlighted. You can change the highlighted drive using the arrow keys.
The Open File Dialog Box
If there are more than 48 files and directories, then the dialog box will only display the
first 48 files and directories. They will be displayed in columns of 12. There will be arrows
on the right hand side to indicate that there are more files available to the right. (As you
move to the right beyond the fourth column, arrows will appear on the left hand side to
indicate that there are more files available to the left.)
Chapter 4 : OOT for DOS
93
In the File Name section, you use the keyboard to enter the name of the file. You may
enter a full or partial path name. You can use Ctrl+Y to erase what has already been typed
and use the standard cursor keys to maneuver in the text. You can enter only legal path
name characters in the File Name section.
For the File and Drive section, the following applies.
Keyboard
Tab
Cycle between the File section and the Drive section.
F8
Cycle between displaying only directories and files ending with ".T" and
displaying all directories and files.
F9
Switch to the directory in which you started DOS OOT.
A letter or number
Highlight the file or the drive that starts with that letter.
Esc
Cancel the dialog without selecting a file.
Enter (with a file highlighted)
Open that file.
Enter (with a directory highlighted)
Change the directory in the Open File dialog box to the selected directory.
Enter (with a drive highlighted)
Change the drive in the Open File dialog box to the selected drive.
Arrow Keys
Move the highlighted file up, down, left or right.
Home / End
Move the highlight to the first / last entry.
Page Up / Page Down
Move the highlight three columns to the left / right.
Mouse
Click on *.T (*.*) Files button
Cycle between displaying only directories and files ending with ".T" and
displaying all directories and files.
Click on Startup Directory button
Switch to the directory in which you started DOS OOT.
Double click on a file
Open that file.
Double click on a directory
Change the directory in the Open File dialog to the selected directory.
94 Object Oriented Turing Reference Manual
Double click on a drive
Change the drive in the Open File dialog to the selected drive.
The Save As Dialog Box
In the Save As dialog box, the file names (not the directory names) in the File section
are grayed-out and may not be selected. If there are more than 48 files and directories, then
the dialog box will only display the first 48 in columns of 12. There will be arrows on the
right hand side to indicate that there are more files available to the right. (As you move to
the right beyond the fourth column, arrows will appear on the left hand side to indicate
that there are more files available to the left.)
In the File Name section, you use the keyboard to enter the name of the file. You may
enter a full or partial path name. You can use Ctrl+Y to erase what has already been typed
and use the standard cursor keys to maneuver in the text. You can enter only legal path
name characters in the File Name section.
For the File and Drive section, the following applies.
Keyboard
Tab
Cycle between the File section, the Drive section and the File Name
section.
F9
Switch to the directory in which you started DOS OOT.
A letter or number
Highlight the file or the drive that starts with that letter.
Esc
Cancel the dialog without giving a file name.
Chapter 4 : OOT for DOS
95
Enter (with a directory highlighted)
Change the directory in the Save As dialog box to the selected directory.
Enter (with a drive highlighted)
Change the drive in the Save As dialog box to the selected drive.
Arrow Keys
Move the highlighted file up, down, left or right.
Home / End
Move the highlight to the first / last entry.
Page Up / Page Down
Move the highlight three columns to the left / right.
Mouse
Click on Startup Directory button
Switch to the directory in which you started DOS OOT.
Double click on a directory
Change the directory in the Save As dialog to the selected directory.
Double click on a drive
Change the drive in the Save As dialog to the selected drive.
The Print Dialog Box
All
This Print dialog box allows you to select the printing options. You can select whether
the program is to be printed with a page header and whether line numbers are to be used
when printing out the program.
Keyboard
Esc or C
Cancel the dialog without printing.
O
Start printing with the selected options.
H
Toggle the Print Page Header option on and off.
L
Toggle the Print Line Numbers option on and off.
Tab or Arrow Keys
Cycle the default button between OK and Cancel.
Enter
Select the current button.
96 Object Oriented Turing Reference Manual
Mouse
Click on a check box
Toggle the option associated with that check box.
Click on OK button
Start printing with the selected options.
Click on Cancel button
Cancel the dialog without printing.
The Memory Info Alert Box
The Memory Info alert box requires no user input. It gives information about the
amount of memory being used by the file in memory and the amount of free space
available for use by DOS OOT. Pressing any key or clicking the mouse button will cause the
Alert box to go away.
The Find Dialog Box
All
The Find dialog box is used to enter a Find string and then search for it. In general, it is
expected that the user enters the string once and then uses Find Again (short cut Ctrl+G) to
step through each occurrence found. Because the Find dialog box has a text field, you must
use Alt+keystroke to toggle the buttons. You can use Ctrl+Y to erase what has already
been typed and use the standard cursor keys to maneuver in the text.
Keyboard
Esc or Alt+C
Cancel the dialog without searching.
Alt+O
Find the next occurrence of the Find string.
Chapter 4 : OOT for DOS
97
Alt+B
Toggle whether to search for the next occurrence later in the text or the
previous occurrence in the text.
Alt+W
Toggle wrap around search mode on and off. If you are searching
downward in wrap around search mode and you reach the end of the file,
find moves to the beginning of the file and continues the search. Likewise,
if you are search backwards and you reach the beginning of the file, the
search continues from the end of the file.
Enter
Select the default button (Cancel if no text has been entered, otherwise
Find the next occurrence).
Mouse
Click on a check box
Toggle the option associated with that check box.
Click on OK button
Find the next occurrence of the Find string.
Click on Cancel button
Cancel the dialog without searching.
The Change Dialog Box
All
The Change dialog box is used to enter a Find string and then search for it. Once
found, you can replace the Find string with the Replace string. Unlike other dialog boxes,
the Change dialog box stays up on the screen until you dismiss it. Using it, you can easily
change a small number of instances of a string, choosing which ones to leave intact and
which ones to change as you see the instances in the file.
To use the Change dialog box, enter in a Find string, then press Tab to move the cursor
to the Replace string. Once there, enter a Replace string. Once this is done, press Enter to
find the first occurrence of the Find string. The screen will scroll and you will see the search
string in the selection. If you want to replace this occurrence of the Find string with the
Replace string, press Enter or click on the Change, then Find button. The selection will be
replaced and the screen will automatically scroll to the next occurrence of the Find string. If
you don't want to replace this occurrence, then press Alt+F or click on the Find Next
button. This will cause the screen to scroll to the next occurrence without changing the
selection.
To change the current selection without going to the next selection, press Alt+H or click
on the Change button.
98 Object Oriented Turing Reference Manual
To change every occurrence of the Find string to the Replace string, press Alt+A or
click on the Change All button.
When finished, press Esc or click the Cancel button in order to dismiss the dialog box.
For either text field, you can use Ctrl+Y to erase what has already been typed and use
the standard cursor keys to maneuver in the text.
Keyboard
Esc
Cancel the dialog without searching.
Tab
Switch between editing the Find string and the Replace string.
Alt+B
Toggle whether to search for the next occurrence later in the text, or the
previous occurrence in the text.
Alt+W
Toggle wrap around search mode on and off. If you are searching
downward in wrap around search mode and you reach the end of the file,
find moves to the beginning of the file and continues the search. Likewise,
if you are search backwards and you reach the beginning of the file, the
search continues from the end of the file.
Enter
Select the default button (Cancel if no text has been entered, Find Next if
no text has yet been found, otherwise Change, then Find).
Alt+F
Find the next occurrence of the Find string.
Alt+C
Change the current selection to the Replace string and then find the next
occurrence of the Find string.
Alt+H
Change the current selection to the Replace string.
Alt+A
Change every occurrence of the Find string to the Replace string.
Mouse
Click on Find Next button
Find the next occurrence of the Find string.
Click on Change, then Find button
Change the current selection to the replace string and then find the next
occurrence of the Find string.
Click on Change button
Change the current selection to the Replace string.
Click on Change All button
Change every occurrence of the Find string to the Replace string.
Click on Cancel button
Cancel the dialog without searching.
Chapter 4 : OOT for DOS
99
Click on a check box
Toggle the option associated with that check box.
Click on a text field
Move the cursor to that text field and to the cursor position.
The Goto Line Box
All
The Goto Line dialog box allows you to jump directly to a particular line in the
program. To use it, enter the line number you wish to jump to and press Enter. The text
field will only allow you to enter numbers. Entering a number larger than the number of
lines in the program will move you to the end of the program. You can use Ctrl+Y to erase
what has already been typed and use the standard cursor keys to maneuver in the text
field.
Keyboard
Esc or Alt+C
Cancel the dialog without doing anything.
Alt+O
Jump to the entered line in the program.
Tab or Arrow Keys
Cycle the default button between OK and Cancel.
Enter
Select the default button.
Mouse
Click on OK button
Jump to the entered line in the program.
Click on Cancel button
Cancel the dialog without doing anything.
The Run with Arguments Dialog Box
100 Object Oriented Turing Reference Manual
The Run with Arguments dialog box is used to enter run-time arguments for the
program about to be executed, and allow you to specify how you want the input and
output to be redirected. To enter the arguments, type the arguments into the Program
Arguments text field. To redirect the input or output, select the radio button (the "( )") from
the Input From: or the Output To: list. This can be done with either keyboard or mouse.
For the text field, you can use Ctrl+Y to erase what has already been typed and use the
standard cursor keys to maneuver in the text.
Note that if you run the program without using Run with Arguments, then
automatically no arguments are selected and the input and output are not redirected. The
Run with Arguments dialog box also remembers the last settings so that you can choose
between running with no arguments and running with a set of arguments and input and
output redirection already set.
Keyboard
Esc or Alt+C
Cancel the dialog without doing anything.
Alt+R
Run the program with the option specified.
Tab
Cycle the default button between Run and Cancel.
Enter
Select the default button.
F3
Set input to come from the keyboard.
F4
Set input to come from a file. A dialog box identical in appearance to the
Open File dialog box appears to allow you to select the input file.
F5
Set input to come from a file, but echo input from the file to the output
device. See Input and Output Redirection. A dialog box identical in
appearance to the Open File dialog box appears to allow you to select the
input file.
F6
Set output to go to the screen.
F7
Set output to go to a file. A dialog box identical in appearance to the Save
As dialog box appears to allow you to specify the output file.
F8
Set output to go to a file and the screen at the same time. See Input and
Output Redirection later in this chapter. A dialog box identical in
appearance to the Save As dialog box appears to allow you to specify the
output file.
F9
Set output to go to the printer.
F10
Set output to go to the screen and a printer at the same time. See Input and
Output Redirection later in this chapter.
Mouse
Click on Run button
Run the program with the option specified.
Chapter 4 : OOT for DOS
101
Click on Cancel button
Cancel the dialog without doing anything.
Click on a radio button
Set the option associated with that radio button.
The Set Main Program Dialog Box
Win DOS Mac
X TOOT Term
The Set Main Program dialog box is used to select the program that you want to be run
every time you select the Run command. As such, it is essentially identical to the Open File
dialog box except that instead of opening a file, that file becomes the main program and
will be run whenever the Run command is given. See the Open File dialog box section for
more information.
The Set Run Directory Dialog Box
The Set Run Directory dialog box is used to select the directory in which you want
your program to execute. Once it is set, any file I/O will use that directory as the base
directory to read and write files. To use this dialog box, you simply change directories as
you would with the Open File or Save As dialog boxes, and then select Set Run Directory
when you are in the appropriate directory by either pressing F7 or clicking on the Set Run
Directory button with the mouse. All files are grayed out and may not be selected.
If there are more than 48 files and directories, then the dialog will only display the first
48 files and directories in columns of 12. There will be arrows on the right hand side to
indicate that there are more files available to the right. (As you move to the right beyond
102 Object Oriented Turing Reference Manual
the fourth column, arrows will appear on the left hand side to indicate that there are more
files available to the left.)
Keyboard
Tab
Cycle between the File section and the Drive section.
F7
Set the current directory displayed as the execution directory.
F9
Switch to the directory in which you started DOS OOT.
A letter or number
Highlight the file or the drive that starts with that letter.
Esc
Cancel the dialog without selecting a file.
Enter (with a file highlighted)
No effect.
Enter (with a directory highlighted)
Change the directory in the Open File dialog to the selected directory.
Enter (with a drive highlighted)
Change the drive in the Open File dialog to the selected drive.
Arrow Keys
Move the highlighted file up, down, left or right.
Home / End
Move the highlight to the first / last entry.
Page Up / Page Down
Move the highlight three columns to the left / right.
Mouse
Click on Set Run Directory button
Set the current directory displayed as the execution directory.
Click on Startup Directory button
Switch to the directory in which you started DOS OOT.
Double click on a file
No effect.
Double click on a directory
Change the directory in the Open File dialog to the selected directory.
Double click on a drive
Change the drive in the Open File dialog to the selected drive.
Chapter 4 : OOT for DOS
103
The Save Executable Dialog Box
Win DOS Mac
X TOOT Term
The Save Executable dialog box allows the user to specify the name of a the compiled
program when compiling a file. The file name for the executable should always in “.EXE”
in order so that the DOS will be able to execute the program when the name is typed.
Otherwise this dialog box is identical to the Save As dialog box. See the Save As dialog box
for more information.
The Compile Options Dialog Box
Win DOS Mac
X TOOT Term
The Compile Options dialog box allows the user to select the certain conditions on how
the compiled program will execute from DOS.
When Ctrl+Break is disabled, the user of the compiled program will not be able to stop
execution of the program except by turning the machine off. Ctrl+Break, Ctrl+C and Ctrl+B
will have no effect.
When No Return to DOS Message is set, when the program stops execution the user
will be immediately returned to the DOS prompt. If the program was in graphics mode, the
display will be set back to text mode erasing the screen. When this option is not selected,
when the program finishes execution, a message is displayed asking the user to press a key
to return to DOS.
When “No Erase before Execution” is set, the program will not erase the screen before
starting execution or after stopping execution. Note that if you switch to graphics mode, the
screen will be erased anyway. This option is most useful for DOS command line utilities.
Programs compiled with DOS OOT will however display the DOS 4G/W (the DOS
extender) banner before execution.
Keyboard
Esc or C
Cancel the dialog without compiling.
O
Start printing with the selected options.
Tab or Arrow Keys
Cycle the default button between OK and Cancel.
104 Object Oriented Turing Reference Manual
B
Toggle the Ctrl+Break Disabled check box.
R
Toggle the No Return to DOS Message check box.
E
Toggle the No Erase Before Execution check box.
Enter
Select the current button.
Mouse
Click on a check box
Toggle the option associated with that check box.
Click on OK button
Compile the program with the options selected.
Click on Cancel button
Cancel the dialog without compiling.
The Pick File
DOS OOT creates a file call TURING.PCK in the directory from which it was started.
This contains the names of the last nine files loaded into DOS OOT for use with the Pick...
menu item. Eliminating the file has no harmful consequences except that the Pick
command will not remember any previous files loaded in.
For single user systems, it may be convenient to have the TURING.PCK file stored
only in one location (the directory that stores TURING.EXE) and have this file used no
matter where you start up DOS OOT. You can change DOS OOT to have this behavior by
using the centralpickfile start up option. For more about start up options, see the next
section.
Start Up Options
There are three places you can specify startup options. These are:
•
•
•
in the TURING.CFG file in the directory where DOSOOT.EXE is found.,
in the TURING.CFG file where the user started DOS OOT,
and in the actual command line used in starting up DOS OOT.
The best way to set a standard set of options for all session of DOS OOT is to put all
the options in the file TURING.CFG in the same directory as TURING.EXE. That way, all
sessions will have the same startup options. You can then customize those options for
individual sessions by making your own TURING.CFG file in the directory in which you
start the session.
Chapter 4 : OOT for DOS
105
To specify a startup option in a TURING.CFG file, the file should list the options, one
per line. It is not necessary to put a "-" or a "/" in front of the option names. For example, a
TURING.CFG file could contain the following line:
printer=LPT2
This would select LPT2 as the default printer. In the command line, options must be
preceded with either a "-" or a "/". For example, you could start DOS OOT up with:
dosoot /alertclr=brown
Here is the complete list of available startup options. Options that have a * beside them
are the default (i.e. are selected if no option is given).
backup *
Automatically save the old file with ".BAK" extension.
nobackup
Don't save the old file.
nopara
Disable the Paragraphing menu item.
noshell
Disable the Shell menu item and the system procedure.
formfeed *
Print a form feed at the end of each print job.
noformfeed
Don't do a form feed at the end of each print job.
tabs *
Convert spaces to tabs when printing and saving.
notabs
Don't convert spaces to tabs when printing and saving.
header *
Print a header at the top of each print job with the date and file name.
noheader
Don't print a header at the top of each print job.
overstrike
Start in overstrike mode.
insert *
Start in insert mode.
mono
Force system to use monochrome display adapter.
graphics
Force DOS OOT to use a graphics display adapter.
checkclr *
Check color number for legality on draw... commands.
nocheckclr
Don't check color number for legality on draw... commands.
bw
Assume a black and white monitor.
clr
Assume a color monitor.
allprog *
In file dialogs, display all files by default.
tprog
In file dialogs, display only files ending with ".T" by default.
arrowcursor *
Use an arrow mouse cursor in the editor.
106 Object Oriented Turing Reference Manual
noarrowcursor Use standard block mouse cursor in editor. Use this option if cursor
appears as a group of Greek letters.
laserjet
Use HP LaserJet control codes when printing. Use this option if you are
using an HP LaserJet that is not configured for Epson emulation.
nolaserjet *
Use standard printer end of line code.
clock *
Display the time in the upper right corner of the screen.
noclock
Don't display the time in the upper right corner of the screen.
logo *
Display the alert at startup that gives the version number.
nologo
Don't display the alert at startup that gives the version number.
specialclear
On certain systems, the cls command doesn't work in MCGA mode. Set
this option to get it to work.
savepick *
Save the pick file when using DOS OOT.
nosavepick
Don't save a TURING.PCK file when using DOS OOT.
mouse *
Allow use of the mouse in the editor.
nomouse
Don't allow use of the mouse in the editor.
useb
Always display drive B, even if there's only one drive.
nouseb
Display drive B, only if there are two drives.
centralpickfile Use and save the TURING.PCK file found in the same directory as
TURING.EXE.
nocentralpickfile * Use and save the TURING.PCK file found in the directory from which
DOS OOT was started up.
nosound
Stop DOS OOT from producing any sound either in the editor or using the
sound commands in the language.
resetmouse
Use this option if the mouse doesn’t seem to be working. Forces the mouse
driver to do a hard reset.
cga
Force DOS OOT to use CGA as the default graphics mode (as it does in
Turing). Use this option for better compatibility with Turing programs.
vesa
Force the DOS OOT graphics library to use VESA graphics commands for
SuperVGA graphics modes. Normally DOS OOT correctly detects the
graphics card in the machine, however in some systems, it can fail (for
example, the ATI Mach64 graphics card). Use this option to get them to
work correctly.
Chapter 4 : OOT for DOS
107
novesa*
Autodetect the graphics adapter and use the appropriate SuperVGA
graphics commands where applicable.
printermargin=n Set the left hand margin on printouts to start printing n spaces to the
right. By default, n = 0.
linesperpage=n Set DOS OOT to print a form feed every n lines. By default, n = 0 which
corresponds to not doing any form feeds.
stack=n
Set DOS OOT to allocate n kilobytes for the stack during execution. By
default n = 32.
line=n
Set DOS OOT to have a line buffer size of n kilobytes. By default,
maxfiles=n
Set the maximum number of files that DOS OOT can read in a directory. By
default n = 150.
printer=name
The name of the printer that DOS OOT will send printouts to (default:
LPT1).
fontdir=name
The name of the directory where the MS Windows fonts are stored (for use
by the Font module) (default: C:\WINDOWS\SYSTEM).
startup=name
The directory that the user will be in when DOS OOT starts up (default:
The current directory).
n = 64.
Note that in the following, colors (clr) can be from 0 to 15, or one of black, blue, green,
cyan, red, magenta, purple, brown, white, grey, gray, brightblue, lightblue, brightgreen,
lightgreen, brightcyan, lightcyan, brightred, lightred, brightmagenta, lightmagenta,
brightpurple, lightpurple, yellow, brightwhite or lightwhite.
Background colors (bkclr), can be from 0 to 7, or one of black, blue, green, cyan, red,
magenta, purple, brown and white.
menuclr=clr
The color of text of menu items (default: black).
menudisabledclr=clr
The color of text of disabled menu items (default: gray).
menuintenseclr=clr
The color of text of intense letters in the menu
(default: brightwhite).
menubklclr=bkclr
The color of the background in the menus (default: white).
textclr=clr
The program text color (default: white).
texthiliteclr=clr
The program highlighted text color (default: brightwhite).
textbkclr=bkclr
The background color of the program (default: black).
textslctbkclr=bkclr
The color of the background in selected text (default: blue).
statusclr=clr
The color of the text in the status line (default: black).
statusbkclr=bkclr
The color of the background of the status line (default: cyan).
108 Object Oriented Turing Reference Manual
messageclr=clr
The color of editor messages (default: yellow).
messagebkclr=bkclr
The background color of the editor messages
(default: black).
dlgclr=clr
The color of text in dialog boxes (default: black).
dlgintenseclr=clr
The color of intense text in dialog boxes (default: brightwhite).
dlgbkclr=bkclr
The background color of dialog boxes (default: white).
dlgdisabledclr=clr
The color of disabled text items in dialog boxes
(default: gray).
dlgborderclr=clr
The color of the border in dialog boxes
(default: brightwhite)
alertclr=clr
The color of the text in alert boxes (default: black).
alertbkclr=bkclr
The background color in alert boxes (default: red).
progerrclr=clr
The color of DOS OOT error messages (default: brightred).
progerrbkclr=bkclr
The background color of the DOS OOT error messages (default:
black).
Compilation to Executable File
DOS OOT has the ability to compile files to produce a standalone executable (a file that
does not need DOS OOT in order to run). To produce the stand alone executable, the user
should select Compile to .EXE from the Run menu, supply a filename that ends in “.EXE”
and compile with the desired compile options (explained previously in The Compile
Options Dialog Box section).
The executables produced are yours alone to distribute or sell as the user sees fit. No
royalties or restrictions apply.
The compiler works by compiling the program to pseudo-code (a sort of generic
assembly language) and then attaching it to the DOS OOT executor. The DOS OOT
executor is fairly large, so users will find that all generated executables are about 400KB in
size (the size of the executor) and then grow slowly after that.
The DOS OOT executor is a Extended DOS application. This means that a DOS
extender must be available to run the program. The DOS extender is called DOS/4GW and
must be included with the executable file. Thus to distribute your program, you must make
a copy of both your executable program, any data files it may use and DOS4GW.EXE.
DOS4GW.EXE can be found in the same directory as DOS OOT and may be freely copied
from there. Note that when running the program, DOS4GW.EXE must be either in the
current directory or in the DOS path.
Note that compilation of a program also compiles all the units (modules and classes)
and include files that the program uses. It is not necessary to distribute any source code
Chapter 4 : OOT for DOS
109
with the executable file. Data files (files opened using the open statement) are not compiled
in and must be included separately.
Input and Output Redirection
DOS OOT has the capability of redirecting program input and output. This enables a
user to write a program that normally reads from the keyboard and writes to the screen to
read input from a file and send output to a file at run-time. This can be used for a variety of
purposes such as testing programs with certain input from a file or saving a copy of output
to a file for later use.
The method for redirecting input or output is to use the Run with Arguments menu
item. When the dialog box comes up, you are given the opportunity to redirect input to
come from the keyboard, file or file with echo. Likewise, you can redirect output to go to
the screen, file, screen and file, printer or screen and printer.
Because a file can hold only textual information, using graphics when sending output
to a file causes a run-time error.
The exact meaning of the redirection command is given below:
Input from:
file
Read input from a specified file instead of the keyboard. Do not
echo the input to the screen.
file with echo
Read input from a specified file instead of the keyboard. Echo the
input to the screen.
Output to:
file
Send output to a specified file instead of the screen. Do not display
any output on the screen.
screen and file
Send output to the screen and a specified file.
printer
Send output to the printer instead of the screen. Do not display
any output on the screen.
screen and printer Send output to the screen and the printer.
Summary
Input from keyboard / Output to screen
Reads input from keyboard. Sends output to the screen.
Input from infile / Output to screen
110 Object Oriented Turing Reference Manual
Reads input from infile. Sends output to the screen. Input from infile is not
seen on the screen.
Input from infile with echo / Output to screen
Reads input from infile. Sends output to the screen. Input from infile is seen
on the screen.
Input from keyboard / Output to outfile
Reads input from keyboard. Sends output to outfile. Input from the keyboard
appears on the screen and not in outfile.
Input from infile / Output to outfile
Reads input from infile. Sends output to outfile. Input from infile is not seen
on the screen or in outfile. The screen is blank while the program is running
because all output is going to outfile.
Input from infile with echo / Output to outfile
Reads input from infile. Sends out put to outfile. Input from infile is seen in
outfile. The screen is blank while the program is running because all output is
going to outfile.
Input from keyboard / Output to screen and outfile
Reads input from keyboard. Sends output to both screen and outfile. Input
from keyboard appears on both the screen and in outfile.
Input from infile / Output to screen and outfile
Reads input from infile. Sends output to both screen and outfile. Input from
infile is not seen on the screen or in outfile.
Input from infile / Output to screen and outfile
Reads input from infile. Sends output to both screen and outfile. Input from
infile appears on both the screen and in outfile.
Input from keyboard / Output to the printer
Reads input from keyboard. Sends output to the printer. Input from the
keyboard appears on the screen and not on the printer.
Input from infile / Output to the printer
Reads input from infile. Sends output to the printer. Input from infile is not
seen on the screen or on the printer. The screen is blank while the program is
running because all output is going to the printer.
Input from infile with echo / Output to the printer
Reads input from infile. Sends out put to the printer. Input from infile is seen
on the printer. The screen is blank while the program is running because all
output is going to the printer.
Chapter 4 : OOT for DOS
111
Input from keyboard / Output to screen and the printer
Reads input from keyboard. Sends output to both screen and the printer.
Input from keyboard appears on both the screen and on the printer.
Input from infile / Output to screen and the printer
Reads input from infile. Sends output to both screen and the printer. Input
from infile is not seen on the screen or on the printer.
Input from infile / Output to screen and the printer
Reads input from infile. Sends output to both screen and the printer. Input
from infile appears on both the screen and on the printer.
112 Object Oriented Turing Reference Manual
Chapter 4 : OOT for DOS
113
Chapter 5
The GUI Module
Introduction
Since the introduction of the Macintosh, graphical user interfaces (GUI) have been
becoming more and more common. Most commercial programs written for either the
Macintosh or Microsoft Windows make use of GUI elements to make their program easier
to use.
For some time, students have been requesting methods of putting GUI elements such as
buttons, check boxes, radio buttons, etc, into their Turing programs. With the Fall 1998
release of OOT, we introduced a new set of predefined subprograms that allow students to
add numerous GUI elements to their programs quickly and easily. These subprograms
allow students to create: buttons, check boxes, radio buttons, sliders, scroll bars, picture
buttons, radio picture buttons, text fields, lines, text labels, and frames.
The entire GUI Library is written in Turing and the source is included with the OOT
distribution. The GUI library is completely procedure-oriented. This means that it is not
necessary to know object-oriented programming or concepts in order to be able to use the
library. Advanced students are welcome (in fact, encouraged) to look at the programs as an
example of a large project written in Turing. We hope that there will be enterprising
students who will be inspired to add new widgets to the library and encourage those who
do so to submit them to Holt Software for possible inclusion into the next version of the
library.
Here is a window with a few widgets.
Some GUI Widgets (Output of Example.dem)
The GUI library is usable by students who understand the concept of subprograms. In
order to use the GUI library, students must write procedures (although they may be as
simple as the student desires). We therefore suggest that teachers introduce students to the
GUI library in a Grade 11 computer science course.
Note: Object Oriented Turing has not changed. It is not a visual building language.
Students wishing to use the GUI library will be writing programs to create and use these
GUI element, not spending their time visually building user interfaces (which may be fun,
but teaches very little). In keeping with the tradition of Turing, the more the students learn
about computer science, the more interesting their programs will be, GUI or no GUI!
Terms
This document will use the term "Turing" to refer to the Turing and Object Oriented
Turing. Thus when we speak of Turing programs, this will mean programs written in
Turing or Object Oriented Turing.
The term "Widget" means any graphical user interface element. A button, a check box
or radio button are all examples of widgets.
The term "Event" means either a keystroke or a mouse button being pressed in a
window.
Example Programs
All example programs shown here are located in the %oot/examples/GUI directory. (In
other words, start in the same directory as WinOOT, move to the examples folder and then
the GUI folder.) All the available GUI widgets have example programs to demonstrate their
use. Examine the file README.TXT for a description of the files.
General Principles of the GUI Library
Here are some general instructions for the use of the GUI library. Read this section before
looking at the specifics of various routines.
•
•
•
All the subprograms for the GUI library are placed in a module called GUI. To call any
of the subprograms, preface the name of the subprogram with "GUI.". For example, the
CreateLabel subprogram would be called using GUI.CreateLabel.
In general, most widgets have a Create subprogram. For example, buttons have a
CreateButton subprogram, radio buttons have a CreateRadioButton subprogram, and so
on. The Create subprogram takes as parameters things such as the location, the size of
the GUI element, and the name of a procedure to be called when the widget is clicked.
This procedure must be declared before the call to the Create subprogram.
For most widgets, there are two forms of the Create subprogram. The Create
subprogram and the CreateFull subprogram. The difference between the two is that the
CreateFull subprogram allows the user to define more parameters that are otherwise set
to default values. For example, the GUI.CreateButton procedure allows the user to
specify the x and y location of the button, the width of the button, the text that appears
in the button, and the procedure to call when the button is clicked. The
GUI.CreateButtonFull routine specifies those and also allows the user to specify the
height of the button (otherwise set to a height that will fit the label), a short cut
keyboard character that allows the user to "press" the button using a keyboard and a
116 Object Oriented Turing Reference Manual
•
•
parameter to allow the user to specify if this button is the "default" button (the one
"pressed" if the user presses the Enter key).
All Create subprograms return an integer. This number is the ID of the widget that has
been created. You need to use this ID if you want to do anything to the widget later,
such as move it, change its size, hide it, and so on. Most simple programs can safely
ignore the widget ID, although they will need to handle the return value from the
function.
After all the widgets have been created, the program must repeatedly call
GUI.ProcessEvent until the function returns true.
% Now process events until the user aborts the program.
loop
exit when GUI.ProcessEvent
end loop
GUI.ProcessEvent checks for user input from the mouse or the keyboard and then
checks to see if the user has clicked on a widget. If the user has, then it responds
appropriately (toggling the check box, pressing the button, etc.) and then if
appropriate, calls the procedure the user supplied in the Create subprogram.
GUI.ProcessEvent returns true when the GUI.Quit has been called, otherwise it returns
false.
•
When a program is finished execution (for example if the user selected "Quit" or "Exit"
from the file menu), it should call the GUI.Quit procedure. This will cause the
GUI.ProcessEvent loop to exit. The program should have any clean up code placed after
the end loop.
Here is a very simple example of a program that puts "Hello" every time a button is
pressed.
% The "Hello" program.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:200;200") % Shrink the run window
% The procedure called when the button is pushed.
procedure PutHello
put "Hello"
end PutHello
% Create the button. The number returned is the ID number of the button.
var b : int := GUI.CreateButton (100, 100, 0, "Say Hello", PutHello)
% Now process events until the user aborts the program.
loop
exit when GUI.ProcessEvent
end loop
Chapter 5 : The GUI Module
117
Here is the output window after the user has pressed the button once.
Output of Hello.dem
Active and Passive Widgets
Widgets come in two forms. Active widgets are ones that respond to keystrokes and
button clicks. Passive widgets do not respond to anything. Examples of passive widgets are
lines, frames, labels, labelled frames and pictures. Passive widgets are generally used to
organize the output window.
Here is an example of a small program that show some passive widgets.
% The "passive" program
% This demonstrates some of the passive widgets such as:
% Lines, Frames, Labelled Frames, Labels and Pictures.
import GUI in "%oot/lib/GUI"
% We'll need a picture for our Picture widget. Most likely
% you would normally have it saved in an external file and
% use Pic.FileNew to read it into a picture. For the example
% program we'll construct it by hand.
Draw.FillOval (50, 50, 50, 50, blue)
Draw.FillBox (17, 17, 83, 83, brightred)
Draw.FillStar (17, 17, 83, 83, brightgreen)
Draw.FillMapleLeaf (37, 37, 63, 63, brightpurple)
var pic := Pic.New (0, 0, 100, 100)
View.Set ("graphics:310;335")
#if _DOS_ then
Draw.Cls
#else
% The background must be gray for indented and exdented
% items to be visible.
GUI.SetBackgroundColor (gray)
#end if
% Now place the widgets.
% Three lines of the different types with labels
var line1 := GUI.CreateLine (70, 10, maxx - 10, 10, GUI.LINE)
var label1 := GUI.CreateLabelFull (60, 10, "Line", 0, 0,
GUI.RIGHT + GUI.MIDDLE, 0)
var line2 := GUI.CreateLine (70, 30, maxx - 10, 30, GUI.INDENT)
var label2 := GUI.CreateLabelFull (60, 30, "Indent", 0, 0,
GUI.RIGHT + GUI.MIDDLE, 0)
var line3 := GUI.CreateLine (70, 50, maxx - 10, 50, GUI.EXDENT)
118 Object Oriented Turing Reference Manual
var label3 := GUI.CreateLabelFull (60, 50, "Exdent", 0, 0,
GUI.RIGHT + GUI.MIDDLE, 0)
% Now place the frames
var frame1 := GUI.CreateFrame (10, 70, 100, 120, GUI.LINE)
var label4 := GUI.CreateLabelFull (10, 70, "Line", 90, 50,
GUI.CENTER + GUI.MIDDLE, 0)
var frame2 := GUI.CreateFrame (110, 70, 200, 120, GUI.INDENT)
var label5 := GUI.CreateLabelFull (110, 70, "Indent", 90, 50,
GUI.CENTER + GUI.MIDDLE, 0)
var frame3 := GUI.CreateFrame (210, 70, 300, 120, GUI.EXDENT)
var label6 := GUI.CreateLabelFull (210, 70, "Exdent", 90, 50,
GUI.CENTER + GUI.MIDDLE, 0)
% Now place the labelled frames
var frame4 := GUI.CreateLabelledFrame (10, 140, 100, 190, GUI.LINE, "Line")
var frame5 := GUI.CreateLabelledFrame (110, 140, 200, 190, GUI.INDENT,
"Indent")
var frame6 := GUI.CreateLabelledFrame (210, 140, 300, 190, GUI.EXDENT,
"Exdent")
% Place the picture
var label7 := GUI.CreateLabel (30, 315, "Picture without merge")
var pic1 := GUI.CreatePicture (30, 210, pic, false)
#if not _DOS_ then
var label8 := GUI.CreateLabel (maxx - 130, 315, "Picture with merge")
var pic2 := GUI.CreatePicture (maxx - 130, 210, pic, true)
#end if
% This loop doesn't do much since none of the widgets have any actions.
loop
exit when GUI.ProcessEvent
end loop
Here is the output window from the program with some labels, a line, a picture, and a
labelled frame.
Chapter 5 : The GUI Module
119
Output of Passive.dem
When an active widget is initialized, usually an action procedure must be specified. This
is the name of a procedure that will be called when the widget is selected. For example, in
the Hello program, the PutHello procedure was specified as the action procedure of the
button. Whenever the button was pressed, the PutHello procedure was called.
Some action procedures have arguments. For example, the action procedure for a slider has
a parameter of the current value. This allows the procedure to use the current value
without having to call a GUI subprogram to get the current slider value.
120 Object Oriented Turing Reference Manual
Keyboard Shortcuts
Several types of widgets can have “shortcuts”. A shortcut is simply a keystroke that has
the same effect as clicking on the widget. When you specify a shortcut to a widget in the
CreateFull procedure for the widget, you must specify a single character. The easiest way to
do this is to use the chr function with the ASCII value of the character to be used as the
shortcut. You can also specify control characters using the “^” notation. For example, the
character Ctrl+F can be expressed as “^F” in Turing.
The following characters cannot be used as shortcuts because the OOT environment
uses them for various purposes (stopping or rerunning programs, and so on.): Ctrl+C,
Ctrl+D, Ctrl+Z, F1, F11 and F12.
Background Color
It is common for windows to have a different background color from the standard
white (or black under DOS). To change the background color of a window (or the screen
under DOS), use the GUI.SetBackgroundColor procedure. This procedure takes one
parameter, the new background color. It redraws the window in the background color and
then redraws all the widgets. It also notifies the widgets about the new background color so
that when the widget is erased or moved, the location of the widget is filled with the new
background color instead of white (or black under DOS).
Note that Microsoft Windows dialog boxes often have a background color of gray. In
order to simulate that, you should give the command GUI.SetBackgroundColor (gray) before
creating widgets.
Several widgets (Canvas, Frame, Labelled Frame, Text Field and Text Box) can have
borders of either type INDENT or EXDENT. These borders give a sort of 3-D appearance to
the widget. However, they require that the background be set to gray.
Here is an example of a small program that creates a Canvas with a 3-D appearance
and then draws circles in the corner.
% The "Canvas1" program.
% Create a canvas and draw four circles on it.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:200;200")
#if not _DOS_ then
% Necessary for a 3-D look for the canvas
GUI.SetBackgroundColor (gray)
#end if
% This procedure is needed as an argument to CreateCanvasFull.
procedure DoNothing (mx, my : int)
end DoNothing
% Create a label for the canvas. We could use CreateLabelFull for more
% precise alignment.
Chapter 5 : The GUI Module
121
var label1 := GUI.CreateLabel (70, 182, "Four Circles")
% Create the canvas. We need to use CreateCanvasFull in order to
% specify the type of border.
var canvas := GUI.CreateCanvasFull (20, 20, 160, 160, GUI.INDENT,
DoNothing, DoNothing, DoNothing)
% Draw the four ovals. Notice that they don't extend off the canvas
% and the co-ordinates they use are relative to the canvas, not the window.
const radius := 70 % Half the width - 10
GUI.DrawFillOval (canvas, 0, 0, radius, radius, brightred)
GUI.DrawFillOval (canvas, 160, 0, radius, radius, brightgreen)
GUI.DrawFillOval (canvas, 0, 160, radius, radius, brightblue)
GUI.DrawFillOval (canvas, 160, 160, radius, radius, brightpurple)
Here is the output window.
Output of Canvas1.dem
Widget Sizes
The size that you specify a widget to be is not necessarily the actual size that the widget
will appear. In fact for many widgets, you can specify a width and height of 0 for the
widget and lets the initializer decide how large the widget should be. Another example is
with check boxes, where if you specify the check box to be right justified, the x and y
coordinates indicate the lower-right corner instead of the lower-left corner as usual. This
122 Object Oriented Turing Reference Manual
means that you may have to do some experimentation to determine where you want the
widgets to be placed. Read the page on each subprogram that you use to find out exactly
what you are specifying with the x, y, width and height parameters.
If you are trying to align widgets together (for example aligning scroll bars with a
canvas), use the GUI.GetX, GUI.GetY, GUI.GetWidth, and GUI.GetHeight functions to
determine the size of the object.
Positioning Text Labels (Aligning Labels with
Widgets)
It is very common to want to align text labels with widgets on the screen. There are a
few tips and tricks to doing so successfully. To align a text label with a widget, it is simply a
matter of using the GUI.CreateLabelFull function with the appropriate x, y, width, height and
alignment arguments.
If you are left or right aligning a label, then generally you will want the x coordinate to
specify the edge to be aligned from and the width parameter should be set to 0. Similarly, if
you are top or bottom aligning a label, then the y coordinate should specify the edge to be
aligned from and the height parameter should be set to 0.
To align a widget horizontally with a widget, choose GUI.CENTER for the horizontal
alignment and the use the x coordinate and width of the widget as the label's x coordinate
and width. You can get the x coordinate and width of a widget using GUI.GetX and
GUI.GetWidth.
Likewise, to align a widget vertically with a widget, choose GUI.MIDDLE for the
vertical alignment and the use the y coordinate and height of the widget as the label's y
coordinate and height. You can get the y coordinate and height of a widget using GUI.GetY
and GUI.GetHeight.
Here is an example illustrating the placement of a label at the center of each of four
sides of a widget called w:
import GUI in "%oot/lib/GUI"
View.Set ("graphics:200;50")
procedure DoNothing (text : string)
end DoNothing
var w : int := GUI.CreateTextField (50, 15, 100, "", DoNothing)
% These following lines are the important part of the program.
var left := GUI.CreateLabelFull (GUI.GetX (w) - 2, GUI.GetY (w),
"Left", 0, GUI.GetHeight (w), GUI.RIGHT + GUI.MIDDLE, 0)
var above := GUI.CreateLabelFull (GUI.GetX (w),
GUI.GetY (w) + GUI.GetHeight (w) + 2, "Above", GUI.GetWidth (w), 0,
GUI.CENTER + GUI.BOTTOM, 0)
var right := GUI.CreateLabelFull (GUI.GetX (w) + GUI.GetWidth (w) + 2,
GUI.GetY (w), "Right", 0, GUI.GetHeight (w), GUI.LEFT + GUI.MIDDLE, 0)
var below := GUI.CreateLabelFull (GUI.GetX (w), GUI.GetY (w) - 2,
"Below", GUI.GetWidth (w), 0, GUI.CENTER + GUI.TOP, 0)
Chapter 5 : The GUI Module
123
Here's the result. Note that the formula for aligning a label with a widget is the same
for any type of widget.
Text Box Aligned with Four Labels
Here's an example illustrating aligning a widget with the top of the window. Notice
that the label is center aligned with x of 0 and width of maxx, and top aligned with a y of
maxy and a height of 0.
var title := GUI.CreateLabelFull (0, maxy, "Title", maxx, 0,
GUI.CENTER + GUI.TOP, 0)
Finally, here's an example illustrating the placement of a label in the center of the
screen. Notice that the label is center aligned with x of 0 and width of maxx, and middle
aligned with a y of 0 and a height of maxy.
var title := GUI.CreateLabelFull (0, 0, "Title", maxx, maxy,
GUI.CENTER + GUI.MIDDLE, 0)
Note that if a label's position or size is changed with GUI.SetPosition, GUI.SetSize or
GUI.SetPositionAndSize, it still retains its alignment with respect to its new x, y, width, and
height values.
Canvases
The canvas is a rather unique widget. It is essentially a drawing surface that you place
in the window. There are calls using a canvas widget that essentially duplicate all the
standard Draw module calls, along with calls corresponding to Font.Draw and various Pic
module calls.
The difference is that the calls using the Canvas widget use (0, 0) to mean the bottom
left corner of the canvas (not the window) and all drawing is clipped to the canvas
(meaning that if you accidentally draw off the canvas, the part of the picture outside the
bounds of the canvas will not appear). One of the most common bugs is to accidentally use
the actual Draw module routines instead of the GUI.Draw routines when drawing in a
canvas. If the drawing is goes outside the bounds of the Canvas, you have made this
mistake.
Another feature of the Canvas widget is that you can specify a procedures to be called
whenever a user clicks in the Canvas widget, drags the mouse with the mouse button down
and then lets go of the mouse button. These procedures allow your program to respond to
mouse activity taking place in the canvas widget.
Here is a program that uses a Canvas to allow the user to draw and a button to allow
the user to erase the drawing.
% The "Draw" program
import GUI in "%oot/lib/GUI"
124 Object Oriented Turing Reference Manual
View.Set ("graphics:300;300")
var oldx, oldy : int
var canvas : int % The drawing canvas.
var clear : int % The clear button.
% Called when the user presses the mouse button in the canvas.
% Sets the initial mouse position.
procedure MouseDown (mx, my : int)
oldx := mx
oldy := my
end MouseDown
% Called as the user drags the mouse with the button down in the canvas.
% Draws a line from the previous mouse position to the current position.
procedure MouseDrag (mx, my : int)
GUI.DrawLine (canvas, oldx, oldy, mx, my, colorfg)
oldx := mx
oldy := my
end MouseDrag
% Called when the mouse button is released.
procedure DoNothing (mx, my : int)
end DoNothing
% Called when the clear button is pressed.
procedure Clear
GUI.DrawCls (canvas)
end Clear
% Create the canvas
canvas := GUI.CreateCanvasFull (10, 30, maxx - 20, maxy - 40, 0,
MouseDown, MouseDrag, DoNothing)
% Create the clear button
clear := GUI.CreateButton (maxx div 2 - 20, 0, 40, "Clear", Clear)
loop
exit when GUI.ProcessEvent
end loop
Here is the output window after the user has drawn some lines.
Output of Draw.dem
Chapter 5 : The GUI Module
125
Multiple Windows
Object Oriented Turing for Windows allows for multiple run windows. This can be
used to add extra functionality to programs, however there are a few issues that must be
understood before multiple windows can be successfully used.
OOT uses the concept of selected windows and active windows. A selected window is
determined by the program and is changed by Window.Select. The selected window is the
window in which all output appears. When a widget is created, it is automatically created
in the selected window.
An active window is last window on which the user clicked. The active window is
shown by having its title bar highlighted. When a user types, all keystrokes are sent to the
active window.
It is entirely possible to have the selected window and the active window be to
different windows.
When you call getch, Mouse.ButtonWait, or any other input routine, OOT checks only the
selected window. The GUI Library works around this by selecting all the windows that
have widgets in them (one at a time, starting with the active window) and checking each
for events.
If you are processing an event from one of several windows, make certain that the
correct window is selected before you output your results. Note that the widgets
automatically select the correct window, so there is no need to change the selected window
before making any calls to the GUI module.
The GUI Library Internals
While it is not necessary to know the internals of the GUI Library to use it, we provide
this brief overview for those who wish to understand the inner workings of the library.
The GUI Library consists of four parts. The only part visible to the user is the GUI
module. This is located in "%oot/lib/GUI", where %oot is the directory in which the OOT
executable is located. The GUI module is essentially a series of procedures that provide a
front end to the Widget Module and the Widget Class Library.
The Widget Module is a module called WidgetModule that consists of a series of
subprograms that cover all the aspects of GUI's that do not pertain to a particular widget.
For example, the procedure to change the background color is here, as well as the
procedure to process an event. It is located in "%oot/lib/GuiClass/wdgtmod.tu"
The GUI Class Library consists of a series of classes arranged in a hierarchy illustrated
in the following figure. Most of the actual Turing code for the GUI Library is located in the
Widget Class Library. Each different type of widget has its own class. Widgets that share
common behavior have the same parent. For example, both the vertical and horizontal
slider have a slider class as a parent. Those classes whose names start with Generic are
abstract classes. They should not be instantiated themselves. They are used to define
common behavior among their subclasses. The Turing source for the classes can be found in
the directory "%oot/lib/GuiClass"
126 Object Oriented Turing Reference Manual
The fourth part is the WidgetGlobals module. This module that consists mostly of global
variables used by the GUI Class Library and the Widget module. It is located in
"%oot/lib/GuiClass/wdgtglob.tu"
Here is an example of how the GUI module works: when you create a button using
GUI.CreateButton, the CreateButton function in the GUI module creates an object of type
ButtonClass. (ButtonClass is found in the Widget Class Library discussed further down). It
then calls the Initialize procedure of the ButtonClass to initialize the button with the
specified parameters. Finally it assigns an ID number to the widget and arranges it in a
table for future reference. Here is another example: when you call a procedure like
GUI.Show, the Show procedure takes the widget ID, looks up the object that it represents,
and then calls the Show procedure of the object.
Students who wish to add new widgets to the GUI library will have to understand the
principles of object oriented programming, as they will be adding a new class to the GUI
Class Library and then adding new subprograms in the GUI module that will call their new
classes. (At the very least, a Create subprogram will be required for the new widget.)
A suggested project would be to create new versions of the ButtonClass, CheckBoxClass
and RadioButtonClass classes that are buttons, check boxes and radio buttons with the new
Windows 95/NT appearance. They could be called the Button95Class, CheckBox95Class and
RadioButton95Class. Properly written, these new classes should inherit from ButtonClass, etc.
and contain only those procedures that differ from the base class (presumably the
procedures that display the widget).
The GUI Class Library Hierarchy
GUI Module Routines Summary
The routines in the GUI module are divided into several different types. There are the
routines to create various widgets, the routines to create menus and menu items, the
routines to do general activities (such as processing an event, changing the background
color, etc.) and the routines that act on various types of widgets.
Here is the list of the routines that create widgets:
CreateButton, CreateButtonFull
Create a button.
CreateCheckBox, CreateCheckBoxFull
Create a check box.
CreateRadioButton, CreateRadioButtonFull
Create a radio button.
CreatePictureButton, CreatePictureButtonFull
Create a picture button.
CreatePictureRadioButton, CreatePictureRadioButtonFull
Create a picture radio button.
CreateHorizontalSlider
Create a horizontal slider.
Chapter 5 : The GUI Module
127
CreateVerticalSlider
Create a vertical slider.
CreateHorizontalScrollBar, CreateHorizontalScrollBarFull
Create a horizontal scroll bar.
CreateVerticalScrollBar, CreateVerticalScrollBarFull
Create a vertical scroll bar.
CreateCanvas, CreateCanvasFull
Create a canvas.
CreateTextField, CreateTextFieldFull
Create a text field.
CreateTextBox, CreateTextBoxFull
Create a text box.
CreateLine
Create a line.
CreateFrame
Create a frame.
CreateLabelledFrame
Create a labelled frame.
CreateLabel, CreateLabelFull
Create a label.
CreatePicture
Create a picture.
Here is the list of routines that create menus and menu items:
CreateMenu, CreateMenuFull
Create a menu.
CreateMenuItem, CreateMenuItemFull
Create a menu item.
Here is the list of general routines:
ProcessEvent
Process a single keyboard or mouse down event.
Quit
Tell the program to exit the event loop.
Refresh
Redraw all the widgets on the screen.
SetBackgroundColour
Change the window's background colour.
SetNullEventHandler
Set the null event handler.
SetKeyEventHandler
Set the keystroke event handler.
SetMouseEventHandler
Set the mouse event handler.
HideMenuBar
Hide the menu bar in the window.
ShowMenuBar
Show the menu bar in the window.
GetEventWidgetID
Get the selected widget's ID (used in a widget's action
procedure).
GetEventWindow
Get the window that the event took place in (used in a widget's
action procedure).
GetEventTime
Get the time (in milliseconds) that the event took place (used in
a widget's action procedure).
GetScrollBarWidth
Return the width of a scroll bar.
GetMenuBarHeight
Return the height of a menu bar.
GetVersion
Return the current version number of the GUI module.
Here is a list of routines that act on the widgets and the sort of widgets they act on.
128 Object Oriented Turing Reference Manual
Show
Display the widget.
All
Hide
Hide the widget.
All
GetX
Return the x coordinate of the
widget's left edge.
All
GetY
Return the y coordinate of the
widget's bottom edge.
All
GetWidth
Return the widget's actual width.
All
GetHeight
Return the widget's actual height.
All
Dispose
Dispose of the widget.
All
SetPosition
Set the widget's position.
All
SetSize
Set the widget's size.
All
SetPositionAndSize
Set the widget's position and size.
All
Enable
Enable the widget to respond to
events.
Active Widgets
Disable
Disable the widget from responding
to events.
Active Widgets
SetLabel
Set the widget's text label.
Button, Check Box,
Radio Button, Label,
Labelled Frame
SetDefault
Make the button the default button.
Button
GetCheckBox
Get whether a check box is filled.
Check Box
SetCheckBox
Set a check box to be filled or not.
Check Box
SelectRadio
Select a radio button.
Radio Button,
Picture Radio Button
GetSliderValue
Return the current value of the
slider.
Slider, Scroll Bar
SetSliderValue
Set the value of the slider.
Slider, Scroll Bar
SetSliderMinMax
Set the slider's minimum and
maximum.
Slider, Scroll Bar
SetSliderSize
Set the slider's length (or height).
Slider, Scroll Bar
SetSliderReverse
Reverse the direction of the slider.
Slider, Scroll Bar
SetScrollAmount
Set the scroll bar's thumb size and
the scroll amount for arrows/page
up and down.
Scroll Bar
Chapter 5 : The GUI Module
129
DrawArc, DrawBox, DrawCls, Routines that perform the same
function as the Draw module for the
DrawDot, DrawFill,
Canvas widget.
DrawFillArc, DrawFillBox,
DrawFillMapleLeaf,
DrawFillOval,
DrawFillPolygon,
DrawFillStar, DrawLine,
DrawMapleLeaf, DrawOval,
DrawPolygon, DrawStar,
DrawText
Canvas
FontDraw
"Font.Draw" for the Canvas widget.
Canvas
PicDraw, PicNew,
PicScreenLoad, PicScreenSave
Routines that perform the same
function as the Pic module for the
Canvas widget.
Canvas
SetXOR
Performs View.Set ("xor") for the
Canvas Widget.
Canvas
SetText
Set the text of a text field.
Text Field
SetSelection
Set the selection in the text field.
Text Field
SetActive
Make the text field the active one
(where keystrokes will go and
where the cursor blinks).
Text Field
ClearText
Clear a text box.
Text Box
AddText
Add text to a text box.
Text Box
AddLine
Add a line of text to a text box.
Text Box
130 Object Oriented Turing Reference Manual
Widgets - Common Routines
All of the procedures in this section can be used with any widget, although some may
have no effect (for example GUI.GetX on a menu item is meaningless).
GUI.Show
Displays a widget. Used in conjunction with Hide to hide and show
widgets. Hidden widgets cannot get events (i.e. respond to keystrokes
or mouse clicks).
GUI.Hide
Hides a widget. Used in conjunction with Show to hide and show
widgets. Hidden widgets cannot get events (i.e. respond to keystrokes
or mouse clicks). If an active text field (see text field) is hidden, then
any keystrokes in the window will be ignored.
GUI.GetX
GUI.GetY
Returns the x (y) coordinate of the left edge of a widget. Note that
this may be different from the x coordinate specified in the widget's
Create call. For example, if a radio button is created with right
justification, the x coordinate in the Create method specifies the right
edge.
Here is a small subprogram that should draw a rectangle entirely
around a widget (i.e. no part of the widget should stick out).
procedure WidgetRect (widgetID : int)
const x : int := GUI.GetX (widgetID)
const y : int := GUI.GetY (widgetID)
const width : int := GUI.GetWidth (widgetID)
const height : int := GUI.GetHeight (widgetID)
Draw.Box (x, y, x + width - 1, y + height - 1, black)
end WidgetRect
GUI.GetWidth
GUI.GetHeight
Returns the actual width (height) of a widget. Note that this may be
different from the width specified in the Create call (especially since
many widgets allow you to specify 0 for the width and let the GUI
module determine the necessary width).
GUI.Dispose
Eliminates a widget. It should be called in order to free up any
memory that the widget might have allocated. Note that you cannot
use the widget after it has been disposed of. If you wish to temporarily
get rid of a widget, consider using the Hide method and then the Show
method when you want to use it again.
GUI.SetPosition
Moves a widget to the specified location. If the widget is visible, it is
moved immediately to the new location. If the widget is hidden, it will
appear at the new location when the Show procedure is called. Note
that the location specified in GUI.SetPosition are the same as in the
Create method. For example, if you had specified a check box to be
right justified in the CreateCheckBoxFull function, then the location in a
call to SetPosition would specify the lower-right corner as opposed to
the lower-left corner.
Chapter 5 : The GUI Module
131
GUI.SetSize
Changes the size of a widget. If the widget is visible, its size is changed
immediately, otherwise the widget will appear in its new size when the
widget is next made visible. Note that the width and height parameters
are not necessarily the actual width and height of the widget. For
example, the TextField widget ignores the height parameter, calculating
the widget's actual height from the height of the text in the TextField.
GUI.SetPositionAndSize
Changes the position and size of the widget
simultaneously. It works the same way as the SetPosition and SetSize
procedures.
Widgets - Buttons
Output of Buttons.dem program
The button widget is used to implement a textual button. When you click on a button,
the button's action procedure is called. If a button is given a short cut, then entering the
keystroke will cause the action procedure to be called. It will not visibly cause the button to
depress.
If a button's width or height is set to zero (or not specified at all), then the button is
shaped to fit the text.
A button can be the default button for a window. If that is the case, then the button will
be drawn with a ticker border around it and if the user presses ENTER, then the button's
action procedure will be called.
When a button is not enabled, the text in the button is grayed out and the button no
longer responds to any mouse clicks or keystrokes until the button is enabled again.
GUI.CreateButton Creates and displays a button. GUI.CreateButton specifies the location,
width, text and action procedure of the button.
GUI.CreateButtonFull Creates and displays a button. GUI.CreateButtonFull also specifies
the height, keyboard shortcut and default status of the button.
GUI.Enable
GUI.Disable
Enables (disables) a button. Disabled buttons have their text grayed
out and cannot get events (i.e. respond to keystrokes or mouse clicks).
GUI.SetLabel
Changes the text of a button.
GUI.SetDefault
Sets the "default status" of a button. If a button is the default button,
then it is drawn with a heavy outline and it is activated when the user
presses ENTER (RETURN on a Macintosh).
132 Object Oriented Turing Reference Manual
Widgets - Check Boxes
Output of ChkBoxes.dem
The check box widget is used to implement a check box that can be set or unset. When
you click on a check box, the status of the check box flips from set to unset and back again
and the check box's action procedure is called with the new status as a parameter. If a check
box is given a short cut, then entering the keystroke will cause the check box to change
status and the action procedure to be called. The new status will be displayed immediately.
A check box's size is not specified during creation. It is determined based on the size of
the text. Instead the user specifies the lower-left corner of the check box (or the lower-right
if the check box is right justified).
When a check box is not enabled, the label beside the check box is grayed out and the
check box no longer responds to any mouse clicks or keystrokes until the check box is
enabled again.
GUI.CreateCheckBox Creates and displays a left aligned (check box to the left of the
label) check box. GUI.CreateCheckBox specifies the location, text and
action procedure of the check box.
GUI.CreateCheckBoxFull
Creates and displays a check box. GUI.CreateCheckBoxFull
also specifies the alignment of the check box (whether the checkbox is
the right or left of the text) and the check box's keyboard shortcut.
GUI.Enable
GUI.Disable
Enables (disables) a check box. Disabled check boxes have their text
grayed out and cannot get events (i.e. respond to keystrokes or mouse
clicks).
GUI.SetLabel
Changes the text of a check box.
GUI.GetCheckBox Returns a check box's status. If the check box is set (has an X in it),
GUI.GetCheckBox returns true, otherwise it returns false.
GUI.SetCheckBox Sets the status of a check box. It calls the check box's action procedure
with the new status and redraws the widget with the new status.
Chapter 5 : The GUI Module
133
Widgets - Radio Buttons
Output from Radios.dem
The radio button widget is used to implement a set of buttons of which one and only
one button must be selected at all times. (Think old-style radio station button. Selecting one
"deselects" the previously-selected station.) When you click on a radio button, any other
radio button that is part of the set is deselected and the radio button's action procedure is
called. If a radio button is given a short cut, then entering the keystroke will cause the radio
button to be selected (and any other radio button in the group to be de-selected) and the
action procedure to be called. The newly-selected or deselected radio buttons will be
displayed immediately.
When a radio button is created, the widget ID of another radio button must be
supplied. A value of zero for the widget ID indicates that this radio button is part of a new
group. The widget ID must be the ID of the last radio button added to the group. Because
radio buttons are almost always placed in groups you can specify –1 for the x and y
coordinates and the radio button will be placed just below the previous radio button and
retain the same alignment. When a group of radio buttons is selected, the first radio button
created in the group will be the selected one. You can change this by using the
GUI.SelectRadio procedure to select a different one.
A radio button's size is not specified during creation. It is determined based on the size
of the text. The user specifies the lower-left corner of the radio button (or the lower-right if
the radio button is right justified).
When a radio button is not enabled, the label beside the radio button is grayed out and
the radio button no longer responds to any mouse clicks or keystrokes until the radio
button is enabled again.
134 Object Oriented Turing Reference Manual
GUI.CreateRadioButton
Creates and displays a left aligned (circle to the left of the
label) radio button. GUI.CreateRadioButton specifies the location, text,
the radio button to be joined to and the action procedure of the radio
button.
GUI.CreateRadioButtonFull
Creates and displays a radio button.
GUI.CreateRadioButtonFull also specifies the alignment of the radio
button (whether the circle is the right or left of the text) and the radio
button's keyboard shortcut.
GUI.Enable
GUI.Disable
Enables (disables) a radio button. Disabled radio buttons have their
text grayed out and cannot get events (i.e. respond to keystrokes or
mouse clicks).
GUI.SetLabel
Changes the text of a radio button.
GUI.SelectRadio
Selects a radio button. The previously-selected radio button is "deselected". The action procedure of the radio button is called.
Widgets - Picture Buttons
Picture button widgets
The picture button widget (hereafter simply called a button) is simply a button with a
picture on it instead of text. The picture must be created by the program beforehand using
Pic.New or Pic.FileNew. The resulting picture can then be used as a parameter to
GUI.CreatePictureButton. In general, pictures should be a maximum of about 30 pixels high
and wide, although there is no built in limit in the GUI library.
When you click on a picture button, the picture button's action procedure is called. If a
picture button is given a short cut, then entering the keystroke will cause the action
procedure to be called. It will not visibly cause the button to depress.
If a button's width or height is set to zero (or not specified at all), then the button is
shaped to fit the picture.
When a picture button is not enabled, the picture button is grayed out and the picture
button no longer responds to any mouse clicks or keystrokes until the button is enabled
again.
GUI.CreatePictureButton
Creates and displays a picture button. The button is
automatically sized to fit the picture. If you need to know the precise
size of the button, use the GUI.GetWidth and GUI.GetHeight functions.
GUI.CreatePictureButton specifies the location, picture id and action
procedure of the button.
GUI.CreatePictureButtonFull Creates and displays a picture button.
GUI.CreatePictureButtonFull also specifies the width, height and
Chapter 5 : The GUI Module
135
keyboard shortcut of the button. It also specifies whether the button
picture should be merged with the background color or not.
GUI.Enable
GUI.Disable
Enables (disables) a picture button. Disabled picture buttons are
grayed out and cannot get events (i.e. respond to keystrokes or mouse
clicks).
Widgets - Picture Radio Buttons
Win DOS Mac
X TOOT Term
Picture radio button widgets
The picture radio button widget (hereafter simply called a button) is simply a picture
button (see Widget - Picture Button) that has the behavior of a radio button. This means
that one and only one picture radio button of a group is selected at any time. A selected
picture radio button is displayed as being pressed.
When you click on a picture button, the previously-selected picture radio button will be
de-selected and the new picture button's action procedure is called. If a picture button is
given a short cut, then entering the keystroke will cause the action procedure to be called and
the picture radio button will be drawn selected.
GUI.CreatePictureRadioButton Creates and displays a picture radio button. The button is
automatically sized to fit the picture. If you need to know the precise
size of the button, use the GUI.GetWidth and GUI.GetHeight functions.
GUI.CreatePictureRadioButton specifies the location, picture id and
action procedure of the button as well as the radio picture button to be
joined to.
GUI.CreatePictureRadioButtonFull
Creates and displays a picture radio button.
GUI.CreatePictureRadioButtonFull also specifies the width, height and
keyboard shortcut of the button. It also specifies whether the button
picture should be merged with the background color or not.
GUI.Enable
GUI.Disable
Enables (disables) a picture radio button. Disabled picture buttons
are grayed out and cannot get events (i.e. respond to keystrokes or
mouse clicks).
Widgets - Sliders
Win DOS Mac
X TOOT Term
Output of Sliders.dem
136 Object Oriented Turing Reference Manual
Sliders are the equivalent of a volume control on a stereo. To control a slider, the user
simply clicks on the control knob and slides the control left and right (up and down for a
vertical slider). Whenever the user slides the control knob, the action procedure of the widget
is called with the current value as a parameter.
The range of values that the slider will give is determined by the min and max
parameters in the Create call. The left side of the slider (bottom for vertical sliders)
represents the minimum value, while the right (top) represents the maximum value.
In some instances, you will want the reverse to be true (right/top is minimum). In that
case, call the GUI.SetSliderReverse procedure to flip the values of the slider.
Sliders always have a fixed height (for horizontal sliders) or width (for vertical sliders).
The length parameter in the Create call specifies how long the slider should be.
GUI.CreateHorizontalSlider Creates and displays a horizontal (left-right) slider.
GUI.CreateHorizontalSlider specifies the location, length, minimum and
maximum values for the slider, the initial value of the slider, and the
action procedure of the slider.
GUI.CreateVerticalSlider
Creates and displays a vertical (up-down) slider.
GUI.CreateVerticalSlider specifies the location, length, minimum and
maximum values for the slider, the initial value of the slider, and the
action procedure of the slider.
GUI.Enable
GUI.Disable
Enables (disables) a slider. Disabled sliders cannot get events
(i.e. respond to mouse clicks).
GUI.GetSliderValue
Returns the current value of a slider.
GUI.SetSliderValue
Sets the value of a slider. It moves the control knob on the slider to
the appropriate location and calls the slider's action procedure with the
new value.
GUI.SetSliderMinMax Sets the minimum and maximum values of a slider. It redraws the
control knob to take into account the new minimum and maximum. If
the current value of the slider is outside the new min/max, then the
value is adjusted appropriately.
GUI.SetSliderSize Changes the length of a slider. Redraws the slider and changes the
position of the control knob to take into account the new size of the
slider.
GUI.SetSliderReverse Sets a slider into (or out of, if already into) "reverse mode".
Normally, a slider is at its minimum value when the control knob is on
the left side (bottom for a vertical slider). This reverses it, so the
minimum value is when the slider is at the right side (top for vertical
sliders) of the track. Calling this routine a second time reverses it back
to normal. This procedure redraws the slider to move the control knob
to its new location.
Chapter 5 : The GUI Module
137
Widgets - Scroll Bars
Win DOS Mac
X TOOT Term
Output of ScrllBrs.dem
Scroll bars are usually used to allow a user to see a piece of a document that cannot be
displayed on the screen in its entirety. The picture above shows the scroll bars appearance.
To control a scroll bar, there are a few choices: the user can click on the thumb (the box in
the scroll bar) and slide it up and down, or the user can click in the scroll bar itself above or
below the thumb (in which case the thumb is moved up or down one "page"), or the user
can click on the up or down arrows at the ends of the scroll bar (in which case the thumb is
moved up one "line"). The programmer defines a page or a line. When the value of the
scroll bar changes, the action procedure of the scroll bar is called with the new value as a
parameter.
The range of values that the scroll bar will give is determined by the min and max
parameters in the Create call. The left side of the scroll bar (bottom for vertical scroll bars)
represents the minimum value, while the right (top) represents the maximum value. There
is also the "thumb size". This represents the range of values that can be seen at once on the
screen.
For example, if you have a window that can display 20 lines of text at once and there
are 100 lines of text, you would set min to 1, max to 100, and thumbSize to 20. The value
returned by the scroll bar would then be the line number of the first line on the screen to be
displayed. When the scroll bar was at its maximum value, it would return 81, since by
doing so, lines 81-100 would be displayed.
When a scroll bar is disabled or the scroll bar's thumb size is greater than the difference
between the minimum and maximum values (i.e. the item being scrolled fits in the
window), the scroll bar is deactivated. The bar is drawn in white rather than gray and the
arrows are grayed out. The scroll bar does not respond to mouse clicks.
In some instances, you will want the minimum and maximum values of the scroll bar to
be reversed (right/top is minimum). In that case, call the GUI.SetSliderReverse procedure to
flip the values of the scroll bar.
Scroll bars always have a fixed height (for horizontal scroll bars) or width (for vertical
scroll bars). To get the scroll bar's width, use the GUI.GetScrollBarWidth function. The length
parameter in the Create call specifies how long the scroll bar should be.
GUI.CreateHorizontalScrollBarCreates and displays a horizontal (left-right) scroll bar.
GUI.CreateHorizontalScrollBar specifies the location, length, minimum
and maximum values for the scroll bar, the initial value of the scroll
bar, and the scroll bar's action procedure.
By default, the arrow increment (the amount the value is changed
when the scrolling arrows are pressed) is set to one. The page
up/down increment (the amount the value is changed when the user
138 Object Oriented Turing Reference Manual
clicks in the bar to the right or left of the thumb) is set to one quarter
the difference between the minimum and the maximum. The "thumb
size" is set to zero (see the description of scroll bars for an explanation
of the thumb size).
GUI.CreateVerticalScrollBar Creates and displays a vertical (up-down) scroll bar.
GUI.CreateVerticalScrollBar specifies the location, length, minimum and
maximum values for the scroll bar, the initial value of the scroll bar,
and the scroll bar's action procedure.
By default, the arrow increment (the amount the value is changed
when the scrolling arrows are pressed) is set to one. The page
up/down increment (the amount the value is changed when the user
clicks in the bar to the right or left of the thumb) is set to one quarter
the difference between the minimum and the maximum. The "thumb
size" is set to zero (see the description of scroll bars for an explanation
of the thumb size).
GUI.CreateHorizontalScrollBarFull
Creates and displays a horizontal (left-right) scroll
bar. GUI.CreateHorizontalScrollBarFull also specifies the arrow
increment, page increment, and thumb size for the scroll bar.
GUI.CreateVerticalScrollBarFull
Creates and displays a horizontal (left-right) scroll
bar. GUI.CreateVerticalScrollBarFull also specifies the arrow increment,
page increment and thumb size for the scroll bar.
GUI.Enable
GUI.Disable
Enables (disables) a scroll bar. Disabled scroll bars cannot get events
(i.e. respond to mouse clicks).
GUI.GetSliderValue
Returns the current value of a scroll bar.
GUI.SetSliderValue
Sets the value of a scroll bar. It moves the control knob on the scroll
bar to the appropriate location and calls the scroll bar's action procedure
with the new value.
GUI.SetSliderMinMax Sets the minimum and maximum values of a scroll bar. It redraws
the control knob to take into account the new minimum and maximum.
If the current value of the scroll bar is outside the new min/max, then
the value is adjusted appropriately.
GUI.SetSliderSize Changes the length of a scroll bar. Redraws the scroll bar and changes
the position of the control knob to take into account the new size of the
scroll bar.
GUI.SetSliderReverse Sets a scroll bar into (or out of, if already into) "reverse mode".
Normally, a scroll bar is at its minimum value when the control knob is
on the left side (bottom for a vertical scroll bar). This reverses it, so the
minimum value is when the scroll bar is at the right side (top for
vertical scroll bars) of the track. Calling this routine a second time
reverses it back to normal. This procedure redraws the scroll bar to
move the control knob to its new location.
GUI.SetScrollBarAmount
Sets a scroll bar's arrow increment, page increment, and
thumb size. Redraws the scroll bar to take into account the new thumb
size.
Chapter 5 : The GUI Module
139
Widgets - Canvases
Win DOS Mac
X TOOT Term
Output of Canvases.dem
A canvas is a drawing surface for use by the program. It differs from just using the
window surface to draw on in that (0, 0) represents the lower-left corner of the canvas and
all drawing is clipped to the canvas. (This means that if you accidentally attempt to draw
outside of the canvas, it will not actually draw beyond the border of the canvas.)
Canvases have procedures that emulate all the procedures in the Draw module as well
as a procedure to emulate Font.Draw, Pic.Draw, Pic.New, Pic.ScreenLoad, and
Pic.ScreenSave.
You can get mouse feedback from a canvas. Using the GUI.CreateCanvasFull method,
you can specify three routines that are called when the mouse button is depressed while
pointing in a canvas. One routine will be called when the user presses the mouse button
down in a canvas. Another routine will be called while the user drags the mouse with the
mouse button down. This routine is repeatedly called whenever the mouse changes
position while the mouse button is down. The last routine is called when the mouse button
is released. All three routines take an x and y parameter, which is the location of the mouse
with respect to the canvas (i.e. (0, 0) is the lower-left corner of the canvas).
GUI.CreateCanvas Creates and displays a canvas. GUI.CreateCanvas specifies the location
and size of the canvas. The canvas will have a line border around it.
GUI.CreateCanvasFull Creates and displays a canvas. GUI.CreateCanvasFull also specifies
the type of border and three procedures to be called when a mouse is
pressed, dragged or released on the canvas.
GUI.Enable
GUI.Disable
Enables (disables) a canvas. Disabled canvases cannot get events
(i.e. respond to mouse clicks). If no mouse routines were specified (i.e.
the canvas was created with GUI.CreateCanvas and not
GUI.CreateCanvasFull) this routine essentially does nothing.
140 Object Oriented Turing Reference Manual
GUI.DrawArc
All these routines draw to a canvas in the same manner as the
GUI.DrawBox
similarly named Draw…, Pic… and Font.Draw subprograms.
GUI.DrawCls
All coordinates are based on the canvas and all drawing is clipped
GUI.DrawDot
to the canvas drawing surface. If the canvas is in "xor mode", all the
GUI.DrawFill
drawing will be done with "xor" set. (See View.Set for more
GUI.DrawFillArc information about "xor".)
GUI.DrawFillBox
GUI.DrawFillMapleLeaf
GUI.DrawFillOval
GUI.DrawFillPolygon
GUI.DrawFillStar
GUI.DrawLine
GUI.DrawBox
GUI.DrawMapleLeaf
GUI.DrawOval
GUI.DrawPolygon
GUI.DrawStar
GUI.DrawText
GUI.FontDraw
GUI.PicDraw
GUI.PicNew
GUI.PicScreenLoad
GUI.PicScreenSave
GUI.SetXOR
Sets the "xor mode" of a canvas. When in "xor mode", all the Draw...
procedures of a canvas are treated as if the View.Set ("xor") statement
had been executed before the Draw procedure.
Widgets - Text Fields
Output of TextFlds.dem
Chapter 5 : The GUI Module
141
A text field is a box for entering one line of text. When the user presses ENTER, the text
field's action procedure is called.
Only one text field is active at a time. The active text field has a blinking cursor (or its
selection highlighted). If a keystroke occurs when a window has an active text field in it, the
keystroke will be directed to the active text field. You can change which text field is active
with the GUI.SetActive procedure or by simply clicking on another text field with the
mouse.
When multiple text fields are created in a window, the first text field created is active
when the program begins.
The current version of the text field does not support cut and paste or keyboard
commands to extend the selection.
Because strings are a maximum of 255 characters, this is the maximum number of
characters in a text field.
The TAB character cycles between different text fields in a window. It cycles through
the text fields in the order in which they were created. BACK TAB (shift+TAB) cycles
through the fields in reverse order.
GUI.CreateTextField Creates and displays a text field. GUI.CreateTextField specifies the
location, width, initial text string, and action procedure of the text field.
The height of the text field is determined by the height of the font used
by the text field. The text field will have a line border around it.
GUI.CreateTextFieldFull
Creates and displays a text field. GUI.CreateTextFieldFull
also specifies the type of border, font for entered text, and kind of input
restriction (integer only, etc.)
GUI.Enable
GUI.Disable
Enables (disables) a text field. Disabled picture buttons are grayed
out and cannot get events (i.e. respond to keystrokes or mouse clicks).
GUI.SetText
Sets the text of a text field. The cursor is set the beginning of the text.
GUI.GetText
Returns the current text of a text field.
GUI.SetSelection
Sets the selection (the selected text) in a text field.
GUI.SetActive
Makes a text field active.
Widgets - Text Boxes
Win DOS Mac
X TOOT Term
Output of TextBxs.dem
A text box is a box used for displaying larger quantities of text. It has both vertical and
horizontal scroll bars to allow the user to scroll through all the text in the box.
142 Object Oriented Turing Reference Manual
GUI.CreateTextBoxCreates and displays a text box. GUI.CreateTextBox specifies the
location and size of the text box. The text box will have a line border
around it.
GUI.CreateTextBoxFullCreates and displays a text box. GUI.CreateTextBoxFull also
specifies the type of border and the font for displayed text.
GUI.ClearText
Clears all the text in a text box.
GUI.AddText
Adds text to the current line of the text box. Does not add a newline
after the text. Equivalent to put text ... This scrolls the text box (if
necessary) so that the added text is now visible. To move the cursor to
the end of the text without adding any extra text, call GUI.AddText
with "" for the text parameter.
GUI.AddLine
Adds text to the current line of the text box followed by a newline.
Equivalent to put text. This scrolls the text box (if necessary) so that the
added text is now visible.
Widgets - Lines
Win DOS Mac
X TOOT Term
Output of Lines.dem
Lines are organizational elements that make the window look better and help organize
the GUI elements.
GUI.CreateLine
Creates and displays a line. GUI.CreateLine specifies the end points of
the line (which must be either vertical or horizontal) and the type of
line.
Widgets - Frames
Win DOS Mac
X TOOT Term
Output from Frames.dem
Frames are organizational elements that make the window look better and help
organize the GUI elements. Frames and labelled frames are the only widgets in which other
widgets can be placed.
Chapter 5 : The GUI Module
143
GUI.CreateFrame Creates and displays a frame. GUI.CreateFrame specifies the
coordinates of the lower-left and upper-right corner of the frame and
the type of border of the frame.
Widgets - Labelled Frames
Win DOS Mac
X TOOT Term
Output from LblFrms.dem
Labelled frames are organizational elements that make the window look better and
help organize the GUI elements. Frames and labelled frames are the only widgets in which
other widgets can be placed.
GUI.CreateLabelledFrame
Creates and displays a labelled frame.
GUI.CreateLabelledFrame specifies the coordinates of the lower-left and
upper-right corner of the frame, the type of border of the frame, and
the text of the frame's label.
GUI.SetLabel
Changes the text of a labelled frame.
Widgets - Labels
Win DOS Mac
X TOOT Term
Output from Labels.dem
Labels are organizational elements that make the window look better and help
organize the GUI elements. They are simply text placed in a window. To aid in aligning text
with various widgets, it is possible to align text in a larger region (as shown in the figure).
GUI.CreateLabel
Creates and displays a label. GUI.CreateLabel specifies the lower-left
corner of the text and the text itself. The system font is used to display
the label.
GUI.CreateLabelFull Creates and displays a label. GUI.CreateLabelFull also specifies the
width, height, alignment, and font for the label. The width and height
are specified for alignment purposes.
GUI.SetLabel
Changes the text of a label.
144 Object Oriented Turing Reference Manual
Widgets - Pictures
Output from Pictures.dem
Pictures are organizational elements that make the window look better and help
organize the GUI elements. They are simply a picture placed in a window. The pictures are
specified using a picture ID from any of the Pic… subprograms.
GUI.CreatePicture Creates and displays a picture. GUI.CreatePicture specifies the location,
picture ID, and whether the picture should be merged with the
background.
Widgets - Menus
Win DOS Mac
X TOOT Term
Output from Menus.dem
Chapter 5 : The GUI Module
145
Menus are used in most modern interfaces. In order to create a full set of menus, you
must create the menu and then create the menu items in that menu. The menus are
automatically added to the menu bar of the selected menu.
Menu items are the individual entries of a menu. To create menus for a window, you
must create a menu, then create the menu items for that menu, then create the next menu,
then the items for that menu, etc. All menu items are automatically added to the last menu
and after the last menu item of the currently selected (not active!) window.
When you select an item in a menu, the action procedure of the item is called. The action
procedure has no parameters.
As of the v1.0 release of the GUI Library, it is an error to create a menu item without
having created a menu first. In future releases it will be possible to create menus and attach
and remove them from menu bars when desired.
Menus and menu items can be enabled and disabled. A disabled menu item is grayed
out. When the user selects the menu, all items in the menu appear disabled and cannot be
selected. A disabled menu item is grayed out when the menu is displayed. The user cannot
select the menu item.
Separators in a menu appear as a solid line across the menu. These are created by
creating a menu item whose text is three dashes "---".
GUI.CreateMenu
Creates and displays a menu. The menu will be added after the other
menus in the menu bar. If there are no previous menus, then a menu
bar is automatically created and the menu added. GUI.CreateMenu
specifies the text that will appear in the menu bar. It is suggested that
the text not have any spaces in it.
GUI.Enable
GUI.Disable
Enables (disables) a menu. Disabled menus are grayed out in the
menu bar. If selected, all the menu items in the menu bar appear
disabled and cannot be selected.
GUI.CreateMenuItem Creates a menu item. GUI.CreateMenuItem specifies the text of the
menu item and the action procedure to be called when the menu item is
selected. The menu item will be added to the last menu after the other
menu items in the menu. If there are no menus defined, an error
results.
GUI.CreateMenuItemFull
Creates a menu item. GUI.CreateMenuItemFull also
specifies a shortcut keystroke.
GUI.Enable
Enables (disables) a menu item. Disabled menu items are grayed out
GUI.Disable
when the menu is displayed and cannot be selected by the user.
Widgets - General Routines
The following procedures are included in the GUI module but do not relate to a specific
widget.
GUI.ProcessEvent This function processes a single event (a mouse button press or a
keystroke). If the event activates a widget, then the action procedure of
the widget is called. To find out which widget was activated and called
the action procedure (necessary if several widgets have the same action
146 Object Oriented Turing Reference Manual
GUI.Quit
procedure), you can call GUI.GetEventWidgetID. To get the exact time
that the event occurred, you can call GUI.GetEventTime. To get the
window in which the event took place, you can call
GUI.GetEventWindow.
If a mouse click occurred, but did not activate any widget, then the
default mouse event handler is called. By default, this does nothing.
However, if you want your program to respond to mouse events that
do not affect a widget, call GUI.SetMouseEventHandler to specify your
own default mouse event handler.
If a keystroke occurred, but did not activate any widget (i.e. it wasn't a
short cut for a widget and there are no text fields in the window) then
the default keystroke handler is called. By default, this does nothing.
However, if you want your program to respond to keystroke events
that do not affect a widget, call GUI.SetKeyEventHandler to specify your
own default key event handler.
If no event occurred, then the null event handler is called. By default,
this does nothing. However, if you want your program to perform
some action repetitively when it is not doing anything else, then call
GUI.SetNullEventHandler to specify your own null event handler. The
null event handler is often used for such things as updating a clock and
making certain that music is playing in the background.
This procedure causes GUI.ProcessEvent to return true. If the program
is structured properly with a
loop
exit when GUI.ProcessEvent
end loop
at the end of the program, then the program will exit the loop after
finishing the current action procedure. This procedure is usually called
from the action procedure of a Quit button or Exit menu item.
GUI.Refresh
This routine redraws all the widgets in the currently-selected window.
This is used when some form of drawing may have overwritten the
widgets in a window. It is used by the GUI Library to redraw all the
widgets when the background color of a window has changed.
GUI.SetBackgroundColor
Changes the background color of the currently-selected
GUI.SetBackgroundColour
window. (Both spellings of color are acceptable.) This does
not change the value of color 0 in the window. Instead it fills the entire
window with the new background color and then redraws all the
widgets. The usual background color outside of white is gray.
GUI.SetNullEventHandler
Sets the new null event handler. The specified procedure
will be called every time GUI.ProcessEvent is called and there is no
keystroke or mouse button pressed.
GUI.SetMouseEventHandler Sets the new default mouse event handler. The specified
procedure will be called every time GUI.ProcessEvent is called and
there is a mouse button pressed which is not handled by any widget.
Chapter 5 : The GUI Module
147
GUI.SetKeyEventHandler
Sets the new default keystroke event handler. The
specified procedure will be called every time GUI.ProcessEvent is called
and there is a keystroke which is not handled by any widget.
GUI.HideMenuBar Hides the menu bar in the selected window. No menu items can be
selected when the menu bar is hidden. (Menu item shortcuts will be
ignored when the menu bar is hidden.)
GUI.ShowMenuBar
Shows the menu bar in the selected window.
GUI.GetEventWidgetID
Returns the widget ID of the widget that was activated by
the mouse button press or the keystroke. This function should only be
called in an action procedure, as it will return –1 when there is no event
that activated a widget being processed.
GUI.GetEventWindow Returns the window ID of the window in which the event (mouse
button or keystroke) took place. This function should only be called in
an action procedure or in a default mouse or keystroke event handler, as
it will return –1 when there is no event being processed.
GUI.GetEventTimeReturns the time in milliseconds when the event (mouse button or
keystroke) took place. This value is the same value that Time.Elapsed
returns if called when the event was processed. This function should
only be called in an action procedure or in a default mouse or keystroke
or null event handler, as it will return –1 when there is no event being
processed.
GUI.DisplayWhenCreated
Sets whether widgets are automatically displayed when
created, or whether GUI.Show must be called first. By default, this is set
to true (widgets are displayed when created). However, there may be
times when you want to create a widget and then make several
additional calls before displaying the widget.
GUI.CloseWindow Closes a window with widgets in it. This procedure automatically
disposes of any widgets in the window and makes certain that the GUI
Library recognizes that the window no longer exists. This procedure
will call Window.Close, so there is no need for the user to do so.
GUI.GetScrollBarWidth
Returns the width of a scroll bar. Useful when placing a
scroll bar widget beneath another object.
GUI.GetMenuBarHeight
Returns the height of the menu bar. Useful when placing
widgets to make certain that they do not overlap the menu.
GUI.GetVersion
Returns the current version of the GUI module. Because the GUI
module is expected to grow, new versions will probably be made
available at Holt Software's web site located at
http://www.holtsoft.com/turing. If you wish to use features that do
not appear in earlier versions of the library, you can have your
program check that the current available version meets the programs
needs. GUI.GetVersion returns an integer from 100 - 999 and is read as
1.00 to 9.99.
148 Object Oriented Turing Reference Manual
Chapter 6
Importing Graphics
into a Turing Program
Introduction
Graphics created outside of Turing can be imported into Turing programs. Typically,
these graphics are created in programs like CorelDraw or Microsoft Paint or downloaded
from the internet.
To import graphics into WinOOT (Object Oriented Turing running under Microsoft
Windows), the graphics must be in BMP file format. BMP is a common MS Windows file
format suitable for images including clipart and digitized photographs. Almost all
conversion programs can convert images to the BMP format.
To import graphics into DOS OOT (Object Oriented Turing running under MS-DOS),
the graphics must be in PCX file format. PCX is a common file format suitable for images
including clipart and digitized photographs. Almost all conversion programs can convert
images to the PCX format.
To import graphics in MacOOT (Object Oriented Turing running on the Macintosh), the
graphics must be in PICT format, which is the standard picture format for the Macintosh.
Almost all Macintosh conversion programs can convert images to PICT format.
Occasionally, pictures will be obtained in a file format that is different from the one
required by Turing. In such a case, a picture format conversion utility is required. There are
a wide variety of freeware and shareware picture format conversion utilities available.
Checking the internet is often the best way to find such programs. The site
www.tucows.com provides large numbers of such programs.
Once the image has been converted into BMP, PCX, or PICT format (depending on
whether a PC or a Macintosh is being used), then the file can be imported into Turing.
Importing BMP Images in WinOOT
WinOOT displays (and saves) BMP images using the Pic.ScreenLoad and
Pic.ScreenSave procedures located in the Pic module.
The BMP format is the standard graphics interchange format for Microsoft Windows.
Almost any graphics program created for MS Windows supports it in one form or another.
Pic.ScreenLoad
The Pic.ScreenLoad procedure has the following format:
Pic.ScreenLoad (filename : string, x, y, mode : int)
This procedure is the equivalent of drawpic, except that it draws from a file instead of
an array. It will display the BMP file, with the lower-left corner of the picture being (x, y).
For full screen pictures, x and y should be 0. The mode parameter must one of picCopy, in
which case the image displays over top of whatever is on the screen, picMerge, which
functions like picCopy except that any occurrence of the background color in the picture is
not drawn to the screen, or picXor, in which case the image is "xor"-ed with the underlying
image on the screen.
If the procedure is unable to open or display the file it will not display an image.
Instead Error.Last will be set to non-zero. It is a good idea to always check Error.Last to
make certain that it is non-zero.
Here is a small example that displays the file "FOREST.BMP" at location (50, 50).
Pic.ScreenLoad ("FOREST.BMP", 50, 50, picCopy)
if Error.Last not= 0 then
put "Error displaying image: ", Error.LastMsg
end if
Pic.ScreenLoad requires that either the file name end with ".BMP" or the file name be
prefaced with "BMP:". For example, if the BMP file was called "FOREST.PCT", then the
Pic.ScreenLoad line would look like:
Pic.ScreenLoad ("BMP:FOREST.PCT", 50, 50, picCopy)
In the current implementation of WinOOT (WinOOT 3.0), Turing remaps the palette of
the image to the palette of the window on the screen. By default, WinOOT run windows
have a palette of only 16 colors. This means that any BMP image drawn to the screen will
be mapped to the base 16 colors, which may render the image less pleasing that desired. To
add more colors to the palette, and thus increase the fidelity of the image, it is necessary to
use RGB.AddColor. If you know the palette of the BMP image, you can add the specific
colors desired. If you want to add a wider general palette to the screen, you can use the
following:
var dummy : int
for red : 0 .. 5
for green : 0 .. 5
for blue : 0 .. 5
dummy := RGB.AddColor (red / 5.0, green / 5.0, blue / 5.0)
end for
end for
end for
150 Object Oriented Turing Reference Manual
Pic.ScreenSave
The Pic.ScreenSave procedure has the following format:
Pic.ScreenSave (x1, y1, x2, y2 : int, filename : string)
This procedure is the equivalent of takepic, except it copies the screen to a file instead
of an array. The section of the screen located in the rectangle specified by (x1, y1) - (x2, y2)
is copied to a BMP file located at filename. As with Pic.ScreenLoad, the file name must end
with ".BMP" or be prefaced with "BMP:".
Here is a small example that writes the entire screen to a file called
"BACKGRND.BMP".
Pic.ScreenSave (0, 0, maxx, maxy, "BACKGRND.BMP")
Importing PICT Images in MacOOT
MacOOT displays (and saves) PICT images using the Pic.ScreenLoad and
Pic.ScreenSave procedures located in the Pic module.
The PICT format is the standard graphics interchange format on the Macintosh. Almost
any graphics program created for the Macintosh supports it in one form or another.
Pic.ScreenLoad
The Pic.ScreenLoad procedure has the following format:
Pic.ScreenLoad (filename : string, x, y, mode : int)
This procedure is the equivalent of drawpic, except that it draws from a file instead of
an array. It will display the PICT file, with the lower-left corner of the picture being (x, y).
For full screen pictures, x and y should be 0. The mode parameter must be either picCopy, in
which case the image displays over top of whatever is on the screen, or picXor, in which
case the image is "xor"-ed with the underlying image on the screen.
If the procedure is unable to open or display the file it will not display an image.
Instead Error.Last will be set to non-zero. It is a good idea to always check Error.Last to
make certain that it is non-zero.
Here is a small example that displays the file "FOREST.PICT" at location (50, 50).
Pic.ScreenLoad ("FOREST.PICT", 50, 50, picCopy)
if Error.Last not= 0 then
put "Error displaying image: ", Error.LastMsg
end if
Pic.ScreenLoad requires that either the file name end with ".PICT" or the file name be
prefaced with "PICT:". For example, if the PICT file was called "FOREST.PCT", then the
Pic.ScreenLoad line would look like:
Chapter 6 : Importing Graphics
151 151
Pic.ScreenLoad ("PICT:FOREST.PCT", 50, 50, picCopy)
Pic.ScreenSave
The Pic.ScreenSave procedure has the following format:
Pic.ScreenSave (x1, y1, x2, y2 : int, filename : string)
This procedure is the equivalent of takepic, except it copies the screen to a file instead
of an array. The section of the screen located in the rectangle specified by (x1, y1) - (x2, y2)
is copied to a PICT file located at filename. As with Pic.ScreenLoad, the file name must end
with ".PICT" or be prefaced with "PICT:".
Here is a small example that writes the entire screen to a file called
"BACKGRND.PICT".
Pic.ScreenSave (0, 0, maxx, maxy, "BACKGRND.PICT")
Importing PCX Images in DOS OOT
Displaying a PCX image in DOS OOT is a two step process. First the PCX image must
be converted to a format that Turing can read in directly (the TM2 format) and then the
Turing program itself must read in the new image.
To convert the PCX image into TM2 format, Turing provides a program called
PCX2TM2.EXE. This program reads in a PCX file and creates a file with the same name but
with the extension ".TM2". Once this is done, the Turing program can call the
Pic.ScreenLoad and Pic.ScreenSave procedures located in the Pic module.
The PCX format is used by PC-Paintbrush (a well known "paint" program). We decided
to allow the interchange from PCX instead of one of the many other graphical standards
because PCX has been around for a long time, it is supported by most draw and paint
programs, and is available as a conversion format from many public domain and
shareware format-conversion programs.
PCX2TM2
To convert the PCX image to a TM2 image, you use the PCX2TM2.EXE program.
PCX2TM2.EXE is a DOS executable (i.e. you run it from the DOS prompt, rather than from
inside Turing.) It has the following format:
pcx2tm2 {<options>} {<video mode>} <pcx file>
If you run PCX2TM2 without any command line arguments, it will ask for a command
line. This allows PCX2TM2 to be used from Windows and menu programs that do not
allow command line arguments.
The <option> is one of three:
–default
This causes the TM2 file generated by PCX2TM2 to use the default
color map. PCX2TM2 will attempt to remap the colors in the PCX
152 Object Oriented Turing Reference Manual
images color palette to provide the closest possible fidelity. The
advantage of such an image is that it can be displayed without
affecting any of the other colors on the display. If you are attempting
to display two or more PCX images at once, this is especially
important.
–debug
Provides some information about the PCX image before it is displayed.
–buffersize This determines the size of the chunks in which the picture is saved.
This number is the size of the takepic buffer in integers. The default
value is 1,000, although you can set it to a maximum of 16,000. Note
that the buffer is temporarily allocated on the stack when the image is
displayed. You must have enough stack space to do this or the
program will halt with a stack overflow message.
In general, the larger the buffer size, the faster the image will be displayed, but the
more memory it will take.
The <video mode> is the mode in which the PCX image will be displayed. If none is
specified, then PCX2TM2 determines the mode from the image. It is possible for it to guess
incorrectly, although it will usually do a good job. If the video mode selected has a different
number of colors than the image does, then PCX2TM2 will attempt to map the colors that
the PCX image uses to the colors available in the video mode selected. Note that displaying
a 256 color image in a 16 color screen mode usually produces a fairly muddy image. TM2
files are video mode specific. This means that if you display the PCX file in a particular
video mode, the screen must be in the same video mode when Pic.ScreenLoad is called or
the Turing program will give an error.
Specifying the video mode is one way to allow several PCX images of different
resolutions to be displayed at once.
MODE
cga
16
h16
ega
v2
vga
mcga
svga
svga1
svga2
svga3
svga4
svga5
svga6
RESOLUTION
320 x 200 4-color CGA
320 x 200 16-color EGA
640 x 200 16-color EGA
640 x 350 16-color EGA
640 x 480 2-color VGA
640 x 480 16-color VGA
320 x 200 256-color MCGA
640 x 480 256-color SVGA
640 x 400 256-color SVGA
640 x 480 256-color SVGA
800 x 600 16-color SVGA
800 x 600 256-color SVGA
1024 x 768 16-color SVGA
1024 x 768 256-color SVGA
TURING MODE
graphics
graphics:16
graphics:h16
graphics:ega
graphics:v2
graphics:vga
graphics:mcga
graphics:svga
graphics:svga1
graphics:svga2
graphics:svga3
graphics:svga4
graphics:svga5
graphics:svga6
Chapter 6 : Importing Graphics
153 153
The last item is the actual name of the file to be converted. The output file will be the
same name as the input file, but with a TM2 extension, hence "pcx2tm2 mypic.pcx"
produces the output file "mypic.tm2".
One thing that you should be aware of is that PCX files can contain a palette. In other
words, they can change the natural mapping of the screen colors (or color table). For
example, an image of a forest might contain 200 different shades of green and no shades of
blue whatsoever.
When such an image is translated by PCX2TM2, the palette information is stored with
it unless the –default option is used. When the image is displayed using Pic.ScreenLoad,
Turing reads in the color palette and changes the color palette used by the display to the
one used by the PCX image. This means that if you are doing other graphics besides
displaying the file, the colors in those graphics may suddenly change.
Note that when you use the –default option, the color fidelity of the image will
bereduced. The forest screne that once had 200 shades of green may now use only the 30
shades of green provided by the default color palette.
If you are displaying more than one PCX image on the screen at the same time, the –
default option is almost required. Otherwise when the second image is displayed, the color
palette will be remapped a second time, causing the colors in the first image to be
completely changed. Without the –default option, a forest screne displayed before a fire
scene may have its entire color palette changed to shades of red when the fire image is
displayed!
One limitation of PCX2TM2 is that it can not process 24-bit color images at this time.
More to the point, Turing does not have any graphics modes that it could use to display
such an image.
Pic.ScreenLoad
The Pic.ScreenLoad procedure has the following format:
Pic.ScreenLoad (filename : string, x, y, mode : int)
This procedure is the equivalent of drawpic, except that it draws from a file instead of
an array. It will display the TM2 file, with the lower-left corner of the picture being (x, y).
For full screen pictures, x and y should be 0. The mode parameter must be either picCopy, in
which case the image displays over top of whatever is on the screen, or picXor, in which
case the image is "xor"-ed with the underlying image on the screen.
If the procedure is unable to open or display the file it will not display an image.
Instead Error.Last will be set to non-zero. It is a good idea to always check Error.Last to
make certain that it is non-zero.
Here is a small example that displays the file "FOREST.TM2" at location (50, 50).
Pic.ScreenLoad ("FOREST.TM2", 50, 50, picCopy)
if Error.Last not= 0 then
put "Error displaying image: ", Error.LastMsg
end if
154 Object Oriented Turing Reference Manual
Pic.ScreenLoad requires that either the file name end with ".TM2" or the file name be
prefaced with "TM2:". For example, if the TM2 file was called "FOREST.PCT", then the
Pic.ScreenLoad line would look like:
Pic.ScreenLoad ("TM2:FOREST.PCT", 50, 50, picCopy)
Pic.ScreenSave
The Pic.ScreenSave procedure has the following format:
Pic.ScreenSave (x1, y1, x2, y2 : int, filename : string)
This procedure is the equivalent of takepic, except it copies the screen to a file instead
of an array. The section of the screen located in the rectangle specified by (x1, y1) - (x2, y2)
is copied to a TM2 file located at filename. As with Pic.ScreenLoad, the file name must end
with ".TM2" or be prefaced with "TM2:".
Here is a small example that writes the entire screen to a file called
"BACKGRND.TM2".
Pic.ScreenSave (0, 0, maxx, maxy, "BACKGRND.TM2")
Chapter 6 : Importing Graphics
155 155
Chapter 6 : Importing Graphics
157 157
Chapter 7
Language Features
Alphabetic listing of all features of Turing.
abs
absolute value function
Syntax
abs ( expn )
Description
The abs function is used to find the absolute value of a number (the expn).
For example, abs (-23) is 23.
Example
This program outputs 9.83.
var x : real := -9.83
put abs ( x ) % Outputs 9.83
Details
The abs function accepts numbers that are either int's or real's. The type of
the result is the same type as the accepted number. The abs function is
often used to see if one number x is within a given distance d of another
number y; for example:
if abs (x - y) <= d then …
See also
predefined unit Math.
Chapter 7 : Language Features 159
addr
address of a variable
Dirty
Syntax
addr ( reference )
Description
The addr attribute is used to find the integer address of a variable or non
scalar constant. This is implementation-dependent. This address may be
used in an indirection operation @.
Example
Set a to be the address of x.
var x : real
Details
var a : addressint := addr ( x )
The value of the address produced by addr is of type addressint, an
integer type whose range is that of the underlying memory addresses.
The concept of an address is implementation-dependent. For example, an
optimizing compiler could determine that a variable does not require space
because the program could be computed without the variable with no
change in output. However, in most implementations, types have a
predictable size and variables of that type occupy that number of bytes in
memory.
See also
the indirection operator @, cheat, explicitIntegerConstant (how to write
hexadecimal constants), and pointer type (in particular unchecked pointer
type). See also sizeof, which returns the size of a variable.
addressint
type
Dirty
Syntax
addressint
Description
The addressint (address integer) type is an integer type whose range of
value is the same as that of the underlying computer. This range is, by its
nature, implementation-dependent. On 32-bit architectures, it is commonly
the same range as nat4 (4-byte natural number).
160 Object Oriented Turing Reference Manual
Record r contains three fields, one of which has type char(28). Variable a is
an integer whose range of values is the same as the addresses of the
underlying computer. This assigns B to the seventh character of a record of
type r which is assumed to be located at absolute address a.
Example
type r :
record
i : int
c28 : char ( 28 )
c11 : char ( 11 )
end record
% An integer
var a : addressint
…
% a is assigned an integer value
r @ ( a ) . c28 ( 7 ) := 'B'
% Use indirection operator @
Although addressint is called an integer type, it is commonly equivalent to
a natural type such as nat4 (for 32-bit machines) or nat2 (for 16-bit
machines).
Details
Be careful not to confuse addressint with pointer types. In low level
languages such as assembler and C, addresses and pointers are the same.
In Turing, however, a pointer is a high level concept that is more abstract
than a machine address. A Turing pointer is a reference to an object, and
the representation of this reference depends upon the implementation. In
current Turing implementations, pointers (which are by default checked)
are represented as a time stamp (a unique number) together with an
address. The time stamp is used to make sure that the pointer actually
locates an object. There are also unchecked pointers. An unchecked
pointer's internal representation is a machine address. You can use type
cheats (a dangerous feature) to translate between addressint and
unchecked pointers. This is meaningful in current implementations.
the indirection operator @, cheat, explicitIntegerConstant (how to write
hexadecimal constants), and pointer type (in particular unchecked pointer
type). See also addr, which returns the address of a variable.
See also
all
all members of a set
Syntax
setTypeName ( all )
Description
Given a set type named S, the set of all of the possible elements of S is
written S (all).
Example
type smallSet : set of 0 .. 2
Chapter 7 : Language Features 161
var x : smallSet := smallSet ( all )
% Set x contains elements 0, 1 and 2
set type for details about sets.
See also
and
operator
Syntax
A and B
Description
The and (boolean) operator yields a result of true if, and only if, both
operands are true. The and operator is a short circuit operator. For
example, if A is false in A and B then B is not evaluated.
Example
var success : boolean := false
%The type is boolean
var continuing := true
…
continuing := continuing and success
Details
The continuing variable is set to true if, and only if, both continuing and
success are true. Since Turing uses short circuit operators, once continuing is
false, success will not be looked at.
The and operator can also be applied to natural numbers. The result is the
natural number that is the bit-wise and of the operands. See nat (natural
number).
Example
This masks out the everything but the lower two bytes of number.
var number : nat := 16#ABCD
var mask : nat := 16#FF
put number and mask
% Outputs 205 (CD16)
See also
boolean (which discusses true/false values), explicitTrueFalseConstant
(which discusses the values true and false), precedence and expn
(expression).
162 Object Oriented Turing Reference Manual
anyclass
the ancestor of all classes
Syntax
anyclass
Description
There is a predefined class called anyclass, which is the root of the
expansion tree. All classes that do not have inherit lists are considered to
be expansions of anyclass. The main purpose of anyclass is to allow
pointers that can locate objects of any class.
Example
Here is the declaration of a pointer p that can locate an object of any class.
var p : pointer to anyclass % Short form: var p : ^ anyclass
var q : pointer to stack
% Short form: var q : ^ stack
new q
% Create a stack object
% Legal because p's class
p := q
% is an ancestor of q's class
Assuming stack is a class, this creates a stack object and places its location in
q and p. The compiler will not allow a call to stack's exported subprograms
using p directly, as in:
p -> push ( 14 )
% ILLEGAL! anyclass has no operations
An assignment from p to q is legal, as in:
q := p % Checks that p locates a stack object (or descendant)
This implies a run time check to make sure that p locates an object that is a
stack (or a descendant of a stack).
Here is a way to call a subprogram exported from stack using p:
stack ( p ) . push ( 14 )
% Checks that p locates a stack object
This checks to see that p locates a stack object (or a descendant) before
calling the stack operation push.
Details
It is legal to create objects of the class called anyclass, but this is not of
much use, because there is nothing you can do with these objects (they
have no operations). It is legal to assign these objects to other objects of the
same class (anyclass), although this accomplishes nothing.
See also
objectclass, which takes a class pointer and produces the class of the object
located by the pointer. This is used for testing to determine the class of the
object located by a pointer.
See also class. See also export list, import list, inherit list, implement list
and implement by list.
Chapter 7 : Language Features 163
arctan
arctangent function (radians)
Syntax
arctan ( r : real ) : real
Description
The arctan function is used to find the arc tangent of a value. The result is
given in radians. For example, arctan (1) is  / 4.
Example
This program prints out the arctangent of 0 through 3 in radians.
for i : 0 .. 12
const arg := i / 4
put "Arc tangent of ", arg, " is ",
arctan ( arg ), " radians"
end for
The formulae for arcsin and arccos are:
Note
arcsin (x) = arctan (sqrt ((x *x ) / (1 - (x *x ))))
arccos (x) = arctan (sqrt ((1 - (x *x )) / (x *x )))
the arctand function which finds the arc tangent of a value with the result
given in degrees. (2 radians are the same as 360 degrees.)
See also
See also predefined unit Math.
arctand
arctangent function (degrees)
Syntax
arctand ( r : real ) : real
Description
The arctand function is used to find the arc tangent of an angle given in
degrees. For example, arctand (0) is 0.
Example
This program prints out the arctangent of values from 0 to 3 in degrees.
for i : 0 .. 12
const arg := i / 4
put "Arc tangent of ", arg, " is ",
arctand ( arg ), " degrees"
end for
164 Object Oriented Turing Reference Manual
The formulae for arcsind and arccosd are:
Note
arcsind (x) = arctand (sqrt ((x *x ) / (1 - (x *x ))))
arccosd (x) = arctand (sqrt ((1 - (x *x )) / (x *x )))
the arctan function which finds the arc tangent of a value with the result
given in radians. (2radians are the same as 360 degrees.)
See also
See also predefined unit Math.
array
type
Syntax
array indexType { , indexType } of typeSpec
Description
An array consists of a number of elements. The typeSpec gives the type of
these elements. There is one element for each item in the (combinations of)
range(s) of the indexType(s). In the following example, the array called
marks consists of 100 elements, each of which is an integer.
Example
var marks : array 1 .. 100 of int
…
var sum : int := 0
for i : 1 .. 100
% Add up the elements of marks
sum := sum + marks ( i )
end for
Details
In the above example, marks(i) is the i-th element of the marks array. We call
i the index or subscript of marks. In Turing, a subscript is surrounded by
parentheses, not by square brackets as is the case in the Pascal language.
Example
The prices array shows how an array can have more than one dimension.
This array has one dimension for the year (1988, 1989 or 1990) and another
for the month (1 .. 12). There are 36 elements of the array, one for each
month of each year.
var price : array 1988 .. 1990, 1 .. 12 of int
…
var sum : int := 0
for year : 1988 .. 1990
% For each year
for month : 1 .. 12
% For each month
sum := sum + price ( year, month )
end for
end for
Chapter 7 : Language Features 165
Details
Each indexType must contain at least one item. The range 1 .. 0, for example,
would not be allowed. Each index type must be a subrange of the integers,
characters (the char type), or of an enumerated type, an (entire)
enumerated type, the char type, the boolean type, or a named type which
is one of these.
Arrays can also be declared in the form
var a : array 1 .. * of typeSpec := init ( … )
The upper bound of a will be computed from the count of the initializing
values. Both var and const arrays can be declared this way. An array
variable/constant declared with "*" as an upper bound must have an
initializing list. Only one dimensional arrays may be declared in this form.
Arrays can be assigned as a whole (to arrays of an equivalent type), but
they cannot be compared.
An array can be initialized in its declaration using init. For details, see var
and const declarations.
Example
In this example, the size of the array is not known until run time.
var howMany : int
get howMany
var height : array 1 .. howMany of real
…read in all the elements of this array…
function total ( a : array 1 .. * of real ) : real
var sum : int := 0
for i : 1 .. upper ( a )
sum := sum + a ( i )
end for
result sum
end total
put "Sum of the heights is ", total ( height )
Details
The ends of the range of a subscript are called the bounds of the array. If
these values are not known until run time, the array is said to be dynamic.
In the above example, height is a dynamic array. Dynamic arrays can be
declared as variables, as in the case for height. However, dynamic arrays
cannot appear inside other types such as records, and cannot be named
types. Dynamic arrays cannot be assigned and cannot be initialized using
init.
In the above example, upper(a) returns the size of a. See also upper and
lower.
In the declaration of an array parameter, the upper bound can be given as
an asterisk (*), as is done in the above example. This means that the upper
bound is taken from that of the corresponding actual parameter (from
height in this example).
You can have arrays of other types, for example arrays of record. If R is an
array of records, then R(i).f is the way to access the f field of the i-th
element of array R.
166 Object Oriented Turing Reference Manual
Details
Arrays can also be made resizeable. This is done using the flexible
keyword. The declaration syntax is:
var name : flexible array indexType { , indexType } of typeSpec
The indices may have compile-time or run-time upper bounds (the lower
bound must be compile-time). The upper bounds can be changed by using:
new name , newUpper1 {,newUpper2}
The existing array entries will retain their values, except that any index
made smaller will have the corresponding array entries lost. Any index
made larger will have the new array entries uninitialized (if applicable).
Additionally, the upper bound (both in the declaration and the new
statement) can be made one less than the lower bound. This effectively
makes an array that contains 0 elements. It can later be increased in size
with another new.
In the current implementation (1999), with a multi-dimensional array with
a non-zero number of total elements, it is a run-time error to change any
but the first dimension (unless one of the new upper bounds is one less
than the corresponding lower bound, giving 0 elements in the array) as the
algorithm to rearrange the element memory locations has not yet been
implemented.
Currently, only variables can be declared in this form. There is no flexible
array parameter type, although a flexible array can be passed to an array
parameter with “*” as the upper bound.
Example
In this example, the array is resized to fit the number of elements in the file.
function getLines (fileName : string) : int
var f, numLines : int
var line : string
open : f, fileName, get
numLines := 0
loop
exit when eof (f)
get : f, line : *
numLines += 1
end loop
close : f
result numLines
end getLines
procedure readFile (var lines : array 1 .. * of string, fileName : string)
var f : int
var line : string
open : f, fileName, get
for i : 1 .. upper (lines)
get : f, lines (i) : *
end for
close : f
end readFile
Chapter 7 : Language Features 167
var lines : flexible array 1 .. 0 of string
new lines, getLines ("text.dat")
readFile (lines, "text.dat")
for i : 1 .. upper (lines)
put lines (i)
end for
assert
statement
Syntax
assert trueFalseExpn
Description
An assert statement is used to make sure that a certain requirement is met.
This requirement is given by the trueFalseExpn. The trueFalseExpn is
evaluated. If it is true, all is well and execution continues. If it is false,
execution is terminated with an appropriate message.
Example
Make sure that n is positive.
assert n > 0
Example
This program assumes that the textFile exists and can be opened, in other
words, that the open will set the fileNumber to a non-zero stream number. If
this is not true, the programmer wants the program halted immediately.
var fileNumber : int
open : fileNumber, "textFile", read
assert fileNumber not= 0
Details
In some Turing systems, checking can be turned off. If checking is turned
off, assert statements may be ignored and as a result never cause
termination.
168 Object Oriented Turing Reference Manual
assignability
Description
of expression to variable
A value, such as 24, is assignable to a variable, such as i, if certain rules are
followed. These rules, given in detail below, are called the assignability
rules. They must be followed in assignment statements as well as when
passing values to non-var parameters.
Example
var i : int
i := 24
% 24 is assignable to i
var width : 0 .. 319
width := 3 * i
% 3 * i is assignable to width
var a : array 1 .. 25 of string
a ( i ) := "Ralph"
% "Ralph" is assignable to a(i)
var name : string ( 20 )
name := a ( i )
% a(i) is assignable to name
…
var b : array 1 .. 25 of string
b := a
% Array a is assignable to b
type personType :
record
age : int
name : string ( 20 )
end record
var r, s : personType
…
s := r
Details
% Record r is assignable to s
The expression on the right of := must be assignable to the variable on the
left. An expression passed to a non-var parameter must be assignable to the
corresponding parameter.
An expression is defined to be assignable to a variable if the two root types
are equivalent or if an integer value is being assigned to a real variable (in
which case the integer value is automatically converted to real). Two types
are considered to be equivalent if they are essentially the same type (see
equivalence for the detailed definition of this term).
In most cases a root type is simply the type itself. The exceptions are
subranges and strings. The root type of a subrange, such as 0 .. 319, is the
type of its bounds (int type in this example). The root type of a string, such
as the type string(9), is the most general string type, namely string.
Chapter 7 : Language Features 169
When a subrange variable, such as width, is used as an expression, for
example on the right side of an assignment statement, its type is considered
to be the root type (integer in this case) rather than the subrange. When an
expression is assigned to a subrange variable such as width, the value (3*i
in this example) must lie in the subrange. Analogously, any string variable
used in an expression is considered to be of the most general type of string.
When a string value is assigned to a string variable, its length must not
exceed the variable's maximum length.
Turing's assignability rule applies to characters and strings in this way. A
char value can be assigned (or passed to an non var parameter) with
automatic conversion to a char(1) variable and vice versa. String values of
length 1 can be assigned to char variables. Character, that is char, values
can be assigned to string variables, yielding a string of length 1. String
values of length n are assignable with automatic conversion to char(n)
variables. Values of type char(n) can be assigned with automatic
conversion to string variables.
Turing's assignability rule applies to pointers to classes in this way. A
pointer that locates an object created as class E, can be assigned to a pointer
to class B only if B is an ancestor of (or the same as) E. For example, a
pointer to an object that is a stackWithDepth can be assigned to a pointer to
stack, where stackWithDepth is a child of stack, but not vice versa. The
pointer nil can be assigned to any pointer variable, but the value nil(C) can
only be assigned to a pointer to an ancestor of C.
Objects of classes can be assigned to each other only if both were created as
the same class.
assignment
statement
An assignmentStatement is:
Syntax
variableReference := expn
Description
An assignment statement calculates the value of the expression (expn) and
assigns that value to the variable (variableReference).
Example
var i : int
i := 24
% Variable i becomes 24
var a : array 1 .. 25 of string
a ( i ) := "Ralph"
% The i-th element of a becomes "Ralph"
…
var b : array 1 .. 25 of string
b := a
% Array b becomes (is assigned) array a
170 Object Oriented Turing Reference Manual
The expression on the right of := must be assignable to the variable on the
left. For example, in the above, any integer value, such as 24, is assignable
to i, but a real value such as 3.14 would not be not assignable to i. Entire
arrays, records and unions can be assigned. For example, in the above,
array a is assigned to array b. See assignability for the exact rules of allowed
assignments.
Details
You cannot assign a new value to a constant (const).
There are short forms that allow you to write assignment statements more
compactly. For example,
i := i + 1
can be shortened to
i += 1
In Turing, there are short forms for combining +, = and * with assignment.
For example, i *= 2 doubles i.
There are also short forms to allow any binary operator to be combined
with assignment. For example, i shl= 2 shifts i by 2 to the left.
begin
statement
A beginStatement is:
Syntax
begin
statementsAndDeclarations
end
Description
A begin statement limits the scope of declarations made within it to the
confines of the begin/end block. In Turing, begin is rarely used, because
declarations can appear wherever statements can appear, and because
every structured statement such as if ends with an explicit end.
Example
begin
var bigArray : array 1 .. 2000 of real
… bigArray will exist only inside this begin statement…
end
Details
In Pascal programs, begin statements are quite common because they are
required for grouping two or more statements, for example, to group the
statements that follow then. This is not necessary in Turing as where ever
you can write a single statement, you can also write several statements.
Chapter 7 : Language Features 171
bind
declaration
A bindDeclaration is:
Syntax
bind [var] id to variableReference
{ , [var] id to variableReference }
Description
The bind declaration creates a new name (or names) for a variable
reference (or references). You are allowed to change the named item only if
you specify var. You can also bind to named non scalar constants.
While variableReference is bound it does not disappear in the scope.
Example
Rename the n-th element of array A so it is called item and then change this
element to 15.
bind var item to A ( n )
item := 15
Details
The scope of the identifier (item above) begins with the bind declaration
and lasts to the end of the surrounding program or statement (or to the end
of the surrounding part of a case or if statement). During this scope, a
change to a subscript (n above) that occurs in the variable reference does
not change the element to which the identifier refers.
You are not allowed to use bind at the outermost level of the main
program (except nested inside statements such as if) or at the outermost
level in a module.
You can also optionally use the register keyword to request that the bind
be done using a machine register. The syntax for bindDeclaration is actually:
bind [var] [register] id to variableReference
{ , [var] [register] id to variableReference }
In the current (1999) implementation, programs are run interpretively
using pseudo-code and the register keyword is ignored.
bits
Syntax
extraction
bits ( expn, subrange )
172 Object Oriented Turing Reference Manual
Description
The bits operator is used to extract a sequence of bits from a natural (nonnegative) number expression. The bits are numbered from right to left as 0,
1, 2 …
Example
Set bits 2 and 1 (third and second from the right) in the variable d to be 01.
We first set b to be the bit string 1100.
type T12 : 1 .. 2
% Use to specify bit range
var d : nat2 := 2#1100
% Two byte natural number
% At this point bits(d, T12) = 2#10
bits ( d, T12 ) := 2#01
% At this point d = 2#1010
Example
Set bit 7 in variable n to be 1. As a result, n will equal 2#10000000.
var n : nat1 := 0
Details
% A one byte variable set to zero
bits (n, 7) := 1
% n now contains the pattern 10000000
The form of subrange must be one of:
(a)
typeSpec
% Subrange type
(b)
compileTimeIntegerExpression
In form (a) the subrange type specifies a range from L to M (for least and
most significant). This is a little confusing because the subrange is written L
.. M with L on the left and M on the right, but in a number, the least
significant bit is on the right and the most significant is on the left. The
subrange type can be either the name of a type, for example T12, or an
explicit subrange, for example 3 .. 7. The values in the explicit subrange
must be compile time values.
Form (b) represents the range n .. n where n is the non-negative value of
the expression. In other words, both L and M equal n. The expression can
be any non-negative integer value or natural number value.
If the expression expn is a variable reference, the bits operation can be
assigned to, but cannot be passed to, a var parameter. For example, in the
above, bits (d, T12) has the value 2#01 assigned to it. For this assignment to
be allowed, the expression expn must be a natural number type (nat, nat1,
nat2 or nat4).
See also
explicitIntegerConstant (for description of constants such as 16#FFFF) and
the following functions that convert one type to another in a machineindependent manner: ord, chr, intstr, strint, natstr, and strnat. See also shr
and shl (shift right and left).
Chapter 7 : Language Features 173
body
Syntax
A bodyDeclaration is one of:
(a)
body procedure procedureId
statementsAndDeclarations
end procedureId
(b)
body function functionId
statementsAndDeclarations
end functionId
body procedure id [(paramDeclaration
{, paramDeclaration })]
statementsAndDeclarations
end id
(c)
(d)
Description
declaration
body function id [ ( [paramDeclaration {,
paramDeclaration } ] ) ] : typeSpec
statementsAndDeclarations
end id
A body declaration is used to resolve either a forward subprogram or a
deferred subprogram.
You declare a procedure or function forward when you want to define its
header but not its body. This is the case when one procedure or function
calls another, which in turn calls the first. This situation is called mutual
recursion. The use of forward is necessary in this case because every item
must be declared before it can be used. The forward declaration must be
followed by a body declaration for the same procedure or function. For
details, see forward declarations.
When a procedure or function in a class is declared to be deferred (or
simply exported from the class), it can be resolved or overridden afterward
by giving its body further down in that class or in descendant classes. The
overriding procedure must use the keyword body. See class or
"implement by" for examples.
Details
You can specify the parameter and return values of the subprogram in the
body declaration. However, the names and types of the parameters and
return values must match the initial declaration exactly, or a warning
results and the parameter list and return values from the body declaration
are ignored.
174 Object Oriented Turing Reference Manual
The example given here is part of a complete Turing program that includes
an explanation of forward declarations.
Example
var token : string
forward procedure expn ( var eValue : real )
import forward term, var token
… other declarations appear here …
body procedure expn
var nextValue : real
term (eValue )
% Evaluate t
loop
% Evaluate { + t}
exit when token not= "+"
get token
term (nextValue)
eValue := eValue + nextValue
end loop
end expn
The syntax of a bodyDeclaration presented above has been simplified by
omitting the optional result identifier, import list, pre and post condition
and init clause. See procedure and function declarations for descriptions
of these omissions.
Details
The "function" or "procedure" token in the declaration is now optional. In
other words the following code fragment is legal
forward procedure p
…
body p
…
end p
boolean
true-false type
Syntax
boolean
Description
The boolean type is used for values that are either true or false. These
true-false values can be combined by various operators such as or and and.
Example
var success : boolean := false
% The type is boolean
var continuing := true
…
success := mark >= 60
continuing := success and continuing
Chapter 7 : Language Features 175
if continuing then …
This type is named after the British mathematician, George Boole, who
formulated laws of logic.
Details
The operators for true and false are and, or, xor, =>, and not. For two
true/false values A and B, these operators are defined as follows:
A and B is true when both are true
A or B is true when either or both are true
A xor B is true when either but not both are true
A => B (A implies B) is true when both are true or when A is false
not A is true when A is false
The and operator has higher precedence than or, so A or B and C means A
or (B and C ).
The operators or, and and => are short circuit operators. For example, if A
is true in A or B, B is not evaluated.
The boolean type can be used as an index to an array.
Details
Example
Declaration of an array with boolean index.
var a : array boolean of int
a (false) := 10
a (true) := 20
Details
The put and get semantics allow put's and get's of boolean values. true
values will be output as "true" and false values will be output as "false".
The only legal input values are "true" and "false", which are case sensitive.
See also
explicitTrueFalseConstant (which discusses the values true and false),
precedence and expn (expression).
break
debugger pause statement
Syntax
break
Description
On systems with a debugger (currently WinOOT and MacOOT), the
environment "pauses" when execution reaches the break statement. While
"pausing" is environment specific, in both WinOOT and MacOOT, the
program stops execution until the user presses the "Continue" button.
While paused, the program variables can be inspected, stack traces done,
etc.
176 Object Oriented Turing Reference Manual
Example
for i : 1 .. 100
put i
break
end for
buttonchoose
switch mouse modes
Syntax
buttonchoose (choice : string)
Description
The buttonchoose procedure is used to change the mode of the mouse. In
Turing, the mouse can either be in "single-button mode" or in "multi-button
mode". In "single-button mode" the mouse is treated as a one button mouse. A
button is considered pressed when any button is pressed and released only
when all buttons have been released.
In Turing, the mouse starts in "single-button mode".
The parameter choice can be one of "singlebutton", "onebutton" (which
switch the mouse into "single-button mode") or "multibutton" (which
switches the mouse into "multi-button mode").
Example
A program that displays the status of the mouse at the top left corner of the
screen.
buttonchoose ("multibutton")
var x, y, button, left, middle, right : int
mousewhere (x, y, button)
left := button mod 10
% left = 0 or 1
middle := (button - left) mod 100 % middle = 0 or 10
% right = 0 or 100
right := button - middle - left
if left = 1 then
put "left button down"
end if
if middle = 10 then
put "middle button down"
end if
if right = 100 then
put "right button down"
end if
Details
Under DOS, the mouse does not start initialized. When the first mouse
command is received (mousehide, mouseshow, mousewhere,
buttonmoved, buttonwait, buttonchoose) Turing attempts to initialize the
mouse. If successful, the mouse cursor will appear at that time.
Chapter 7 : Language Features 177
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
buttonmoved and buttonwait to get mouse events saved in a queue. See
also mousewhere to get the current status of mouse button(s).
See also
See also predefined unit Mouse.
buttonmoved
has a mouse event occurred
Syntax
buttonmoved (motion : string) : boolean
Description
The buttonmoved function indicates whether there is a mouse event of the
appropriate type on the mouse queue. Events are either "up", "down",
"updown" or "downup" events (although the "downup" and "updown" are
the same event).
The parameter motion must be one of "up", "down", "updown" or
"downup". If an event of the type requested is in the queue, buttonmoved
returns true. If the event is not in the queue, then buttonmoved returns
false.
In "single-button mode" (where the mouse is treated like a one-button
mouse), a "down" event occurs whenever all the buttons are up and a
button is pressed. An "up" event takes place when the last button is
released so that no buttons remain pressed.
In "multi-button mode", a "down" event occurs whenever any button is
pressed, and an "up" event occurs whenever any button is released.
Example
This program draws random circles on the screen until the user clicks the
mouse button, whereupon is starts drawing random boxes. Clicking the
mouse button switches between the two.
var circles: boolean := true
loop
var x, y, radius, clr: int
if buttonmoved ("down") then
var buttonnumber, buttonupdown : int
buttonwait ("down", x, y, buttonnumber, buttonupdown)
circles := not circles
end if
randint (x, 0, maxx)
randint (y, 0, maxy)
randint (radius, 0, 100)
178 Object Oriented Turing Reference Manual
randint (clr, 0, maxcolor)
if circles then
drawfilloval (x, y, radius, radius, clr)
else
drawfillbox (x, y, x + radius, y + radius, clr)
end if
end loop
Example
This is an example demonstrating how to check for both character and
mouse input at the same time.
var ch : string (1)
var x, y, btnnum, btnupdown : int
loop
if hasch then
getch (ch)
locate (1, 1)
put "The character entered is a: ", ch
end if
if buttonmoved ("down") then
buttonwait ("down", x, y, btnnum, btnupdown)
locate (1, 1)
put "The button was clicked at position: ", x, ", ",y
end if
end loop
Details
buttonmoved can be thought of as the mouse equivalent of hasch in that
they both check for something in a queue and both return immediately.
Under DOS, the mouse does not start initialized. When the first mouse
command is received (mousehide, mouseshow, mousewhere,
buttonmoved, buttonwait, buttonchoose), Turing attempts to initialize the
mouse. If the initialization is successful, the mouse cursor will appear.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
See also
buttonmoved to get mouse events saved in the queue. See also
buttonchoose to switch between "single-button mode" and "multi-button
mode".
See also predefined unit Mouse.
Chapter 7 : Language Features 179
buttonwait
get a mouse event procedure
Syntax
buttonwait (motion : string,
var x, y, buttonnumber, buttonupdown : int)
Description
The buttonwait procedure gets information about a mouse event and
removes it from the queue.
The parameter motion must be one of "up", "down", "updown" or
"downup". If an event of the type requested is in the queue, buttonwait
returns instantly. If there isn't such an event, buttonwait waits until there
is one and then returns (much like getch handles keystrokes).
In "single-button mode" (where the mouse is treated like a one-button
mouse), a "down" event occurs whenever all the buttons are up and a
button is pressed. An "up" event takes place when the last button is
released so that no buttons remain pressed.
In "multi-button mode", a "down" event occurs whenever any button is
pressed, and an "up" event occurs whenever any button is released.
The parameters x and y are set to the position of the mouse cursor when
the button was pressed. The parameter buttonnumber is set to 1 when in
"single-button mode". In "multi-button mode", it is set to 1 if the left button
was pressed, 2 if the middle button was pressed, and 3 if the right button
was pressed. The parameter buttonupdown is set to 1, if a button was
pressed and 0 if a button was released.
Example
This program draws lines. It starts a line where the user presses down and
continues to update the line while the mouse button is held down. When
the button is released, the line is permanently draw and the user can draw
another line.
var x, y, buttonnumber, buttonupdown, buttons : int
var nx, ny : int
loop
buttonwait ("down", x, y, buttonnumber, buttonupdown)
nx := x
ny := y
loop
drawline (x, y, nx, ny, 0) % Erase previous line
exit when buttonmoved ("up")
mousewhere (nx, ny, buttons)
drawline (x, y, nx, ny, 1) % Draw line to position
end loop
buttonwait ("up", nx, ny, buttonnumber, buttonupdown)
drawline (x, y, nx, ny, 2) % Draw line to final position
180 Object Oriented Turing Reference Manual
end loop
Example
This is an example demonstrating how to check for both character and
mouse input at the same time.
var ch : string (1)
var x, y, btnnum, btnupdown : int
loop
if hasch then
getch (ch)
locate (1, 1)
put "The character entered is a: ", ch
end if
if buttonmoved ("down") then
buttonwait ("down", x, y, btnnum, btnupdown)
locate (1, 1)
put "The button was clicked at position: ", x, ", ",y
end if
end loop
Details
buttonwait can be thought of as the mouse equivalent of getch in that they
both read something in a queue and both wait until they get the thing
they're looking for.
Under DOS, the mouse does not start initialized. When the first mouse
command is received (mousehide, mouseshow, mousewhere,
buttonmoved, buttonwait, buttonchoose), Turing attempts to initialize the
mouse. If the initialization is successful, the mouse cursor will appear.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
See also
buttonwait to see if an appropriate event is in the queue. See also
buttonchoose to switch between "single-button mode" and "multi-button
mode".
See also predefined unit Mouse.
case
Syntax
selection statement
A caseStatement is:
case expn of
{ label compileTimeExpn {, compileTimeExpn } :
statementsAndDeclarations }
Chapter 7 : Language Features 181
[ label :
statementsAndDeclarations ]
end case
Description
A case statement is used to choose among a set of statements (and
declarations). One set is chosen and executed and then execution continues
just beyond end case.
The expression (expn) following the keyword case is evaluated and used to
select one of the alternatives (sets of declarations and statements) for
execution. The selected alternative is the one having a label value equaling
the case expression. If none are equal and there is a final label with no
expression, that alternative is selected.
Example
Output a message based on value of mark.
case mark of
label 9, 10 : put "Excellent"
label 7, 8 :
put "Good"
label 6 :
put "Fair"
label :
put "Poor"
end case
Example
Output a message based on value of name.
case name of
label "horse", "cow" : put "Farm animal"
label "tiger", "lion" : put "Jungle animal"
label "cat", "dog" :
put "Pet"
label :
put "Unknown animal"
end case
Details
The case expression is required to match one of the labels. If it does not,
there must be a final label with no expression. Label expressions must
have values known at compile time. All label values must be distinct. The
case expression and the label values must have the same equivalent type,
which must be an integer, char, boolean, an enum type or strings.
catenation (+)
Syntax
joining together strings
A catenation is:
stringExpn + stringExpn
182 Object Oriented Turing Reference Manual
Description
Two strings (stringExpns), char or char(n) values can be joined together
(catenated) using the + operator.
Example
var lastName, wholeName : string
lastName := "Austere"
wholeName := "Nancy" + " " + lastName
% The three strings Nancy, a blank and Austere
% catenated together to make the string
% "Nancy Austere". This string becomes the
% value of wholeName
Details
The length of a string catenation is limited to 255 characters.
Catenation is sometimes called concatenation.
Catenation can also be applied to char and char(n) values. See char and
char(n). If either operand, s or t in s + t, is a string or a dynamic char(n)
(length not known at compile time), the result type is string. Otherwise
(when both s and t are char or non-dynamic char(n)) the result type is
char(n).
The result of catenation is considered to be a compile time value if both
operands are compile time values.
If both operands have the type char or char(n) neither of which is a
dynamic char(n), the result is of type char(n), which is also of a non
dynamic type. This allows the creation of very long char(n) values that can
effectively span line boundaries using catenation to join lines. If either
operand is a dynamic type or a string type, the catenation produces a
string, whose length is limited to 255 characters.
See also
substrings (for separating a strings into parts), repeat (for making repeated
catenations), string type, length, and index (to determine where one string
is located inside another).
See also string, char, char(n), explicitStringConstant, explicitCharConstant,
substring and length.
ceil
real-to-integer function
Syntax
ceil ( r : real ) : int
Description
Returns the smallest integer greater than or equal to r.
Chapter 7 : Language Features 183
Details
The ceil (ceiling) function is used to convert a real number to an integer.
The result is the smallest integer that is greater than or equal to r. In other
words, the ceil function rounds up to the nearest integer. For example, ceil
(3) is 3, ceil (2.25) is 3 and ceil (-8.43) is -8.
See also
See also the floor and round functions.
char
type
Syntax
char
Description
Each variable whose type is a char contains a single character, such as the
letter A, the digit 3 or the special character &.
Example
Count characters until a period is found. Character c is read using a get
statement and is compared to the explicit character constant '.'.
var c : char
var counter := 0
loop
exit when eof
% Read a single character
get c
exit when c = '.' % Single quotes for char constant
counter := counter + 1
end loop
put counter, " characters before the period"
Example
Count capital letters. This example illustrates the use of the char type as the
subscript type for the frequency array, the use of character variable c as a
subscript, and the use of d as a for counter that ranges across the letters A
to Z.
var frequency : array 'A' .. 'Z' of nat
for d : 'A' .. 'Z'
frequency ( d ) := 0
end for
% Tabulate use of capital letters
loop
exit when eof
var c : char
% Read one character
get c
if c >= 'A' and c <= 'Z' then
frequency ( c ) := frequency ( c ) + 1
end if
end loop
% Print frequency of capital letters
for d : 'A' .. 'Z'
put d, " ", frequency ( d )
184 Object Oriented Turing Reference Manual
end for
Details
The type string (or char(n)) is used instead of char when more than one
character needs to be stored, such as the string of characters Henry. Unless
the program needs to be quite efficient, it is usually easier to use the string
type. See also the char(n) type, which always stores exactly n characters.
The char type differs from the string(1) type in the following way: char
always represents exactly one character, while string(1) can represent
either the null string or a string containing one character. The char type is
similar to the char(1) type in that both contain at most one character.
The char type is an index type and can be used, for example, as subscripts,
for ranges and case labels. For example, this declaration
var charCounts : array char of int
creates an array whose subscripts are characters.
The char type is a scalar type, which implies that its parameters are passed
by value, instead of by reference (which is the case for char(n) and string).
Values of the char type can be assigned and they can be compared for both
equality and ordering. Explicit char constants are written as a character
surrounded by single quotes, for example, 'A'. For details, including how to
write control characters, see explicitCharConstant.
Characters can be read and written by get and put statements.
There are 256 char values, corresponding to the distinct patterns in an 8-bit
byte. This allows the patterns eos (internal value 0) and uninitchar (internal
value 128) to be char values (these patterns are not allowed in the string
type; see the string type). All 256 patterns are used, so there is no pattern
left to be the "uninitialized value". Uninitialized checking is not done for
the char type.
The ord and chr functions convert between the char values and their
corresponding numeric representation in a byte. See ord and chr.
In general, you can freely intermix the values of the types char, char(n) and
string. This means that catenation (+), comparisons, length and substrings
can be applied to any of these types. See char(n) for details about
conversions between char, char(n) and string.
char(n)
Syntax
type
char ( numberOfCharacters )
Chapter 7 : Language Features 185
Description
Each variable whose type is a char(n) contains exactly n characters.
Example
Canadian postal codes contain six characters, for example, M4V 1Y9. This
is represented in a char(6) variable:
var postalCode : char ( 6 ) := 'M4V1Y9'
Details
Explicit constants for the char(n) type use single quotes as in 'M4V1Y9', as
opposed to explicit string constants which use double quotes, as in
"Nancy". A single character single quoted character, such as 'A', is
considered to have the type char instead of char(n), but since these two
types can be assigned to each other and compared to each other, this fact
has little consequence.
The type char(n) is generally more difficult to use than the string type,
which is favored for most simple programs. The type char(n) has the
advantage that it is efficient in terms of both space and time. In particular,
it is represented as n bytes in the computer's memory. By contrast, the
string type must use extra space (a trailing zero byte in current
implementations) to represent the current length and allocates space for the
maximum value it can hold.
The form of numberOfCharacters is one of:
(a) expn % Integer value
(b) *
% Only in subprogram parameters
The first form determines n. If the expression is a run time value, the type
is considered to be dynamic char(n). The value of n must be at least 1. The
second form is used only for subprogram parameters and uses the length
of the actual parameter. This too, is considered to be a dynamic char(n)
type. Dynamic char(n) types can only be passed to char(*) parameters.
Dynamic char(n) types have the same restrictions as dynamic arrays. This
implies they cannot be assigned as a whole and cannot appear in record
and union types.
An implementation may impose a limit, recommended to be at least 32767,
on the length n.
Values of the char(n) type can be assigned and they can be compared for
both equality and for ordering, but only if they have the same length and
they are not dynamic (i.e. the length must be known at compile time).
Values of the char(n) type can be read and written by get and put
statements.
The char(n) type is a nonscalar, which implies that its parameters are
always passed by reference (by means of an implicit pointer).
As is true for the char type, all 256 possible values of an 8-bit byte are
allowed for each character in char(n) type. There is no pattern left to be
used for the "initialized value", so there is no uninitialized checking for
char(n).
186 Object Oriented Turing Reference Manual
In general, you can freely intermix the values of the types char, char(n) and
string. This means that catenation (+), comparisons, length and substrings
can be applied to any of these types. See catenation and substring. If two
non dynamic char(n) values (or char values) are catenated, the result is a
char(n) value. If either are dynamic, it is a string value. This implies that
very long char(n) values can be created by catenating them together, for
example to initialize a char(n) variable.
A char value can be assigned (or passed to an non var parameter) with
automatic conversion to a char(1) variable and vice versa. String values of
length 1 can be assigned to char variables. Character (char) values can be
assigned to string variables, yielding a string of length 1. String values of
length n are assignable with automatic conversion to char(n) variables.
Values of type char(n) can be assigned with automatic conversion to string
variables.
When comparing two char(n) values, as in s > t, if both are non-dynamic
and of the same length, they are compared without converting to strings. If
either are dynamic, they are converted to strings and then compared.
the char type which is much like char(1). See also the string type.
See also
cheat
type cheating
Dangerous
A typeCheat is one of:
Syntax
(a)
(b)
(c)
Description
cheat ( targetType, expn [ : sizeSpec ] )
# expn
id : cheat typeSpec
A type cheat interprets the representation (bits) of one type as another
type. Type cheats are dirty (machine-dependent) and sometimes
dangerous (arbitrary corruption) and should be used only by programmers
who know the underlying computer representation of values.
Form (b) is a short form type cheat in which the target type is a natural
number.
Form (c) is used as a parameter in a subprogram declaration. It causes
whatever is passed in to the parameter to be interpreted as typeSpec.
Example
The character 'B' is assigned to variable i, whose type is considered to be
char (although it is really int1).
var i : int1
% One byte integer
cheat ( char, i ) := 'B'
Chapter 7 : Language Features 187
This assignment is equivalent (on byte oriented computers) to either of the
following:
i := cheat ( int1, 'B' )
i := ord ( 'B' )
Details
The form of targetType must be one of:
(a)
[ id . ] typeId
(b)
int, int1, int2 or int4
(c)
nat, nat1, nat2 or nat4
(d)
boolean
(e)
char [ ( numberOfCharacters ) ]
(f)
string [ ( maximumLength ) ]
(g)
addressint
In form (a) the beginning identifier id must be the name of a module,
monitor or class that exports the typeId. Each of numberOfCharacters and
maximumLength must be compile time integer expressions.
If the expn in a type cheat is a variable reference and the sizeSpec is omitted,
the type cheat is considered to be a variable whose type is targetType. This
allows, for example, the type cheat to be assigned to, as in:
cheat ( char, i ) := 'B'
If the expn is a value that is not a variable reference, or if sizeSpec is present,
the type cheat is an expression value whose type is targetType.
The sizeSpec is a compile time integer expression giving the size of the
expn's value. It can be specified only for integer or natural number values
(where it must be 1, 2 or 4) or real values (where it must be 4 or 8).
A type cheat is carried out in two steps. The first step converts the value if
necessary to the size given by sizeSpec. The second step, which involves no
generated code, interprets the value as the target type.
The prefix operator # is a short form for a class of type cheats. It interprets
its argument as a natural number. In general, # expn is the same as cheat
(natn, expn) where n is determined as follows. If the expn is a variable or
expression of size 1, 2 or 4, n is the size of the item, otherwise n is 4.
Example
Set the second character of d so it has the numeric representation 24. In
general, if c is a character, then #c = ord(c). Note that #c can have a number
value assigned to it, but ord(c) cannot.
var d : char ( 3 )
Example
#d ( 2 ) := 24
% Same as d(2) := chr(24)
The notation 16#FFFF means FFFF in base 16, which is 32767 in base 10 and
is 16 1's in a row in base 2. This same pattern is the two's complement
representation of the value -1 in a 2-byte integer.
var i : int2
#i := 16#FFFF
188 Object Oriented Turing Reference Manual
% Equivalent to i := -1
The following example prints out a string located at addressint myAddr.
Example
procedure PrintString (str : cheat string)
put str
end PrintString
var myAddr : addressint
...
% Assigned a value to myAddr
PrintString (myAddr) % myAddr will be treated as a string
Details
An implementation may prohibit certain type cheats. Memory alignment
requirements may render some type cheats unfeasible. It is dangerous to
consider a value to have a targetType larger than the value's type. An
implementation may prohibit certain type cheats on register scalar items.
See also
explicitIntegerConstant (for description of constants such as 16#FFFF) and
the following functions that convert one type to another in a machineindependent manner: ord, chr, intstr, strint, natstr, and strnat.
checked
Description
chr
compiler directive
Unchecked means that certain run time tests, which take place by default,
can be eliminated, usually to make the program more efficient at the risk of
unreliability. The keyword checked, used as a statement, requests that the
disabling of checking, previously requested by the keyword unchecked, be
re-enabled. See unchecked for details and an example.
integer-to-character function
Syntax
chr ( i : int ) : char
Description
The chr function is used to convert an integer to a character. The character
is the i-th character of the ASCII sequence of characters (except on the IBM
mainframe, which uses the EBCDIC sequence.) For example, chr (65) is
"A".
The ord function is the inverse of chr, so for any character c,:
chr (ord (c)) = c.
Chapter 7 : Language Features 189
ord, intstr and strint functions.
See also
class
declaration
A classDeclaration is:
Syntax
[ monitor ]
class id
[ inherit inheritItem ]
[ implement implementItem ]
[ implement by implementByItem ]
[ import [ var ] importItem {, [ var ] importItem } ]
[ export [ howExport ] id {, [ howExport ] id } ]
statementsAndDeclarations
end id
Description
A class declaration defines a template for a package of variables, constants,
types, subprograms, etc. The name of the class (id) is given in two places,
just after class and just after end. Items declared inside the class can be
accessed outside of the class only if they are exported. Items from outside
the class that are to be used in the class, need to be imported (unless they
are predefined or pervasive). Instances (objects) of a class are created using
the new statement. Each object is essentially a module located by a pointer.
Example
This class is a template for creating objects, each of which is a stack of
strings. (See the module description for the corresponding module that
implements a single stack of strings.)
class stackClass % Template for creating individual stacks
export push, pop
var top : int := 0
var contents : array 1 .. 100 of string
procedure push ( s : string )
top := top + 1
contents ( top ) := s
end push
190 Object Oriented Turing Reference Manual
procedure pop ( var s : string )
s := contents ( top )
top := top - 1
end pop
end stackClass
var p: pointer to stackClass % Short form: var p: ^stackClass
new stackClass, p
% Short form: new p
p -> push ( "Harvey" )
var name : string
p -> pop ( name )
% This sets name to be Harvey
Pointer p is used to locate individual objects of the class. The new
statement creates one of these objects. The statement
p -> push ( "Harvey" )
is a short form for:
stackClass (p) . push ("Harvey")
This inserts the string Harvey into the stack object located by p.
Details
The new statement is used to create objects of a class. Many instances of a
class can exist at a given time, each located by a pointer. The free statement
is used to destroy objects that are no longer of use. Turing does not support
garbage collection (automatic recovery of space belonging to inaccessible
objects).
See modules for a discussion of importing, exporting and related concepts.
When an object is created by new, its initialization code is executed. In this
example, the object's top variable is set to 0. As is true in modules, an
exported subprogram of an object's class cannot be called until the object is
completely initialized.
You are not allowed to create variables of a class, as in:
var s : stack
% Not legal!
If the monitor keyword is present (just before class), the objects are
monitors. This means that only one process at a time can be active in the
object. See monitor and process.
Inherit lists are used to specify inheritance. See inherit list. Implement and
implement-by lists provide a special kind of expansion which supports the
separation of an interface from its implementation. See implement list and
implement-by list. A class cannot contain both an inherit and an
implement list.
Class declarations can be nested inside modules and monitors but cannot
be nested inside other classes or inside procedures or functions. A class
must not contain a bind as one of its (outermost) declarations. A return
statement cannot be used as one of the (outermost) statements in a class.
A class cannot export variables (or run time constants) as unqualified
(because each object has a distinct set of variables).
Chapter 7 : Language Features 191
The syntax of a classDeclaration presented above has been simplified by
leaving out pre, invariant and post clauses. The full syntax which supports
pre, invariant and post is the same as that for modules. The initialization
of classes is the same as that for modules. See module.
Example
We will give an example in which a subprogram in one class overrides the
corresponding subprogram in a class that is being inherited. The example
is based on a program that implements a file system inside an operating
system. All files have open, close, read and write operations. Some files,
called Device files, also have an operation called ioCtl (input/output
control). The kind of file determines the implementation method. Here is
the expansion (inheritance) hierarchy among the classes of files.
File
Headers for open,
close, read, write
close, read, write
Text File
Device
Bodies for open,
close, read, write
close, read, write
Header for ioCtl
Tape
Disk
Bodies for open, close,
Bodies for open, close,
read, write, ioCtl
read, write, ioCtl
close, read, write
close, read, write
The class called File gives the interface to all possible kinds of files. The
TextFile class implements files that are text (ASCII characters). The Device
class gives the interface to all files that have the ioCtl operation in addition
to open, close, read and write. The Tape and Disk classes implement files that
are actually physical tapes or disks. Here is the declaration of the File class:
class File
export open, close, read, write
deferred procedure open ( … parameters for open … )
deferred procedure close ( … parameters for close … )
deferred procedure read ( … parameters for read … )
deferred procedure write ( … parameters for write … )
end File
192 Object Oriented Turing Reference Manual
The TextFile class implements the File interface by giving variables
declarations and procedure bodies for ASCII files:
class TextFile
inherit File
var internalTextFileData :
… internal data for text files …
body procedure open
… body for open for text files …
end open
… bodies for close, read and write procedures for text files…
end TextFile
Objects to represent individual text files are created using the new
statement:
var textFilePtr : ^ TextFile
% Pointer will locate a text file object
new textFilePtr
% Create a text file object
textFilePtr -> read ( … actual parameters … ) % Read text file
The Device class adds the ioCtl procedure to the File interface.
class Device
inherit File
export ioCtl
deferred procedure ioCtl ( … parameters for ioCtl … )
end Device
The Disk class provides data and procedures to implement a file that is
actually a disk (the Tape class is analogous):
class Disk
inherit Device
var internalDiskFileData : … internal data for disk files
body procedure open
… body for open …
end open
… bodies for close, read, write and ioCtl procedures for disk …
end Disk
A pointer that can locate any kind of File object is declared this way:
var filePtr : ^ File
This may locate, for example, a TextFile:
filePtr := textFilePtr
This assignment is allowed because filePtr's corresponding class (File) is an
ancestor of textFilePtr's corresponding class (TextFile). It is guaranteed that
the object now located by filePtr supports a version of all the operations of
a File (open, close, read and write).
Chapter 7 : Language Features 193
When we call a procedure in the object located by filePtr, the actual
procedure called will depend upon the object:
filePtr -> read ( … actual parameters … )
For example, if filePtr currently locates a Disk file, this will call the read
procedure from the Disk class. This is an example of dynamic binding in
which the version of read to be used is selected at run time and this choice
is based on the object located by filePtr. This is called polymorphism, because
File objects can have more than one form.
Example
As another example, consider class C, which contains headers and bodies
for functions f and g. C exports functions f and g. There is also a class D,
which inherits from C. Class D contains a body that overrides the body for
g. D also contains a header and body for function h. D exports function h.
Pointer p has been declared to locate an object of class C, but at runtime p
locates an object of class D. When p is used to call f, by means of p->f, the
body of f, which appears in C, is invoked. When p is used to call g, by
means of p->g, g’s overriding body in D is invoked. Any attempt to use p to
call h is illegal because p can only be used to call functions that are
exported from C.
class C
export f, g
procedure f
put "C’s f"
end f
procedure g
put "C’s g"
end g
end C
class D
inherit C
% Inherit f and g
body procedure g
% Overrides g in C
put "*** D’s g ***"
end g
procedure h
put "*** D’s h ***"
end h
end D
var p : pointer to C
% p can point to any descendant of C
new D, p
% p locates an object of class D
% Outputs "C’s f"
p -> f
% Outputs "*** D’s g ***"
p -> g
% Causes error "‘h’ is not in export list of ‘C’"
p -> h
See also
module, monitor and unit. See also import list, export list, implement list,
implement by list, and inherit list. See also deferred subprogram. See also
anyclass and objectclass.
194 Object Oriented Turing Reference Manual
clock
millisecs used procedure
Syntax
clock ( var c : int )
Description
The clock statement is used to determine the amount of time since a
program (process) started running. Variable c is assigned the number of
milliseconds since the program started running.
Example
This program tells you how much time it has used.
var timeRunning : int
clock ( timeRunning )
put "This program has run ", timeRunning, " milliseconds"
Details
On IBM PC compatibles, this is the total time since the Turing system was
started up. The hardware resolution of duration is in units of 55
milliseconds. For example, clock(i) may be off by as much as 55
milliseconds.
On Apple Macintoshes, this is the total time since the machine was turned
on. The hardware resolution of duration is in units of 17 milliseconds
(1/60-th of a second).
See also
delay, time, sysclock, wallclock and date statements.
See also predefined unit Time.
close
Syntax
file statement
A closeStatement is:
close : fileNumber
Description
In Turing, files are read and written using a fileNumber. In most cases, this
number is given a value using the open statement, which translates a file
name, such as "Master", to a file number, such as 5. When the program is
finished using the file, it disconnects from the file using the close
statement.
Example
This program illustrates how to open, read and then close a file.
Chapter 7 : Language Features 195
var fileName : string := "Master"% Name of file
var fileNo : int
% Number of file
var inputVariable : string ( 100 )
open : fileNo, fileName, read
…
read : fileNo, inputVariable
…
close : fileNo
In a Turing implementation, there will generally be a limit on the number
of currently open files. This limit will typically be around 10. To avoid
exceeding this limit, a program that uses many files one after another
should close files that are no longer in use.
Details
If a program does not close a file, the file will be automatically closed when
the program finishes.
There is an older and still acceptable version of close that has this syntax:
close ( fileNumber : int )
See also
cls
the open, get, put, read, write, seek and tell statements.
clear screen graphics procedure
Syntax
cls
Description
The cls (clear screen) procedure is used to blank the screen. The cursor is
set to the top left (to row 1, column 1).
Details
In "graphics" mode all pixels are set to color number 0, so the screen is
displayed in background color. In screen mode, the screen is set to the text
background color.
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See setscreen
for details.
See also predefined unit Draw and Text.
196 Object Oriented Turing Reference Manual
collection
declaration
A collectionDeclaration is one of:
Syntax
(a)
(b)
var id { , id } : collection of typeSpec
var id { , id } : collection of forward typeId
Description
A collection declaration creates a new collection (or collections). A
collection can be thought of as an array whose elements are dynamically
created (by new) and deleted (by free). Elements of a collection are
referred to by the collection's name subscripted by a pointer. See also new,
free and pointer.
Example
Create a collection that will represent a binary tree.
var tree : collection of
record
name : string ( 10 )
left, right : pointer to tree
end record
var root : pointer to tree
new tree, root
tree ( root ) . name := "Adam"
Details
The statement "new C,p" creates a new element in collection C and sets p to
point at i. If there is no more memory space for the element, though, p is set
to nil ( C ), which is the null pointer for collection C. The statement "free
C,p" deletes the element of C pointed to by p and sets p to nil ( C ). In each
case, p is passed as a var parameter and must be a variable of the pointer
type of C.
The keyword forward (form b above) is used to specify that the typeId of
the collection elements will be given later in the collection's scope. The later
declaration must appear at the same level (in the same list of declarations
and statements) as the original declaration. This allows cyclic collections,
for example, when a collection contains pointers to another collection,
which in turn contains pointers to the first collection. In this case, the typeId
is the name of the type that has not yet been declared; typeId cannot be
used until its declaration appears. A collection whose element type is
forward can be used only to declare pointers to it until the type's
declaration is given.
Suppose pointer q is equal to pointer p and the element they point to is
deleted by "free C,p". We say q is a dangling pointer because it seems to
locate an element, but the element no longer exists. A dangling pointer is
considered to be an uninitialized value. It cannot be assigned, compared,
used as a collection subscript, or passed to free.
Chapter 7 : Language Features 197
Collections cannot be assigned, compared, passed as parameters, bound to,
or named by a const declaration. Collections must not be declared in
procedures, functions, records or unions.
The same short forms for classes can be also used for collections. These
include omission of the collection name in new, free and nil together with
the ^ and -> notations. Pointers to types (see pointer) can also be used,
which are often more convenient to use than collections.
The syntax of a collectionDeclaration presented above has been simplified by
leaving out unchecked collections. With this feature, a collectionDeclaration
is one of:
(a) var id { , id } : [ unchecked ] collection of typeSpec
(b) var id { , id } : [ unchecked ] collection of forward typeId
When unchecked is specified, the checking to verify that pointers actually
locate elements is removed. This checking is done using a "time stamp"
attached to each element and pointer, and making sure that these match
with each other. When unchecked is specified, the execution is dangerous,
but faster and smaller, and the pointers become simply machine addresses
(as in C ).
color
text color graphics procedure
Syntax
color ( Color : int )
Description
The color procedure is used to change the currently active color. This is the
color of characters that are to be put on the screen. The alternate spelling is
colour.
Example
This program prints out the message "Bravo" three times, each in a
different color.
setscreen ( "graphics" )
for i : 1 .. 3
color ( i )
put "Bravo"
end for
Example
This program prints out a message. The color of each letter is different
from the preceding letter. For letter number i the color number is i mod
maxcolor + 1. This cycles repeatedly through all the available colors.
setscreen ( "screen" )
const message := "Happy New Year!!"
198 Object Oriented Turing Reference Manual
for i : 1 .. length ( message )
color ( i mod maxcolor + 1 )
put message ( i ) ..
end for
See setscreen for the number of colors available in the various "graphics"
modes.
Details
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See setscreen
for details.
colorback, whatcolor, whattextcolor and maxcolor.
See also
See also predefined unit Text.
colorback
background color procedure
Syntax
colorback ( Color : int )
Description
The colorback procedure is used to change the current background color.
The alternate spelling is colourback.
In "screen" mode on IBM PC compatibles, colorback sets the background
color to one of the colors numbered 0 to 7. This is the color that surrounds
characters when they are put onto the screen. On UNIX dumb terminals,
colorback(1) turns on highlighting and colorback(0) turns it off. On other
systems, this procedure may have no effect.
In "graphics" mode on IBM PC compatibles, colorback is used to associate a
color with pixel color number 0, which is considered to be the background
color. Using colorback immediately changes the color being displayed for
all pixels with color number 0.
Example
Since this program is in "screen" mode, changing the background color has
no immediately observable effect. When the message "Greetings" is output,
the background surrounding each letter will be in color number 2.
setscreen ( "screen" )
…
colorback ( 2 )
put "Greetings"
Example
Since this program is in "graphics" mode, changing the background color
immediately changes the colors of all pixels whose color number is 0.
setscreen ( "graphics" )
…
Chapter 7 : Language Features 199
colorback ( 2 )
Details
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See setscreen
for details
See also
color and whatcolorback.
See also predefined unit Text.
comment
Description
remark statement
A comment is a remark to the reader of the program, which the computer
ignores. The most common form of comment in Turing starts with a
percent sign (%) and continues to the end of the current line; this is called
an end-of-line comment. There is also the bracketed comment, which begins
with the /* and ends with */ and which can continue across line
boundaries.
Example
% This is an end-of-line comment
var x : real
% Here is another end-of-line comment
const s := "Hello"
/* Here is a bracketed comment that
lasts for two lines */
const pi := 3.14159
Details
In the BASIC language, comments are called remarks and start with the
keyword REM. In Pascal, comments are bracketed by (* and *).
comparisonOperator
Syntax
A comparisonOperator is one of:
(a)
(b)
(c)
<
>
=
% Less than
% Greater than
% Equal
200 Object Oriented Turing Reference Manual
(d)
(e)
(f)
Description
<= % Less than or equal; subset
>= % Greater than or equal; superset
not= % Not equal
A comparison operator is placed between two values to determine their
equality or ordering. For example, 7 > 2 is true and so is "Adam" < "Cathy".
The comparison operators can be applied to numbers as well as to
enumerated types. They can also be applied to strings to determine the
ordering between strings (see the string type for details). Arrays, records,
unions and collections cannot be compared. Boolean values (true and
false) can be compared only for equality (= and not=); the same is true of
pointer values. Set values can be compared using <= and >=, which are the
subset and superset operators. The not= operator can be written as ~=.
Comparisons among classes is also supported (see class). If C and D are
classes, C <= D means D is a descendant of (inherits from) C. See class.
See also
See also infix operators and precedence of operators. See also the int, real,
string, set, boolean and enum types. See also string comparison.
Concurrency
Description
All
This unit contains the predefined procedures that deal with concurrency. It
contains one predefined function, although conceptually it contains three
other subprograms.
All routines in the Concurrency module are exported unqualified. (This
means you can call the entry points directly.)
Entry Points
empty*
Returns true if no processes are waiting on the condition
queue.
getpriority*
Returns the priority of the current process.
setpriority*
Sets the priority of the current process.
simutime
Returns the number of simulated time units that have
passed.
* Part of the language, conceptually part of the Concurrency unit.
Chapter 7 : Language Features 201
Concurrency.empty
All
Syntax
empty ( variableReference ) : boolean
Description
The empty function is used in a concurrent program. It returns true if the
variableReference, which must be a condition variable, has no processes
waiting for it. Processes join the queue of a condition variable by executing
the wait statement, and are awakened by the signal statement.
Status
Part of the language and only conceptually part of the Concurrency unit.
This means that you can only call the function by calling empty, not by
calling Concurrency.empty.
See also
condition, wait, signal, fork and monitor.
Concurrency.getpriority
All
Syntax
getpriority : nat
Description
The getpriority function returns the priority of an executing process in a
concurrent program. A smaller value means a faster speed.
Status
Part of the language and only conceptually part of the Concurrency unit.
This means that you can only call the function by calling getpriority, not by
calling Concurrency.getpriority.
See also
setpriority, fork and monitor.
202 Object Oriented Turing Reference Manual
Concurrency.setpriority
All
Syntax
setpriority ( p : nat )
Description
The setpriority procedure is used to set the priority of a process in a
concurrent program. This priority cannot be counted on to guarantee
critical access to shared variables. A smaller value of p means increased
speed. The argument to setpriority may be limited to the range 0 to
2**15 – 1.
Status
Part of the language and only conceptually part of the Concurrency unit.
This means that you can only call the function by calling setpriority, not by
calling Concurrency.setpriority.
getpriority, fork and monitor.
See also
Concurrency.simutime
All
Syntax
simutime : int
Description
The simutime function returns the number of simulated time units that
have passed since program execution began.
Details
Simulated time only passes when all process are either paused or waiting.
This simulates the fact that CPU time is effectively infinitely faster than
"pause" time.
Example
This prints out the simulated time passing between two processes. This will
print out 3, 5, 6, 9, 10, 12, 15, 15, 18, 20, 21, ...
process p (t : int)
loop
pause t
put simutime
end loop
end p
fork p (3)
Chapter 7 : Language Features 203
fork p (5)
Exported unqualified.
Status
This means that you can call the function by calling simutime or by calling
Concurrency.simutime.
condition
declaration
A conditionDeclaration is:
Syntax
var id { , id } : [ array indexType {, indexType } of ]
[ conditionOption ] condition
Description
A condition is essentially a queue of sleeping processes. It is used in a
concurrent program to allow processes to block themselves (by the wait
statement) and later to be awakened (by the signal statement). A condition
variable, which can occur only inside a monitor (a special kind of module
that handles concurrency) or monitor class, is used by the wait and signal
statements for putting processes to sleep and later waking them up.
Example
The processes use this monitor to gain exclusive access to a resource. A
process wanting to use the resource calls the request entry point and is
blocked until the resource is free. When the process is finished with the
resource, it calls the release entry point. This monitor is essentially a binary
semaphore in which the semaphore's P operation is the request and the V is
the release.
monitor resource
export request, release
var available : boolean := true
var nowAvailable : condition
procedure request
if not available then
wait nowAvailable % Go to sleep
end if
assert available
% Allocate resource
available := false
end request
procedure release
assert not available
% Resource is allocated
% Free the resource
available := true
signal nowAvailable % Wake up one process
% If any are sleeping
204 Object Oriented Turing Reference Manual
end release
end resource
process worker
loop
…
resource.request
… use resource …
resource.release
end loop
end worker
fork worker
Details
% Block until available
% Activate one worker
fork worker
% Activate another worker
A conditionOption is one of:
(a)
priority
(b)
deferred
(c)
timeout
The priority option requires that the corresponding wait statements
include priorities. Options (b) and (c) declare deferred conditions. A signal
to a deferred condition causes the signaled process to become ready to
enter the monitor when the monitor becomes inactive. The signaling
process continues running in the monitor. A signal to an immediate (non
deferred) condition causes the signaled process to begin running in the
monitor immediately. The signaling process waits to re-enter the monitor
when the monitor becomes inactive. All conditions in a device monitor
must be deferred (or timeout).
A timeout option means the signaling is deferred and that an extra
parameter to the wait statement must give a timeout interval. If a process
waits longer than its interval, it is automatically signaled. Beware that the
empty function can be non-repeatable when applied to timeout conditions.
For example, empty(c) may not be equal to empty(c) in a single expression.
In the current (1999) version of Turing, the time for time outs is measured
in simulation time rather than real time. See the pause statement.
Conditions cannot be named as types, cannot be contained in records,
unions or collections and cannot be declared in statements (such as begin
or loop) or in subprograms. They can only be declared in monitors and
monitor classes.
There is no guaranteed order of progress among awakened deferred
processes, processes signaling immediate conditions, and processes
attempting to enter an active monitor.
Note that conditionOption must precede the keyword condition.
See also
wait and signal. See also monitor and fork. See also empty. See also
pause.
Chapter 7 : Language Features 205
Win DOS Mac
Config
Description
X TOOT Term
This unit contains the predefined subprograms that deal with getting
configuration information about the machine and environment on which
the program is being run. It exists in order to allow users to obtain
information about the system that may only be available at run time.
All routines in the Config module are exported qualified (and thus must be
prefaced with "Config.").
Entry Points
Display
Returns information about the display currently attached.
Lang
Returns information about the language environment that
the program is currently running within.
Machine
Returns information about the hardware on which the
program is running.
Win DOS Mac
Config.Display
X TOOT Term
Syntax
Config.Display (displayCode : int) : int
Description
Config.Display returns information about the display (or displays)
attached to the computer. The parameter displayCode determines what sort
of information is passed back. displayCode has a number of possible values,
all summarized by a set of predefined constants.
At the time of this writing, the following constants were defined:
cdScreenHeight
return the height of the screen in pixels.
cdScreenWidth
return the width of the screen in pixels.
cdMaxNumColors
return the maximum number of colors supported
by the display.
cdMaxNumColours
return the maximum number of colors supported
by the display.
cdMaxNumPages
return the maximum number of display pages
supported by the display (DOS OOT only).
206 Object Oriented Turing Reference Manual
This program prints the screen width and height.
Example
const width : int := Config.Display (cdScreenWidth)
const height: int := Config.Display (cdScreenHeight)
put "The screen width is ", width, " the screen height is ", height
Details
On the Macintosh, it’s possible to have multiple displays attached to a
single computer. To get information about the extra displays, you can call
Config.Display with any of the first four constants above plus one, two,
three, etc. This will return the height, width or maximum number of colors
for the second, third and beyond displays.
Example
This program prints the screen width and height of the second display on a
Macintosh.
const width : int := Config.Display (cdScreenWidth + 1)
const height: int := Config.Display (cdScreenHeight + 1)
put "The second display size is ", width, " x ", height
Status
Exported qualified.
This means that you can only call the function by calling Config.Display,
not by calling Display.
See Also
View.SetActivePage, View.SetVisiblePage for details about multiple
pages (clMaxNumPages).
Win DOS Mac
Config.Lang
X TOOT Term
Syntax
Config.Lang (langCode : int) : int
Description
Config.Lang returns information about the language and the limitations of
the implementation that the program is currently running. The parameter
langCode determines what sort of information is passed back. langCode has
a number of possible values, all summarized by a set of predefined
constants.
At the time of this writing, the following constants were defined:
clRelease
return the current release number of the
environment (e.g. 7.02 = 702).
clLanguageVersion
return the current version number of the
language (e.g. 1.81 = 181).
Chapter 7 : Language Features 207
clMaxNumStreams
return the maximum number of I/O streams
(used by the open and close statements) that can
be opened at once.
clMaxNumDirStreams
return the maximum number of directory streams
that can be opened at once.
clMaxNumRunTimeArgs return the maximum number of run-time
arguments.
This program prints the current environment version.
Example
const version : int := Config.Lang (clLanguageVersion)
put "The language version number is ", version
Exported qualified.
Status
This means that you can only call the function by calling Config.Lang, not
by calling Lang.
Win DOS Mac
Config.Machine
X TOOT Term
Syntax
Config.Machine (machineCode : int) : int
Description
Config.Machine returns information about the machine that the program
is currently running on. The parameter machineCode determines what sort
of information is passed back. machineCode has a number of possible
values, all summarized by a set of predefined constants.
At the time of this writing, the following constants were defined:
Example
cmProcessor
return an encoding of the processor number.
cmFPU
return 1 if there is an FPU installed, 0 if not.
cmOS
return the current version number of the operating
system (e.g. 6.07 = 607).
This program prints whether the machine has an FPU or not.
if Config.Machine (cmFPU) = 1 then
put "The machine has an FPU installed"
else
put "The machine does not have an FPU installed"
end if
Status
Exported qualified.
This means that you can only call the function by calling Config.Machine,
not by calling Machine.
208 Object Oriented Turing Reference Manual
const
constant declaration
A constantDeclaration is:
Syntax
const id [ : typeSpec ] := initializingValue
Description
A const declaration creates a name id for a value.
Example
const c := 3
const s := "Hello"
% The type of s is string
const x := sin ( y ) ** 2
const a : array 1..3 of int := init ( 1, 2, 3 )
const b : array 1..3 of int := a
const c : array 1..2, 1..2 of int := init ( 1, 2, 3, 4 )
% So c(1,1)=1, c(1,2)=2, c(2,1)=3, c(2,2)=4
Details
The initializing value can be an arbitrary value or else a list of items
separated by commas inside init ( … ). The syntax of initializingValue is:
a.
expn
b.
init ( initializingValue, initializingValue )
Each init ( … ) corresponds to an array, record or union value that is being
initialized. These must be nested for initialization of nested types. In the
Pascal language, constants must have values known at compile time;
Turing has no such restriction.
When the typeSpec is omitted, the variable's type is taken to be the (root)
type of the initializing expression, for example, int or string. The typeSpec
cannot be omitted for dynamic arrays or when the initializing value is of
the form init ( … ). The values inside init ( … ) must be known at compile
time.
The keyword pervasive can be inserted just after const. When this is done,
the constant is visible inside all subconstructs of the constant's scope.
Without pervasive, the constant is not visible inside modules, monitors or
classes unless explicitly imported. Pervasive constants need not be
imported. You can abbreviate pervasive as an asterisk (*).
You can also optionally use the register keyword to request that the
constant be placed in a machine register. The syntax for
constantDeclaration is actually:
const [pervasive] [register] id [ : typeSpec ] := initializingValue
In the current (1999) implementation, programs are run interpretively
using pseudo-code, which has no machine registers, and the register
keyword is ignored. See also register for restrictions on the use of register
constants.
Chapter 7 : Language Features 209
constantReference
use of a constant
A constantReference is:
Syntax
constantId { componentSelector }
Description
In a Turing program, a constant is declared and given a name (constantId)
and then used. Each use is called a constant reference.
If the constant is an array, record or union, its parts (components) can be
selected using subscripts and field names (using componentSelectors). The
form of a componentSelector is one of:
(a)
( expn {, expn} )
(b)
. fieldId
Form (a) is used for subscripting (indexing) arrays. The number of array
subscripts must be the same as in the array's declaration. Form (b) is used
for selecting a field of a record or union. Component selectors are used in
the same manner as variable references. See variableReference for details. See
also const declaration and explicitConstant.
Example
var radius : real
% Constant declaration
const pi := 3.14159
…
put "Area is: ", pi * radius **2
cos
% pi is a constant reference.
cosine function (radians)
Syntax
cos ( r : real ) : real
Description
The cos function is used to find the cosine of an angle given in radians. For
example, cos (0) is 1.
Example
This program prints out the cosine of /6, 2/6, 3/6, up to 12/6 radians.
const pi := 3.14159
for i : 1 .. 12
const angle := i * pi / 6
210 Object Oriented Turing Reference Manual
put "Cos of ", angle, " is ", cos ( angle )
end for
cosd function which finds the cosine of an angle given in degrees. (2
radians are the same as 360 degrees.)
See also
See also predefined unit Math.
cosd
cosine function (degrees)
Syntax
cosd ( r : real ) : real
Description
The cosd function is used to find the cosine of an angle given in degrees.
For example, cosd (0) is 1.
Example
This program prints out the cosine of 30, 60, 90, up to 360 degrees.
for i : 1 .. 12
const angle := i * 30
put "Cos of ", angle, " is ", cosd ( angle )
end for
See also
cos function which finds the cosine of an angle given in radians.
(2radians are the same as 360 degrees.)
See also predefined unit Math.
date
procedure
Syntax
date ( var d : string )
Description
The date statement is used to determine the current date. Variable d is
assigned a string in the format "dd mmm yy", where mmm is the first 3
characters of the month, e.g., "Apr". For example, if the date is Christmas
1989, d will be set to "25 Dec 89".
Example
This program greets you and tells you the date.
Chapter 7 : Language Features 211
var today : string
date ( today )
put "Greetings!! The date today is ", today
Details
Be warned that on some computers, such as IBM PC compatibles or Apple
Macintoshes, the date may not be set correctly in the operating system; in
that case, the date procedure will give incorrect results.
See also
delay, clock, sysclock, wallclock and time statements.
See also predefined unit Time.
declaration
create a variable
A declaration is one of:
Syntax
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
Description
variableDeclaration
constantDeclaration
typeDeclaration
bindDeclaration
procedureDeclaration
functionDeclaration
moduleDeclaration
classDeclaration
processDeclaration
monitorDeclaration
conditionDeclaration
A declaration creates a new name (or names) for a variable, constant, type,
procedure, function, module, class, process, monitor, or condition. These
names are called identifiers, where id is the abbreviation for identifier.
Example
var width : int
const pi := 3.14159
type range : 0 .. 150
% Variable declaration
% Constant declaration
% Type declaration
procedure greet
put "Hello world"
% Procedure declaration
end greet
212 Object Oriented Turing Reference Manual
Ordinarily, each new name must be distinct from names that are already
visible; that is, redeclaration is not allowed. There are certain exceptions to
this rule, for example, names of parameters and fields of records can be the
same as existing visible variables. Variables declared inside a subprogram
(a procedure and function) are allowed to be the same as variables global
to (outside of) the subprogram.
Details
The effect of a declaration (its scope) lasts to the end of the construct in
which the declaration occurs; this will be the end of the program, the end
of the surrounding procedure, function or module, the end of a loop, for,
case or begin statement, or the end of the then, elsif, or else clause of an if
statement, or the end of the case statement alternative.
A name must be declared before it can be used; this is called the DBU
(Declaration Before Use) rule. The exceptions to this rule use the keyword
forward, as in import lists and in collection declarations.
A declaration can appear any place a statement can appear. This differs from
the Pascal language, in which declarations are allowed only at the
beginning of the program or at the beginning of a procedure or function.
Each declaration can optionally be followed by a semicolon (;).
There are certain restrictions on the placement of declarations. Procedures
and functions cannot be declared inside other procedures and functions
nor inside statements (for example, not inside an if statement). A bind
declaration cannot appear at the outer level of either the main program or a
module. A condition declaration can appear only inside a monitor.
Processes cannot be declared inside procedures, functions, monitors or
classes. Classes cannot be declared inside classes. However, modules and
monitors can be declared inside classes and vice versa. Monitors can be
declared inside modules, not vice versa.
deferred
subprogram declaration
A deferredDeclaration is:
Syntax
deferred subprogramHeader
Description
A procedure or function is declared to be deferred when you want to be
able to override the subprogram in an expansion. The procedure or
function must be in a module, monitor or class.
Example
The display procedure is deferred in this class of stacks to allow various
ways of graphically displaying the stack on the screen:
class stack
Chapter 7 : Language Features 213
export push, pop
… local declarations …
… declarations of the push and pop procedures …
deferred procedure display ( howbig : int )
end stack
An expansion to the stack class can give a body for display, as in:
class stackWithSimpleDisplay
body procedure display
% ( howbig : int )
… graphically display the stack on the screen …
end display
end stackWithSimpleDisplay
The following creates a stack that can be displayed and displays it:
var p : ^stackWithSimpleDisplay
new p
…
p -> display ( 25 )
% Display the stack on the screen
A deferred procedure is resolved by giving its body. This can be done in the
scope (module, monitor or class) containing the deferred declaration
(following the deferred declaration) or in any expansion of that scope.
Only one resolution per scope is allowed. Unresolved subprograms can be
called, but they immediately abort.
Details
All exported subprograms are implicitly deferred and can be overridden in
expansions.
During initialization of a module, monitor or object of a class, deferred
subprograms (including exported subprograms) cannot be called. This
restriction prevents accessing an object before it is fully initialized.
A deferred declaration must not appear in the main program.
module, monitor and class. See also export list, import list, inherit list,
implement list and implement by list.
See also
delay
procedure
Syntax
delay ( duration : int )
Description
The delay statement is used to cause the program to pause for a given
time. The time duration is in milliseconds.
Example
This program prints the integers 1 to 10 with a second delay between each.
for i : 1 .. 10
214 Object Oriented Turing Reference Manual
put i
delay ( 1000 )% Pause for 1 second
end for
Details
On IBM PC compatibles, the hardware resolution of duration is in units of
55 milliseconds. For example, delay(500) will delay the program by about
half a second, but may be off by as much as 55 milliseconds.
On Apple Macintoshes, the hardware resolution of duration is in units of
17 milliseconds (1/60th of a second). For example, delay(500) will delay
the program by about half a second, but may be off by as much as 17
milliseconds.
See also
sound, clock, sysclock, wallclock, time and date statements.
See also predefined unit Time.
Win DOS Mac
Dir
Description
X TOOT Term
This unit contains the predefined subprograms that deal with directories.
You can use these subprograms to list the contents of directories, create
directories, change directories and return the current directory.
All routines in the Dir module are exported qualified (and thus must be
prefaced with "Dir.").
Entry Points
See also
Open
Opens a directory stream in order to get a listing of the
directory contents.
Get
Gets the next file name in the directory listing.
GetLong
Gets the next file name and other information in the
directory listing.
Close
Closes the directory stream.
Create
Creates a new directory.
Delete
Deletes a directory (must be empty).
Change
Changes the current execution directory.
Current
Returns the current execution directory.
Files unit for an explanation of the different ways of specifying a path
name of a file or directory under the different operating systems.
Chapter 7 : Language Features 215
Win DOS Mac
Dir.Change
X TOOT Term
Syntax
Dir.Change (directoryPathName : string)
Description
Dir.Change changes the execution directory to that specified by the
parameter directoryPathName. This is the equivalent of doing a cd in UNIX.
On a Windows or DOS machine, specifying a drive in the
directoryPathName parameter causes the drive to become the default drive
(unlike the DOS cd command).
Details
If the Dir.Change call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program changes to the directory called /usr/west and then lists the
current directory.
Dir.Change ("/usr/west")
if Error.Last = eNoError then
put "Directory changed"
else
put "Did not change the directory."
put "Error: ", Error.LastMsg
end if
put "The current execution directory is ", Dir.Current
Status
Exported qualified.
This means that you can only call the function by calling Dir.Change, not
by calling Change.
Win DOS Mac
Dir.Close
Syntax
X TOOT Term
Dir.Close (streamNumber : int)
216 Object Oriented Turing Reference Manual
Description
Dir.Close is part of a series of four subprograms that help users get
directory listings. Dir.Close is used to close a directory stream number
opened by Dir.Open. After the directory stream number is closed, it can
not be used with Dir.Get or Dir.GetLong.
Details
If the Dir.Close call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
This program prints a listing of all the files in the directory datafiles.
Example
var streamNumber : int
var fileName : string
streamNumber := Dir.Open ("datafiles")
assert streamNumber > 0
loop
fileName := Dir.Get (streamNumber)
exit when fileName = ""
put fileName
end loop
Dir.Close (streamNumber)
Exported qualified.
Status
This means that you can only call the function by calling Dir.Close, not by
calling Close.
Win DOS Mac
Dir.Create
X TOOT Term
Syntax
Dir.Create (directoryPathName : string)
Description
Dir.Create is used to create the directory specified by the parameter
directoryPathName. This is the equivalent of doing a mkdir in DOS or
UNIX. On the Macintosh, it creates a folder.
Details
If the Dir.Create call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program creates the directory called information.
Dir.Create ("information")
if Error.Last = eNoError then
put "Directory created"
else
put "Did not create the directory."
put "Error: ", Error.LastMsg
Chapter 7 : Language Features 217
end if
Exported qualified.
Status
This means that you can only call the function by calling Dir.Create, not by
calling Create.
Win DOS Mac
Dir.Current
X TOOT Term
Syntax
Dir.Current : string
Description
Dir.Current returns the full path name of the current execution directory.
This is the equivalent of doing a pwd in UNIX.
Details
If the Dir.Current call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program changes to the directory called /usr/west and then lists the
current directory.
Dir.Change ("/usr/west")
if Error.Last = eNoError then
put "Directory changed"
else
put "Did not change the directory."
put "Error: ", Error.LastMsg
end if
put "The current execution directory is ", Dir.Current
Status
Exported qualified.
This means that you can only call the function by calling Dir.Current, not
by calling Current.
Win DOS Mac
Dir.Delete
Syntax
X TOOT Term
Dir.Delete (directoryPathName : string)
218 Object Oriented Turing Reference Manual
Description
Dir.Delete is used to delete the directory specified by the parameter
directoryPathName. This is the equivalent of doing a rmdir in DOS or UNIX.
On the Macintosh, it removes a folder.
Dir.Delete will fail if it attempts delete a directory that has files in it.
If the Dir.Delete call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Details
This program deletes the directory called information.
Example
Dir.Delete ("information")
if Error.Last = eNoError then
put "Directory delete"
else
put "Did not delete the directory."
put "Error: ", Error.LastMsg
end if
Exported qualified.
Status
This means that you can only call the function by calling Dir.Delete, not by
calling Delete.
Win DOS Mac
Dir.Get
X TOOT Term
Syntax
Dir.Get (streamNumber : int) : string
Description
Dir.Get is part of a series of four subprograms that help users get directory
listings. Dir.Get is used to get the file names in the directory. Each time the
function is called, it returns the next file name in the directory. The names
are not sorted. When there are no more file names in the directory, Dir.Get
returns the empty string.
Details
If the Dir.Get call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program prints a listing of all the files in the directory datafiles.
var streamNumber : int
var fileName : string
streamNumber := Dir.Open ("datafiles")
assert streamNumber > 0
loop
fileName := Dir.Get (streamNumber)
Chapter 7 : Language Features 219
exit when fileName = ""
put fileName
end loop
Dir.Close (streamNumber)
Status
Exported qualified.
This means that you can only call the function by calling Dir.Get, not by
calling Get.
Win DOS Mac
Dir.GetLong
X TOOT Term
Syntax
Dir.GetLong (streamNumber : int, var entryName : string,
var size, attribute, fileTime : int)
Description
Dir.GetLong is part of a series of four subprograms that help users get
directory listings. Dir.GetLong is used to get the names and assorted
information of the files in the directory. Each time the function is called, it
returns the name and information of the next file in the directory. The
names are not sorted. When there are no more file names in the directory,
Dir.GetLong returns the empty string in the entryName parameter.
The size parameter is the size of the file in bytes. The attribute parameter
has its individual bits set as follows (the individual bits can be extracted
using the bits operator):
Bit 0 – ootAttrDir – set to 1 if entry is a directory.
Bit 1 – ootAttrRead – set to 1 if the program can read the file.
Bit 2 – ootAttrWrite – set to 1 if the program can write the file.
Bit 3 – ootAttrExecute – set to 1 if the program can execute the file.
Bit 4 – ootAttrHidden – set to 1 if the entry if a hidden file (PC, Mac).
Bit 5 – ootAttrSystem – set to 1 if the entry is a system file (PC only).
Bit 6 – ootAttrVolume – set to 1 if the entry is a volume name (PC only).
Bit 7 – ootAttrArchive – set to 1 if the entry has archive bit set (PC only).
Bit 8 – ootAttrOwnerRead – set to 1 if the file owner can read it.
Bit 9 – ootAttrOwnerWrite – set to 1 if the file owner can write it.
Bit 10 – ootAttrOwnerExecute – set to 1 if the file owner can execute it.
Bit 11 – ootAttrGroupRead – set to 1 if the file owner’s group members
can read it.
Bit 12 – ootAttrGroupWrite – set to 1 if the file owner’s group members
can write it.
Bit 13 – ootAttrGroupExecute – set to 1 if the file owner’s group
members can execute it.
Bit 14 – ootAttrGroupRead – set to 1 if anyone can read it.
Bit 15 – ootAttrGroupWrite – set to 1 if anyone can write it.
220 Object Oriented Turing Reference Manual
Bit 16 – ootAttrGroupExecute – set to 1 if anyone can execute it.
Bits 8-16 are only really applicable to multiple user systems like UNIX.
although they are set properly on any OOT system.
The ootAttr... constants are defined in the File unit. They correspond to the
values of attribute if a specified bit is set. For example, ootAttrSystem is the
value of the attribute parameter if bit 5 is set to 1. You can and or or these
constants to get combinations of specific file attributes.
The fileTime is the time of last modification of the file. It is returned as the
number of seconds since 00:00:00 GMT 1/1/1970. To convert this to a
string, use Time.SecDate
Details
Example
If the Dir.GetLong call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
This program prints a listing of all the files in the directory datafiles.
var streamNumber : int
var fileName : string
var size, attribute, fileTime : int
streamNumber := Dir.Open ("datafiles")
assert streamNumber > 0
loop
Dir.GetLong (streamNumber, fileName, size, attribute, fileTime)
exit when fileName = ""
put fileName, " ", Time.SecDate (fileTime)
end loop
Dir.Close (streamNumber)
Example
This program prints a listing of the attributes of all the files in the current
directory.
var streamNumber : int
var fileName : string
var size, attribute, fileTime : int
streamNumber := Dir.Open (Dir.Current)
assert streamNumber > 0
loop
Dir.GetLong (streamNumber, fileName, size, attribute, fileTime)
exit when fileName = ""
put fileName, " "..
if (attribute and ootAttrDir) not= 0 then
put "Directory "..
end if
if (attribute and ootAttrRead) not= 0 then
put "Readable "..
end if
if (attribute and ootAttrWrite) not= 0 then
put "Writeable "..
end if
if (attribute and ootAttrExecute) not= 0 then
put "Executable "..
end if
Chapter 7 : Language Features 221
put ""
end loop
Dir.Close (streamNumber)
Exported qualified.
Status
This means that you can only call the function by calling Dir.GetLong, not
by calling GetLong.
Win DOS Mac
Dir.Open
X TOOT Term
Syntax
Dir.Open (directoryPathName : string) : int
Description
Dir.Open is part of a series of four subprograms that help users get
directory listings. Dir.Open returns a directory stream number if the
directory could be opened. This stream number can be used to get file
names and information using the Dir.Get and Dir.GetLong subprograms.
After getting the listing, the user should call Dir.Close.
Details
If the Dir.Open call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program prints a listing of all the files in the current directory.
var streamNumber : int
var fileName : string
streamNumber := Dir.Open (Dir.Current)
assert streamNumber > 0
loop
fileName := Dir.Get (streamNumber)
exit when fileName = ""
put fileName
end loop
Dir.Close (streamNumber)
Status
Exported qualified.
This means that you can only call the function by calling Dir.Open, not by
calling Open.
222 Object Oriented Turing Reference Manual
div
integer truncating division operator
Syntax
div
Description
The div operator divides one number by another and produces the integer
result, truncated in the direction of zero. For example, 7 div 2 produces 3
and -7 div 2 produces -3.
Example
In this example, eggCount is the total number of eggs. The first put
statement determines how many dozen eggs there are. The second put
statement determines how many extra eggs there are beyond the last
dozen.
var eggCount : int
get eggCount
put "You have ", eggCount div 12, " dozen eggs"
put "You have ", eggCount mod 12, " left over"
See also
infix operators, precedence of operators and the mod operator.
Win DOS Mac
Draw
Description
X TOOT Term
This unit contains the predefined subprograms that deal with drawing
pixel graphics to the screen.
All routines in the Draw unit are exported qualified (and thus must be
prefaced with "Draw.").
Entry Points
Cls
Clears the screen to color 0.
Dot
Draws a dot.
Line
Draws a line.
Box
Draws a box.
FillBox
Draws a filled box.
Oval
Draws an oval.
Chapter 7 : Language Features 223
FillOval
Draws a filled oval.
Arc
Draws an arc.
FillArc
Draws a filled arc or a wedge.
Polygon
Draws a polygon.
FillPolygon
Draws a filled polygon.
MapleLeaf
Draws a maple leaf.
FillMapleLeaf Draws a filled maple leaf.
Star
Draws a star.
FillStar
Draws a filled star.
Fill
Does a flood fill.
Text
Draws text as graphics
Win DOS Mac
Draw.Arc
X TOOT Term
Syntax
Draw.Arc ( x, y, xRadius, yRadius : int,
initialAngle, finalAngle, Color : int )
Description
The Draw.Arc procedure is used to draw an arc whose center is at (x, y).
This is just like Draw.Oval, except that you must also give two angles,
initialAngle and finalAngle, which determine where to start and stop
drawing. Zero degrees is "three o'clock", 90 degrees is "twelve o'clock", etc.
The horizontal and vertical distances from the center to the arc are given by
xRadius and yRadius.
Example
This program draws a semicircle (actually, an approximation to a
semicircle) whose center is (midx,0) the bottom center of the screen, using
color number 1. The maxx and maxy functions are used to determine the
maximum x and y values on the screen.
View.Set ( "graphics" )
const midx := maxx div 2
Draw.Arc ( midx, 0, maxy, maxy, 0, 180, 1 )
224 Object Oriented Turing Reference Manual
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Arc, not by
calling Arc.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.Box
X TOOT Term
Syntax
Draw.Box ( x1, y1, x2, y2, Color : int )
Description
The Draw.Box procedure is used to draw a box on the screen with bottom
left and top right corners of (x1, y1) to (x2, y2) using the specified Color.
Example
This program draws a large box, reaching to each corner of the screen
using color number 1. The maxx and maxy functions are used to determine
the maximum x and y values on the screen. The point (0,0) is the left
bottom of the screen and (maxx, maxy) is the right top.
View.Set ( "graphics" )
Draw.Box ( 0, 0, maxx, maxy, 1 )
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Box, not by
calling Box.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Chapter 7 : Language Features 225
Win DOS Mac
Draw.Cls
X TOOT Term
Syntax
Draw.Cls
Description
The Draw.Cls (clear screen) procedure is used to blank the screen. The
cursor is set to the top left (to row 1, column 1).
Details
In "graphics" mode all pixels are set to color number 0, so the screen is
displayed in background color. In screen mode, the screen is set to the text
background color.
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See View.Set
for details.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Cls, not by
calling Cls.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Win DOS Mac
Draw.Dot
X TOOT Term
Syntax
Draw.Dot ( x, y, Color : int )
Description
The Draw.Dot procedure is used to color the dot (pixel) at location (x, y)
using the specified Color.
226 Object Oriented Turing Reference Manual
Example
This program randomly draws dots with random colors. The maxx, maxy
and maxcolor functions give the maximum x, y and color values.
View.Set ( "graphics" )
var x, y, c : int
loop
x := Rand.Int (0, maxx)
y := Rand.Int (0, maxy)
clr := Rand.Int (0, maxcolor)
Draw.Dot ( x, y, c )
% Random x
% Random y
% Random color
end loop
Details
The screen should be in a "graphics" mode. If the screen is not in a "graphics"
mode, it will automatically be set to "graphics" mode. See View.Set for
details.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Dot, not by
calling Dot.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Win DOS Mac
Draw.Fill
Syntax
X TOOT Term
Draw.Fill ( x, y : int, fillColor, borderColor : int )
Chapter 7 : Language Features 227
Description
The Draw.Fill procedure is used to color in a figure that is on the screen.
Starting at (x, y), the figure is filled with fillColor to a surrounding border
whose color is borderColor.
Example
This program draws an oval with x and y radius of 10 in the center of the
screen using color 1. Then the oval is filled with color 2. The maxx and
maxy functions are used to determine the maximum x and y values on the
screen.
View.Set ( "graphics" )
const midx := maxx div 2
const midy := maxy div 2
drawoval ( midx, midy, 10, 10, 1 )
Draw.Fill ( midx, midy, 2, 1 )
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Warning: In Turing for IBM PC compatibles, Draw.Fill fails to completely
fill in some complicated figures that contain "islands" within them
surrounded by the borderColor.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Fill, not by
calling Fill.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Win DOS Mac
Draw.FillArc
X TOOT Term
Syntax
Draw.FillArc ( x, y, xRadius, yRadius : int,
initialAngle, finalAngle, Color : int )
Description
The Draw.FillArc procedure is used to draw a filled arc whose center is at
(x, y). It then fills in the pie-shaped wedge using the specified Color. To
outline a filled arc, use Draw.FillArc with the Color parameter set to the fill
color and then Draw.Arc with the Color parameter set to the border color.
For initialAngle and finalAngle, which determine the edges of the wedge,
zero degrees is "three o'clock" and 90 degrees is "twelve o'clock", etc. The
horizontal and vertical distances from the center to the arc are given by
xRadius and yRadius.
228 Object Oriented Turing Reference Manual
This program draws a filled semicircle (actually, an approximation to a
semicircle) whose center is (midx,0) the bottom center of the screen, using
color number 1. The maxx and maxy functions are used to determine the
maximum x and y values on the screen.
Example
View.Set ( "graphics" )
const midx := maxx div 2
Draw.FillArc ( midx, 0, maxy, maxy, 0, 180, 1 )
On the PC, Draw.FillArc fills the pie-shaped wedge by using a "flood" fill
and is thus subject to all the conditions of a flood fill.
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Exported qualified.
Status
This means that you can only call the function by calling Draw.FillArc, not
by calling FillArc.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.FillBox
X TOOT Term
Syntax
Draw.FillBox ( x1, y1, x2, y2, Color : int )
Description
The Draw.FillBox procedure is used to draw a filled box on the screen with
bottom left and top right corners of (x1, y1) to (x2, y2) filled using the
specified Color. To get a box outlined in a different color, use Draw.FillBox
with the Color parameter set to the fill color and then call Draw.Box with
the Color parameter set to the border color.
Example
This program will fill the bottom half of the screen with color 1 and then
outline it in color 2. The maxx and maxy functions are used to determine
the maximum x and y values on the screen. The point (0,0) is the left
bottom of the screen and (maxx, maxy) is the right top.
View.Set ( "graphics" )
Draw.FillBox ( 0, 0, maxx, maxy div 2, 1 )
Draw.Box ( 0, 0, maxx, maxy div 2, 2 )
Chapter 7 : Language Features 229
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.FillBox, not
by calling FillBox.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.FillMapleLeaf
X TOOT Term
Syntax
Draw.FillMapleLeaf ( x1, y1, x2, y2, Color : int )
Description
The Draw.FillMapleLeaf procedure is used to draw a filled maple leaf on
the screen bounded by a rectangle with bottom left and top right corners of
(x1, y1) to (x2, y2) and filled using the specified Color. To get a maple leaf
outlined in a different color, use Draw.FillMapleLeaf with the Color
parameter set to the fill color and then call Draw.MapleLeaf with the Color
parameter set to the border color. If y1 is greater than y2, then the
mapleleaf is drawn upside down.
(x2, y2)
(x2, y2)
(x1, y1)
Example
(x1, y1)
This program will draw two maple leaves beside each other. The first will
be outlined in color 1 and filled in color 2. The second maple leaf will be
upside down and both filled and outlined in color 3.
View.Set ( "graphics" )
Draw.FillMapleLeaf ( 0, 0, 100, 100, 1 )
Draw.MapleLeaf ( 0, 0, 100, 100, 2 )
Draw.FillMapleLeaf ( 150, 100, 250, 0, 3 )
230 Object Oriented Turing Reference Manual
The Draw.FillMapleLeaf procedure is useful for drawing the Canadian
flag.
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Exported qualified.
Status
This means that you can only call the function by calling
Draw.FillMapleLeaf, not by calling FillMapleLeaf.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.FillOval
X TOOT Term
Syntax
Draw.FillOval ( x, y, xRadius, yRadius, Color : int )
Description
The Draw.FillOval procedure is used to draw a filled oval whose center is
at (x, y). The horizontal and vertical distances from the center to the oval
are given by xRadius and yRadius. To get an oval outlined in a different
color, use Draw.FillOval with the Color parameter set to the fill color and
then call Draw.Oval with the Color parameter set to the border color.
Example
This program draws a large filled oval that just touches each edge of the
screen using color number 1. The maxx and maxy functions are used to
determine the maximum x and y values on the screen. The center of the
oval is at (midx, midy), which is the middle of the screen.
View.Set ( "graphics" )
const midx := maxx div 2
const midy := maxy div 2
Draw.FillOval ( midx, midy, midx, midy, 1 )
Details
Ideally, a circle is drawn when xRadius = yRadius. In fact, the aspect ratio
(the ratio of height to width of pixels displayed on the screen) of the IBM
PC compatibles is not 1.0, so this does not draw a true circle. In CGA
graphics mode this ratio is 5 to 4.
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Chapter 7 : Language Features 231
Exported qualified.
Status
This means that you can only call the function by calling Draw.FillOval,
not by calling FillOval.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.FillPolygon
X TOOT Term
Syntax
Draw.FillPolygon ( x, y : array 1 .. * of int, n : int, Color : int )
Description
The Draw.FillPolygon procedure is used to draw a filled polygon with n
points. The polygon is described by the points (x(1), y(1)) to (x(2), y(2)) to
(x(3), y(3)) and so on to (x(n), y (n)). The polygon will be drawn and filled
with Color.
To get an polygon outlined in a different color, use Draw.FillPolygon with
the Color parameter set to the fill color and then call Draw.Polygon with
the Color parameter set to the border color.
Example
This program will create a filled octagon and display it in color 1 and then
outline it in color 3.
View.Set ( "graphics" )
var x : array 1..8 of int := init (100, 100, 135, 185,
220, 220, 185, 135)
var y : array 1..8 of int := init (100, 150, 185, 185,
150, 100, 65, 65)
Draw.FillPolygon (x, y, 8, 1)
Draw.Polygon (x, y, 8, 3)
Details
The PC allows a maximum of 256 points. As well, Draw.FillPolygon can
fail (due to lack of memory). If failure occurs, it will try to draw an outline
of the polygon. If that also fails, it will not draw anything.
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling
Draw.FillPolygon, not by calling FillPolygon.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
232 Object Oriented Turing Reference Manual
Win DOS Mac
Draw.FillStar
X TOOT Term
Syntax
Draw.FillStar ( x1, y1, x2, y2, Color : int )
Description
The Draw.FillStar procedure is used to draw a filled five pointed star on
the screen bounded by a rectangle with bottom left and top right corners of
(x1, y1) to (x2, y2) and filled using the specified Color. To get a star outlined
in a different color, use Draw.FillStar with the Color parameter set to the
fill color and then call Draw.Star with the Color parameter set to the border
color. If y1 is greater than y2, then the star is drawn upside down.
(x2, y2)
(x2, y2)
(x1, y1)
Example
(x1, y1)
This program will draw two stars beside each other. The first will be
outlined in color 1 and filled in color 2. The second star will be upside
down and both filled and outlined in color 3.
View.Set ( "graphics" )
Draw.FillStar ( 0, 0, 100, 100, 1 )
Draw.Star ( 0, 0, 100, 100, 2 )
Draw.FillStar( 150, 100, 250, 0, 3 )
Details
The Draw.FillStar procedure is useful for drawing the American flag.
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.FillStar, not
by calling FillStar.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Chapter 7 : Language Features 233
Win DOS Mac
Draw.Line
X TOOT Term
Syntax
Draw.Line ( x1, y1, x2, y2, Color : int )
Description
The Draw.Line procedure is used to draw a line on the screen from (x1, y1)
to (x2, y2) using the specified Color.
Example
This program draws a large X, reaching to each corner of the screen using
color number 1. The maxx and maxy functions are used to determine the
maximum x and y values on the screen. The point (0,0) is the left bottom of
the screen, (maxx, maxy) is the right top, etc.
View.Set ( "graphics" )
% First draw a line from the left bottom to right top
Draw.Line ( 0, 0, maxx, maxy, 1)
% Now draw a line from the left top to right bottom
Draw.Line ( 0, maxy, maxx, 0, 1 )
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Line, not by
calling Line.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Win DOS Mac
Draw.MapleLeaf
X TOOT Term
Syntax
Draw.MapleLeaf ( x1, y1, x2, y2, Color : int )
Description
The Draw.MapleLeaf procedure is used to draw a maple leaf on the screen
bounded by a rectangle described by the bottom left and top right corners
of (x1, y1) to (x2, y2) using the specified Color. If y1 is greater than y2, then
the maple leaf is drawn upside down.
234 Object Oriented Turing Reference Manual
(x2, y2)
(x1, y1)
(x2, y2)
(x1, y1)
This program will draw two maple leaves beside each other. The first will
be in color 1 and the second maple leaf will be upside down and in color 2.
Example
View.Set ( "graphics" )
Draw.MapleLeaf ( 0, 0, 100, 100, 1 )
Draw.MapleLeaf ( 150, 100, 250, 0, 2 )
Details
The Draw.MapleLeaf procedure is useful for drawing the Canadian flag.
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.MapleLeaf,
not by calling MapleLeaf.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Win DOS Mac
Draw.Oval
X TOOT Term
Syntax
Draw.Oval ( x, y, xRadius, yRadius, Color : int )
Description
The Draw.Oval procedure is used to draw an oval whose center is at (x, y).
The horizontal and vertical distances from the center to the oval are given
by xRadius and yRadius.
Chapter 7 : Language Features 235
This program draws a large oval that just touches each edge of the screen
using color number 1. The maxx and maxy functions are used to determine
the maximum x and y values on the screen. The center of the oval is at
(midx, midy), which is the middle of the screen.
Example
View.Set ( "graphics" )
const midx := maxx div 2
const midy := maxy div 2
Draw.Oval ( midx, midy, midx, midy, 1 )
Ideally, a circle is drawn when xRadius = yRadius. In fact, the aspect ratio
(the ratio of height to width of pixels displayed on the screen) of the IBM
PC compatibles is not 1.0, so this does not draw a true circle. In CGA
graphics mode this ratio is 5 to 4.
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Exported qualified.
Status
This means that you can only call the function by calling Draw.Oval, not
by calling Oval.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.Polygon
X TOOT Term
Syntax
Draw.Polygon ( x, y : array 1 .. * of int, n : int, Color : int )
Description
The Draw.Polygon procedure is used to draw a polygon with n points. A
line is drawn in Color from the point (x(1), y(1)) to (x(2), y(2)) to (x(3), y(3))
and so on. After drawing the line to (x(n), y (n)), a line will be drawn back
to (x(1), y(1)), closing the polygon. The Draw.Polygon procedure is
equivalent to:
for i : 1 .. n - 1
Draw.Line (x (i), y(i), x (i + 1), y (i + 1), Color)
end for
Draw.Line (x (n), y (n), x (1), y (1), Color)
Example
This program will create an octagon and display it in color 1.
View.Set ( "graphics" )
var x : array 1..8 of int := init (100, 100, 135, 185,
220, 220, 185, 135)
var y : array 1..8 of int := init (100, 150, 185, 185,
236 Object Oriented Turing Reference Manual
150, 100, 65, 65)
Draw.Polygon (x, y, 8, 1)
The IBM PC limits Draw.Polygon to a maximum of 256 points.
Details
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Exported qualified.
Status
This means that you can only call the function by calling Draw.Polygon,
not by calling Polygon.
View.Set, maxx, maxy and the various procedures in the Draw unit.
See also
Win DOS Mac
Draw.Star
X TOOT Term
Syntax
Draw.Star ( x1, y1, x2, y2, Color : int )
Description
The Draw.Star procedure is used to draw a star on the screen bounded by
a rectangle described by the bottom left and top right corners of (x1, y1) to
(x2, y2) using the specified Color. If y1 is greater than y2 then the star is
drawn upside down.
(x2, y2)
(x2, y2)
(x1, y1)
Example
(x1, y1)
This program will draw two stars beside each other. The first star will be in
color 1 and the second star will be upside down and in color 2.
View.Set ( "graphics" )
Draw.Star ( 0, 0, 100, 100, 1 )
Draw.Star ( 150, 100, 250, 0, 2 )
Chapter 7 : Language Features 237
Details
The Draw.Star procedure is useful for drawing the American flag.
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Draw.Star, not by
calling Star.
See also
View.Set, maxx, maxy and the various procedures in the Draw unit.
Win DOS Mac
Draw.Text
X TOOT Term
Syntax
Draw.Text (txtStr : string, x, y, fontID, Color : int)
Description
Draw.Text is used to actually draw text in a specified font. The textStr
parameter contains the string to be drawn. The x and y paramters are the
location of the lower left hand corner of the text to be displayed. The fontID
parameter is the number of the font in which the text is to be drawn. The
Color parameter is used to specify the color in which the text is to appear.
Note that the text that appears is completely unrelated to the text that
appears using put. Draw.Text is a graphics command and thus does not
use or affect the cursor location.
The text drawn by the Draw.Text procedure does not erase the
background.
Details
If Draw.Text is passed an invalid font ID, a fatal error occurs. If the
Draw.Text call fails for other (non-fatal) reasons then Error.Last will return
a non-zero value indicating the reason for the failure. Error.LastMsg will
return a string which contains the textual version of the error.
Details
Draw.Text is identical to Font.Draw. It is placed here for consistency with
other pixel graphics drawing routines.
238 Object Oriented Turing Reference Manual
The program draws a phrase in red surrounded by a box in bright blue.
Example
var font : int
font := Font.New ("serif:12")
assert font1 > 0
var width : int:= Font.Width ("This is in a serif font", font)
var height, ascent, descent, internalLeading : int
Font.Sizes (font, height, ascent, descent, internalLeading)
Draw.Text ("This is in a serif font", 50, 30, font, red)
Draw.Box (50, 30 + descent, 50 + width, 30 + height, brightblue)
Font.Free (font)
Status
Exported qualified.
This means that you can only call the function by calling Draw.Text, not by
calling Text.
See Also
Font module for more information about selecting the font to be displayed.
drawpic
graphics procedure
Pixel graphics only
Syntax
drawpic ( x, y : int, buffer : array 1 .. * of int, picmode : int )
Description
The drawpic procedure is used to copy of a rectangular picture onto the
screen. The left bottom of the picture is placed at (x, y). In the common
case, the buffer was initialized by calling takepic. The values of picmode
are:
0: Copy actual picture on screen.
1: Copy picture by XORing it onto the screen.
XORing a picture onto the screen twice leaves the screen as it was (this is a
convenient way to move images for animation). XORing a picture onto a
background effectively superimposes the picture onto the background.
Details
See takepic for an example of the use of drawpic and for further
information about buffers for drawing pictures.
The screen should be in a "graphics" mode. See the setscreen procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
See also
takepic and sizepic. See also setscreen, maxx, maxy and the various
draw… procedures.
See also predefined unit Draw and Pic.
Chapter 7 : Language Features 239
empty
condition function
Syntax
empty ( variableReference ) : boolean
Description
The empty function is used in a concurrent program. It returns true if the
variableReference, which must be a condition variable, has no processes
waiting for it. Processes join the queue of a condition variable by executing
the wait statement, and are awakened by the signal statement.
See also
condition, wait, signal, fork and monitor.
See also predefined unit Concurrency.
enum
enumerated type
An enumeratedType is:
Syntax
enum ( id { , id } )
Description
The values of an enumerated type are distinct and increasing. They can be
thought of as the values 0, 1, 2 and so on, but arithmetic is not allowed with
these values.
Example
type color : enum ( red, green, blue )
var c : color := color . red
var d : color := succ ( c )
Details
% d becomes green
Each value of an enumerated type is the name of the type followed by a
dot followed by the element's name, for example, color.red. Enumerated
values can be compared for equality and for ordering. The succ and pred
functions can be used to find the value following or preceding a given
enumerated value. The ord function can be used to find the enumeration
position of a value, for example, ord (color.red) is 0.
Enumerated types cannot be combined with integers or with other
enumerated types.
240 Object Oriented Turing Reference Manual
It is illegal to declare an "anonymous" enum. The only legal declaration for
an enum is in a type declaration. For example, the following is now illegal:
Details
var a : array enum (red, green, blue ) of int
Given that there is no (easy) way of generating an enum value without it
being a named type, this should not impact any but the most bizarre code.
The "put" and "get" statement semantics have been expanded to allow put's
and get's of enum values. The values printed and input are the element
names themselves, case sensitive. For example, for
Details
type colors : enum ( red, green, blue )
var c : colors := colors . red
put c
% outputs "red" (without the quotes)
enumeratedValue
enumerated value
An enumeratedValue is:
Syntax
enumeratedTypeId . enumeratedId
Description
The values of an enumerated type are written as the type name
(enumeratedTypeId) followed by a dot followed by one of the enumerated
values of the type (enumeratedId).
Example
In this example, color.red is an enumeratedValue.
type color : enum ( red, green, blue )
var c : color := color . red
var d : color := succ ( c )
% d becomes green
Details
The above description has been simplified by ignoring the possibility that
the enum type can be exported from a module. If this is the case, each use
of one of the enumerated values outside of module M must be preceded by
the module name and a dot, as in M.color.red.
See also
the enum type and explicitConstant.
Chapter 7 : Language Features 241
eof
end-of-file function
Syntax
eof ( streamNumber : int ) : boolean
Description
The eof (end of file) function is used to determine if there is any more
input. It returns true when there are no more characters to be read. The
parameter and its parentheses are omitted when referring to the standard
input (usually this is the keyboard); otherwise the parameter specifies the
number of a stream. The stream number has been determined (in most
cases) by an open statement.
Example
This program reads and outputs all the lines in the file called "info".
var line : string
var fileNumber : int
open : fileNumber, "info", get
loop
exit when eof (fileNumber)
get : fileNumber, line : *
put line
end loop
Details
See also the description of the get statement, which gives more examples of
the use of eof. See also the open and read statements.
When the input is from the keyboard, the user can signal end-of-file by
typing control-Z on a PC (or control-D on UNIX). If a program tests for eof
on the keyboard, and the user has not typed control-Z (or control-D) and
the user has typed no characters beyond those that have been read, the
program must wait until the next character is typed. Once this character is
typed, the program knows whether it is at the end of the input, and returns
the corresponding true or false value for eof.
242 Object Oriented Turing Reference Manual
equivalence
Description
of types
Two types are equivalent to each other if they are essentially the same types
(the exact rules are given below). When a variable is passed to a var formal
parameter, the types of the variable and the formal parameter must be
equivalent because they are effectively the same variable. When an
expression is assigned to a variable, their types must be equivalent, except
for special cases. For example, Turing allows you to assign an integer
expression to a real variable (see assignability for details).
Example
var j : int
var b : array 1 .. 25 of string
type personType :
record
age : int
name : string ( 20 )
end record
procedure p ( var i : int, var a : array 1 .. 25 of string,
var r : personType )
… body of procedure p, which modifies each of i, a and r …
end p
var s : personType
% Procedure call to p
p ( j, b, s )
% i and j have the equivalent type int
% Arrays a and b have equivalent types
% Records r and s have equivalent types
Details
Two types are defined to be equivalent if they are:
(a) the same standard type (int, real, boolean or string),
(b) subranges with equal first and last values,
(c) arrays with equivalent index types and equivalent component types,
(d) strings with equal maximum lengths,
(e) sets with equivalent base types, or
(f) pointers to the same collection; in addition,
(g) a declared type identifier is also equivalent to the type it names (and to
the type named by that type, if that type is a named type, etc.)
(h) both char,
(i)
both char(n) with the same length,
Chapter 7 : Language Features 243
(j) both procedure types, with corresponding equivalent parameter types
and corresponding var or non-var of the parameters,
(k) both function types, with corresponding equivalent parameter types
and corresponding var or non-var of the parameters and equivalent
result types,
(l) both pointer types to the same class or equivalent type and both are
checked or unchecked.
Each separate instance of a record, union or enumerated type (written out
using one of the keywords record, union or enum) creates a distinct type,
equivalent to no other type. By contrast, separate instances of arrays,
strings, subranges and sets are considered equivalent if their parts are
equal and equivalent.
Opaque type T, exported from a module, monitor or class M as opaque, is
a special case of equivalence. Outside of M this type is written M.T, and is
considered to be distinct from all other types. By contrast, if type U is
exported non-opaque, the usual rules of equivalence apply. The parameter
or result type of an exported procedure or function or an exported constant
is considered to have type M.T outside of M if the item is declared using
the type identifier T. Outside of M, the opaque type can be assigned, but
not compared.
It is not required that subprogram types have the same names and
parameter names to be equivalent. They also do not require the same
factoring of parameters across their types, as in i, j: int instead of i: int, j:
int.
erealstr
real-to-string function
Syntax
erealstr ( r : real,
width, fractionWidth, exponentWidth : int ) :string
Description
The erealstr function is used to convert a real number to a string; for
example, erealstr (2.5e1, 10, 3, 2)="b2.500e+01" where b represents a blank.
The string (including exponent) is an approximation to r, padded on the
left with blanks as necessary to a length of width.
The width must be a non-negative int value. If the width parameter is not
large enough to represent the value of r, it is implicitly increased as
needed.
244 Object Oriented Turing Reference Manual
The fractionWidth parameter is the non-negative number of fractional digits
to be displayed. The displayed value is rounded to the nearest decimal
equivalent with this accuracy. In the case of a tie, the value is rounded to
the larger of the two values.
The exponentWidth parameter must be non-negative and give the number
of exponent digits to be displayed. If exponentWidth is not large enough to
represent the exponent, more space is used as needed. The string returned
by erealstr is of the form:
{blank}[-]digit.{digit}e sign digit {digit}
where sign is a plus or minus sign. The leftmost digit is non-zero, unless all
the digits are zeros.
The erealstr function approximates the inverse of strreal, although roundoff errors keep these from being exact inverses.
See also
frealstr, realstr, strreal, intstr and strint functions.
Win DOS Mac
Error
Description
X TOOT Term
This unit contains the predefined subprograms that deal with errors
returned from other predefined subprograms.
All routines in the Error unit are exported qualified (and thus must be
prefaced with "Error.").
The constants representing the possible errors returned by this module can
be found in the ErrorNum module.
Entry Points
Last
Returns the (integer) error code produced by the last call to
a predefined subprogram.
LastMsg
Returns the error string produced by the last call to a
predefined subprogram.
LastStr
Returns the string version of the error constant produced
by the last call to a predefined subprogram.
Msg
Returns the string that corresponds to a specified error
code.
Str
Returns the string version of the error constant that
corresponds to a specified error code.
Trip
This causes execution to halt with returning the specified
error.
Chapter 7 : Language Features 245
Win DOS Mac
Error.Last
X TOOT Term
Syntax
Error.Last : int
Description
Error.Last is a function that returns the error code set by the last called
predefined subprogram. If there is no error, then it returns eNoError (which
is 0). If there is an error, you can use Error.LastMsg to obtain a textual form
of the error or Error.LastStr to obtain a string version of the error constant.
The fact that Error.Last is not eNoError does not necessarily mean that the
previous predefined function failed or failed completely. Error.Last also
returns a number of warning codes. For example, if a user specifies a
number larger than maxcolor for the color parameter of the Draw.Line
procedure, the line is still drawn, only in color maxcolor. However,
Error.Last will return a code that warns the user of the fact.
Example
This program creates the directory called information. If the creation fails, it
prints out the error number and an error message.
Dir.Create ("information")
if Error.Last = eNoError then
put "Directory created"
else
put "Did not create the directory."
put "Error Number: ", Error.Last
put "Error Message: ", Error.LastMsg
put "Error Constant: ", Error.LastStr
end if
Status
Exported qualified.
This means that you can only call the function by calling Error.Last, not by
calling Last.
Win DOS Mac
Error.LastMsg
Syntax
Error.LastMsg : string
246 Object Oriented Turing Reference Manual
X TOOT Term
Description
Error.LastMsg is a function that returns the error message set by the last
called predefined subprogram. If there is no error, then it returns the
empty string. If there is an error, you can use Error.Last to obtain the error
code.
The fact that Error.LastMsg is not "" does not necessarily mean that the
previous predefined function failed or failed completely. Error.LastMsg
also returns a number of warning messages. For example, if a user specifies
a number larger than maxcolor for the color parameter of the Draw.Line
procedure, the line is still drawn, only in color maxcolor. However,
Error.LastMsg will return a message that indicates that the color was out of
range
This program creates the directory called information. If the creation fails, it
prints out the error number and an error message.
Example
Dir.Create ("information")
if Error.Last = eNoError then
put "Directory created"
else
put "Did not create the directory."
put "Error Number: ", Error.Last
put "Error Message: ", Error.LastMsg
end if
Status
Exported qualified.
This means that you can only call the function by calling Error.LastMsg,
not by calling LastMsg.
Win DOS Mac
Error.LastStr
X TOOT Term
Syntax
Error.LastStr : string
Description
Error.LastStr is a function that returns the string version of the error code
set by the last called predefined subprogram (i.e. it would return the string
"eDrawClrNumTooLarge" for using a color greater than maxcolor in a
Draw command). If there is no error then it returns the empty string. If
there is an error, you can use Error.Last to obtain the actual error code.
Chapter 7 : Language Features 247
The fact that Error.LastStr is not "" does not necessarily mean that the
previous predefined function failed or failed completely. Error.LastStr also
returns a number of error codes for warning messages. For example, if a
user specifies a number larger than maxcolor for the color parameter of the
Draw.Line procedure, the line is still drawn, only in color maxcolor.
However, Error.LastStr will return a string version of the error code that
indicates that the color was out of range.
You can take a look at the error constants defined by looking at the unit
ErrorNum which contains all defined error codes.
This program creates the directory called information. If the creation fails, it
prints out the error number and an error message.
Example
Dir.Create ("information")
if Error.Last = eNoError then
put "Directory created"
else
put "Did not create the directory."
put "Error Number: ", Error.Last
put "Error Constant: ", Error.LastStr
end if
Status
Exported qualified.
This means that you can only call the function by calling Error.LastStr, not
by calling LastStr.
Win DOS Mac
Error.Msg
X TOOT Term
Syntax
Error.Msg (errorCode : int): string
Description
Error.Msg is a function that returns the error message related to a specified
error code. If the error code is eNoError, or if there is no such error code, it
returns the empty string. If there is such an error, it returns the textual
message associated with that error.
Example
This program prints out the error message associated with
eFsysFileNotFound (“File not found”).
put Error.Msg (eFsysFileNotFound)
Status
Exported qualified.
This means that you can only call the function by calling Error.Msg, not by
calling Msg.
248 Object Oriented Turing Reference Manual
Win DOS Mac
Error.Str
X TOOT Term
Syntax
Error.Str (errorCode : int): string
Description
Error.Str is a function that returns the error message related to a specified
error code. If the error code is eNoError or if there is no such error code, it
returns the empty string. If there is such an error, it returns the textual
message associated with that error.
Example
This program prints out the string "eFsysFileNotFound".
put Error.Str (eFsysFileNotFound)
Status
Exported qualified.
This means that you can only call the function by calling Error.Str, not by
calling Str.
Win DOS Mac
Error.Trip
X TOOT Term
Syntax
Error.Trip (errorCode : int)
Description
Error.Trip is a procedure that causes execution of the program to abort
immediately with an error message that corresponds to the error code
passed in. The program will abort even if the error code passed in is
normally a warning.
Error codes that do not correspond to recognized errors will cause an abort
with the error message “Unknown Error #n” where n is the error passed
in.
You can find a list of constants for the legal error codes in the module
ErrorNum. Any call to Error.Trip should use a constant found in the
ErrorNum module.
Example
This program aborts execution with the message “File not found”.
put Error.Trip (eFsysFileNotFound)
Status
Exported qualified.
Chapter 7 : Language Features 249
This means that you can only call the function by calling Error.Trip, not by
calling Trip.
ErrorNum
Description
All
This unit contains all the constants representing errors used by the Error
module.
All constants in the ErrorNum module are exported unqualified. (This
means you can use the constants directly without having to use the
qualifier "ErrorNum.".)
Exceptions
Description
All
This unit contains all the constants corresponding to exception numbers in
Turing for use in building exception handlers.
All constants in the Exceptions module are exported unqualified. (This
means you can use the constants directly without having to use the
qualifier "Exceptions.".)
exit
Syntax
statement
An exitStatement is one of:
(a)
(b)
exit when trueFalseExpn
exit
250 Object Oriented Turing Reference Manual
Description
An exit statement is used to stop the execution of a loop or for statement.
Form (a) is the most common. Here, the true/false expression is evaluated.
If it is true, the loop is terminated and execution jumps down and
continues just beyond end loop or end for. If it is false, the loop keeps on
repeating. Form (b) always causes the loop to terminate. This form is
almost always used inside another conditional statement such as if.
Example
Input names until finding Jones.
var name : string
loop
get name
exit when name = "Jones"
end loop
Exit statements must occur only inside loop or for statements. An exit takes
you out of the closest surrounding loop or for. The only other ways to
terminate a loop or for is by return (in a procedure or in the main program,
in which case the entire procedure or main program is terminated) or by
result (in a function, in which case the entire function is terminated and a
result value must be supplied).
Details
The form "exit when trueFalseExpn" is equivalent to "if trueFalseExpn then
exit end if".
exp
exponentiation function
Syntax
exp ( r : real ) : real
Description
The exp function is used to find e to the power r, where e is the natural
base and r is the parameter to exp. For example, exp (0) returns 1 and exp
(1) returns the value of e.
Example
This program prints out the exponential values of 1, 2, 3, … up to 100.
for i : 1 .. 100
put "Exponential of ", i, " is ", exp ( i )
end for
See also
ln (natural logarithm function).
See also predefined unit Math.
Chapter 7 : Language Features 251
explicitCharConstant
character literal
Syntax
An explicitCharConstant is a sequence of characters surrounded by single
quotation marks, for example, 'Renzo'.
Example
In the following, the explicit character constants are 'H' and 'Hi'.
var c : char := 'H'
Details
var d : char ( 2 ) := 'Hi'
An explicit character constant must contain at least one character. If it
contains exactly one character, as in 'A', its type is char. If it contains two or
more characters (n characters), as 'Width', its type is char(n). The difference
between the char and char(1) types is rarely of significance, but does make
a difference in declarations without an explicit type, for example:
var c := 'H'
% Type is char
var d := 'Hi' % Type is char ( 2 )
var e := "H" % Type is string
The backslash \ is used in explicit string and char(n) constants to specify
special values, for example, '\T' is the tab character. Similarly, the carat ^ is
used to specify ASCII control characters, for example, '^H' is the ASCII
backspace. See explicitStringConstants for details.
Explicit character constants cannot cross line boundaries. To represent a
constant that is longer than a line, break it into two or more strings on
separate lines and use + (catenation) to join the individual strings. See
catenation.
An explicit character constant may be limited in length by the
implementation. We recommend that this limitation be at least 32767.
Explicit character constants, but not strings, are allowed to contain the
character internal values 0 (called eos for end of string) and 128 (called
uninitchar, used as the uninitialized string value).
explicitConstant
Syntax
An explicitConstant is one of:
252 Object Oriented Turing Reference Manual
literal
(a)
(b)
(c)
(d)
(e)
explicitStringConstant % e.g.: "Hello world"
explicitIntegerConstant % e.g.: 25
explicitRealConstant
% e.g.: 51.8
explicitTrueFalseConstant % e.g.: true
explicitCharConstant
% e.g.: 'Hi'
Description
An explicitConstant gives its value directly. For example, the value of the
explicit constant 25 is twenty-five.
Example
In the following, the explicit constants are "Hello world", 3.14159 and 2.
Note that pi is a named constant rather than an explicit constant.
put "Hello world"
var diameter : real
const pi := 3.14159
diameter := pi * r ** 2
var x := diameter
Details
In some programming languages, explicit constants are called literals or
literal values, because they literally (explicitly) give their values.
See also
explicitStringConstant, explicitIntegerConstant, explicitRealConstant,
explicitTrueFalseConstant and explicitCharConstant. See also enumeratedValue.
explicitIntegerConstant
integer literal
Syntax
An explicitIntegerConstant is a sequence of one or more decimal digits (0 to
9) optionally preceded by a plus or minus sign. This is an alternate form
that specifies a number base (such as base 2 or base 16).
Example
In the following, the explicit integer constants are 0, 115 and 5.
var count : int := 0
const height := 115
…
count := height - 5
Details
In current implementations of Turing, the range of the int (integer) type is
from -2147483647 to 2147483647. In other words, the maximum size of
integer is 2**31 - 1. This is the range that fits into four bytes, with one
pattern left over (the largest negative 4-byte number) to represent the
uninitialized value. See maxint.
Chapter 7 : Language Features 253
Values can be written in base 2 or 16 or any other base in the range 2 to 36
(36 because there are 10 digits and 26 letters). This form begins with the
base, such as 16, then #, and then the value written in that base, for
example, 16#A has the value 10. The letters a, b, c … represent the digit
values 10, 11, 12 … Capital letters A, B, C … can be used instead of lower
case. Here are some examples.
2#1
=1
(Base 2)
2#11
=3
(Base 2)
16#a
= 10
(Base 16)
16#FF
= 255
(Base 16)
16#FFFF
= 32767 (Base 16)
8#10
=8
(Base 8)
Here is an example of using these:
const maxnat1 := 16#FF
% Largest 1-byte natural number
const maxint2 := 16#7FFF
% Largest 2-byte integer
You should be careful to avoid confusion about patterns such as 16#FFFF.
It is tempting to think that this is the value –1, because the bit pattern (2byte two's complement internal representation) for –1 is the same as the bit
pattern for 16#FFFF = 32767. However, the value (as opposed to the
internal representation) of –1 and 32767 are different.
See also
int, maxint (the largest integer value), nat (positive values only) and intn
(n-byte integers). See also intstr and natstr which convert integer and
natural number values to corresponding character strings in any base, for
example intstr (4, 0, 2) = "100".
explicitRealConstant
real literal
Syntax
An explicitRealConstant consists of an optional plus or minus sign, a
significant digits part, and an exponent part.
Example
In the following, the explicit real constants are 0.0 and 2.93e3.
var temperature : real := 0.0
const speed := 2.93e3
Details
% Value is 2,930.0
The significant digits part (or fractional part) of an explicit real constant
consists of a sequence of one or more digits (0 to 9) optionally containing a
decimal point (a period). The decimal point is allowed to follow the last
digit as in 16. or to precede the first digit, as in .25.
254 Object Oriented Turing Reference Manual
The exponent part consists of the letter e or E followed optionally by a plus
or minus sign followed by one or more digits. For example, in -9.837e-3 the
exponent part is e-3. The value of -9.837e-3 is -9.837 times 0.001.
If the significant figures part contains a decimal point, then the exponent
part is not required.
explicitStringConstant
string literal
Syntax
An explicitStringConstant is a sequence of characters surrounded by
quotation marks.
Example
In the following, the explicit string constants are "Hello world", "" and "273
O'Reilly Ave.".
var name : string := "Hello world"
name := ""
% Null string, containing zero characters
var address : string := "273 O'Reilly Ave."
Details
Within an explicit string constant (and within an explicit character
constant), the back slash \ is used to represent certain other characters as
follows:
\"
quotation mark character
\n or \N
end of line character
\t or \Ttab character
\f or \Fform feed character
\r or \R
return character
\b or \B
backspace character
\e or \E
escape character
\d or \D
delete character
\\
backslash character
For example, put "One\nTwo" will output One on one line and Two on the
next. In an explicit character constant (which is surrounded by single
quotes, as in 'John'), the backslash is not required before a double quote ",
but it is required before a single quote ', as in these two constants:
'John said "Hello" to you'
'Don\'t cry'.
You can use the caret ^ to specify ASCII control characters, for example:
Chapter 7 : Language Features 255
'^H'
ASCII backspace character
The caret specifies that the top three bits of the character are set to zero. For
any character c, the following is true:
'^c' = chr (ord ('c') & 2#11111)
However if c is the question mark, as in '^?', the bits are not turned off.
Explicit string constants cannot cross line boundaries. To represent a string
that is longer than a line, break it into two or more strings on separate lines
and use catenation (+) to join the individual strings.
An explicit string constant can contain at most 255 characters (this is in implementation
constraint).
String values are not allowed to contain characters with the code values of
0 or 128; these character values are called eos (end of string) and uninitchar
(uninitialized character). These are reserved by the implementation to
mark the end of a string value and to see if a string variable has been
initialized.
explicitTrueFalseConstant
Syntax
An explicitTrueFalseConstant is one of:
(a)
(b)
Example
boolean literal
true
false
The following determines if string s contains a period. After the for
statement, found will be true if there is a period in s.
var found : boolean := false
for i : 1 .. length ( s )
if s = "." then
found := true
end if
end for
Details
true/false values are called boolean values. A boolean variable, such as
found in the above example, can have a value of either true or false.
See also
boolean type.
256 Object Oriented Turing Reference Manual
expn
expression
An expn is one of:
Syntax
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
Description
explicitConstant
% e.g.: 25
variableReference
% e.g.: width
constantReference
% e.g.: pi
expn infixOperator expn % e.g.: 3 + width
prefixOperator expn
% e.g.: - width
( expn )
% e.g.: ( width - 7)
substring
% e.g.: s (3 .. 5)
functionCall
% e.g.: sqrt ( 25 )
setConstructor
% e.g.: modes (4, 3)
enumeratedValue
% e.g.: color . red
An expression (expn) returns a value; in the general case, this may involve
a calculation, such as addition, as in the expression:
3 + width
Example
put "Hello world"
var diameter : real
const pi := 3.14159
diameter := pi * r ** 2
var x := diameter
Details
% "Hello world" is an expn
% 3.14159 is an expn
% pi * r ** 2 is an expn
% diameter is an expn
In the simplest case, an expression (expn) is simply an explicit constant
such as 25 or "Hello world". A variable by itself is considered to be an
expression when its value is used. This is the case above, where the value
of diameter is used to initialize x. More generally, an expression contains an
operator such as + and carries out an actual calculation. An expression may
also be a substring, function call, set constructor or enumerated value. For
details, see the descriptions of these items.
The Turing infix operators are: +, -, *, /, div, mod, **, <, >, =, <=, >=, not=,
not, and, or, =>, in, not in, shr (shift right), shl (shift left), and xor
(exclusive or). For details, see infixOperator. The Turing prefix operators are
+, - and not, ^ (pointer following) and # (see cheat). For details see prefix
operator.
See also
precedence of operators, as well as the int, real, string and boolean types.
Chapter 7 : Language Features 257
export
list
An exportList is:
Syntax
export [ howExport ] id {, [ howExport ] id }
Description
An export list is used to specify those items declared in a module, monitor
or class that can be used outside of it. Items that are declared inside a
module, monitor or class but not exported cannot be accessed outside of it.
Example
In this example, the procedures names pop and push are exported from the
stack module. These two procedures are called from outside the module on
the last and third from last lines of the example. Notice that the word stack
and a dot must precede the use of these names. Since top and contents were
not exported, they can be accessed only from inside the module.
module stack
export push, pop
var top : int := 0
var contents : array 1..100 of string
procedure push … end push
procedure pop … end pop
end stack
stack . push ( "Harvey" )
var name : string
stack . pop ( name )
Details
% This sets name to Harvey
Procedures, functions, variables, constants and types can be exported.
Modules, monitors or classes canot be exported. Parentheses are allowed
around the items in an export list, as in:
export ( push, pop )
The following syntax specifies that each exported identifier can optionally
be preceded by the keywords var, unqualified, pervasive and opaque. Of
these, only opaque is available in Turing proper.
The form of howExport is:
{ exportMethod }
The form of exportMethod is one of:
(a)
var
(b)
unqualified
(c)
pervasive
(d)
opaque
258 Object Oriented Turing Reference Manual
The keyword var means that the exported variable can be changed outside of the exporting
module, monitor or class. This keyword applies only to exported variables.
For example, if string variable name is exported var from module M, name
can be changed from outside of M by M.name := "Surprise!".
The keyword unqualified means that references to the exported item do
not need to be prefixed by the name of the exporting item. For example, if
module M exports procedure p unqualified, a call to p outside of M can be
simply p instead of the usual M.p. A class cannot export variables or
dynamic constants unqualified (because each object of the class has its own
copies of these). The only things a class can export unqualified are types
and compile time constants. The keyword unqualified can be abbreviated
to ~. which is pronounced as "not dot".
The keyword pervasive, which is only meaningful if unqualified is also
present, specifies that the exported item is to be visible in subsequent
scopes, in other words that it is not necessary to import it into internal
modules, monitors and classes.
The keyword opaque, which can only precede type names, specifies that
outside the module, monitor or class, the type is considered to be distinct
from all other types. This means, for example, that if the type is an array, it
cannot be subscripted outside of the module. See module declaration for
an example that uses opaque types. In most cases, classes are preferable to
opaque types.
Exported subprograms are considered to be deferred, meaning that
expansions are allowed to override these subprograms. See also deferred
subprograms. These can be overridden using the keyword body before the
resolving subprogram body.
A class cannot export items from its parent or it parent's ancestors. All
exported item must be declared in the current class.
Details
You can export all from a module, monitor or a class. This means that
every sibmle that is legal to export is exported. You may also qualify the
all, as in export opaque unqualified pervasive all where the qualifiers are
added to each export item (if it makes sense).
If all is specified as the export item, no other item may be specified. Also,
and all export affects only the module, monitor or class that it is given in.
Any inheriting or implementing module, monitor or class does not export
all unless they also specify it.
See also
unit, module, monitor and class. See also import list, inherit clause,
implement clause, implement by clause and deferred subprogram.
Chapter 7 : Language Features 259
external
declaration
Dangerous
An externalDeclaration is one of:
Syntax
(a) external [ overrideName ] subprogramHeader
(b) external [ addressSpec ] var id [ : typeSpec ] [ :=expn ]
Description
An external declaration is used to access variables or subprograms that are
written in other languages or which require special linkage. This feature is
implementation-dependent and dangerous and may cause arbitrary data
or program corruption. From an interpretive environment such as Turing,
this provides linkage to items that are part of the Turing system. For
compiled versions of Turing, the linkage would be by means of a standard,
operating system-specific linkage editor.
Details
In form (a) the optional overrideName must be an explicit string constant,
such as "printf". If it is omitted, the external name is the name in the
subprogramHeader. See subprogramHeader.
The current implementation does not support form (b). This form is
documented here in case a future version supports it. The addressSpec is a
compile time expression (its value must fit in the range of the addressint
type) or is a compile time string value. If the addressSpec is omitted, the
identifier is the name of an external variable. This name represents an
implementation-dependent method of locating a variable. At least one of
typeSpec or expn must be present.
Declaring variables at absolute addresses is useful for device management
in computer architectures with memory mapped device registers. External
variables declared to be int or nat will by default be checked for
initialization. To avoid this check, declare them to be int4 or nat4.
Example
Place variable ttyData at hexadecimal location 9001 and assign it the
character A.
external 16#9001 var ttyData : char
ttyData := 'A'
Example
% Character A is assigned to hex location 9001
Access an external integer variable named ERRFLAG.
external var ERRFLAG : int
if ERRFLAG = 0 then …
Example
Access an integer variable which is called y in this program but is called x
externally.
external "x" var y : int
260 Object Oriented Turing Reference Manual
Declare drawcircle to be a procedure that is externally known as circle.
Example
external "circle" procedure drawcircle ( x, y, r, Color : int )
false
boolean value (not true)
Syntax
false
Description
A boolean (true/false) variable can be either true or false (see boolean
type).
Example
var found : boolean := false
var word : int
for i : 1 .. 10
get word
found := found or word = "gold"
end for
if found = true then
put "Found 'gold' in the ten words"
end if
The line if found=true then can be simplified to if found then with no
change to the meaning of the program.
Details
fetcharg
fetch argument function
Syntax
fetcharg ( i : int ) : string
Description
The fetcharg function is used to access the i-th argument that has been
passed to a program from the command line. For example, if the program
is run from the Turing environment using
:r file1 file2
then fetcharg(2) will return "file2". If a program called prog.x is run under
UNIX using this command:
prog.x file1 file2
Chapter 7 : Language Features 261
the value of fetcharg(2) will similarly be "file2".
The nargs function, which gives the number of arguments passed to the
program, is usually used together with the fetcharg function. Parameter i
passed to fetcharg must be in the range 0 .. nargs.
The 0-th argument is the name of the running program.
This program lists its own name and its arguments.
Example
put "The name of this program is : ", fetcharg ( 0 )
for i : 1 .. nargs
put "Argument ", i, " is ", fetcharg ( i )
end for
nargs
See also
Win DOS Mac
File
Description
X TOOT Term
This unit contains the predefined subprograms that deal with file
manipulation on a whole-file basis (as opposed to manipulating the data in
the file using open and close, etc.). These routines allow you to rename,
copy and delete files, as well as get information about a file and get the free
space on disk available for a file.
All routines in the File module are exported qualified (and thus must be
prefaced with "File.").
Entry Points
Details
Exists
Returns whether a file exists.
Status
Gets information about a file such as size, modification
date, etc.
Copy
Copies a file to another location.
Rename
Renames a file or directory.
Delete
Deletes a file.
DiskFree
Gets the free space on the disk upon which a file or
directory resides.
On the PC, a path name of a file or a directory can use either the forward
slash or backward slash to separate directory names. The drive must be
followed by a colon. Thus the following are legal path names:
x:\students\west\example.t
c:/turing/test.t
262 Object Oriented Turing Reference Manual
/west/binary.t (uses the default drive).
On the Macintosh, a path name of a file or directory can use the standard
Macintosh format of Volume Name:Directory Name:Directory Name:File
Name or the Unix format of /Volume Name/Directory Name/Directory
Name/File Name. Note that the names can have spaces in them.
HSA:Applications:Turing Files:example.t
/HSA/Applications/Turing Files/example.t
On UNIX systems, the path name must correspond to the UNIX standard
of using a forward slash between parts of the path.
/export/home/west/turing/example.t
In general, you can achieve the greatest portability by using the UNIX
standard for use in path names, as all Turing systems support it.
Win DOS Mac
File.Copy
X TOOT Term
Syntax
File.Copy (srcPathName, destPathName : string)
Description
File.Copy copies a file named by the srcPathName parameter to the file
named by the destPathName parameter. The copy can be between different
disks or file systems.
Details
The source file name must be an actual file. This procedure will not copy
directories.
If the File.Copy call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program copies the file "/usr/west/example" to "/usr/holt/testcase".
File.Copy ("/usr/west/example", "/usr/holt/testcase")
if Error.Last = eNoError then
put "File copied"
else
put "Did not copy the file."
put "Error: ", Error.LastMsg
end if
Status
Exported qualified.
Chapter 7 : Language Features 263
This means that you can only call the function by calling File.Copy, not by
calling Copy.
Win DOS Mac
File.Delete
X TOOT Term
Syntax
File.Delete (filePathName : string)
Description
File.Delete is used to delete the file specified by the parameter
filePathName. This is the equivalent of doing a del in DOS or rm in UNIX.
Details
If the File.Delete call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
This program deletes the file called information.
Example
File.Delete ("information")
if Error.Last = eNoError then
put "File delete"
else
put "Did not delete the file."
put "Error: ", Error.LastMsg
end if
Status
Exported qualified.
This means that you can only call the function by calling File.Delete, not
by calling File.
Win DOS Mac
File.DiskFree
X TOOT Term
Syntax
File.DiskFree (pathName : string) : int
Description
File.DiskFree gets the number of bytes for the disk upon which pathName
resides. The pathName parameter can specify either a file or a directory. If it
is the empty string, then File.DiskFree returns the number of bytes of free
disk space on the disk upon which the execution directory resides.
264 Object Oriented Turing Reference Manual
Details
If the File.DiskFree call fails, then it returns –1. Also Error.Last will return
a non-zero value indicating the reason for the failure. Error.LastMsg will
return a string which contains the textual version of the error.
Example
This program prints out the amount of space on the A: drive on a PC and in
the execution directory.
var bytesFree : int
bytesFree := File.DiskFree ("A:\\")
if bytesFree = –1 then
put "Can't get free space on drive A:."
put "Error: ", Error.LastMsg
else
put "There are ", bytesFree , " bytes free on drive A:"
end if
bytesFree := File.DiskFree ("")
if bytesFree = –1 then
put "Can't get free space on default directory."
put "Error: ", Error.LastMsg
else
put "There are ", bytesFree , " bytes free on the default dir"
end if
Exported qualified.
Status
This means that you can only call the function by calling File.DiskFree, not
by calling DiskFree.
Win DOS Mac
File.Exists
X TOOT Term
Syntax
File.Exists (pathName : string) : boolean
Description
File.Exists returns true if a file by the name of pathName exists. It will
return true if pathName is a file or a directory.
Details
If the File.Exists returns false, you can examine Error.Last for more
information (i.e. whether the path failed or the file was simply not found).
Example
This program loops until the user types in a path name that either doesn’t
already exist or is allowed to be overwritten.
var pathName : string
var choice : string
loop
put "Enter file name to write results to" ..
get pathName
Chapter 7 : Language Features 265
if File.Exists (pathName) then
put "Overwrite ", pathName, "?" ..
get choice
exit when choice = "y"
else
exit
end if
end loop
Exported qualified.
Status
This means that you can only call the function by calling File.Exists, not by
calling Exists.
Win DOS Mac
File.Rename
X TOOT Term
Syntax
File.Rename (srcPathName, destName : string)
Description
File.Copy renames a file or directory named by the srcPathName parameter
to the destName parameter. The destName parameter must be a name only.
In other words File.Rename can't move a file between different directories.
Details
If the File.Rename call fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
This program renames the file "/usr/west/example" to "testcase"
File.Rename ("/usr/west/example", "testcase")
if Error.Last = eNoError then
put "File renamed"
else
put "Did not rename the file."
put "Error: ", Error.LastMsg
end if
Status
Exported qualified.
This means that you can only call the function by calling File.Rename, not
by calling Rename.
266 Object Oriented Turing Reference Manual
Win DOS Mac
File.Status
X TOOT Term
Syntax
File.Status (pathName : string, var size, attribute, fileTime : int)
Description
File.Status is used to get assorted information about a file or directory.
When the function is called with a specified pathName, it returns the
information about the file in the other parameters.
The size parameter is the size of the file in bytes.
The attribute parameter has its individual bits set as exactly as the attribute
parameter in Dir.GetLong subprogram does. See Dir.GetLong for the list
of attribute constants.
The fileTime is the time of last modification of the file. It is returned as the
number of seconds since 00:00:00 GMT 1/1/1970. To convert this to a
string, use Time.SecDate.
Details
If the File.Status call fails, size, attribute and fileTime are all set to –1.
Error.Last will return a non-zero value indicating the reason for the failure.
Error.LastMsg will return a string which contains the textual version of the
error.
Example
This program prints an approximation of the UNIX “ls -l” command for the
file "~/mail".
const pathName : string := "~/mail"
var size, attribute, fileTime : int
File.Status (pathName, size, attribute, fileTime )
if Error.Last = eNoError then
put pathName, " ", Time.SecDate (fileTime)
if (attribute and ootAttrDir) not= 0 then
put "d" ..
else
put "–" ..
end if
if (attribute and ootAttrOwnerRead) not= 0 then
put "r" ..
else
put "–" ..
end if
if (attribute and ootAttrOwnerWrite) not= 0 then
put "w" ..
else
put "–" ..
end if
if (attribute and ootAttrOwnerExecute) not= 0 then
put "x" ..
Chapter 7 : Language Features 267
else
else
put "–" ..
end if
if (attribute and ootAttrGroupRead) not= 0 then
put "r" ..
else
put "–" ..
end if
if (attribute and ootAttrGroupWrite) not= 0 then
put "w" ..
else
put "–" ..
end if
if (attribute and ootAttrGroupExecute) not= 0 then
put "x" ..
else
put "–" ..
end if
if (attribute and ootAttrGroupRead) not= 0 then
put "r" ..
else
put "–" ..
end if
if (attribute and ootAttrGroupWrite) not= 0 then
put "w" ..
else
put "–" ..
end if
if (attribute and ootAttrGroupExecute) not= 0 then
put "x" ..
else
put "–" ..
end if
put size : 8, " ", Time.SecDate (fileTime), " ", pathName
put "Unable to get file information"
put "Error: ", Error.LastMsg
end if
Status
Exported qualified.
This means that you can only call the function by calling File.Status, not by
calling Status.
flexible
Syntax
array initialization
flexible array indexType { , indexType } of typeSpec
268 Object Oriented Turing Reference Manual
Description
The flexible keyword allows an array to be resized using new at a later
point in time. The indices may have compile-time or run-time upper
bounds (the lower bound must be compile-time). The upper bounds can be
changed by using:
new name , newUpper1 {,newUpper2}
The existing array entries will retain their values, except that any index
made smaller will have the corresponding array entries lost. Any index
made larger will have the new array entries uninitialized (if applicable).
Additionally, the upper bound (both in the declaration and the new
statement) can be made one less than the lower bound. This effectively
makes an array that contains 0 elements. It can later be increased in size
with another new.
In the current implementation (1999), with a multi-dimensional array with
a non-zero number of total elements, it is a run-time error to change any
but the first dimension (unless one of the new upper bounds is one less
than the corresponding lower bound, giving 0 elements in the array) as the
algorithm to rearrange the element memory locations has not yet been
implemented.
Currently, only variables can be declared in this form. There is no flexible
array parameter type, although a flexible array can be passed to an array
parameter with “*” as the upper bound.
Example
See array for an example of flexible.
See also
array and new.
floor
real-to-integer function
Syntax
floor ( r : real ) : int
Description
Returns the largest integer that is less than or equal to r.
Details
The floor function is used to convert a real number to an integer. The result
is the largest integer that is less than or equal to r. In other words, the floor
function rounds down to the nearest integer. For example, floor (3) is 3,
floor (2.75) is 2 and floor (-8.43) is -9.
See also
ceil and round functions.
Chapter 7 : Language Features 269
Font
Description
This unit contains the predefined subprograms that deal with fonts. Using
these routines, you can display text in a selected font name, size and style
on the screen. Note that output in a particular font is treated as graphics
output.
All routines in the Font module are exported qualified (and thus must be
prefaced with "Font.").
Details
There is a default font. You can draw in and obtain information about the
default font by passing fontDefaultID to Font.Draw, Font.Width and
Font.Sizes. The default font is the same font as is used by put in the output
window.
Entry Points
New
Selects a particular font name, size and style for a new
font.
Free
Frees up the font created by using New.
Draw
Draws text in a given font.
Width
Gets the width in pixels of a particular piece of text in a
specified font.
Sizes
Gets the height and various leadings of a specified font.
Name
Returns the name of the specified font.
StartName
Prepares to list all available fonts,
GetName
Gets the next font name.
GetStyle
Gets all the available styles for a specified font.
StartSize
Prepares to list all available sizes for a specified font and
style.
GetSize
Gets the next font size.
270 Object Oriented Turing Reference Manual
Win DOS Mac
Font.Draw
X TOOT Term
Syntax
Font.Draw (txtStr : string, x, y, fontID, Color : int)
Description
Font.Draw is used to actually draw text in a specified font. The textStr
parameter contains the string to be drawn. The x and y parameters are the
location of the lower left hand corner of the text to be displayed. The fontID
parameter is the number of the font in which the text is to be drawn. The
Color parameter is used to specify the color in which the text is to appear.
Note that the text that appears is completely unrelated to the text that
appears using put. Font.Draw is a graphics command and thus does not
use or affect the cursor location.
The text drawn by the Font.Draw procedure does not erase the
background.
Details
If Font.Draw is passed an invalid font ID, a fatal error occurs. If the
Font.Draw call fails for other (non-fatal) reasons, then Error.Last will
return a non-zero value indicating the reason for the failure. Error.LastMsg
will return a string which contains the textual version of the error.
Example
The program prints out several phrases in a variety of fonts.
var font1, font2, font3, font4 : int
font1 := Font.New ("serif:12")
assert font1 > 0
font2 := Font.New ("sans serif:18:bold")
assert font2 > 0
font3 := Font.New ("mono:9")
assert font3 > 0
font4 := Font.New ("Palatino:24:bold,italic")
assert font4 > 0
Font.Draw ("This is in a serif font", 50, 30, font1, red)
Font.Draw ("This is in a sans serif font", 50, 80, font2, brightblue)
Font.Draw ("This is in a mono font", 50, 130, font3, colorfg)
Font.Draw ("This is in Palatino (if available)", 50, 180, font4, green)
Font.Free (font1)
Font.Free (font2)
Font.Free (font3)
Font.Free (font4)
Status
Exported qualified.
This means that you can only call the function by calling Font.Draw, not by
calling Draw.
Chapter 7 : Language Features 271
Win DOS Mac
Font.Free
X TOOT Term
Syntax
Font.Free (fontID : int)
Description
Font.Free is used to release a font that is no longer needed. There is a limit
to the number of fonts that may be defined at any one time. By having a
Font.Free for every Font.New, the number of simultaneously defined fonts
is kept to a minimum.
Details
If Font.Free is passed an invalid font ID, a fatal error occurs. If the
Font.Free call fails for other (non-fatal) reasons, Error.Last will return a
non-zero value indicating the reason for the failure. Error.LastMsg will
return a string which contains the textual version of the error.
Example
The program prints out several phrases in a variety of fonts.
var font1, font2, font3, font4 : int
font1 := Font.New ("serif:12")
assert font1 > 0
font2 := Font.New ("sans serif:18:bold")
assert font2 > 0
font3 := Font.New ("mono:9")
assert font3 > 0
font4 := Font.New ("Palatino:24:Bold,Italic")
assert font4 > 0
Font.Draw ("This is in a serif font", 50, 30, font1, red)
Font.Draw ("This is in a sans serif font", 50, 80, font2, brightblue)
Font.Draw ("This is in a mono font", 50, 130, font3, colorfg)
Font.Draw ("This is in Palatino (if available)", 50, 180, font4, green)
Font.Free (font1)
Font.Free (font2)
Font.Free (font3)
Font.Free (font4)
Status
Exported qualified.
This means that you can only call the function by calling Font.Free, not by
calling Free.
272 Object Oriented Turing Reference Manual
Win DOS Mac
Font.GetName
X TOOT Term
Syntax
Font.GetName : string
Description
Font.GetName is used to get the next font available on the system. By
using Font.StartName and then calling Font.GetName repeatedly, you can
get the names of all the fonts available to the program.
Font.StartName must be called before any calls to Font.GetName. After
that, Font.GetName returns the list of the font names, one per call. When
there are no more sizes, Font.GetName returns the empty string.
Once the name of a font is known, it’s possible to list the available styles
(using Font.GetStyle) and the available sizes (using Font.StartSize and
Font.GetSize) for that font.
The program lists all the fonts available on the system.
Example
var fontName : string
Font.StartName
loop
fontName := Font.GetName
exit when fontName = ""
put fontName
end loop
Status
Exported qualified.
This means that you can only call the function by calling Font.GetName,
not by calling GetName.
Win DOS Mac
Font.GetSize
X TOOT Term
Syntax
Font.GetSize : int
Description
Font.GetSize is used to get the next size in the list of available font sizes for
a particular font name and style.
Font.StartSize must be called before any calls to Font.GetSize. After that,
Font.GetSize returns the list of sizes, one per call. When there are no more
sizes, Font.GetSize returns 0.
Chapter 7 : Language Features 273
Some fonts are “scalable”. This means that the computer can scale the fonts
to fit any given size. (Under Microsoft Windows and the Apple Macintosh,
TrueType and PostScript fonts are scalable with the appropriate utilities.)
In this case, Font.GetSize returns -1.
Example
See Font.StartSize for a program that lists all the fonts, styles and sizes
available on the system.
Status
Exported qualified.
This means that you can only call the function by calling Font.GetSize, not
by calling GetSize.
Win DOS Mac
Font.GetStyle
X TOOT Term
Syntax
Font.GetStyle (fontName : string,
var bold, italic, underline : boolean): string
Description
Font.GetStyle is used to get the styles available on the system for a
specified font. bold, italic and underline are set to true if bold, italic or
underline versions of the font are available. Once the styles available for a
font are known, it’s possible to get the sizes available for each style by
using Font.StartSize and Font.GetSize.
Example
The program lists all the fonts and their styles available on the system.
var fontName : string
var bold, italic, underline : boolean
Font.StartName
loop
fontName := Font.GetName
exit when fontName = ""
Font.GetStyle (fontName, bold, italic, underline)
put fontName : 30 ..
if bold then
put "bold " ..
end if
if italic then
put "italic " ..
end if
if underline then
put "underline " ..
end if
end loop
Status
Exported qualified.
274 Object Oriented Turing Reference Manual
This means that you can only call the function by calling Font.GetStyle,
not by calling GetStyle.
Win DOS Mac
Font.Name
X TOOT Term
Syntax
Font.Name (fontID : int) fontName : string
Description
Font.Name is used to get the name of a font that is being used. The string
that is returned can be used to determine which font is actually being used
for the default fonts “serif”, “sans serif” and “mono”.
Example
The program prints out the fonts used for “serif”, “sans serif” and “mono”.
var serifFont, sansSerifFont, monoFont : int
serifFont := Font.New ("serif:12")
assert serifFont > 0
sansSerifFont := Font.New ("sans serif:12")
assert sansSerifFont > 0
monoFont := Font.New ("mono:12")
assert monoFont > 0
put "serif = ", Font.Name (serifFont)
put "sans serif = ", Font.Name (sansSerifFont)
put "mono = ", Font.Name (monoFont)
Font.Free (serifFont)
Font.Free (sansSerifFont)
Font.Free (monoFont)
Status
Exported qualified.
This means that you can only call the function by calling Font.Name, not
by calling Name.
Win DOS Mac
Font.New
Syntax
X TOOT Term
Font.New (fontSelectStr : string) : int
Chapter 7 : Language Features 275
Description
Font.New is used to obtain a font for drawing. The fontSelectStr parameter
specifies the name, size and style of the font. Font.New returns a font
identifier which is then used by the Font.Draw procedure to draw text in
the selected font.
The format for the fontSelectStr parameter is "Family:Size:Style". Each
element is separated by a colon. The ":Style" is optional. If left out, the text
appears in the standard face for the font.
Family is the name of the font, such as "Times", "Helvetica", etc. The name
must match an existing font on the system. Because one does not
necessarily know which fonts will be available and names for the same font
change between different systems (i.e Times, Times-Roman, etc.), OOT
defines three family names that will be mapped as closely as possible to
fonts that exist on the system.
•
"serif" is used for a serifed body font. This will usually be mapped
to Times-Roman.
•
"sans serif" is used for a non-serifed display font. This will usually
be mapped to Helvetica.
•
"mono" is used for a mono spaced font. This will usually be
mapped to Courier.
Size is the point size in which the text should appear. If the number is
larger or smaller than can be created on a given system, the system will
return the font of the largest or smallest size available and set Error.Last.
Under WinOOT, the size parameter may also have the form height x width
where height and width are the pixel height and width desired. What is
returned is the font scaled in order to fit into the width and height
requested. The font name must be a scaleable font for this to succeed.
example
fontID := Font.New ("Ariel:18x12:Italic")
Style is the font style in which the text should appear. It can be one of
"bold", "italic" or "underline". You can also have "bold,italic" and any other
combination.
Details
The DOS OOT font module uses Microsoft Windows font files (those
labelled with the “.FON” file name suffix. If you don’t have MS Windows
installed and thus have no font files, you will be limited to two fonts
detailed below.
In order to find the font file, DOS OOT must be provided with the name of the directory
that the “.FON” files are located in. By default, this is the
“C:\WINDOWS\SYSTEM” directory. If the fonts are located there,
nothing further need be done. However, if they are located in a different
directory, then the directory must be specified by using the “-fontdir=“
start up option. The directory is specified either as a command line
argument or in the “turing.cfg” file (see Chapter Four) as
“–fontdir=<dirname>”. For example, if one had all the fonts placed in the
directory D:\FONTS, you would start DOS OOT with:
DOSOOT –fontdir=D:\FONTS
276 Object Oriented Turing Reference Manual
There are two fonts that are always available, regardless of the presence or
absence of MS Window fonts. These are the default font (font number 0)
and a font called “vector”. In DOS OOT, the default font is 8 pixels in
height and width. The “vector” font is a completely scalable font. To use it,
you must specify both the height and the width of the font when calling
Font.New. The Font is drawn with line segments, rather than scaled from a
bitmap and looks somewhat simplistic. However, the font can be scaled as
necessary for use in programs. To set a version of the font to be 20 pixels
high and 12 pixels wide, one would call:
var fontID : int := Font.New ("vector:20x12")
DOS OOT does not support any other scalable fonts. Specifically, it does
not support TrueType fonts at the time of writing. DOS OOT does not
support any font styles. Any call to Font.New will have the style ignored.
Details
If the Font.New call fails, then it returns 0. Also Error.Last will return a
non-zero value indicating the reason for the failure. Error.LastMsg will
return a string which contains the textual version of the error.
It is quite possible for Error.Last to be set, even if the call succeeds.
Font.New will report success even if unable to successfully match the
requested font with the available resources. A font will be set that matches
as closely as possible the requested font and Last.Error will be set to
indicate that some substitutions were required.
Example
The program prints out several phrases in a variety of fonts.
Status
var font1, font2, font3, font4 : int
font1 := Font.New ("serif:12")
font2 := Font.New ("sans serif:18:bold")
font3 := Font.New ("mono:9")
font4 := Font.New ("Palatino:24:Bold,Italic")
assert font1 > 0 and font2 > 0 and font3 > 0 and font4 > 0
Font.Draw ("This is in a serif font", 50, 30, font1, red)
Font.Draw ("This is in a sans serif font", 50, 80, font2, brightblue)
Font.Draw ("This is in a mono font", 50, 130, font3, colorfg)
Font.Draw ("This is in Palatino (if available)", 50, 180, font4, green)
Font.Free (font1)
Font.Free (font2)
Font.Free (font3)
Font.Free (font4)
Exported qualified.
This means that you can only call the function by calling Font.New, not by
calling New.
Chapter 7 : Language Features 277
Win DOS Mac
Font.Sizes
X TOOT Term
Description
Font.Sizes is used to get the metrics of a particular font. The various parts
of the metric are illustrated below. Note that you can calculate the external
leading by subtracting the ascent and descent from the height.
descent
ascent
Font.Sizes (fontID : int, var height, ascent, descent,
internalLeading : int)
height
Syntax
Details
Example
j
external leading
internal leading
Baseline
If Font.Sizes is passed an invalid font ID, a fatal error occurs. If the
Font.Sizes call fails for other (non-fatal) reasons, the metrics for the default
font will be returned. As well, Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
The program gets information about 24pt Bold Italic Palatino.
var fontID, height, ascent, descent, internalLeading, : int
var externalLeading : int
fontID := Font.New ("Palatino:24:bold,italic")
Font.Sizes (fontID, height, ascent, descent, internalLeading)
externalLeading := height – ascent – descent
put "The height of the font is ", height, " pixels"
put "The ascent of the font is ", ascent, " pixels"
put "The descent of the font is ", descent, " pixels"
put "The internal leading of the font is ", internalLeading, " pixels"
put "The external leading of the font is ", externalLeading, " pixels"
Font.Free (fontID)
Status
Exported qualified.
This means that you can only call the function by calling Font.Sizes, not by
calling Sizes.
278 Object Oriented Turing Reference Manual
Win DOS Mac
Font.StartName
X TOOT Term
Syntax
Font.StartName
Description
Font.StartName is used to start the listing of all the fonts available on the
system. This procedure is called before making calls to Font.GetName to
get the name of the fonts available. Once the name of a font is known, it’s
possible to list the available styles (using Font.GetStyle) and the available
sizes (using Font.StartSize and Font.GetSize).
Example
The program lists all the fonts available on the system.
var fontName : string
Font.StartName
loop
fontName := Font.GetName
exit when fontName = ""
put fontName
end loop
Status
Exported qualified.
This means that you can only call the function by calling Font.StartName,
not by calling StartName.
Win DOS Mac
Font.StartSize
X TOOT Term
Syntax
Font.StartSize (fontName, fontStyle : string)
Description
Font.StartSize is used to start a listing of all the sizes for a particular font
name and style.
The fontName parameter should be an actual font name (as opposed to the
default names of “serif”, etc). You can get a list of the font names by using
the Font.StartName and Font.GetName subprograms. The fontStyle
parameter should be in the same format as would appear in the Font.New
procedure.
Chapter 7 : Language Features 279
The program lists all the fonts, styles and sizes available on the system.
Example
var fontName : string
var bold, italic, underline : boolean
var size : int
var styles : array boolean, boolean, boolean of string :=
init ("", "underline", "italic", "italic, underline", "bold",
"bold,italic", "bold,italic,underline")
Font.StartName
loop
fontName := Font.GetName
exit when fontName = ""
Font.GetStyle (fontName, bold, italic, underline)
for b : false .. bold
for i : false .. italic
for u : false .. underline
put fontName : 30, styles (b, i, u) : 22 ..
Font.StartSize (fontName, styles (b, i, u) )
loop
size := Font.GetSize
exit when size = 0
if size = –1 then put "scalable " ..
else put size, " " ..
end if
end loop
put ""
end for
end for
end for
end loop
Status
"bold,underline",
Exported qualified.
This means that you can only call the function by calling Font.StartSize,
not by calling StartSize.
Win DOS Mac
Font.Width
X TOOT Term
Syntax
Font.Width (txtStr : string, fontID : int) : int
Description
Font.Width is used to obtain the width in pixels that a specified string will
take to draw in a specified font. The textStr parameter is the string. The
fontID parameter is the font in which the string would be drawn.
280 Object Oriented Turing Reference Manual
If Font.Width is passed an invalid font ID, a fatal error occurs. If the
Font.Width call fails for other (non-fatal) reasons, the width for string in
the default font will be returned. As well, Error.Last will return a non-zero
value indicating the reason for the failure. Error.LastMsg will return a
string which contains the textual version of the error.
Details
The program gets information about 24pt Bold Italic Palatino.
Example
var width : int
fontID := Font.New ("Palatino:24:Bold,Italic")
width := Font.Width ("Test String", fontID)
put "The width of \"Test String\" is ", width, " pixels"
Font.Free (fontID)
Status
Exported qualified.
This means that you can only call the function by calling Font.Width, not
by calling Width.
for
Syntax
statement
A forStatement is:
for [decreasing] [id ] : first .. last [by increment]
statementsAndDeclarations
end for
Description
The statements and declarations in a for statement are repeatedly executed.
In the first iteration, the identifier is assigned the value of first. With each
additional iteration, the identifier increases by 1 (or by increment, if the by
clause is present). The loop stops executing when adding 1 (or increment) to
the identifier would cause the identifier to exceed last. first and last must be
integer values (or else enumerated or char values). If you specify
decreasing, then the identifier decreases by 1 (or by increment) each time
through.
Increment must be a positive integer value. When the by clause is present,
the for loop terminates as soon as the identifier would become greater than
last, unless decreasing is present. If decreasing is present, the loop
terminates when the identifier would become less than last.
Details
The identifier is checked before it is added to (or subtracted from). This
means that the loop
for i : 1 .. maxint
will not cause an overflow.
Chapter 7 : Language Features 281
Example
Output 1, 2, 3 to 10.
for i : 1 .. 10
put i
Example
Output 1, 3, 5, 7 and 9.
for i : 1 .. 10 by 2
put i
Example
end for
end for
Output 10, 9, 8, down to 1.
for decreasing j : 10 .. 1
put j
end for
Example
Output 10, 6, and 2.
for decreasing j : 10 .. 1 by 4
put j
end for
Example
Output 1.
for j : 1 .. 10 by 20
put j
Example
Output nothing.
for j : 5 .. 2
put j
Details
end for
end for
The for statement declares the counting identifier (a separate declaration
should not be given for i or j). The scope of this identifier is restricted to the
for statement.
If first is a value beyond last, there will be no repetitions (and no error
message). The counting identifier is always increased (or decreased) by 1
or increment if the by clause is present. Executing an exit statement inside a
for statement causes a jump to just beyond end for. You are not allowed to
change the counting variable (for example, you are not allowed to write i :=
10).
The counting identifier can be omitted. In this case, the statement is just as
before, except that the program cannot use the value of the identifier.
If decreasing is not present, first .. last can be replaced by the name of a
subrange type, for example by dozen, declared by:
type dozen : 1..12
Procedures, functions and modules cannot be declared inside a for
statement. Just preceding the statements and declarations, you are allowed
to write an "invariant clause" of the form:
invariant trueFalseExpn
282 Object Oriented Turing Reference Manual
This clause is equivalent to: assert trueFalseExpn.
fork
statement
Dirty parts
A forkStatement is:
Syntax
fork processId [ ( [ expn { , expn } ] ) ]
[ : reference [ , expn [ , reference ] ] ]
Description
A fork activates (starts the concurrent execution of) a process declaration.
If the process has parameters, a parenthesized list of expressions (expns)
must follow the process' name (processId).
Example
This program initiates (forks) two concurrent processes, one of which
repeatedly outputs Hi and the other Ho. The resulting output is an
unpredictable sequence of Hi's and Ho's, as greetings executes twice
concurrently, one instance with its word set to Hi and the other with its
word set to Ho.
process greetings ( word : string )
loop
put word
end loop
end greetings
fork greetings ( "Hi" )
fork greetings ( "Ho" )
Details
See procedure declaration for details about parameters. The first optional
reference in the fork statement must be a boolean variable reference. The
fork sets this to true if the process is actually activated. If this fails to occur
(probably because stack space could not be allocated), this reference is set to
false. If the fork fails but this reference is omitted, an exception occurs. See
exception handlers.
The optional expn specifies the number of bytes for the process' stack; this
overrides the optionally given stack size in the process declaration. The
second optional reference must be a variable reference with the type
addressint. See addressint. This variable is set to identify the process
activation. This reference has the implementation-dependent meaning of
locating the process' internal descriptor.
In this explanation of the fork statement, we have up to this point ignored
the possibility of processes exported from modules. If the process is being
forked from outside of a module from which it has been exported, the
syntax of the fork statement is:
Chapter 7 : Language Features 283
fork moduleId . procedureId [ ( expn {, expn } ) ] …
In other words, the module's name and a dot must precede the process'
name.
forward
subprogram declaration
A forwardDeclaration is:
Syntax
forward subprogramHeader
[ import importItem {, importItem } ]
Description
A procedure or function is declared to be forward when you want to
define its header but not its body. This is the case when one procedure or
function calls another, which in turn calls the first; this situation is called
mutual recursion. The use of forward is necessary in this case, because every
item must be declared before it can be used.
Example
This example program evaluates an input expression e of the form t { + t }
where t is of the form p { * p } and p is of the form (e ) or an explicit real
expression. For example, the value of 1.5 + 3.0 * (0.5 + 1.5) halt is 7.5.
var token : string
forward procedure expn ( var eValue : real )
forward procedure term ( var tValue : real )
forward procedure primary ( var pValue: real )
body procedure expn
var nextValue : real
% Evaluate t
term ( eValue )
loop
% Evaluate { + t}
exit when token not= "+"
get token
term ( nextValue )
eValue := eValue + nextValue
end loop
end expn
body procedure term
var nextValue : real
primary (tValue )
% Evaluate p
loop
% Evaluate { * p}
exit when token not= "*"
get token
284 Object Oriented Turing Reference Manual
primary ( nextValue )
tValue := tValue + nextValue
end loop
end term
body procedure primary
if token = "(" then
get token
% Evaluate (e)
expn ( pValue )
assert token = ")"
else
% Evaluate "explicit real"
pValue := strreal ( token )
end if
get token
end primary
get token
var answer : real
expn ( answer )
Details
% Start by reading first token
% Scan and evaluate input expression
put "Answer is ", answer
Following a forward procedure or function declaration, the body of the
procedure must be given at the same level (in the same sequence of
statements and declarations as the forward declaration). This is the only
use of the keyword body. See also body.
Any procedure or function that is declared using forward requires an
import list. In this list, imported procedures or functions that have not yet
appeared must be listed as forward. For example, the import list for expn is
import forward term … Before a procedure or function can be called,
before its body appears, and before it can be passed as a parameter, its
header as well as headers of procedures or functions imported directly or
indirectly by it must have appeared.
The keyword forward is also used in collection and type declarations.
See also
collections and type declarations.
frealstr
real-to-string function
Syntax
frealstr ( r : real, width, fractionWidth : int ) : string
Description
The frealstr function is used to convert a real number to a string. For
example, frealstr (2.5e1, 5, 1)="b25.0" where b represents a blank. The
string is an approximation to r, padded on the left with blanks as necessary
to a length of width.
Chapter 7 : Language Features 285
The number of digits of the fraction to be displayed is given by
fractionWidth.
The width must be non-negative. If the width parameter is not large enough
to represent the value of r, it is implicitly increased as needed.
The fractionWidth must be non-negative. The displayed value is rounded to
the nearest decimal equivalent with this accuracy. In the case of a tie, the
value is rounded to the next larger value. The result string is of the form:
{blank} [-]{digit}. {digit}
If the leftmost digit is zero, then it is the only digit to the left of the decimal
point.
The frealstr function approximates the inverse of strreal, although roundoff errors keep these from being exact inverses.
the erealstr, realstr, strreal, intstr and strint functions.
See also
free
statement
A freeStatement is:
Syntax
free [ collectionOrClassId, ]
pointerVariableReference
Description
A free statement destroys (deallocates) an element that has been allocated
by a new statement.
Example
Using a collection, declare a list of records and allocate one of these
records. Then deallocate the record.
var list : collection of
record
contents : string ( 10 )
next : pointer to list
% Short form: next : ^ list
end record
% Short form: var next : ^ list
var first : pointer to list
new list, first
% Allocate an element of list; place its location in first
% Short form: new first
…
free list, first % Deallocate the element of list located by first
% Short form: free first
286 Object Oriented Turing Reference Manual
The free statement sets the pointer variable to the nil value. See the new
statement for examples of allocating elements of classes and values of
types. It the pointer locates a type, the collectionOrClassId in the free
statement must be omitted.
Details
An imported class can have one of its objects destroyed (by the free
statement) only if the class is imported var.
The collectionOrClassId is optional in the free statement.
class and collection declarations, the pointer type, the new statement and
the nil value.
See also
function
declaration
A functionDeclaration is:
Syntax
function id [ ( [paramDeclaration {, paramDeclaration } ] ) ]
: typeSpec
statementsAndDeclarations
end id
Description
A function declaration creates (but does not run) a new function. The name
of the function (id) is given in two places, just after function and just after
end.
Example
function doubleIt ( x : real ) : real
result 2.0 * x
end doubleIt
put doubleIt ( 5.3 )
Details
% This outputs 10.6
The set of parameters declared with the function are called formal
parameters. For example, in the doubleIt function, x is a formal parameter.
A function is called (invoked) by a function call which consists of the
function's name followed by the parenthesized list of actual parameters (if
any). For example, doubleIt (5.3) is a call having 5.3 as an actual parameter.
If there are no parameters and no parentheses, the call does not have
parentheses. The keyword function can be abbreviated to fcn. See also
functionCall and procedureDeclaration.
Each actual non-var parameter must be assignable to the type of its
corresponding formal parameter. See also assignability.
Chapter 7 : Language Features 287
A function must finish by executing a result statement, which produces the
function's value. In the above example, the result statement computes and
returns the value 2.0 * x.
In principle, a function (1) should not change any variables outside of itself
(global variables) or (2) should not have var parameters. In other words, it
should have no side effects. The original implementation prevented (1) and
(2) and thereby prevented function side effects. Current implementations
of Turing do not enforce this restriction.
The upper bounds of arrays and strings that are parameters may be
declared to be an asterisk (*), meaning the bound is that of the actual
parameter. See paramDeclaration for details about parameters.
Procedures and functions cannot be declared inside other procedures and
functions.
The syntax of a functionDeclaration presented above has been simplified by leaving out the
optional result identifier, import list, pre condition, init clause, post
condition and exception handler. The full syntax is
function [ pervasive ] id
[ ( [ paramDeclaration {,paramDeclaration } ] ) ]
[ resultId ] : typeSpec
[ pre trueFalseExpn ]
[ init id := expn {, id := expn } ]
[ post trueFalseExpn ]
[ exceptionHandler ]
statementsAndDeclarations
end id
See also
import list, pre condition, init clause, post condition and exceptionHandler
for explanations of these additional features.
See also pervasive. The resultId is the name of the result of the function and
can be used only in the post condition.
A function must be declared before being called; to allow for mutually
recursive procedures and functions, there are forward declarations with
later declaration of the procedure or function body. See forward and body
declarations for explanations.
You declare parameterless functions using an empty parameter list. When
this is done, a call to the function must include an empty parameter list.
functionCall
Syntax
A functionCall is:
functionId [ ( [ expn { , expn } ] ) ]
288 Object Oriented Turing Reference Manual
Description
A function call is an expression that calls (invokes or activates) a function.
If the function has parameters, a parenthesized list of expressions (expns)
must follow the function's name (functionId).
Example
This function takes a string containing a blank and returns the first word in
the string (all the characters up to the first blank).
function firstWord ( str : string ): string
for i : 1 .. length ( str )
if str ( i ) = " " then
result str ( 1 .. i - 1)
end if
end for
end firstWord
put "The first word is: ", firstWord ("Henry Hudson")
% The output is Henry.
Details
The parameter declared in the header of a function, is a formal parameter,
for example, str above is a formal parameter. Each expression in the call is
an actual parameter, for example, sample above is an actual parameter.
Each actual parameter passed to its non-var formal parameter must be
assignable to that parameter (see assignability for details). See also
functionDeclaration and procedureDeclaration.
In this explanation of functionCall, we have up to this point ignored the
possibility of functions exported from modules. If the function is being
called from outside of a module from which it has been exported, the
syntax of the functionCall is:
moduleId . functionId [ ( expn {, expn } ) ]
In other words, the module or monitor name and a dot must precede the
function's name. If the function is being called from outside of a class from
which it has been exported, the syntax of the functionCall is one of:
(a) classId ( p ) . functionId [ ( [ expn {, expn } ] ) ]
(b)
p -> functionId [ ( [ expn {, expn } ] ) ]
In these p must be a pointer value that locates an object in the class. Form
(b) is a short form for form (a).
See also
class.
get
Syntax
file statement
A getStatement is:
Chapter 7 : Language Features 289
get [ : streamNumber , ] getItem { , getItem }
Description
The get statement inputs each of the getItems. Ordinarily, the output comes
from the keyboard. However, if the streamNumber is present, the input
comes from the file specified by the stream number (see the open statement
for details). Also, input can be redirected so it is taken from a file rather
than the keyboard. Check the documentation on the environment for
instructions on doing so.
The syntax of a getItem is one of:
(a) variableReference
(b) skip
(c) variableReference : *
(d)
variableReference : widthExpn
These items are used to support three kinds of input:
(1) token and character oriented input: supported by forms (a) and (b),
(2) line oriented input: supported by form (c), and
(3) character oriented input: supported by form (d).
Examples of these will be given, followed by detailed explanations.
Example
Token-oriented input.
var name, title : string
var weight : real
% If input is Alice, it is input into name
get name
get title
% If input is "A lady", A lady is input
var weight
Example
Line-oriented input.
var query : string
Example
get query : * % Entire line is input into query
Character-oriented input.
var code : string
Details
% If input is 9.62, it is input into weight
get code : 2
% Next 2 characters are input into code.
A token is a sequence of characters surrounded by white space, where white
space is defined as the characters: blank, tab, form feed, new line, and
carriage return as well as end-of-file. The sequence of characters making
up the token are either all non-white space or else the token must be a
quoted string (an explicit string constant). When the variableReference in
form (a) is a string, integer, real, intn, natn, or realn. Turing skips white
space, reads a token into the variableReference, and then skips white space
(stopping at the beginning of the next line).
If the variableReference is a string, the token is assigned to the variable (if the
token is quoted, the quotation marks are first removed). See the examples
involving name and title above. If the variableReference is an integer or a real,
the token is converted to be numeric before being assigned to the variable.
See the example involving weight above.
290 Object Oriented Turing Reference Manual
When the input is coming from the keyboard, no input is done until Return
is typed. The line that is input may contain more than one token. Any
tokens that are not input by one get statement will remain to be input by
the next get statement.
Turing has been modified so that token-oriented input now also skips
white space following the token, but does not skip beyond the beginning of
the next line. This change implies that form (b) is usually not needed, as
skip was used to skip white space after the token.
Form (a) supports char and char(n). If the type is char, exactly one
character is read, with no skipping of white space before or after. This
character may be, for example, a blank or a carriage return. If the type is
char(n), exactly n characters are read, with no skipping of white space.
Example
Inputting char and char(n) types using form (a). The statement get c:1 is not
legal, because length specification is not allowed with character variables.
var c : char
var d : char ( 3 )
get c
% Read one character.
get d
% Read three characters
Details
Form (a) supports enumerated types. If the type is an enumerated type,
then the token read in must be one of the elements of the enumerated type.
Example
Inputting an enumerated type using form (a). The statement get c:1 is not
legal, because length specification is not allowed with enumerated
variables.
type colors : enum (red, blue, green)
var c : colors
get c
% Read one of red, green or blue
Details
Form (a) supports boolean. If the type is an boolean type, then the token
read in must be one of "true" or "false"
Example
Inputting a boolean type using form (a). The statement get c:1 is not legal,
because length specification is not allowed with boolean variable.
var tf : boolean
get tf
% Read one of true or false
Details
In form (b) of getItem, skip causes white space in the input to be skipped
until non-white space (a token) or the end-of-file is reached. This is used
when the program needs to determine if there are more tokens to be input.
To determine if there are more tokens to be read, the program should first
skip over any possible white space (such as a final new line character) and
then test to see if eof (end-of-file) is true. This is illustrated in this example:
Example
Using token-oriented input, input and then output all tokens. This example
gives what used to be the standard way of reading tokens up to end of file.
With the new meaning of form (a) for reading tokens, the get skip line can
be omitted. This omission is possible because the line get word now
automatically skips white space following the input value, up to the
beginning of the next line.
Chapter 7 : Language Features 291
var word : string
loop
% Skip over any white space
get skip
exit when eof
% Are there more characters?
get word
% Input next token
put word
% Output the token
end loop
In the above and the next example, if the input has been redirected so that
it is from a file, eof becomes true exactly when there are no more characters
to be read. If the input is coming from the keyboard, you can signal eof by
typing control-Z (on a PC) or control-D (on UNIX).
Details
In form (c) of getItem, the variableReference is followed by :* which implies
line-oriented input. This form causes the entire line (or the remainder of
the current line) to be read. In this case the variable must be a string (not an
integer or real). The new line character at the end of the line is discarded. It
is an error to try to read another line when you are already at the end of the
file. The following example shows how to use line-oriented input to read
all lines in the input.
Example
Using line-oriented input, input and then output all lines.
var line : string
loop
% Are there more characters?
exit when eof
get line : *
% Read entire line
put line
end loop
Details
In form (d) of getItem, the variableReference is followed by
: widthExpn
which specifies character-oriented input. This form causes the specified
number (widthExpn) of characters to be input (or all of the remaining
characters if not enough are left). If no characters remain, the null string is
read and no warning is given. In this form, the new line character is
actually input into the variableReference (this differs from line-oriented input
which discards new line characters). The following example shows how to
use character-oriented input to read each character of the input. Form (d)
can be used with string and char(n) variables, but not with char, int or any
other type.
Example
Using character-oriented input, input and then output all characters.
var ch : string ( 1 )
loop
% Are there more characters?
exit when eof
get ch : 1
% Read one character
put ch ..
% Output the character, which
% may be a new line character
end loop
Example
Using character-oriented input, input two characters.
var d : char ( 3 ) := 'abc'
292 Object Oriented Turing Reference Manual
get d : 2
% Read two character (replace 'ab')
read statement, which provides binary file input.
See also
getch
get character procedure
Syntax
getch ( var ch : string ( 1 ) )
Description
The getch procedure is used to input a single character without waiting for
the end of a line. The parameter ch is set to the next character in the
keyboard buffer (the oldest not-yet-read character).
Example
This program contains a procedure called getKey which causes the program
to wait until a key is pressed.
setscreen ("graphics")
procedure getKey
var ch : string (1)
getch (ch)
end getKey
for i : 1 .. 1000
put i : 4, " Pause till a key is pressed"
getKey
end for
Details
The screen should be in a "screen" or "graphics" mode. See the setscreen
procedure for details. If the screen is not in one of these modes, it will
automatically be set to "screen" mode.
On IBM PC’s some keys, such as the left arrow key, insert key, delete key,
and function keys do not produce ordinary character values. These
keystrokes are returned by getch as their "scan code" with 128 added to
them, unless the scan code already has a value of 128 or greater. This
provides a unique value for every key on the keyboard. See Appendix D
for these codes.
See also
hasch (has character) procedure which is used to see if a character has been
typed but not yet read.
See also predefined unit Input.
Chapter 7 : Language Features 293
getchar
get character function
Syntax
getchar : char
Description
The getchar function is used to input a single character without waiting for
the end of a line. The next character in the keyboard buffer (the oldest notyet-read character) is returned.
Example
This program contains a procedure called getKey which causes the program
to wait until a key is pressed.
setscreen ("graphics")
procedure getKey
var ch : char
ch := getchar
end getKey
for i : 1 .. 1000
put i : 4, " Pause till a key is pressed"
getKey
end for
Details
The screen should be in a "screen" or "graphics" mode. See the setscreen
procedure for details. If the screen is not in one of these modes, it will
automatically be set to "screen" mode.
On IBM PC’s some keys, such as the left arrow key, insert key, delete key,
and function keys do not produce ordinary character values. These
keystrokes are returned by getchar as their "scan code" with 128 added to
it, unless the scan code already has a value of 128 or greater. This provides
a unique value for every key on the keyboard. See Appendix D for these
codes.
See also
hasch (has character) procedure which is used to see if a character has been
typed but not yet read.
See also predefined unit Input.
294 Object Oriented Turing Reference Manual
Win DOS Mac
getenv
get environment function
X TOOT Term
Syntax
getenv ( symbol : string ) : string
Description
The getenv function is used to access the environment string whose name
is symbol. These strings are determined by the shell (command processor)
or the program that caused your program to run. See also the nargs and
fetcharg functions.
Example
On a PC, this retrieves the environment variable USERLEVEL and prints
extra instructions if USERLEVEL had been set to NOVICE. USERLEVEL
can be set to NOVICE with the command SET USERLEVEL = NOVICE in
the autoexec.bat file or in any batch file.
const userLevel : string
userLevel := getenv ("USERLEVEL")
if userLevel = "NOVICE" then
…
% put a set of instructions
end if
See also
See also predefined unit Sys.
Win DOS Mac
getpid
get process id function
X TOOT Term
Syntax
getpid : int
Description
The getpid function is used to determine the I.D. (number) that identifies
the current operating system task (process). Beware that there are
processes, activated by the fork statement, that are independent of the
operating systems tasks.
Under UNIX, the number is used, for example, for creating a unique name
of a file.
See also
nargs, fetcharg and getenv.
See also predefined unit Sys.
Chapter 7 : Language Features 295
getpriority
function
Syntax
getpriority : nat
Description
The getpriority function returns the priority of an executing process in a
concurrent program. A smaller value means a faster speed.
See also
setpriority, fork and monitor.
See also predefined unit Concurrency.
Win DOS Mac
GUI
Description
X TOOT Term
This unit contains the predefined subprograms for creating and using a
GUI (Graphical User Interface). Elements of the GUI include buttons, check
boxes, text boxes, scroll bars, menus, etc.
For a general introduction to the the GUI module, see Chapter 5 of this
book.
All routines in the GUI module are exported qualified (and thus must be
prefaced with "GUI.").
Win DOS Mac
GUI.AddLine
Syntax
X TOOT Term
GUI.AddLine ( widgetID : int, text : string )
296 Object Oriented Turing Reference Manual
Description
GUI.AddLine adds text and a newline to the current line of the text box
specified by widgetID. It is essentially equivalent to put text in the text box.
GUI.AddLine scrolls the text box (if necessary) so that the added text is
now visible. The widgetID parameter must be the widget id of a text box.
The text parameter is the text to be added to the text box.
Example
The following creates a text box and puts the numbers from 1 to 25 in it.
import GUI in "%oot/lib/GUI"
var boxID : int := GUI.CreateTextBox (50, 50, 200, 200)
for i : 1 .. 25
GUI.AddLine (boxID, intstr (i))
end for
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling GUI.AddLine,
not by calling AddLine.
GUI.CreateTextBox.
See also
Win DOS Mac
GUI.AddText
X TOOT Term
Syntax
GUI.AddText ( widgetID : int, text : string )
Description
GUI.AddText adds text to the current line of the text box specified by
widgetID. It does not add a newline after the text. It is essentially equivalent
to put text ... in the text box. GUI.AddLine scrolls the text box (if
necessary) so that the added text is now visible. The widgetID parameter
must be the widget id of a text box. The text parameter is the text to be
added to the text box.
Details
To force a text box to scroll to the end of the text without adding any extra
text, call GUI.AddText with "" (the null string) for the text parameter.
Example
The following creates a text box and puts the numbers from 1 to 26
followed by the appropriate letter of the alphabet in it.
import GUI in "%oot/lib/GUI"
var boxID : int := GUI.CreateTextBox (50, 50, 200, 200)
for i : 1 .. 26
GUI.AddText (boxID, intstr (i))
GUI.AddText (boxID, " ")
Chapter 7 : Language Features 297
GUI.AddLine (boxID, chr (64 + i))
end for
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling GUI.AddText, not
by calling AddText.
See also
GUI.CreateTextBox.
Win DOS Mac
GUI.Alert[2,3,Full]
Syntax
X TOOT Term
GUI.Alert ( title, msg : string )
GUI.Alert2 ( title, msg1, msg2 : string )
GUI.Alert3 ( title, msg1, msg2, msg3 : string )
GUI.AlertFull ( title : string,
msg : array 1 .. * of string, button : string )
Description
Displays a dialog box with the string specified by msg in it. There is a single
button labelled OK which dismisses the dialog and resumes execution. The
title parameter specifies the window title under Microsoft Windows. On
the Apple Macintosh, there is no title, so do not assume the user will see
the title. The dialog box is centered on the screen.
The GUI.Alert2 and GUI.Alert3 procedures allow the user to specify a two
or three line message respectively. The GUI.AlertFull procedure allows the
user to specify any number of lines of text in the string array specified by
msg as well as the text in the dismissal button. Any empty strings at the
end of the array are not displayed.
Note: This function is not available in the current version of the GUI
Procedure Library (shipping with WinOOT 3.1, DOS OOT 2.5 and
MacOOT 1.5). It is documented here for use with future shipping version
of Object Oriented Turing. It is likely to be implemented in the version of
Object Oriented Turing released in September 2000. Check the release
notes that are found in the on-line help to find out if this function is now
available.
298 Object Oriented Turing Reference Manual
Example
The following program asks the user for the name of a file puts up an alert
dialog box if it fails.
import GUI in "%oot/lib/GUI"
var fileName : string
var streamNumber : int
loop
fileName := GUI.SaveFile ("Save As")
open : streamNumber, fileName, put
exit when streamNumber > 0
GUI.Alert ("Open Failure", "\"" + fileName +
"\" could not be opened")
end loop
Example
The following program asks the user for the name of a file puts up a more
complete alert dialog box if it fails.
import GUI in "%oot/lib/GUI"
var fileName : string
var streamNumber : int
loop
fileName := GUI.SaveFile ("Save As")
open : streamNumber, fileName, put
exit when streamNumber > 0
GUI.Alert2 ("Open Failure",
"\"" + fileName + "\" could not be opened.",
"Reason: " + Error.LastMsg)
end loop
Example
The following program fragment displays an alert with four lines of text
and a button that says "Abort".
var message : array 1 .. 10 of string
for i : 1 .. 10
message (i) := ""
end for
…
message (1) := "The program must now quit"
message (2) := "becasue of an unrecoverable error."
message (3) := "A Read Error occurred while reading"
message (4) := "file \"" + fileName + "\"."
message (5) := Error.LastMsg
GUI.AlertFull ("Error", message, "Abort")
Status
Exported qualified.
This means that you can only call the function by calling GUI.Alert, not by
calling Alert.
Chapter 7 : Language Features 299
Win DOS Mac
GUI.Choose[Full]
X TOOT Term
Syntax
GUI.Choose ( title, msg1, msg2, msg3 : string,
btn1, btn2, btn3 : string ) : int
GUI.ChooseFull ( title : string,
msg : array 1 .. * of string,
btn1, btn2, btn3 : string, defaultBtn : int ) : int
Description
Displays a dialog box with text and from one to three buttons. The user
selects a button to dismiss the dialog. The number of the button pressed is
returned by the function. The dialog box is centered on the screen.
The title parameter specifies the title in the window bar of the dialog box.
The Apple Macintosh does not have a title bar, so do not assume that the
user will see the string in the title parameter. The message is specified by
strings in msg1, msg2 and msg3 for GUI.Choose and the string array
message for GUI.ChooseFull. In each case, empty strings at the end of the
list of strings are ignored. The btn1, btn2, and btn3 parameters specify the
text to appear in the buttons. If the text is an empty string (""), the button is
not displayed.
The function returns the button number from one to three that was chosen.
The defaultBtn parameter in GUI.ChooseFull specifies which, if any, button
should be the default button. The default button is selected if the user
presses Enter. If the default button is 0, then no button is highlighted as the
default button.
Note: This function is not available in the current version of the GUI
Procedure Library (shipping with WinOOT 3.1, DOS OOT 2.5 and
MacOOT 1.5). It is documented here for use with future shipping version
of Object Oriented Turing. It is likely to be implemented in the version of
Object Oriented Turing released in September 2000. Check the release
notes that are found in the on-line help to find out if this function is now
available.
Example
The following program asks if the user wants coffee or tea and set
wantsCoffee appropriately.
import GUI in "%oot/lib/GUI"
var wantsCoffee : boolean
var choice : int := GUI.Choose ("Beverage Choice",
"Do you want coffee or tea?", "", "", "Coffee", "Tea", "")
if choice = 1 then
wantsCoffee := true
else
wantsCoffee := false
end if
300 Object Oriented Turing Reference Manual
The following program asks the user whether they want to save their work,
don't save their work or Cancel.
Example
import GUI in "%oot/lib/GUI"
% Returns false if cancelling operation
procedure CheckUnsavedWork : boolean
var message : array 1 .. 3 of string
message (1) := "Changes to " + fileName + " have not been "
message (2) := "saved. Unsaved work will be lost. Do you "
message (3) := "want to save before quitting."
var choice : int := GUI.ChooseFull ("Save Before Quit",
message, "Save", "Don't Save", "Cancel", 1)
if choice = 1 then
SaveWork
elsif choice = 3 then
return false
end if
return true
end CheckUnsavedWork
Exported qualified.
Status
This means that you can only call the function by calling GUI.Choose, not
by calling Choose.
Win DOS Mac
GUI.ClearText
X TOOT Term
Syntax
GUI.ClearText ( widgetID : int )
Description
Clears all the text in a text box specified by widgetID. The widgetID
parameter must be the widget id of a text box.
Example
The program lists 25 numbers in a text box. Every time the button is
pressed, it clears the text box and prints the next 25 numbers.
import GUI in "%oot/lib/GUI"
var boxID, buttonID, start : int
start := 1
procedure PrintTwentyFive
GUI.ClearText (boxID)
for i : start .. start + 24
GUI.AddLine (boxID, intstr (i))
end for
start += 25
end PrintTwentyFive
boxID := GUI.CreateTextBox (50, 50, 200, 200)
Chapter 7 : Language Features 301
buttonID := GUI.CreateButton (50, 5, 0, "Next 25", PrintTwentyFive)
PrintTwentyFive
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling GUI.ClearText,
not by calling ClearText.
GUI.CreateTextBox.
See also
Win DOS Mac
GUI.CloseWindow
X TOOT Term
Syntax
GUI.CloseWindow ( window : int )
Description
Closes a window with widgets in it. This procedure automatically disposes
of any widgets in the window and makes certain that the GUI Library
recognizes that the window no longer exists. This procedure will call
Window.Close, so there is no need for the user to do so.
Example
The program opens up a window with two buttons. If the button labelled
"Close and Open" is pressed, the window is closed and a new window with
two buttons is opened in a random location on the screen.
import GUI in "%oot/lib/GUI"
const screenWidth : int := Config.Display (cdScreenWidth)
const screenHeight : int := Config.Display (cdScreenHeight)
const titleBarHeight : int := 32
const windowEdgeSize : int := 13
const windowWidth : int := 150
const windowHeight : int := 100
var windowID, windowNumber, closeButton, quitButton : int := 0
procedure CloseAndOpen
if windowID not= 0 then
GUI.CloseWindow (windowID)
end if
windowNumber += 1
var xPos : int := Rand.Int (0, screenWidth – windowWidth –
windowEdgeSize)
var yPos : int := Rand.Int (0, screenHeight – windowHeight –
titleBarHeight)
windowID := Window.Open ("title:Window #" +
intstr (windowNumber) + ",graphics:" +
intstr (windowWidth) + ";" + intstr (windowHeight) +
302 Object Oriented Turing Reference Manual
",position:" + intstr (xPos) + ";" + intstr (yPos))
closeButton := GUI.CreateButton (10, 60, 130,
"Close And Open", CloseAndOpen)
quitButton := GUI.CreateButton (10, 10, 130, "Quit", GUI.Quit)
end CloseAndOpen
CloseAndOpen
loop
exit when GUI.ProcessEvent
end loop
GUI.CloseWindow (windowID)
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CloseWindow, not by calling CloseWindow.
Win DOS Mac
GUI.CreateButton[Full]
Syntax
X TOOT Term
GUI.CreateButton ( x, y, width : int, text : string,
actionProc : procedure x () ) : int
GUI.CreateButtonFull ( x, y, width : int, text : string,
actionProc : procedure x (), height : int,
shortcut : char, default : boolean ) : int
Description
Creates a button and returns the button's widget ID.
The button widget is used to implement a textual button. When you click
on a button, the button's action procedure is called. If a button is given a
short cut, then entering the keystroke will cause the action procedure to be
called. It will not visibly cause the button to depress.
Chapter 7 : Language Features 303
Two Buttons
The x and y parameters specify the lower-left corner of the button. The
width parameter specifies the width of the button. If width is less than the
size necessary to display the button, the button is automatically enlarged to
fit the text. The text parameter specifies the text to appear in the button.
The actionProc parameter is the name of a procedure that is called when the
button is pressed.
For GUI.CreateButtonFull, the height parameter specifies the height of the button. If height
is less than the size necessary to display the button, the button is
automatically enlarged to fit the text. The shortcut parameter is the
keystroke to be used as the button's shortcut. The default parameter is a
boolean indicating whether the button should be the default button. If
there is already a default button, and default is set to true, then this button
becomes the new default button.
Example
The following program creates two buttons, one which draws a random
circle on the screen and one which quits the program.
import GUI in "%oot/lib/GUI"
procedure DrawRandomCircle
var r : int := Rand.Int (20, 50)
var x : int := Rand.Int (r, maxx – r)
var y : int := Rand.Int (r, maxy – r)
var c : int := Rand.Int (0, maxcolor)
Draw.FillOval (x, y, r, r, c)
% In case we drew over the buttons, redraw them.
GUI.Refresh
end DrawRandomCircle
View.Set ("graphics:300;200")
var draw : int := GUI.CreateButtonFull (50, 10, 0, "Draw Circle",
DrawRandomCircle, 0, '^D', true)
var quitBtn : int := GUI.CreateButton (200, 10, 0, "Quit", GUI.Quit)
loop
304 Object Oriented Turing Reference Manual
exit when GUI.ProcessEvent
end loop
Details
When GUI.CreateButton or GUI.CreateButtonFull is called, the newly
created button will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
If a button's width or height is set to zero (or not specified at all), then the
button is shaped to fit the text.
A button can be the default button for a window. The default button is
drawn with a thicker border around it. If the user presses ENTER in a
window with a default button, the default button's action procedure is
called.
When a button is not enabled, the text in the button is grayed out and the
button no longer responds to any mouse clicks or keystrokes until the
button is enabled again.
Details
The following GUI subprograms can be called with a button as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.SetLabel, GUI.SetDefault
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateButton, not by calling CreateButton.
See also
GUI.SetLabel for changing the button's text and GUI.SetDefault for
setting the default button in a window.
Win DOS Mac
GUI.CreateCanvas[Full]
Syntax
X TOOT Term
GUI.CreateCanvas ( x, y, width, height : int ) : int
GUI.CreateCanvasFull ( x, y, width, height : int,
border : int,
mouseDown : procedure x (mx, my : int),
mouseDrag : procedure x (mx, my : int),
mouseUp : procedure x (mx, my : int) ) : int
Chapter 7 : Language Features 305
Description
Creates a canvas and returns the canvas' widget ID.
A canvas is a drawing surface for use by the program. It differs from just
using the window surface to draw on in that (0, 0) represents the lower-left
corner of the canvas and all drawing is clipped to the canvas. (This means
that if you accidently attempt to draw outside of the canvas, it will not
actually draw beyond the border of the canvas.)
Canvases have procedures that emulate all the procedures in the Draw
module as well as a procedure to emulate Font.Draw, Pic.Draw, Pic.New,
Pic.ScreenLoad and Pic.ScreenSave.
You can get mouse feedback from a canvas. Using the CreateCanvasFull
method, you can specify three routines that are called when the mouse
button is depressed while pointing in a canvas. One routine will be called
when the user presses the mouse button down in a canvas. Another routine
will be called while the user drags the mouse with the mouse button down.
This routine is repeatedly called whenever the mouse changes position
while the mouse button is down. The last routine is called when the mouse
button is released. All three routines take an x and y parameter, which is
the location of the mouse with respect to the canvas (i.e. (0, 0) is the lowerleft corner of the canvas).
Output of Canvases.dem
The x and y parameters specify the lower-left corner of the canvas. The
width and height parameters specify the width and height of the canvas.
For GUI.CreateCanvasFull, the border parameter specifies the type of
border that surrounds the canvas and is one of 0, GUI.LINE, GUI.INDENT
or GUI.EXDENT. A border of 0 is the default and is the same as GUI.LINE.
GUI.INDENT and GUI.EXDENT only display properly if the background
colour has been set to gray using GUI.SetBackgroundColor. GUI.INDENT
makes the canvas appear indented or recessed. GUI.EXDENT makes the
canvas appear to stand out from the window.
306 Object Oriented Turing Reference Manual
The mouseDown parameter is a procedure called when the user presses the
mouse button in the canvas. The mouseDrag parameter is a procedure
called when the user drags the mouse while the mouse button is still
pressed. The mouseUp parameter is a procedure called when the user
releases the mouse button. The parameters to all three are the x and y
location of the mouse where the button was pressed (dragged/released).
The coordinates are given with respect to the canvas (i.e. (0, 0) is the lowerleft corner of the canvas).
Example
The following program draws 10 random stars in the canvas.
import GUI in "%oot/lib/GUI"
var canvas : int := GUI.CreateCanvas (10, 10, maxx – 20, maxy – 20)
for i : 1 .. 10
var x : int := Rand.Int (0, maxx – 20)
var y : int := Rand.Int (0, maxy – 20)
var c : int := Rand.Int (0, maxcolor)
GUI.DrawFillOval (canvas, x, y, 20 20, c)
end for
Details
When GUI.CreateCanvas or GUI.CreateCanvasFull is called, the newly
created canvas will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
The border of the canvas is just outside the drawing surface, so GUI.GetWidth and
GUI.GetHeight will return slight larger values than width and height.
When the canvas is disabled, clicking the mouse in the canvas does not call
any of the mouseDown, mouseDrag, or mouseUp procedures. The appearance
of the canvas does not change.
Details
The following GUI subprograms can be called with a button as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.Draw…, GUI.FontDraw, GUI.Pic…, GUI.SetXOR
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateCanvas, not by calling CreateCanvas.
See also
GUI.Draw…, GUI.FontDraw, GUI.Pic…, and GUI.SetXOR for drawing
on a canvas.
Chapter 7 : Language Features 307
Win DOS Mac
GUI.CreateCheckBox[Full]
Syntax
X TOOT Term
GUI.CreateCheckBox ( x, y : int, text : string,
actionProc : procedure x (filled : boolean) ) : int
GUI.CreateCheckBoxFull ( x, y : int, text : string,
actionProc : procedure x (filled : boolean),
alignment : int, shortcut : char ) : int
Description
Creates a check box (with accompanying text) and returns the check box's
widget ID.
The check box widget is used to implement a check box that can be set or
unset. When you click on a check box, the status of the check box flips from
set to unset and back again and the check box's action procedure is called
with the new status as a parameter. If a check box is given a short cut, then
entering the keystroke will cause the check box to change status and the
action procedure to be called. The new status will be displayed immediately.
Two Check Boxes
The x and y parameters specify the lower-left corner of the check box
(unless alignment is set to GUI.RIGHT, in which case they specify the lowerright corner of the check box). The text parameter specifies the text (or
label) beside the check box. The actionProc parameter is the name of a
procedure that is called when the status of the check box changes. The
actionProc procedure must have one boolean parameter which is the new
status of the check box. In GUI.CreateCheckBox, the check box's text is
always to the right of the actual check box. In GUI.CreateCheckBoxFull,
the text can be set to the right or left of the check box with the alignment
parameter.
308 Object Oriented Turing Reference Manual
For GUI.CreateCheckBoxFull, the alignment parameter specifies the position of the check
box in relation to the text as well as the meaning of the x and y parameters.
The alignment parameter is one of 0, GUI.LEFT, or GUI.RIGHT. An
alignment of 0 is the default and is the same as GUI.LEFT. GUI.LEFT means
the actual box in the check box appears to the left of the check box's label
and (x, y) specifies the lower-left corner. An alignment of GUI.RIGHT means
that the actual box appears to the right of the check box's label and (x, y)
specifies the lower-right corner of the check box. The shortcut parameter is
the keystroke to be used as the button's shortcut. The default parameter is a
boolean indicating whether the button should be the default button. If
there is already a default button, and default is set to true, then this button
becomes the new default button.
A check box's size is not specified during creation. It is determined based
on the size of the text. Instead the user specifies the lower-left corner of the
check box (or the lower-right if the check box is right justified).
Example
The following program creates two buttons, one which draws a random
circle on the screen and one which quits the program
import GUI in "%oot/lib/GUI"
procedure DoNothing (status : boolean)
end DoNothing
View.Set ("graphics:300;100")
var cb1 : int := GUI.CreateCheckBox (10, 10, "Check Box 1",
DoNothing)
var cb2 : int := GUI.CreateCheckBoxFull (200, 10, "Check Box 2",
DoNothing, GUI.RIGHT, '2')
GUI.SetCheckBox (cb2, true)
var quitBtn : int := GUI.CreateButton (230, 10, 0, "Quit", GUI.Quit)
loop
exit when GUI.ProcessEvent
end loop
var cb1Status : boolean := GUI.GetCheckBox (cb1)
var cb2Status : boolean := GUI.GetCheckBox (cb2)
if cb1Status then
put "Check box 1: filled"
else
put "Check box 1: empty"
end if
if cb2Status then
put "Check box 2: filled"
else
put "Check box 2: empty"
end if
Details
When GUI.CreateButton or GUI.CreateButtonFull is called, the newly
created check box will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
Chapter 7 : Language Features 309
When a check box is not enabled, the label beside the check box is grayed
out and the check box no longer responds to any mouse clicks or
keystrokes until the check box is enabled again.
Details
The following GUI subprograms can be called with a check box as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.SetLabel, GUI.GetCheckBox, GUI.SetCheckBox
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateCheckBox, not by calling CreateCheckBox.
See also
GUI.SetLabel for changing the chec box's text and GUI.GetCheckBox and
GUI.SetCheckBox for reading and setting the check box's state.
Win DOS Mac
GUI.CreateFrame
X TOOT Term
Syntax
GUI.CreateFrame ( x1, y1, x2, y2, kind : int ) : int
Description
Creates a frame and returns the frame's widget ID.
A frame is a box drawn around other GUI widgets to make the window
look better and help organize the GUI elements.
Three Types of Frames With a Label in Each Frame
310 Object Oriented Turing Reference Manual
Frames are the only GUI widgets that can have other widgets placed
within them. Frames are passive widgets, meaning that they do not
respond to button clicks or keystrokes.
The x1 and y1 parameters specify the lower-left corner of the frame and the
x2 and y2 parameters specify the upper-right corner of the frame. The kind
parameter specifies the type of frame. This is one of 0, GUI.LINE,
GUI.INDENT, or GUI.EXDENT. A kind of 0 is the default and is the same
as GUI.LINE.
GUI.INDENT and GUI.EXDENT only display properly if the background
colour has been set to gray using GUI.SetBackgroundColor. GUI.INDENT
makes the contents frame appear indented or recessed. GUI.EXDENT
makes the contents of the frame appear to stand out from the window.
Example
The following program draws three frames in the window and draws a
label in each one.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:250;90")
GUI.SetBackgroundColor (gray)
var lineFrame, indentFrame, exdentFrame : int
var lineLabel, indentLabel, exdentLabel : int
lineFrame := GUI.CreateFrame (10, 10, 80, 70, 0)
indentFrame := GUI.CreateFrame (90, 10, 160, 70, GUI.INDENT)
exdentFrame := GUI.CreateFrame (170, 10, 240, 70, GUI.EXDENT)
% Label the lines.
lineLabel := GUI.CreateLabelFull (10, 10, "Line", 70, 60,
GUI.CENTER + GUI.MIDDLE, 0)
indentLabel := GUI.CreateLabelFull (90, 10, "Indent", 70, 60,
GUI.CENTER + GUI.MIDDLE, 0)
exdentLabel := GUI.CreateLabelFull (170, 10, "Exdent", 70, 60,
GUI.CENTER + GUI.MIDDLE, 0)
Details
When GUI.CreateFrame is called, the newly created frame will be
displayed immediately unless GUI.DisplayWhenCreated has been called
with the display parameter set to false.
A frame widget is a passive widget and cannot be enabled or disabled.
Details
The following GUI subprograms can be called with a frame as the widgetID
parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateFrame, not by calling CreateFrame.
Chapter 7 : Language Features 311
GUI.CreateHorizontalScrollBar[Full]
Syntax
GUI.CreateHorizontalScrollBar ( x, y, size : int,
min, max, start : int,
actionProc : procedure x (value : int) ) : int
GUI.CreateHorizontalScrollBarFull ( x, y, size : int,
min, max, start : int,
actionProc : procedure x (value : int),
arrowInc, pageInc, thumbSize : int) : int
Description
Creates a horizontal (left-right) scroll bar and returns the scroll bar's
widget ID.
A scroll bar is a widget that allows users to see a piece of a document that
cannot be displayed on the screen in its entirety. The picture below shows a
horizontal scroll bar. To control a scroll bar, there are a few choices: the
user can click on the thumb (the box in the scroll bar) and slide it left or
right, or the user can click in the scroll bar itself to the left or right of the
thumb (in which case the thumb is moved up or down one "page"), or the
user can click on the left or right arrows at the ends of the scroll bar (in
which case the thumb is moved left or right one "arrow increment").
A Horizontal Scroll Bar
The programmer defines a page or an arrow increment. When the value of
the scroll bar changes, the action procedure of the scroll bar is called with the
new value as a parameter. The action procedure should then redraw the
contents using the new value of the scroll bar.
The range of values that the scroll bar will give is determined by the min
and max parameters in the Create call. The left side of the scroll bar
represents the minimum value, while the right represents the maximum
value. There is also the "thumb size". This represents the range of values
that can be seen at once on the screen.
312 Object Oriented Turing Reference Manual
By default, the arrow increment (the amount the value is changed when the scrolling
arrows are pressed) is set to one. The page increment (the amount the
value is changed when the user clicks in the bar to the right or left of the
thumb) is set to one quarter the difference between the minimum and the
maximum. The "thumb size" is set to zero (see the description of scroll bars
for an explanation of the thumb size).
The x and y parameters specify the lower-left corner of the scroll bar. The
size parameter specifies the length of the scroll bar (including the arrows)
in pixels. The min and max parameters are the minimum and maximum
values returned by the scroll bar. The start parameter is the initial value of
the scroll bar and should be between min and max inclusive. The actionProc
parameter is the name of a procedure that is called when the value of the
scroll bar is changed. The parameter to the action procedure is the current
value of the scroll bar.
The following program creates a horizontal scroll bar. Whenever the scroll
bar's value is changed, a message is displayed in the window.
Example
import GUI in "%oot/lib/GUI"
View.Set ("graphics:300;60")
var scrollBar : int
procedure ScrollBarMoved (value : int)
Text.Locate (2, 3)
put "Horizontal Scroll Bar: ", value : 4
end ScrollBarMoved
scrollBar := GUI.CreateHorizontalScrollBar (10, 10, 250,
50, 150, 50, ScrollBarMoved)
loop
exit when GUI.ProcessEvent
end loop
Description
For GUI.CreateHorizontalScrollBarFull, the arrowInc parameter specifies
the arrow increment (the amount the scroll bar's value is changed when the
scroll arrows are pressed). The pageInc specifies the page increment (the
amount the scroll bar's value is changed when the user clicks in the page
left/right section of the scroll bar). The thumbSize parameter specifies the
"thumb size". (See the scroll bar explanation for more detail on a scroll bar's
"thumb size").
For example, if you have a window that can display 20 lines of text at once
and there are 100 lines of text, you would set min to 1, max to 100 and
thumbSize to 20. The value returned by the scroll bar would then be the line
number of the first line on the screen to be displayed. When the scroll bar
was at its maximum value, it would return 81, since by doing so, lines 81100 would be displayed.
Example
Here is an example program that scrolls a large picture over a smaller
window.
% The "ScrollPic" program.
import GUI in "%oot/lib/GUI"
Chapter 7 : Language Features 313
var h, v : int
% The scroll bars.
var canvas : int
% The canvas.
var pic : int
% The picture.
const width : int := 220
% The width of the canvas.
procedure ScrollPic (ignore : int)
% Get the current value of the scroll bars.
var x : int := GUI.GetSliderValue (h)
var y : int := GUI.GetSliderValue (v)
GUI.PicDraw (canvas, pic, –x, –y, picCopy)
end ScrollPic
pic := Pic.FileNew ("Forest.bmp")
if pic <= 0 then
put "Error loading picture: ", Error.LastMsg
return
end if
View.Set ("graphics:265;265")
canvas := GUI.CreateCanvas (15, 15 + GUI.GetScrollBarWidth,
width, width)
% Note the frame of the canvas is:
% (14, 14 + ScrollbarWidth) - (235, 235 + ScrollbarWidth)
h := GUI.CreateHorizontalScrollBarFull (14, 14,
221, 0, Pic.GetWidth (pic) , 0, ScrollPic, 3, 100, width)
v := GUI.CreateVerticalScrollBarFull (235,
14 + GUI.GetScrollBarWidth, 221, 0, Pic.GetHeight (pic),
Pic.GetHeight (pic), ScrollPic, 3, 100, width)
ScrollPic (0) % Draw the picture initially
loop
exit when GUI.ProcessEvent
end loop
Details
In some instances, you will want the the minimum and maximum values of
the scroll bar to be reversed (right/top is minimum). In that case, call the
GUI.SetSliderReverse procedure to flip the values of the scroll bar.
Scroll bars always have a fixed height (for horizontal scroll bars) or width
(for vertical scroll bars). To get a scroll bar's width, use the
GUI.GetScrollBarWidth function.
When GUI.CreateHorizontalScrollBar or GUI.CreateHorizontalScrollBarFull is called,
the newly created scroll bar will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a scroll bar is not enabled, the gray in the bar is set to white and the
thumb is not displayed. The scroll bar no longer responds to any mouse
clicks until the scroll bar is enabled again.
314 Object Oriented Turing Reference Manual
The following GUI subprograms can be called with a scroll bar as the
widgetID parameter:
Details
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.GetSliderValue, GUI.SetSliderValue,
GUI.SetSliderMinMax,GUI.SetSliderSize,
GUI.SetSliderReverse, GUI.SetScrollAmount
Exported qualified.
Status
This means that you can only call the function by calling
GUI.CreateHorizontalScrollBar, not by calling
CreateHorizontalScrollBar.
See also
GUI.GetSliderValue and GUI.SetSliderValue for reading and setting the
value of a scroll bar, GUI.SetSliderMinMax for changing the minimum
and maximum values of a scroll bar, and GUI.SetScrollAmount for
changing the scrolling increments and thumb size of a scroll bar. See also
GUI.SetSliderSize for setting the length of a scroll bar and
GUI.SetSliderReverse for reversing the sense of a scroll bar.
Win DOS Mac
GUI.CreateHorizontalSlider
X TOOT Term
Syntax
GUI.CreateHorizontalSlider ( x, y, length : int,
min, max, start : int,
actionProc : procedure x (value : int) ) : int
Description
Creates a horizontal (left-right) slider and returns the slider's widget ID.
A slider is a widget that allows the user to set a continuous set of values. It
has a real-life equivalent in things such as a stereo volume control.
A Horizontal Slider
Chapter 7 : Language Features 315
To control a slider, the user clicks on the slider box and drags it back and
forth. Every time the value changes, a procedure is called with the new
value as a parameter.
The range of values that the slider will give is determined by the min and
max parameters in the Create call. The left side of the slider represents the
minimum value, while the right represents the maximum value.
The x and y parameters specify the lower-left corner of the slider track.
This means that the slider actually extends above and below this point (and
slightly to the left of it to take into account the rounded end of the track).
The length parameter specifies the length of the track in pixels. (You can
use GUI.GetX, GetY, GetWidth, and GetHeight to get the exact
dimensions of the slider.) The min and max parameters are the minimum
and maximum values returned by the slider. The start parameter is the
initial value of the slider and should be between min and max inclusive.
The actionProc parameter is the name of a procedure that is called when the
value of the slider is changed. The parameter to the action procedure is the
current value of the slider.
Example
The following program creates a horizontal slider. Whenever the slider's
value is changed, a message is displayed in the window.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:300;60")
var slider : int
procedure SliderMoved (value : int)
Text.Locate (2, 3)
put "Horizontal Slider: ", value : 4
end SliderMoved
slider := GUI.CreateHorizontalSlider (10, 10, 250,
50, 150, 50, SliderMoved )
loop
exit when GUI.ProcessEvent
end loop
Details
In some instances, you will want the the minimum and maximum values of
the slider to be reversed (right is minimum). In that case, call the
GUI.SetSliderReverse procedure to flip the values of the slider.
Sliders always have a fixed height (for horizontal sliders) or width (for
vertical sliders).
When GUI.CreateHorizontalSlideror GUI.CreateHorizontalSliderFull is
called, the newly created slider will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a slider is not enabled, the appearance does not change. However,
the slider no longer responds to any mouse clicks until it is enabled again.
Details
The following GUI subprograms can be called with a slider as the widgetID
parameter:
316 Object Oriented Turing Reference Manual
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.GetSliderValue, GUI.SetSliderValue,
GUI.SetSliderMinMax, GUI.SetSliderSize,
GUI.SetSliderReverse
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateHorizontalSlider, not by calling CreateHorizontalSlider.
See also
GUI.GetSliderValue and GUI.SetSliderValue for reading and setting the
value of a slider, GUI.SetSliderMinMax for changing the minimum and
maximum values of a slider. See also GUI.SetSliderSize for setting the
length of a slider and GUI.SetSliderReverse for reversing the sense of a
slider.
Win DOS Mac
GUI.CreateLabel[Full]
Syntax
X TOOT Term
GUI.CreateLabel ( x, y : int, text : string ) : int
GUI.CreateLabelFull ( x, y : int, text : string,
width, height, alignment, fontID : int ) : int
Description
Creates a label and returns the label's widget ID.
The label widget is used to display text. It can be used to display text in a
variety of fonts and sizes. Label widgets can also be aligned in a variety of
ways.
Chapter 7 : Language Features 317
Three Labels
The x and y parameters specify the lower-left corner of the area in which
the text will be drawn. For GUI.CreateLabel, this is the lower-left corner of
the text. The text parameter specifies the text of the label.
For GUI.CreateLabelFull, the width and height parameters specify the area
in which the label is to appear. This is used for alignment purposes. See the
program below for an example of aligning the text to different corners of
the window. The alignment parameter specifies the alignment of the text in
the text area. This value is the sum of horizontal alignment and the vertical
alignment. The horizontal alignment is one of 0, GUI.LEFT, GUI.CENTER,
or GUI.RIGHT. A horizontal alignment of 0 is the default and is the same as
the alignment of GUI.LEFT. The vertical alignment is one of 0, GUI.TOP,
GUI.MIDDLE, or GUI.BOTTOM. A horizontal alignment of 0 is the default
and is the same as the alignment of GUI.BOTTOM. These alignments align
the text in various ways in the text area. The fontID parameter specifies the
font ID of the font to be used in the text field. The font ID is received from a
Font.New call. Do not call Font.Free for this font ID until the label has been
disposed of by calling GUI.Dispose.
By using the fondID parameter, labels can be have any size or typeface.
Labels are passive widgets, meaning that they do not respond to button
clicks or keystrokes.
Example
The following program creates three labels, one with the default alignment,
the other two aligned to appear in the center and upper-right corner of the
window.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:300;100")
var lowerLeft : int := GUI.CreateLabel (0, 0, "Lower-Left")
var center : int := GUI.CreateLabelFull (0, 0, "Center", maxx, maxy,
GUI.MIDDLE + GUI.CENTER, 0)
var upperRight : int := GUI.CreateLabelFull (0, 0, "Upper-Right",
maxx, maxy, GUI.RIGHT + GUI.TOP, 0)
Details
When GUI.CreateLabel or GUI.CreateLabelFull is called, the newly
created label will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
A frame widget is a passive widget and cannot be enabled or disabled.
Details
The following GUI subprograms can be called with a label as the widgetID
parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.GetSliderValue, GUI.SetSliderValue,
GUI.SetSliderMinMax, GUI.SetLabel
318 Object Oriented Turing Reference Manual
Exported qualified.
Status
This means that you can only call the function by calling GUI.CreateLabel,
not by calling CreateLabel.
See also
GUI.SetLabel for changing the label's text.
Win DOS Mac
GUI.CreateLabelledFrame
X TOOT Term
Syntax
GUI.CreateLabelledFrame ( x1, y1, x2, y2, kind : int,
text : string ) : int
Description
Creates a labelled frame and returns the frame's widget ID.
A labelled frame is a box with a text label drawn around other GUI
widgets to make the window look better and help organize the GUI
elements.
Three Types of Labelled Frames
Frames and labelled frames are the only GUI widgets that can have other
widgets placed within them. Labelled frames are passive widgets, meaning
that they do not respond to button clicks or keystrokes.
The x1 and y1 parameters specify the lower-left corner of the frame and the
x2 and y2 parameters specify the upper-right corner of the frame. (The text
will extend above the frame.) The kind parameter specifies the type of
frame. This is one of 0, GUI.LINE, GUI.INDENT or GUI.EXDENT. A kind of
0 is the default and is the same as GUI.LINE.
GUI.INDENT and GUI.EXDENT only display properly if the background
colour has been set to gray using GUI.SetBackgroundColor. GUI.INDENT
makes the contents frame appear indented or recessed. GUI.EXDENT
makes the contents of the frame appear to stand out from the window.
Chapter 7 : Language Features 319
Example
The following program draws three frames in the window.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:250;90")
GUI.SetBackgroundColor (gray)
var lineFrame, indentFrame, exdentFrame : int
lineFrame := GUI.CreateFrame (10, 10, 80, 70, 0, "Line")
indentFrame := GUI.CreateFrame (90, 10, 160, 70, GUI.INDENT,
"Indent")
exdentFrame := GUI.CreateFrame (170, 10, 240, 70, GUI.EXDENT,
"Exdent")
Details
When GUI.CreateLabelledFrame is called, the newly created labelled
frame will be displayed immediately unless GUI.DisplayWhenCreated
has been called with the display parameter set to false.
A labelled frame widget is a passive widget and cannot be enabled or
disabled.
Details
The following GUI subprograms can be called with a labelled frame as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.SetLabel
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateLabelledFrame, not by calling CreateLabelledFrame.
See also
GUI.SetLabel for changing the frame's text.
Win DOS Mac
GUI.CreateLine
X TOOT Term
Syntax
GUI.CreateLine ( x1, y1, x2, y2, kind : int ) : int
Description
Creates a line and returns the line's widget ID.
Lines are generally used to separate parts of a window. A line is used to
make the window look better and help organize the GUI elements.
320 Object Oriented Turing Reference Manual
Three Types of Lines
Lines are passive widgets, meaning that they do not respond to button
clicks or keystrokes.
The x1 and y1 parameters specify one end-point of the line and the x2 and
y2 parameters specify the other end point. The line must either be
horizontal or vertical (i.e. x1 must equal x2 or y1 must equal y2). The kind
parameter specifies the type of line. This is one of 0, GUI.LINE,
GUI.INDENT or GUI.EXDENT. A kind of 0 is the default and is the same as
GUI.LINE.
GUI.INDENT and GUI.EXDENT only display properly if the background
colour has been set to gray using GUI.SetBackgroundColor. GUI.INDENT
makes the line appear indented or recessed. GUI.EXDENT makes the line
appear to stand out from the window.
Example
The following program draws three lines with three labels in the window.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:180;100")
GUI.SetBackgroundColor (gray)
var line indentLine, exdentLine : int
var lineLabel, indentLabel, exdentLabel : int
line := GUI.CreateLine (30, 20, 30, 90, 0)
indentLine := GUI.CreateLine (90, 20, 90, 90, GUI.INDENT)
exdentLine := GUI.CreateLine (150, 20, 150, 90, GUI.EXDENT)
% Label the lines.
lineLabel := GUI.CreateLabelFull (30, 15, "Line", 0, 0,
GUI.CENTER + GUI.TOP, 0)
indentLabel := GUI.CreateLabelFull (90, 15, "Indent", 0, 0,
GUI.CENTER + GUI.TOP, 0)
exdentLabel := GUI.CreateLabelFull (150, 15, "Exdent", 0, 0,
GUI.CENTER + GUI.TOP, 0)
Details
When GUI.CreateLine is called, the newly created line will be displayed
immediately unless GUI.DisplayWhenCreated has been called with the
display parameter set to false.
A line widget is a passive widget and cannot be enabled or disabled.
Chapter 7 : Language Features 321
Details
The following GUI subprograms can be called with a line as the widgetID
parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize
Status
Exported qualified.
This means that you can only call the function by calling GUI.CreateLine,
not by calling CreateLine.
Win DOS Mac
GUI.CreateMenu
X TOOT Term
Syntax
GUI.CreateMenu ( name : string ) : int
Description
Creates a menu and returns the menu's widget ID. The menu will be added
after the other menus in the menu bar. If there are no previous menus, then
a menu bar is automatically created and the menu added.
The name parameter specifies the text that appears in the menu bar.
A Menu With an Item Selected
Menus are used in most modern interfaces. In order to create a full set of
menus, you must create the menu and then create the menu items in that
menu. The menus are automatically added to the menu bar of the selected
menu.
322 Object Oriented Turing Reference Manual
As of the v1.0 release of the GUI Library, it is an error to create a menu
item without having created a menu first. In future releases it will be
possible to create menus and attach and remove them from menu bars
when desired.
Example
The following program creates a series of menus with menu items in them.
It then disables the second menu.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:250;150")
var first, second : int
% The menus.
var item : array 1 .. 12 of int
% The menu items.
var name : array 1 .. 12 of string (20) :=
init ("Quit", "---", "A", "B", "---", "C", "D",
"Disable B Menu Item", "Enable B Menu Item", "---",
"Disable Second Menu", "Enable Second Menu")
procedure MenuSelected
for i : 1 .. 12
if item (i) = GUI.GetEventWidgetID then
Text.Locate (maxrow, 1)
put name (i) + " selected
" ..
end if
end for
end MenuSelected
procedure DisableB
GUI.Disable (item (4))
end DisableB
procedure EnableB
GUI.Enable (item (4))
end EnableB
procedure DisableFirst
GUI.Disable (first)
end DisableFirst
procedure EnableFirst
GUI.Enable (first)
end EnableFirst
% Create the menus
first := GUI.CreateMenu ("First")
item (1) := GUI.CreateMenuItem (name (1), GUI.Quit)
for cnt : 2 .. 7
item (cnt) := GUI.CreateMenuItem (name (cnt),
MenuSelected)
end for
second := GUI.CreateMenu ("Second")
item (8) := GUI.CreateMenuItem (name (8), DisableB)
item (9) := GUI.CreateMenuItem (name (9), EnableB)
item (10) := GUI.CreateMenuItem (name (10), MenuSelected)
Chapter 7 : Language Features 323
item (11) := GUI.CreateMenuItem (name (11), DisableFirst)
item (12) := GUI.CreateMenuItem (name (12), EnableFirst)
loop
exit when GUI.ProcessEvent
end loop
Details
When a menu is not enabled, the text in the menu bar is grayed out and
clicking on the menu does not cause the menu to appear.
Details
The following GUI subprograms can be called with a menu as the widgetID
parameter:
GUI.Show, GUI.Hide, GUI.Dispose, GUI.Enable, GUI.Disable
Exported qualified.
Status
This means that you can only call the function by calling GUI.CreateMenu,
not by calling CreateMenu.
See also
GUI.CreateMenuItem for adding items to a menu. See also
GUI.ShowMenuBar and GUI.HideMenuBar for showing and hiding the
menu bar.
Win DOS Mac
GUI.CreateMenuItem[Full]
Syntax
X TOOT Term
GUI.CreateMenuItem ( name : string,
actionProc : procedure x () ) : int
GUI.CreateMenuItemFull ( name : string,
actionProc : procedure x (), shortCut : char,
addNow : boolean) : int
Description
Creates a menu item and returns the menu item's widget ID.
Menu items are the individual entries of a menu. To create menus for a
window, you must create a menu, then create the menu items for that
menu, then create the next menu, etc. All menu items are automatically
added to the last menu and after the last menu item of the currently
selected (not active!) window.
The menu item will be added to the last menu after the other menu items
in the menu. If there are no menus defined, an error results.
324 Object Oriented Turing Reference Manual
The name parameter specifies the text that is to appear. A name of three
dashes ("---") creates a separator across the menu. The actionProc parameter
specifies the name of a procedure to be called when user the selects the
menu item from the menu.
For GUI.CreateMenuItemFull, the shortCut parameter specifies the
keystroke to be used as the menu item's shortcut. If no shortcut is desired,
then '\0' can be used. The addNow parameter has no effect in the current
version of the GUI Library. In future versions, it will allow you to create
menu items that can then be added to a menu later in the program.
Examples
See the example for GUI.CreateMenu.
Details
When a menu item is not enabled, the text of the menu item is grayed out
and clicking on the menu item does not cause the menu to appear.
Details
The following GUI subprograms can be called with a menu as the widgetID
parameter:
GUI.Show, GUI.Hide, GUI.Dispose, GUI.Enable, GUI.Disable
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateMenuItem, not by calling CreateMenuItem.
Win DOS Mac
GUI.CreatePicture
X TOOT Term
Syntax
GUI.CreatePicture ( x, y, picture : int,
mergePic : boolean ) : int
Description
Creates a picture and returns the picture's widget ID.
The picture widget is used to display a picture. It can be used to display a
picture either merged into the background or not.
The x and y parameters specify the lower-left corner of the picture. The
picture parameter specifies the picture ID of the picture. The picture ID is
received from a Pic.New or Pic.FileNew call. Do not call Pic.Free for this
picture ID until the button has been disposed of by calling GUI.Dispose.
The mergePic parameter is a boolean that specifies whether anything that
was the background colour in the picture (usually colour 0) should be set
to the background colour of the window.
A picture widget is a passive widget and cannot be enabled or disabled.
Chapter 7 : Language Features 325
Two Pictures
Example
The following program draws two pictures, merged and not merged.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:230;135")
% We'll need to create a picture for our Picture widget. Normally
% an external file (and Pic.FileNew) would be used.
Draw.FillOval (50, 50, 50, 50, blue)
Draw.FillBox (17, 17, 83, 83, brightred)
Draw.FillStar (17, 17, 83, 83, brightgreen)
Draw.FillMapleLeaf (37, 37, 63, 63, brightpurple)
var pic : int := Pic.New (0, 0, 100, 100)
var picture1, picture2 : int
var label1, label2 : int
GUI.SetBackgroundColor (gray)
label1 := GUI.CreateLabel (15, 5, "Picture (no merge)")
picture1 := GUI.CreatePicture (10, 25, pic, false)
label2 := GUI.CreateLabel (135, 5, "Picture (merge)")
picture2 := GUI.CreatePicture (120, 25, pic, true)
Details
When GUI.CreatePicture is called, the newly created picture will be
displayed immediately unless GUI.DisplayWhenCreated has been called
with the display parameter set to false.
A picture widget is a passive widget and cannot be enabled or disabled.
Details
The following GUI subprograms can be called with a picture as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize
326 Object Oriented Turing Reference Manual
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreatePicture, not by calling CreatePicture.
Win DOS Mac
GUI.CreatePictureButton[Full]
Syntax
X TOOT Term
GUI.CreatePictureButton ( x, y, picture : int,
actionProc : procedure x () ) : int
GUI.CreatePictureButtonFull ( x, y, picture : int,
actionProc : procedure x (), width, height : int,
shortcut : char, mergePic : boolean ) : int
Description
Creates a picture button and returns the button's widget ID.
Picture buttons behave like buttons (see GUI.CreateButton) except that
instead of text on the button, a picture specified by the user is displayed on
the button. The picture button widget responds to mouse clicks and
keystrokes in the same manner as a regular button widget.
The picture must be created by the program beforehand using Pic.New or
Pic.FileNew. The resulting picture can then be used as a parameter to
GUI.CreatePictureButton. In general, pictures should be a maximum of
about 50 pixels high and wide, although there is no built-in limit in the GUI
library.
The x and y parameters specify the lower-left corner of the picture button.
The picture parameter specifies the picture ID of the picture to be displayed
on the button. (Note that, in general, this picture should be fairly small.)
The picture ID is received from a Pic.New or Pic.FileNew call. Do not call
Pic.Free for this picture ID until the button has been disposed of by calling
GUI.Dispose. The actionProc parameter specifies the name of a procedure
that is called when the picture button is pressed.
For GUI.CreatePictureButtonFull, the width and height parameters specify
the width and height of the button. If they are set to 0, then the picture
radio button is automatically sized to fit the picture. If you need to know
the precise size of the button, use the GUI.GetWidth and GUI.GetHeight
functions. If width and height are larger than the picture, the picture is
centered in the button. The shortCut parameter is the keystroke to be used
as the button's shortcut. The mergePic parameter specifies whether
anything that was the background colour in the picture (usually colour 0)
should be set to the background colour of the button (which is usually
gray). This defaults to true for CreatePictureButton.
Chapter 7 : Language Features 327
Two Picture Buttons
Example
The following program displays five picture buttons which output a
message when pressed.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:100;70")
const size : int := 25
const border : int := 3
% The buttons size.
var starButton, mapleButton, starPic, mapleLeafPic : int
procedure StarPressed
Text.Locate (1, 1)
put "Star Pressed "
end StarPressed
procedure MaplePressed
Text.Locate (1, 1)
put "Maple Pressed "
end MaplePressed
% Create the pictures.
% The star.
Draw.Star (border, border, border + size, border + size, black)
Draw.Star (border + 1, border + 1, border + size – 1,
border + size – 1, black)
Draw.FillStar (border + 2, border + 2, border + size – 2,
border + size – 2, brightred)
starPic := Pic.New (0, 0, 2 * border + size, 2 * border + size)
% The mapleleaf.
Draw.FillBox (border, border, border + size, border + size, white)
Draw.MapleLeaf (border, border, border + size, border + size, black)
Draw.MapleLeaf (border + 1, border + 1, border + size – 1,
border + size – 1, black)
Draw.FillMapleLeaf (border + 2, border + 2, border + size – 2,
border + size – 2, brightred)
mapleLeafPic := Pic.New (0, 0, 2 * border + size, 2 * border + size)
% Create the picture buttons.
Draw.Cls
starButton := GUI.CreatePictureButton (10, 10, starPic, StarPressed)
328 Object Oriented Turing Reference Manual
mapleButton := GUI.CreatePictureButton (55, 10, mapleLeafPic,
MaplePressed)
loop
exit when GUI.ProcessEvent
end loop
Details
When GUI.CreatePictureButton or GUI.CreatePictureButtonFull is called,
the newly created picture will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a picture button is not enabled, the picture button is grayed out and
the picture button no longer responds to any mouse clicks or keystrokes
until the button is enabled again.
Details
The following GUI subprograms can be called with a picture button as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreatePictureButton, not by calling CreatePictureButton.
GUI.CreatePictureRadioButton[Full]
Syntax
GUI.CreatePictureRadioButton ( x, y, picture : int,
joinID : int, actionProc : procedure x () ) : int
GUI.CreatePictureRadioButtonFull (x, y, picture :
int,
joinID : int, actionProc : procedure x (),
width, height : int, shortcut : char,
mergePic : boolean ) : int
Description
Creates a picture radio button and returns the button's widget ID.
Picture radio buttons behave like picture buttons (see
GUI.CreatePictureButton) except that they have the "radio" property. That
is, one of the buttons in the radio group is always selected, and if another
button in the group is selected, the previously selected button is
unselected.
Chapter 7 : Language Features 329
A common example is the buttons on a paint program that indicate the
current shape being painted. The maple leaf button is currently selected. If
the star button is selected by the user, the maple leaf button becomes
unselected. For picture buttons, the selected button appears depressed.
Four Picture Radio Buttons with the Maple Leaf Seledcted
A radio group is created by first creating a single radio button. To add another button to
the group, a second radio button is created specifying the first radio button
in the joinID parameter. Subsequent radio buttons are added, each
specifying a previous member of the group in the joinID parameter.
The picture must be created by the program beforehand using Pic.New or
Pic.FileNew. The resulting picture can then be used as a parameter to
GUI.CreatePictureButton. In general, pictures should be a maximum of
about 50 pixels high and wide, although there is no built-in limit in the GUI
library.
The x and y parameters specify the lower-left corner of the picture radio
button. If these are both –1 and joinID is not zero, then the button will be
placed directly below the previous picture radio button in the group. The
picture parameter specifies the picture ID of the picture to be displayed on
the button. (Note that, in general, this picture should be fairly small.) The
picture ID is received from a Pic.New or Pic.FileNew call. Do not call
Pic.Free for this picture ID until the button has been disposed of by calling
GUI.Dispose. The joinID parameter specifies a member of the radio group
that this widget should join. A joinID of 0 sepecifies this radio button is not
a member of any group. The actionProc parameter specifies the name of a
procedure that is called when the picture button is pressed.
330 Object Oriented Turing Reference Manual
For GUI.CreatePictureRadioButtonFull, the width and height parameters
specify the width and height of the button. If they are set to 0, then the
picture radio button is automatically sized to fit the picture. If you need to
know the precise size of the button, use the GUI.GetWidth and
GUI.GetHeight functions. If width and height are larger than the picture,
the picture is centered in the button. The shortCut parameter is the
keystroke to be used as the button's shortcut. The mergePic parameter
specifies whether anything that was the background colour in the picture
(usually colour 0) should be set to the background colour of the button
(which is usually gray). This defaults to true for CreatePictureRadioButton.
Example
The following program creates and displays for picture radio buttons.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:100;70")
const size : int := 25
const border : int := 3
% The buttons size.
var starButton, mapleButton, circleButton, squareButton : int
var starPic, mapleLeafPic, circlePic, squarePic : int
procedure StarPressed
Text.Locate (1, 1)
put "Star Pressed "
end StarPressed
procedure MaplePressed
Text.Locate (1, 1)
put "Maple Pressed "
end MaplePressed
procedure CirclePressed
Text.Locate (1, 10)
put "Circle Pressed"
end CirclePressed
procedure SquarePressed
Text.Locate (1, 10)
put "Square Pressed"
end SquarePressed
% Create the pictures.
% The star.
Draw.Star (border, border, border + size, border + size, black)
Draw.Star (border + 1, border + 1, border + size – 1,
border + size – 1, black)
Draw.FillStar (border + 2, border + 2, border + size – 2,
border + size – 2, brightred)
starPic := Pic.New (0, 0, 2 * border + size, 2 * border + size)
% The mapleleaf.
Draw.FillBox (border, border, border + size, border + size, white)
Draw.MapleLeaf (border, border, border + size, border + size, black)
Draw.MapleLeaf (border + 1, border + 1, border + size – 1,
Chapter 7 : Language Features 331
border + size – 1, black)
Draw.FillMapleLeaf (border + 2, border + 2, border + size – 2,
border + size – 2, brightred)
mapleLeafPic := Pic.New (0, 0, 2 * border + size, 2 * border + size)
% The circle.
const radius : int := size div 2
Draw.FillBox (border, border, border + size, border + size, white)
Draw.Oval (border + radius, border + radius, radius, radius, black)
Draw.Oval (border + radius, border + radius, radius – 1, radius – 1,
black)
Draw.FillOval (border + radius, border + radius, radius – 2,
radius – 2, brightred)
circlePic := Pic.New (0, 0, 2 * border + size, 2 * border + size)
% The square.
Draw.FillBox (border, border, border + size, border + size, white)
Draw.Box (border, border, border + size, border + size, black)
Draw.Box (border + 1, border + 1, border + size – 1,
border + size – 1, black)
Draw.FillBox (border + 2, border + 2, border + size – 2,
border + size – 2, brightred)
squarePic := Pic.New (0, 0, 2 * border + size, 2 * border + size)
% Create the picture buttons.
Draw.Cls
starButton := GUI.CreatePictureButton (10, maxy – 40, starPic,
0, StarPressed)
mapleButton := GUI.CreatePictureButton (–1, –1, mapleLeafPic,
starButton, MaplePressed)
circleButton := GUI.CreatePictureRadioButton ( –1, –1, circlePic,
mapleButton, CirclePressed)
squareButton := GUI.CreatePictureRadioButton ( –1, –1, squarePic,
circleButton, SquarePressed)
loop
exit when GUI.ProcessEvent
end loop
Details
When GUI.CreatePictureRadioButton or
GUI.CreatePictureRadioButtonFull is called, the newly created picture
will be displayed immediately unless GUI.DisplayWhenCreated has been
called with the display parameter set to false.
When a picture radio button is not enabled, the picture radio button is
grayed out and the picture button no longer responds to any mouse clicks
or keystrokes until the button is enabled again.
Details
The following GUI subprograms can be called with a picture radio button
as the widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.SelectRadio
332 Object Oriented Turing Reference Manual
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreatePictureRadioButton, not by calling CreatePictureRadioButton.
See also
GUI.SelectRadio for selecting a picture radio button in a program. See also
GUI.CreatePictureButton and GUI.CreateRadioButton for information on
picture buttons and radio buttons.
Win DOS Mac
GUI.CreateRadioButton[Full]
Syntax
X TOOT Term
GUI.CreateRadioButton ( x, y : int, text : string,
joinID : int, actionProc : procedure x () ) : int
GUI.CreateRadioButtonFull ( x, y : int, text : string,
joinID : int, actionProc : procedure x (),
alignment : int, shortCut : char ) : int
Description
Creates a radio button and returns the radio button's widget ID.
A slider is a widget that allows the user to select one of a set of values. It
has a real-life equivalent in the old car stereos where a single station is
selected at a time. That is, one of the buttons in the radio group is always
selected, and if another button in the group is selected, the previously
selected button is unselected.
Six Radio Buttons in Two Groups
Chapter 7 : Language Features 333
A radio group is created by first creating a single radio button. To add
another button to the group, a second radio button is created specifying the
first radio button in the joinID parameter. Subsequent radio buttons are
added, each specifying a previous member of the group in the joinID
parameter.
The x and y parameters specify the lower-left corner of the radio button (unless alignment is
set to GUI.RIGHT, in which case they specify the lower-right corner of the
radio button). If these are both –1 and joinID is not zero, then the button
will be placed directly below the previous radio button in the group. The
text parameter specifies the text (or label) beside the radio button. The
joinID parameter specifies a member of the radio group that this widget
should join. A joinID of 0 sepecifies this radio button is not a member of
any group. The actionProc parameter is the name of a procedure that is
called when the radio button is selected. In GUI.CreateRadioButton, the
radio button's text is always to the right of the actual radio button. In
GUI.CreateRadioButtonFull, the text can be set to the right or left of the
radio button with the alignment parameter.
For GUI.CreateRadioButtonFull, the alignment parameter specifies the
position of the radio button in relation to the text as well as the meaning of
the x and y parameters. The alignment parameter is one of 0, GUI.LEFT, or
GUI.RIGHT. An alignment of 0 is the default and is the same as GUI.LEFT.
GUI.LEFT means the actual box in the check box appears to the left of the
check box's label and (x, y) specify the lower-left corner. An alignment of
GUI.RIGHT means that the actual box appears to the right of the radio
button's label and (x, y) specify the lower-right corner of the radio button.
The shortcut parameter is the keystroke to be used as the button's shortcut.
A radio button's size is not specified during creation. It is determined
based on the size of the text. Instead the user specifies the lower-left corner
of the radio button (or the lower-right if the radio button is right justified).
Example
The following program creates six radio buttons in two groups.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:350;80")
var radio : array 1 .. 6 of int % The radio button IDs.
procedure RadioPressed
Text.Locate (1, 1)
put "Radio Button " ..
for i : 1 .. 6
if radio (i) = GUI.GetEventWidgetID then
put i ..
end if
end for
put " Selected"
end RadioPressed
radio (1) := GUI.CreateRadioButton (15, maxy – 35,
"Radio Button 1", 0, RadioPressed)
radio (2) := GUI.CreateRadioButton ( –1, –1, "Radio Button 2",
radio (1), RadioPressed)
334 Object Oriented Turing Reference Manual
radio (3) := GUI.CreateRadioButton ( –1, –1, "Radio Button 3",
radio (2), RadioPressed)
radio (4) := GUI.CreateRadioButtonFull (maxx – 15, maxy – 35,
"Radio Button A (Shortcut: 'a')", 0, RadioPressed,
GUI.RIGHT, 'a')
radio (5) := GUI.CreateRadioButtonFull ( –1, –1,
"Radio Button B (Shortcut: 'b')", radio (4), RadioPressed,
GUI.RIGHT, 'b')
radio (6) := GUI.CreateRadioButtonFull ( –1, –1,
"Radio Button C (Shortcut: 'c')", radio (5), RadioPressed,
GUI.RIGHT, 'c')
loop
exit when GUI.ProcessEvent
end loop
Details
When a group of radio buttons is selected, the first radio button created in
the group will be the selected one. You can change this by using the
GUI.SelectRadio procedure to select a different one.
When GUI.CreateRadioButton or GUI.CreateRadioButtonFull is called,
the newly created picture will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a radio button is not enabled, the radio button is grayed out and the
radio button no longer responds to any mouse clicks or keystrokes until the
button is enabled again.
Details
The following GUI subprograms can be called with a radio button as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.SetLabel, GUI.SelectRadio
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateRadioButton, not by calling CreateRadioButton.
See also
GUI.SelectRadio for selecting a radio button in a program. See also
GUI.SetLabel for changing the radio button's text.
Chapter 7 : Language Features 335
Win DOS Mac
GUI.CreateTextBox[Full]
Syntax
X TOOT Term
GUI.CreateTextBox ( x, y, width, height : int ) : int
GUI.CreateTextBoxFull ( x, y, width, height : int,
border, fontID : int ) : int
Description
Creates a text box and returns the text box's widget ID.
A text box is a box used for displaying text. It has scroll bars that activate
when text appears outside the border of the text box. The user cannot
directly select, edit or modify the text in the text box.
The x and y parameters specify the lower-left corner of the area in which
the text will be drawn. The width and height parameters specify the width
and height of the text drawing area The text box border is just outside the
text drawing area. Because of this, GUI.GetX and GUI.GetY will return a
value slightly smaller than x and y and GUI.GetWidth and
GUI.GetHeight will return values slightly larger than width and height.
For GUI.CreateTextBox, the border around the text box is always a line.
For GUI.CreateTextBoxFull, the type of border is specified by the border
parameter. The border parameter is one of 0, GUI.LINE, GUI.INDENT, or
GUI.EXDENT. A border of 0 is the default and is the same as GUI.LINE.
GUI.INDENT and GUI.EXDENT only display properly if the background
colour has been set to gray using GUI.SetBackgroundColor. GUI.INDENT
makes the text box appear indented or recessed. GUI.EXDENT makes the
text box appear to stand out from the window. The fontID parameter
specifies the font ID of the font to be used in the text box. The font ID is
received from a Font.New call. Do not call Font.Free for this font ID until the
label has been disposed of by calling GUI.Dispose.
By using the fondID parameter, text boxes can have any size or typeface.
336 Object Oriented Turing Reference Manual
A text box displaying the contents of a file.
Example
The following program displays the contents of a file in a text box.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:300;300")
const fileNameToBeViewed : string := "TextBxs.DEM"
var textBox : int
% The Text Field ID.
var title : int
% The label for the title.
var f : int
% The stream number of the file.
var line : string
% Lines to be read from the file.
% Open the file.
open : f, fileNameToBeViewed, get
if f = 0 then
put "Unable to open " + fileNameToBeViewed + " : ",
Error.LastMsg
return
end if
Chapter 7 : Language Features 337
% Set background color to gray for indented text box.
GUI.SetBackgroundColor (gray)
% Create the title label and text box.
title := GUI.CreateLabelFull (20, 280, fileNameToBeViewed, 250, 0,
GUI.CENTER, 0)
textBox := GUI.CreateTextBoxFull (10, 10, 280, 265,
GUI.INDENT, 0)
% Read the file and place it in the text box.
loop
exit when eof (f)
get : f, line : *
GUI.AddLine (textBox, line)
end loop
close : f % Close the file.
loop
exit when GUI.ProcessEvent
end loop
Details
When GUI.CreateTextBox or GUI.CreateTextBoxFull is called, the newly
created picture will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
A text box widget is a passive widget and cannot be enabled or disabled.
Details
The following GUI subprograms can be called with a text box as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.AddLine, GUI.AddText, GUI.ClearText
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateTextBox, not by calling CreateTextBox.
See also
GUI.AddLine, GUI.AddText for adding text to the text box. See also
GUI.ClearText for cleaing the text box.
338 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.CreateTextField[Full]
Syntax
X TOOT Term
GUI.CreateTextField ( x, y, width : int, text : string,
actionProc : procedure x (text : string) ) : int
GUI.CreateTextFieldFull ( x, y, width : int,
text : string,
actionProc : procedure x (text : string),
border, fontID, inputKind : int) : int
Description
Creates a text field and returns the text field's widget ID.
A text field is used to create a line of text that can be edited by the user.
The user can use the mouse to select part of the text and can enter text into
the text field.
If one or more text fields are enabled in a window, then one of the text
fields will be active. This means that when any keystrokes are entered into
the window, the active text field will receive the keystrokes. The active text
field can be changed using the GUI.SetActive procedure.
The x and y parameters specify the lower-left corner of the area in which
the text will be drawn. The text field border is just outside the text drawing
area. The width parameter specifies the width of the text drawing area. The
height of the text field is determined by the height of the font used by the
text field. The border of the text field is just outside the text drawing area,
so GUI.GetWidth will return values slightly larger than width. The
actionProc parameter specifies the name of the procedure to be called when
the user presses ENTER (RETURN on a Macintosh) when the text field is
active. The parameter is the current text in the text field.
For GUI.CreateTextField, the border around the text field is always a line.
For GUI.CreateTextFieldFull, the type of border is specified by the border
parameter. The border parameter is one of 0, GUI.LINE, GUI.INDENT, or
GUI.EXDENT. A border of 0 is the default and is the same as GUI.LINE.
GUI.INDENT and GUI.EXDENT only display properly if the background
colour has been set to gray using GUI.SetBackgroundColor. GUI.INDENT
makes the text field appear indented or recessed. GUI.EXDENT makes the
text field appear to stand out from the window. The fontID parameter
specifies the font ID of the font to be used in the text field. The font ID is
received from a Font.New call. Do not call Font.Free for this font ID until the
label has been disposed of by calling GUI.Dispose. The inputKind
parameter specifies the type of input accepted by the text field. This is one
of 0, GUI.ANY, GUI.INT, or GUI.REAL. An input type of 0 is the default
and is the same as GUI.ANY. GUI.ANY allows any type of input in the text
Chapter 7 : Language Features 339
field. GUI.INTEGER only allows positive integer input in the text field.
GUI.REAL allows any real number input in the text field. Note that using
GUI.INTEGER or GUI.REAL does not guarantee that the text field string can be converted
to an integer or a real. The text could be a null string, and for GUI.REAL
could be part of a number such as the string "–" or "1.25E" both of which
are illegal numbers. (To check the conversion, use the strintok or strrealok
functions before calling strint or strreal.)
Two Text Fields
Example
The following program creates a text field and echoes it on the screen when
the user presses ENTER.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:200;100")
var nameTextField, addressTextField : int % The Text Field IDs.
procedure NameEntered (text : string)
GUI.SetSelection (addressTextField, 0, 0)
GUI.SetActive (addressTextField)
end NameEntered
procedure AddressEntered (text : string)
GUI.SetSelection (nameTextField, 0, 0)
GUI.SetActive (nameTextField)
end AddressEntered
GUI.SetBackgroundColor (gray)
var quitButton := GUI.CreateButton (52, 5, 100, "Quit", GUI.Quit)
nameTextField := GUI.CreateTextFieldFull (50, 70, 100, "",
NameEntered, GUI.INDENT, 0, 0)
addressTextField := GUI.CreateTextFieldFull (50, 40, 100, "",
AddressEntered, GUI.INDENT, 0, 0)
var nameLabel := GUI.CreateLabelFull (45, 70, "Name", 0, 0,
GUI.RIGHT, 0)
var addressLabel := GUI.CreateLabelFull (45, 40, "Address", 0, 0,
GUI.RIGHT, 0)
340 Object Oriented Turing Reference Manual
loop
exit when GUI.ProcessEvent
end loop
GUI.Dispose (quitButton)
colorback (gray)
Text.Locate (maxrow – 1, 1)
put "Name = ", GUI.GetText (nameTextField)
put "Address = ", GUI.GetText (addressTextField) ..
Details
Only one text field is active at a time. The active text field has a blinking
cursor (or its selection highlighted). If a keystroke occurs when a window
has an active text field in it, the keystroke will be directed to the active text
field. You can change which text field is active with the GUI.SetActive
procedure or by simply clicking on another text field with the mouse.
When multiple text fields are created in a window, the first text field
created is active when the program begins.
The current version of the text field does not support cut and paste or
keyboard commands to extend the selection.
Because strings are a maximum of 255 characters, this is the maximum
number of characters in a text field.
The TAB character cycles between different text fields in a window. It
cycles through the text fields in the order in which they were created.
BACK TAB (shift+TAB) cycles through the fields in reverse order.
Details
When GUI.CreateTextField or GUI.CreateTextFieldFull is called, the
newly created picture will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a text field is not enabled, the text field cannot be made active and
the text in the field cannot be edited.
Details
The following GUI subprograms can be called with a text field as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.GetText, GUI.SetText, GUI.SetSelection, GUI.SetActive
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateTextField, not by calling CreateTextField.
See also
GUI.GetText and GUI.SetText for reading and setting the text in the text
field. See also GUI.SetSelection for setting the selected area of the text. See
also GUI.SetActive for making the text field active.
Chapter 7 : Language Features 341
GUI.CreateVerticalScrollBar[Full]
Win DOS Mac
X TOOT Term
Syntax
GUI.CreateVerticalScrollBar ( x, y, size : int,
min, max, start : int,
actionProc : procedure x (value : int) ) : int
GUI.CreateVerticalScrollBarFull ( x, y, size : int,
min, max, start : int,
actionProc : procedure x (value : int),
arrowInc, pageInc, thumbSize : int) : int
Description
Creates a vertical (up-down) scroll bar and returns the scroll bar's widget
ID.
A scroll bar is a widget that allows users to see a piece of a document that
cannot be displayed on the screen in its entirety. The picture below shows a
vertical scroll bar. To control a scroll bar, there are a few choices: the user
can click on the thumb (the box in the scroll bar) and slide it up and down,
or the user can click in the scroll bar itself above or below the thumb (in
which case the thumb is moved up or down one "page"), or the user can
click on the up or down arrows at the ends of the scroll bar (in which case
the thumb is moved up or down one "arrow increment" or "line").
A Vertical Scroll Bar
342 Object Oriented Turing Reference Manual
The programmer defines a page or an arrow increment. When the value of the scroll bar
changes, the action procedure of the scroll bar is called with the new value as
a parameter. The action procedure should then redraw the contents using the
new value of the scroll bar.
The range of values that the scroll bar will give is determined by the min
and max parameters in the Create call. The left side of the scroll bar
represents the minimum value, while the right represents the maximum
value. There is also the "thumb size". This represents the range of values
that can be seen at once on the screen.
By default, the arrow increment (the amount the value is changed when
the scrolling arrows are pressed) is set to one. The page increment (the
amount the value is changed when the user clicks in the bar to the right or
left of the thumb) is set to one quarter the difference between the minimum
and the maximum. The "thumb size" is set to zero (see the description of
scroll bars for an explanation of the thumb size).
The x and y parameters specify the lower-left corner of the scroll bar. The
size parameter specifies the length of the scroll bar (including the arrows)
in pixels. The min and max parameters are the minimum and maximum
valies returned by the scroll bar. The start parameter is the initial value of
the scroll bar and should be between min and max inclusive. The actionProc
parameter is the name of a procedure that is called when the value of the
scroll bar is changed. The parameter to the action procedure is the current
value of the scroll bar.
Example
The following program creates a vertical scroll bar. Whenever the scroll
bar's value is changed, a message is displayed in the window.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:85;200")
var scrollBar : int
procedure ScrollBarMoved (value : int)
Text.Locate (9, 7)
put "Scroll"
Text.Locate (10, 8)
put "Bar"
Text.Locate (11, 8)
put value : 3
end ScrollBarMoved
scrollBar := GUI.CreateVerticalScrollBar (10, 10, 180,
50, 150, 50, ScrollBarMoved)
loop
exit when GUI.ProcessEvent
end loop
Chapter 7 : Language Features 343
Description
For GUI.CreateVerticalScrollBarFull, the arrowInc parameter specifies the
arrow increment (the amount the scroll bar's value is changed when the
scroll arrows are pressed). The pageInc specifies the page increment (the
amount the scroll bar's value is changed when the user clicks in the page
left/right section of the scroll bar). The thumbSize parameter specifies the
"thumb size". See the scroll bar explanation for more detail on a scroll bar's
"thumb size".
For example, if you have a window that can display 20 lines of text at once
and there are 100 lines of text, you would set min to 1, max to 100 and
thumbSize to 20. The value returned by the scroll bar would then be the line
number of the first line on the screen to be displayed. When the scroll bar
was at its maximum value, it would return 81, since by doing so, lines 81100 would be displayed.
Example
For an example program that scrolls a large picture over a smaller window,
see GUI.CreateHorizontalScrollBar.
Details
In some instances, you will want the the minimum and maximum values of
the scroll bar to be reversed (right/top is minimum). In that case, call the
GUI.SetSliderReverse procedure to flip the values of the scroll bar.
Scroll bars always have a fixed height (for horizontal scroll bars) or width
(for vertical scroll bars). To get the scroll bar's width, use the
GUI.GetScrollBarWidth function.
When GUI.CreateVerticalScrollBar or GUI.CreateVerticalScrollBarFull is
called, the newly created scroll bar will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a scroll bar is not enabled, the gray in the bar is set to white and the
thumb is not displayed. The scroll bar no longer responds to any mouse
clicks until the scroll bar is enabled again.
Details
The following GUI subprograms can be called with a scroll bar as the
widgetID parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.GetSliderValue, GUI.SetSliderValue,
GUI.SetSliderMinMax, GUI.SetSliderSize,
GUI.SetSliderReverse, GUI.SetScrollAmount
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateVerticalScrollBar, not by calling CreateVerticalScrollBar.
See also
GUI.GetSliderValue and GUI.SetSliderValue for reading and setting the
value of a scroll bar, GUI.SetSliderMinMax for changing the minimum
and maximum values of a scroll bar, and GUI.SetScrollAmount for
changing the scrolling increments and thumb size of a scroll bar. See also
GUI.SetSliderSize for setting the length of a scroll bar and
GUI.SetSliderReverse for reversing the sense of a scroll bar.
344 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.CreateVerticalSlider
X TOOT Term
Syntax
GUI.CreateVerticalSlider ( x, y, length : int,
min, max, start : int,
actionProc : procedure x (value : int) ) : int
Description
Creates a vertical (up-down) slider and returns the slider's widget ID.
A slider is a widget that allows the user to set a continuous set of values. It
has a real-life equivalent in things such as a stereo volume control.
A Vertical Slider
To control a slider, the user clicks on the slider box and drags it back and
forth. Every time the value changes, a procedure is called with the new
value as a parameter.
The range of values that the slider will give is determined by the min and
max parameters in the Create call. The left side of the slider represents the
minimum value, while the right represents the maximum value.
The x and y parameters specify the lower-left corner of the slider track. This means that the
slider actually extends above and below this point (and slightly to the left
of it to take into account the rounded end of the track). The length
parameter specifies the length of the track in pixels. (You can use
GUI.GetX, GetY, GetWidth, and GetHeight to get the exact dimensions of
the slider.) The min and max parameters are the minimum and maximum
valies returned by the slider. The start parameter is the initial value of the
slider and should be between min and max inclusive. The actionProc
parameter is the name of a procedure that is called when the value of the
slider is changed. The parameter to the action procedure is the current value
of the slider.
Example
The following program creates a vertical slider. Whenever the slider's
value is changed, a message is displayed in the window.
Chapter 7 : Language Features 345
import GUI in "%oot/lib/GUI"
View.Set ("graphics:85;200")
var slider : int
procedure SliderMoved (value : int)
Text.Locate (9, 7)
put "Slider"
Text.Locate (10, 9)
put value : 3
end SliderMoved
slider := GUI.CreateVerticalSlider (20, 10, 180,
50, 150, 50, SliderMoved )
loop
exit when GUI.ProcessEvent
end loop
Details
In some instances, you will want the the minimum and maximum values of
the slider to be reversed (right is minimum). In that case, call the
GUI.SetSliderReverse procedure to flip the values of the slider.
Sliders always have a fixed height (for horizontal sliders) or width (for
vertical sliders).
When GUI.CreateVerticalSlider or GUI.CreateVerticalSliderFull is
called, the newly created slider will be displayed immediately unless
GUI.DisplayWhenCreated has been called with the display parameter set
to false.
When a slider is not enabled, the appearance does not change. However,
the slider no longer responds to any mouse clicks until it is enabled again.
Details
The following GUI subprograms can be called with a slider as the widgetID
parameter:
GUI.Show, GUI.Hide, GUI.Enable, GUI.Disable, GUI.Dispose,
GUI.GetX, GUI.GetY, GUI.GetWidth, GUI.GetHeight,
GUI.SetPosition, GUI.SetSize, GUI.SetPositionAndSize,
GUI.GetSliderValue, GUI.SetSliderValue,
GUI.SetSliderMinMax, GUI.SetSliderSize,
GUI.SetSliderReverse
Status
Exported qualified.
This means that you can only call the function by calling
GUI.CreateVerticalSlider, not by calling CreateVerticalSlider.
See also
GUI.GetSliderValue and GUI.SetSliderValue for reading and setting the
value of a slider, GUI.SetSliderMinMax for changing the minimum and
maximum values of a slider. See also GUI.SetSliderSize for setting the
length of a slider and GUI.SetSliderReverse for reversing the sense of a
slider.
346 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.Disable
X TOOT Term
Syntax
GUI.Disable ( widgetID : int )
Description
Disables a widget specified by widgetID.
Used in conjunction with GUI.Enable to enable and disable widgets.
Disabled widgets generally are "grayed out" to visually depict their
disabled status.
Disabled widgets do not respond to keystrokes or mouse clicks.
Example
The three color radio buttons are enabled only when the color check box is
selected.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:100;100")
var colorCheckBox, redRadio, greenRadio, blueRadio : int
procedure DoNothing
end DoNothing
procedure ColorCheckBoxProc (filled : boolean)
if filled then
GUI.Enable (redRadio)
GUI.Enable (greenRadio)
GUI.Enable (blueRadio)
else
GUI.Disable (redRadio)
GUI.Disable (greenRadio)
GUI.Disable (blueRadio)
end if
end ColorCheckBoxProc
colorCheckBox := GUI.CreateCheckBox (10, 80,
"Use Color", ColorCheckBoxProc)
redRadio := GUI.CreateRadioButton (33, 60, "Red", 0, DoNothing)
greenRadio := GUI.CreateRadioButton (–1, –1, "Green",
redRadio, DoNothing)
blueRadio := GUI.CreateRadioButton (–1, –1, "Blue",
greenRadio, DoNothing)
ColorCheckBoxProc (false)
loop
exit when GUI.ProcessEvent
end loop
Chapter 7 : Language Features 347
The following types of widgets can be enabled or disabled:
Details
Buttons, Check Boxes, Radio Buttons, Picture Buttons,
Picture Radio Buttons, Horizontal Scroll Bars, Horizontal Sliders,
Canvases, Text Fields, Vertical Scroll Bars, Vertical Sliders
Exported qualified.
Status
This means that you can only call the procedure by calling GUI.Disable,
not by calling Disable.
GUI.Enable.
See also
Win DOS Mac
GUI.Dispose
X TOOT Term
Syntax
GUI.Dispose ( widgetID : int )
Description
Eliminates the widget specified by widgetID.
If the widget is visible, it is immediately made invisible before being
deleted. It should be called in order to free up any memory that the widget
might have allocated. Note that you cannot use the widget after it has been
disposed of. If you wish to temporarily get rid of a widget, consider using
the Hide method and then the Show method when you want to use it again.
Example
The following program waits for the Quit button to be pressed. When it is,
the Quit button is deleted and a message is displayed in the center of the
screen.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:150;100")
var button, message : int
button := GUI.CreateButton (20, 40, 0, "Quit", GUI.Quit)
loop
exit when GUI.ProcessEvent
end loop
GUI.Dispose (button)
message := GUI.CreateLabelFull (0, 0, "Done", maxx, maxy,
GUI.CENTER + GUI.MIDDLE, 0)
Status
Exported qualified.
This means that you can only call the procedure by calling GUI.Dispose ,
not by calling Dispose .
348 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.Draw...
Syntax
X TOOT Term
GUI.DrawArc ( widgetID, x, y, xRadius, yRadius : int,
initialAngle, finalAngle, Color : int )
GUI.DrawBox ( widgetID, x1, y1, x2, y2, Color : int )
GUI.DrawCls ( widgetID : int )
GUI.DrawDot ( widgetID, x, y, Color : int )
GUI.DrawFill ( widgetID, x, y : int,
fillColor, borderColor : int)
GUI.DrawFillArc ( widgetID, x, y : int,
xRadius, yRadius : int,
initialAngle, finalAngle, Color : int )
GUI.DrawFillBox ( widgetID, x1, y1, x2, y2 : int,
Color : int )
GUI.DrawFillMapleLeaf ( widgetID, x1, y1 : int,
x2, y2, Color : int )
GUI.DrawFillOval ( widgetID, x, y : int,
xRadius, yRadius : int, Color : int)
GUI.DrawFillPolygon ( widgetID : int,
x, y : array 1 .. * of int, n : int, Color : int )
GUI.DrawFillStar ( widgetID, x1, y1, x2, y2 : int,
Color : int )
GUI.DrawLine ( widgetID, x1, y1, x2, y2, Color : int )
GUI.DrawMapleLeaf ( widgetID, x1, y1, x2, y2 : int,
Chapter 7 : Language Features 349
Color : int )
GUI.DrawOval ( widgetID, x, y : int,
xRadius, yRadius, Color : int )
GUI.DrawPolygon ( widgetID : int,
x, y : array 1 .. * of int, n : int, Color : int )
GUI.DrawStar ( widgetID, x1, y1, x2, y2, Color : int )
GUI.DrawText ( widgetID : int, textStr : string,
x, y : int, fontID, Color : int )
Description
Performs a Draw… command to the canvas specified by widgetID.
All of these routines are essentially the same as the similarly-named
procedures in the Draw module. All coordinates are based on the canvas
and all drawing is clipped to the canvas drawing surface. If the canvas is in
"xor mode", all the drawing will be done with "xor" set. (See View.Set for
more information about "xor".)
The widgetID must specify a canvas widget.
Example
See GUI.CreateCanvas for an example of GUI.DrawFillOval.
Status
Exported qualified.
This means that you can only call the procedures by calling GUI.Draw…,
not by calling Draw….
See also
GUI.CreateCanvas.
Win DOS Mac
GUI.Enable
X TOOT Term
Syntax
GUI.Enable ( widgetID : int )
Description
Enables a disabled widget specified by widgetID.
Used in conjunction with GUI.Disable to enable and disable widgets.
Disabled widgets generally are "grayed out" to visually depict their
disabled status.
Disabled widgets do not respond to keystrokes or mouse clicks.
350 Object Oriented Turing Reference Manual
Example
See GUI.Disable for an example of GUI.Enable.
Details
The following types of widgets can be enabled or disabled:
Buttons, Check Boxes, Radio Buttons, Picture Buttons,
Picture Radio Buttons, Horizontal Scroll Bars, Horizontal Sliders,
Canvases, Text Fields, Vertical Scroll Bars, Vertical Sliders
Status
Exported qualified.
This means that you can only call the procedure by calling GUI.Enable, not
by calling Enable.
See also
GUI.Disable.
Win DOS Mac
GUI.FontDraw
X TOOT Term
Syntax
GUI.FontDraw ( widgetID : int, textStr : string,
x, y, fontID, Color : int )
Description
Performs a Font.Draw command to the canvas specified by widgetID.
This routine is essentially the same as the Font.Draw procedure in the Font
module. All coordinates are based on the canvas and all drawing is clipped
to the canvas drawing surface. If the canvas is in "xor mode", all the
drawing will be done with "xor" set. (See View.Set for more information
about "xor".)
The widgetID must specify a canvas widget.
Status
Exported qualified.
This means that you can only call the procedure by calling GUI.FontDraw,
not by calling FontDraw.
See also
GUI.CreateCanvas.
Chapter 7 : Language Features 351
Win DOS Mac
GUI.GetCheckBox
X TOOT Term
Syntax
GUI.GetCheckBox ( widgetID : int ) : boolean
Description
Returns the status of the check box specified by widgetID. If the check box
is set (has an X in it), GetCheckBox returns true, otherwise it returns false.
Example
See GUI.CreateCheckBox for an example of GUI.GetCheckBox.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.GetCheckBox, not by calling GetCheckBox.
GUI.CreateCheckBox.
See also
Win DOS Mac
GUI.GetEventTime
X TOOT Term
Syntax
GUI.GetEventTime : int
Description
Returns the time in milliseconds when the event (mouse button or
keystroke) took place. This value is the same value as Time.Elapsed returns
if called when the event was processed. This function should only be called
in an action procedure or in a default mouse, keystroke, or null event
handler, as it will return –1 when there is no event being processed.
This event can be used as a timer for various functions such as determining
whether a single click or a double click of the mouse button took place or
for timing keyboard input.
Example
The following program times the interval between two button presses.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:300;100")
var startTime, startButton, finishButton : int
procedure Start
startTime := GUI.GetEventTime
end Start
352 Object Oriented Turing Reference Manual
procedure Finish
Text.Locate (1, 1)
put "The time between button pressed is ",
GUI.GetEventTime – startTime, " msecs"
GUI.Quit
end Finish
startButton := GUI.CreateButton (10, 10, 110, "Click First", Start)
finishButton := GUI.CreateButton (180, 10, 110, "Click Second", Finish)
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling
GUI.GetEventTime , not by calling GetEventTime .
GUI.ProcessEvent.
See also
Win DOS Mac
GUI.GetEventWidgetID
X TOOT Term
Syntax
GUI.GetEventWidgetID : int
Description
Returns the widget ID of the widget that was activated by the mouse
button press or the keystroke. This function should only be called in an
action procedure, as it will return –1 when there is no event that activated a
widget being processed.
This function is used when a several buttons use the same action procedure
to determine which button was pressed.
Example
The following program prints a message stating which button was selected.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:150;210")
var buttonNames : array 1 .. 5 of string := init ("Red", "Green",
"Blue", "Yellow", "Purple")
var buttons : array 1 .. 5 of int
procedure ButtonPush
for i : 1 .. 5
if GUI.GetEventWidgetID = buttons (i) then
Text.Locate (1, 1)
Chapter 7 : Language Features 353
put buttonNames (i), " selected"
end if
end for
end ButtonPush
for i : 1 .. 5
buttons (i) := GUI.CreateButton (10, 210 – 40 * i, 110,
buttonNames (i), ButtonPush)
end for
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling
GUI.GetEventWidgetID, not by calling GetEventWidgetID.
GUI.ProcessEvent.
See also
Win DOS Mac
GUI.GetEventWindow
X TOOT Term
Syntax
GUI.GetEventWindow : int
Description
Returns the window ID of the window in which the event (mouse button or
keystroke) took place. This function should only be called in an action
procedure or in a default mouse or keystroke event handler, as it will return
–1 when there is no event being processed.
This function is commonly used when several windows share the same
layout. The same buttons in each window point to the same action
procedures. To determine which button was actually pressed, the function is
called to get the window.
Example
The following program creates four windows in a row, each with a button
that, when pressed, causes a star to be drawn in that window.
import GUI in "%oot/lib/GUI"
procedure DrawStar
var windowID : int := GUI.GetEventWindow
Window.Select (windowID)
Draw.FillStar (25, 40, 175, 190, Rand.Int (10, 15))
end DrawStar
for i : 0 .. 3
var window : int := Window.Open ("graphics:200;200")
% Place window above task bar, across from previous one.
354 Object Oriented Turing Reference Manual
Window.SetPosition (window, 220 * i, 27)
var button : int := GUI.CreateButton (5, 5, 190, "Draw Star",
DrawStar)
end for
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling
GUI.GetEventWindow, not by calling GetEventWindow.
GUI.ProcessEvent.
See also
Win DOS Mac
GUI.GetHeight
X TOOT Term
Syntax
GUI.GetHeight ( widgetID : int ) : int
Description
Returns the actual height of a widget. Note that this may be different from
the height specified in the Create call (especially since many widgets do not
specify a height. The GUI module determines the actual height).
This function is used in conjunction with GUI.GetX, GUI.GetY and
GUI.GetWidth to determine the bounds of a widget. The entire widget
should always fit in the box (GUI.GetX, GUI.GetY) - (GUI.GetX +
GUI.GetWidth – 1, GUI.GetY + GUI.GetHeight – 1)
The position and size of a widget is known only after it has been drawn to
the screen. Attempting to get the location or dimesions of the widget may
cause an uninitialized variable error.
Example
The following procedure draws a red box around the widget specified by
widgetID.
import GUI in "%oot/lib/GUI"
procedure BoxWidget (widgetID : int)
var x, y, width, height : int
x := GUI.GetX (widgetID)
y := GUI.GetY (widgetID)
width := GUI.GetWidth (widgetID)
height := GUI.GetHeight (widgetID)
Draw.Box (x – 1, x – 1, x + width, y + height, red)
Draw.Box (x – 2, x – 2, x + width + 1, y + height + 1, red)
end BoxWidget
Chapter 7 : Language Features 355
var title : int := GUI.CreateLabel (20, 20, "Frame This!")
BoxWidget (title)
Exported qualified.
Status
This means that you can only call the function by calling GUI.GetHeight,
not by calling GetHeight.
GUI.GetX, GUI.GetY, and GUI.GetWidth.
See also
Win DOS Mac
GUI.GetMenuBarHeight
X TOOT Term
Syntax
GUI.GetMenuBarHeight : int
Description
Returns the height of the menu bar. Useful when drawing or placing
widgets to make certain that they don't overlap the menu bar.
Example
The following program draws a red box in the window just belowe the
menu bar.
import GUI in "%oot/lib/GUI"
var menu : int := GUI.CreateMenu ("File")
var item : int := GUI.CreateMenuItem ("Quit", GUI.Quit)
Draw.FillBox (0, 0, maxx, maxy – GUI.GetMenuBarHeight – 2,
brightred)
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling
GUI.GetMenuBarHeight, not by calling GetMenuBarHeight.
See also
GUI.CreateMenu.
356 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.GetScrollBarWidth
X TOOT Term
Syntax
GUI.GetScrollBarWidth : int
Description
Returns the width of a scroll bar. Useful when placing a scroll bar widget
beneath or beside another widget or object.
Example
See the ScrollPic program in GUI.CreateHorizontalScrollBarFull for an
example of GUI.GetScrollBarWidth.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.GetScrollBarWidth, not by calling GetScrollBarWidth.
See also
GUI.CreateHorizontalScrollBar and GUI.CreateVerticalScrollBar.
Win DOS Mac
GUI.GetSliderValue
X TOOT Term
Syntax
GUI.GetSliderValue ( widgetID : int ) : int
Description
Returns the current value of a slider or scroll bar specified by widgetID. The
widgetID must specify either scroll bar or a slider (horizontal or vertical).
Example
See the ScrollPic program in GUI.CreateHorizontalScrollBarFull for an
example of GUI.GetSliderValue.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.GetSliderValue, not by calling GetSliderValue.
See also
GUI.SetSliderValue for setting a slider or scroll bar's value. See also
GUI.CreateHorizontalScrollBar, GUI.CreateVerticalScrollBar,
GUI.CreateHorizontalSlider, and GUI.CreateVerticalSlider.
Chapter 7 : Language Features 357
Win DOS Mac
GUI.GetText
X TOOT Term
Syntax
GUI.GetText ( widgetID : int ) : string
Description
Returns the current text of a text field specified by widgetID. The widgetID
must specify a text field widget.
Example
See GUI.CreateTextField for an example of GUI.GetText.
Status
Exported qualified.
This means that you can only call the function by calling GUI.GetText, not
by calling GetText.
GUI.SetText for setting the text in a text field. See also
GUI.CreateTextField.
See also
Win DOS Mac
GUI.GetVersion
X TOOT Term
Syntax
GUI.GetVersion : int
Description
Returns the current version of the GUI Procedure Library. Because the GUI
Procedure Library is expected to grow, new versions will probably be
made available at our web site http://www.holtsoft.com/turing. If you
wish to use features that do not appear in earlier versions of the library,
you can have your program check that the current available version meets
the programs needs. GUI.GetVersion returns an integer from 100 - 999
and is read as 1.00 to 9.99.
Example
The following program fragment immediately exits if OOT does not
support version 1.1 of the GUI Procedure Library
import GUI in "%oot/lib/GUI"
if GUI.GetVersion < 110 then
put "You must update to at least version 1.1 of the"
put "GUI Procedure Library to use this program."
return
end if
358 Object Oriented Turing Reference Manual
Details
In version 1.00 (shipped with MacOOT 1.5), GUI.GetVersion did not exist.
Status
Exported qualified.
This means that you can only call the function by calling GUI.GetVersion,
not by calling GetVersion.
Win DOS Mac
GUI.GetWidth
X TOOT Term
Syntax
GUI.GetWidth ( widgetID : int ) : int
Description
Returns the actual width of a widget. Note that this may be different from
the width specified in the Create call (especially since some widgets do not
specify a width. The GUI module determines the actual width).
This function is used in conjunction with GUI.GetX, GUI.GetY and
GUI.GetHeight to determine the bounds of a widget. The entire widget
should always fit in the box (GUI.GetX, GUI.GetY) - (GUI.GetX +
GUI.GetWidth – 1, GUI.GetY + GUI.GetHeight – 1)
The position and size of a widget is known only after it has been drawn to
the screen. Attempting to get the location or dimesions of the widget may
cause an uninitialized variable error.
Example
The following procedure draws a red box around the widget specified by
widgetID.
import GUI in "%oot/lib/GUI"
procedure BoxWidget (widgetID : int)
var x, y, width, height : int
x := GUI.GetX (widgetID)
y := GUI.GetY (widgetID)
width := GUI.GetWidth (widgetID)
height := GUI.GetHeight (widgetID)
Draw.Box (x – 1, x – 1, x + width, y + height, red)
Draw.Box (x – 2, x – 2, x + width + 1, y + height + 1, red)
end BoxWidget
var title : int := GUI.CreateLabel (20, 20, "Frame This!")
BoxWidget (title)
Status
Exported qualified.
This means that you can only call the function by calling GUI.GetWidth,
not by calling GetWidth.
See also
GUI.GetX, GUI.GetY, and GUI.GetHeight.
Chapter 7 : Language Features 359
Win DOS Mac
GUI.Get{X,Y}
X TOOT Term
GUI.GetX ( widgetID : int ) : int
Syntax
GUI.GetY ( widgetID : int ) : int
Description
Returns the x coordinate of the left edge of a widget. Note that this may be
different from the x coordinate specified in the widget's Create call. For
example, if a radio button is created with right justification, the x
coordinate in the Create method specifies the right edge while GUI.GetX
will return the x coordinate of the left edge.
This function is used in conjunction with GUI.GetWidth and
GUI.GetHeight to determine the bounds of a widget. The entire widget
should always fit in the box (GUI.GetX, GUI.GetY) - (GUI.GetX +
GUI.GetWidth – 1, GUI.GetY + GUI.GetHeight – 1)
The position and size of a widget is known only after it has been drawn to
the screen. Attempting to get the location or dimesions of the widget may
cause an uninitialized variable error.
Example
The following procedure draws a red box around the widget specified by
widgetID.
import GUI in "%oot/lib/GUI"
procedure BoxWidget (widgetID : int)
var x, y, width, height : int
x := GUI.GetX (widgetID)
y := GUI.GetY (widgetID)
width := GUI.GetWidth (widgetID)
height := GUI.GetHeight (widgetID)
Draw.Box (x – 1, x – 1, x + width, y + height, red)
Draw.Box (x – 2, x – 2, x + width + 1, y + height + 1, red)
end BoxWidget
var title : int := GUI.CreateLabel (20, 20, "Frame This!")
BoxWidget (title)
Status
Exported qualified.
This means that you can only call the function by calling GUI.GetX, not by
calling GetX.
See also
GUI.GetHeight and GUI.GetWidth.
360 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.Hide
X TOOT Term
Syntax
GUI.Hide ( widgetID : int )
Description
Hides a widget specified by widgetID. Used in conjunction with Show to
hide and show widgets. Hidden widgets cannot get events (i.e. respond to
keystrokes or mouse clicks). If an active text field (see text field) is hidden,
then any keystrokes in the window will be ignored.
In most cases where a widget is to appear, then disappear, then appear
again, it is advised to create the widget once and hide it until it is to
appear, whereupon GUI.Show is called. When the user is finished with the
widget, the widget is hidden using GUI.Hide. This saves the overhead of
creating and disposing of the same widget several times.
Example
See GUI.SetDisplayWhenCreated for an example of GUI.Hide.
Status
Exported qualified.
This means that you can only call the function by calling GUI.Hide, not by
calling Hide.
See also
GUI.Show.
Win DOS Mac
GUI.HideMenuBar
X TOOT Term
Syntax
GUI.HideMenuBar
Description
Hides the menu bar in the selected window. No menu items can be
selected when the menu bar is hidden. (Menu item shortcuts are ignored
while the menu bar is hidden.)
Example
See GUI.SetMouseEventHandler for an example of GUI.HideMenuBar.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.HideMenuBar, not by calling HideMenuBar.
Chapter 7 : Language Features 361
GUI.ShowMenuBar. See also GUI.CreateMenu.
See also
Win DOS Mac
GUI.OpenFile
X TOOT Term
Syntax
GUI.OpenFile ( title : string ) : string
Description
Displays an "Open File" dialog box to obtain the name of an already
existing file. The caption (a window title under MS Windows, a string in a
Macintosh dialog box) is specified by the title parameter. The function uses
a dialog box specific to the operating system the program is being run on.
If the user did not choose a file (i.e. hit the Cancel button in the dialog), the
function returns "" (the empty string).
Note: This function is not available in the current version of the GUI
Procedure Library (shipping with WinOOT 3.1, DOS OOT 2.5 and
MacOOT 1.5). It is documented here for use with future shipping version
of Object Oriented Turing. It is likely to be implemented in the version of
Object Oriented Turing released in September 2000. Check the release
notes that are found in the on-line help to find out if this function is now
available.
Example
The following program asks the user for the name of a file and then echoes
the contents of it.
import GUI in "%oot/lib/GUI"
var fileName, line : string
var streamNumber : int
fileName := GUI.OpenFile ("Choose a Text File")
open : streamNumber, fileName, get
assert streamNumber > 0
loop
exit when eof (streamNumber)
get : streamNumber, line : *
put line
end loop
close : streamNumber
Status
Exported qualified.
This means that you can only call the function by calling GUI.OpenFile,
not by calling OpenFile.
362 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.OpenFileFull
X TOOT Term
Syntax
GUI.OpenFileFull ( title, filter : string,
startDir : string ) : string
Description
Displays an "Open File" dialog box to obtain the name of an already
existing file. The caption (a window title under MS Windows, a string in a
Macintosh dialog box) is specified by the title parameter. The list of files
shown is specified by the filter parameter. The initial directory to be
displayed is specified by the startDir parameter. The function uses a dialog
box specific to the operating system the program is being run on.
The filter parameter is a file name suffix that should be displayed. Multiple
suffixes can be specified by separating them with commas. If the user
specifies the empty string for filter, then all the files in the directory are
displayed. If the startDir parameter is empty, or if it specifies a non-existent
directory, then the current directory is displayed in the "Open File" dialog
box.
If the user did not choose a file (i.e. hit the Cancel button in the dialog), the
function returns "" (the empty string).
Note: This function is not available in the current version of the GUI
Procedure Library (shipping with WinOOT 3.1, DOS OOT 2.5 and
MacOOT 1.5). It is documented here for use with future shipping version
of Object Oriented Turing. It is likely to be implemented in the version of
Object Oriented Turing released in September 2000. Check the release
notes that are found in the on-line help to find out if this function is now
available.
Example
The following program asks the user to select a file ending in ".txt". The
initial directory is the root directory of the C drive.
import GUI in "%oot/lib/GUI"
var fileName : string
fileName := GUI.OpenFileFull ("Choose a Text File", "txt", "C:\\")
Details
If a suffix is placed in single quotes, it will be ignored on all but the Apple
Macintosh, where it will specify a Macintosh file type.
Example
The example makes the dialog box display all files ending in ".txt" or ".text"
on all systems but the Macintosh. On the Apple Macintosh, only files of file
type 'TEXT' will be displayed.
fileName := GUI.OpenFileFull ("Open", "txt,text,'TEXT'", "")
Chapter 7 : Language Features 363
Status
Exported qualified.
This means that you can only call the function by calling
GUI.OpenFileFull, not by calling OpenFileFull.
Win DOS Mac
GUI.Pic...
Syntax
X TOOT Term
GUI.PicDraw ( widgetID : int, picID, x, y, mode : int )
GUI.PicNew ( widgetID : int, x1, y1, x2, y2 : int ) : int
GUI.PicScreenLoad ( widgetID : int, fileName : string,
x, y, mode : int )
GUI.PicScreenSave ( widgetID : int, x1, y1, x2, y2 : int,
fileName : string )
Description
Performs a Pic… command to the canvas specified by widgetID.
All of these routines are essentially the same as the similarly-named
procedures in the Pic module. All coordinates are based on the canvas and
all drawing is clipped to the canvas drawing surface.
Example
See the ScrollPic program in GUI.CreateHorizontalScrollBarFull for an
example of GUI.PicDraw.
Status
Exported qualified.
This means that you can only call the procedures by calling GUI.Pic…, not
by calling Pic….
See also
GUI.CreateCanvas.
Win DOS Mac
GUI.ProcessEvent
Syntax
GUI.ProcessEvent : boolean
364 Object Oriented Turing Reference Manual
X TOOT Term
Description
This function processes a single event (a mouse button press or a
keystroke). If the event activates a widget, then the action procedure of the
widget is called.
The function returns false until GUI.Quit is called. It then returns true.
In order for the widgets to function once placed, the GUI.ProcessEvent
must be called continually. Without a call to GUI.ProcessEvent, the
widgets will appear, but will not react to mouse clicks or keystrokes.
Almost all programs involving the GUI Procedure Library have the
following code fragment in the program. This code fragment is often called
the event loop.
loop
exit when GUI.ProcessEvent
end loop
The loop runs continuously until GUI.Quit is called, whereupon
GUI.ProcessEvent will return true and the loop will exit. The rest of the
program is reached through the action procedures that are called when the
user interacts with various widgets.
Details
It is usually desirable to allow the user some way of quitting the program
without having to abort it. This can be done most simply by adding a Quit
button and placing it in an appropriate location.
Example
Here is program that does nothing but wait for the user to press the quit
button.
import GUI in "%oot/lib/GUI"
var quitButton : int := GUI.CreateButton (10, 10, 0, "Quit", GUI.Quit)
loop
exit when GUI.ProcessEvent
end loop
Details
To find out which widget was activated and called the action procedure
(necessary if several widgets have the same action procedure), you can call
GUI.GetEventWidgetID. To get the exact time that the event occurred,
you can call GUI.GetEventTime. To get the window in which the event
took place, you can call GUI.GetEventWindow.
If a mouse click occured, but did not activate any widget, then the default mouse event
handler is called. By default, this does nothing. However, if you want your
program to respond to mouse events that do not affect a widget, call
GUI.SetMouseEventHandler to specify your own default mouse event
handler.
If a keystroke occurred, but did not activate any widget (i.e. it wasn't a
short cut for a widget and there are no text fields in the window) then the
default keystroke handler is called. By default, this does nothing. However,
if you want your program to respond to keystroke events that do not affect
a widget, call GUI.SetKeyEventHandler to specify your own default key
event handler.
Chapter 7 : Language Features 365
If no event occurred, then the null event handler is called. By default, this
does nothing. However, if you want your program to perform some action
repetetively when it is not doing anything else, then call
GUI.SetNullEventHandler to specify your own null event handler. The
null event handler is often used for such things as updating a clock and
making certain that music is playing in the background.
Exported qualified.
Status
This means that you can only call the procedures by calling
GUI.PProcessEvent, not by calling ProcessEvent.
GUI.GetEventWidgetID, GUI.GetEventTime, and
GUI.GetEventWindow for obtaining information about an event in an
action procedure. See also GUI.SetMouseEventHandler,
GUI.SetKeyEventHandler and GUI.SetNullEventHandler for handling
mouse, keyboard an d null events in the event loop. See also GUI.Quit for
information on exitting the event loop.
See also
Win DOS Mac
GUI.Quit
X TOOT Term
Syntax
GUI.Quit
Description
This procedure causes GUI.ProcessEvent to return true. If the program is
structured properly with a
loop
exit when GUI.ProcessEvent
end loop
at the end of the program, then the program will exit the loop after
finishing the current action procedure. This procedure is usually called from
the action procedure of a Quit button or Exit menu item.
Example
Here is program that does nothing but wait for the user to press the quit
button or type the letter 'Q', 'q', 'X', or 'x'.
import GUI in "%oot/lib/GUI"
procedure KeyHandler (ch : char)
if ch = 'Q' or ch = 'q' or ch = 'X' or ch = 'x' then
GUI.Quit
end if
end KeyHandler
var quitButton : int := GUI.CreateButton (10, 10, 0, "Quit", GUI.Quit)
GUI.SetKeyEventHandler (KeyHandler)
366 Object Oriented Turing Reference Manual
loop
exit when GUI.ProcessEvent
end loop
put "Done!"
Status
Exported qualified.
This means that you can only call the procedures by calling GUI.Quit, not
by calling Quit.
See also
GUI.ProcessEvent.
Win DOS Mac
GUI.Refresh
X TOOT Term
Syntax
GUI.Refresh
Description
This routine redraws all the widgets in the currently-selected window. This
is used when some form of drawing may have overwritten the widgets in a
window.
It is often used when there is some possibility that the widgets may have
been drawn over. For example, a program that places buttons on top of a
background image should call GUI.Refresh when the image is changed.
Details
GUI.Refresh is used by the GUI Library to redraw all the widgets when
the background colour of a window has changed.
Status
Exported qualified.
This means that you can only call the procedures by calling GUI.Refresh,
not by calling Refresh.
Win DOS Mac
GUI.SaveFile
Syntax
X TOOT Term
GUI.SaveFile ( title : string ) : string
Chapter 7 : Language Features 367
Description
Displays an "Save File" dialog box to obtain the name of a file. The caption
(a window title under MS Windows, a string in a Macintosh dialog box) is
specified by the title parameter. The function uses a dialog box specific to
the operating system the program is being run on.
If the user did not choose a file (i.e. hit the Cancel button in the dialog), the
function returns "" (the empty string).
Note: This function is not available in the current version of the GUI
Procedure Library (shipping with WinOOT 3.1, DOS OOT 2.5 and
MacOOT 1.5). It is documented here for use with future shipping version
of Object Oriented Turing. It is likely to be implemented in the version of
Object Oriented Turing released in September 2000. Check the release
notes that are found in the on-line help to find out if this function is now
available.
Example
The following program asks the user for the name of a file and then writes
the numbers 1 to 10 in it.
import GUI in "%oot/lib/GUI"
var fileName : string
var streamNumber : int
fileName := GUI.SaveFile ("Choose a Text File")
open : streamNumber, fileName, put
assert streamNumber > 0
for i : 1 .. 10
put : streamNumber, i
end loop
close : streamNumber
Status
Exported qualified.
This means that you can only call the function by calling GUI.SaveFile, not
by calling SaveFile.
Win DOS Mac
GUI.SaveFileFull
Syntax
X TOOT Term
GUI.SaveFileFull ( title, filter : string,
startDir : string ) : string
368 Object Oriented Turing Reference Manual
Description
Displays an "Save File" dialog box to obtain the name of an already existing
file. The caption (a window title under MS Windows, a string in a
Macintosh dialog box) is specified by the title parameter. The list of files
shown is specified by the filter parameter. The initial directory to be
displayed is specified by the startDir parameter. The function uses a dialog
box specific to the operating system the program is being run on.
The filter parameter is a file name suffix that should be displayed. Multiple
suffixes can be specified by separating them with commas. If the user
specifies the empty string for filter, then all the files in the directory are
displayed. If the startDir parameter is empty, or if it specifies a non-existent
directory, then the current directory is displayed in the "Open File" dialog
box.
If the user did not choose a file (i.e. hit the Cancel button in the dialog), the
function returns "" (the empty string).
Note: This function is not available in the current version of the GUI
Procedure Library (shipping with WinOOT 3.1, DOS OOT 2.5 and
MacOOT 1.5). It is documented here for use with future shipping version
of Object Oriented Turing. It is likely to be implemented in the version of
Object Oriented Turing released in September 2000. Check the release
notes that are found in the on-line help to find out if this function is now
available.
Example
The following program segment asks the user for the name of a file ending
in ".txt". The initial directory is the root directory of the C drive.
var fileName : string := GUI.SaveFileFull ("Choose a Text File",
"txt", "C:\\")
Details
If a suffix is placed in single quotes, it will be ignored on all but the Apple
Macintosh, where it will specify a Macintosh file type.
Example
The following program segment asks the user for the name of a file. It
displays files of type 'TEXT'. The initial directory is the "Turing Programs"
directory on the "Macintosh HD" volume.
var fileName : string := GUI.SaveFileFull ("Choose a Text File",
"'TEXT'", "Macintosh HD:Turing Programs")
Status
Exported qualified.
This means that you can only call the function by calling GUI.SaveFileFull,
not by calling SaveFileFull.
Chapter 7 : Language Features 369
Win DOS Mac
GUI.SelectRadio
X TOOT Term
Syntax
GUI.SelectRadio ( widgetID : int )
Description
Selects a radio button specified by widgetID. The previously-selected radio
button is "de-selected". The action procedure of the radio button is called.
Example
The following program creates siz radio buttons. Selecting one of the
buttons cause the bottom two radio buttons to become selected.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:350;110")
var radio : array 1 .. 6 of int
% The radio button IDs.
procedure RadioPressed
Text.Locate (1, 1)
for i : 1 .. 6
if radio (i) = GUI.GetEventWidgetID then
put "Radio Button " , i, " Selected"
end if
end for
end RadioPressed
procedure Select
GUI.SelectRadio (radio (3))
GUI.SelectRadio (radio (6))
end Select
radio (1) := GUI.CreateRadioButton (15, maxy – 35,
"Radio Button 1", 0, RadioPressed)
radio (2) := GUI.CreateRadioButton ( –1, –1, "Radio Button 2",
radio (1), RadioPressed)
radio (3) := GUI.CreateRadioButton ( –1, –1, "Radio Button 3",
radio (2), RadioPressed)
radio (4) := GUI.CreateRadioButtonFull (maxx – 15, maxy – 35,
"Radio Button 4", 0, RadioPressed, GUI.RIGHT, GUI.NONE)
radio (5) := GUI.CreateRadioButtonFull ( –1, –1, "Radio Button 5",
radio (4), RadioPressed, GUI.RIGHT, GUI.NONE)
radio (6) := GUI.CreateRadioButtonFull ( –1, –1, "Radio Button 6",
radio (5), RadioPressed, GUI.RIGHT, GUI.NONE)
loop
var selectButton : int := GUI.CreateButton (15, 10, 100,
"Select Bottom Buttons", Select)
var quitButton : int := GUI.CreateButton (maxx – 15 – 100, 10, 100,
"Quit", GUI.Quit)
exit when GUI.ProcessEvent
end loop
370 Object Oriented Turing Reference Manual
Status
Exported qualified.
This means that you can only call the procedures by calling
GUI.SelectRadio, not by calling SelectRadio.
See also
GUI.CreateRadioButton and GUI.CreatePictureRadioButton.
Win DOS Mac
GUI.SetActive
X TOOT Term
Syntax
GUI.SetActive ( widgetID : int )
Description
Makes a text field specified by widgetID active. If the text field is not in an
active window, then the text field will become active when the window
does. If another text field was active in the window, it is deactivated.
Example
See GUI.CreateTextFieldfor an example of GUI.SetActive.
Status
Exported qualified.
This means that you can only call the procedures by calling GUI.SetActive,
not by calling SetActive.
See also
GUI.CreateTextField.
Win DOS Mac
GUI.SetBackgroundColor
X TOOT Term
Syntax
GUI.SetBackgroundColor ( Color : int )
Description
Changes the background colour of the currently-selected window to the
color specified by Color. This does not change the value of color 0 in the
window. Instead it fills the entire window with the new background color
and then redraws all the widgets.
For indented and extended items, the background color is assumed to be
set to gray.
The alternate spelling is GUI.SetBackgroundColour
Chapter 7 : Language Features 371
Example
See GUI.CreateFrame for an example of GUI.SetBackgroundColour.
Status
Exported qualified.
This means that you can only call the procedures by calling
GUI.SetBackgroundColor, not by calling SetBackgroundColor.
Win DOS Mac
GUI.SetCheckBox
X TOOT Term
Syntax
GUI.SetCheckBox ( widgetID : int, status : boolean )
Description
Sets the status of a check box specified by widgetID. If status is true, the
check box is filled (marked with an 'X'). If status is false, the check box is set
empty. GUI.SetCheckBox calls the check box's action procedure with the
new status and redraws the widget with the new status.
Example
See GUI.CreateCheckBox for an example of GUI.SetCheckBox.
Status
Exported qualified.
This means that you can only call the procedures by calling
GUI.SetCheckBox, not by calling SetCheckBox.
See also
GUI.CreateCheckBox.
Win DOS Mac
GUI.SetDefault
X TOOT Term
Syntax
GUI.SetDefault ( widgetID : int, default : boolean )
Description
Sets the "default status" of a button specified by widgetID. If a button is the
default button, then it is drawn with a heavy outline and it is activated
when the user presses ENTER (RETURN on a Macintosh).
Only one button can be the default button per window. If a button is set to
be the default button, then the previous default button has its "default
status" removed.
372 Object Oriented Turing Reference Manual
Example
See GUI.CreateTextField for an example of GUI.SetCheckBox.
Status
Exported qualified.
This means that you can only call the procedures by calling
GUI.SetDefault, not by calling SetDefault.
GUI.CreateButton.
See also
Win DOS Mac
GUI.SetDisplayWhenCreated
X TOOT Term
Syntax
GUI.SetDisplayWhenCreated ( display : boolean )
Description
By default, whenever a widget is created with a GUI.Create… procedure,
the widget instantly appears. Sometimes, this is not the desired behaviour.
For example, if several widgets are to occupy the same location with only
one being displayed at a time, then it is desirable not to have the widget
appear when first created.
If a widget is not displayed when created, then GUI.Show must be called to
make the widget visible.
If the display parameter is true, then widgets are displayed immediately
upon creation. If the display parameter is set to false, then the widget is not
made visible on creation and GUI.Show must be called to display the
widget.
Example
The following program toggles the visiblility of the frame when the button
is pushed. The frame starts out invisible.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:150;100")
var visible : boolean := false
var button, frame : int
procedure Toggle
if visible then
GUI.Hide (frame)
else
GUI.Show (frame)
end if
visible := not visible
end Toggle
button := GUI.CreateButton (25, 40, 0, "Toggle Frame", Toggle)
GUI.SetDisplayWhenCreated (false)
Chapter 7 : Language Features 373
frame := GUI.CreateFrame (10, 10, 140, 90, 0)
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling
GUI.SetDisplayWhenCreated, not by calling SetDisplayWhenCreated.
Win DOS Mac
GUI.SetKeyEventHandler
X TOOT Term
Syntax
GUI.SetKeyEventHandler (
keyEventHandler : procedure x (ch : char) )
Description
Sets the new default keystroke event handler. The keyEventHandler
parameter is the name of a procedure that is called every time
GUI.ProcessEvent is called and there is a keystroke which is not handled
by any widget. The ch parameter in the keyEventHandler is the keystroke
typed by the user.
This procedure is often used to allow for more than one shortcut character
for a given button.
Example
The following program draws a star or quits depening on the button. The
Draw Star button can be activated by clicking on it or typing 'D', 'd', 'S', 's'
or Ctrl+S. The Quit button can be activate by typing 'Q', 'q' or Ctrl+Q. The
Draw Star button is also the default button. It is activated whenever the
user presses ENTER.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:220;200")
procedure DrawStar
Draw.FillStar (25, 40, 175, 190, Rand.Int (10, 15))
end DrawStar
procedure KeyHandler (ch : char)
if ch = 'q' or ch = '^Q' then
Draw.Cls
GUI.Quit
elsif ch = 'd' or ch = '^d' or ch = 'S' or ch = 's' or ch = '^s' then
DrawStar
end if
end KeyHandler
GUI.SetKeyEventHandler (KeyHandler)
var button : int := GUI.CreateButtonFull (5, 5, 100, "Draw Star",
DrawStar, 0, 'D', true)
374 Object Oriented Turing Reference Manual
var quitButton : int := GUI.CreateButtonFull (115, 5, 100, "Quit",
GUI.Quit, 0, 'Q', false)
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling
GUI.SetKeyEventHandler, not by calling SetKeyEventHandler.
GUI.ProcessEvent.
See also
Win DOS Mac
GUI.SetLabel
X TOOT Term
Syntax
GUI.SetLabel ( widgetID : int, text : string )
Description
Changes the text of a widget specified by widgetID to text. This procedure
can accept a button, check box, radio button, label, or a labelled frame
widget as the widgetID parameter.
In most cases, if the text will not fit in the widget's current size, the widget
will be resized to fit the text. If the widget was made larger to fit the text
and then the text is changed, the widget will be resized as appropriate for
the original width specified and the new text.
Example
The following program changes the text in the button whenever a
keystroke occurs. When the text is changed back to "Quit", the button
assumes a width of 100 again.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:220;50")
var short : boolean := true
var button : int
procedure KeyHandler (ch : char)
if short then
GUI.SetLabel (button, "Press This Button to Quit")
else
GUI.SetLabel (button, "Quit")
end if
short := not short
end KeyHandler
GUI.SetKeyEventHandler (KeyHandler)
button := GUI.CreateButton (10, 5, 100, "Quit", GUI.Quit)
Chapter 7 : Language Features 375
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling GUI.SetLabel, not
by calling SetLabel.
GUI.CreateButton, GUI.CreateCheckBox, GUI.CreateRadioButton,
GUI.CreateLabel, and GUI.CreateLabelledFrame.
See also
Win DOS Mac
GUI.SetMouseEventHandler
X TOOT Term
Syntax
GUI.SetMouseEventHandler (
mouseEventHandler : procedure x (mx, my : int) )
Description
Sets the new default mouse event handler. The mouseEventHandler
parameter is the name of a procedure that is called every time
GUI.ProcessEvent is called and there is a mouse button down which is not
handled by any widget. The mx and my parameters in the
mouseEventHandler are the location of mouse when the button was pressed.
This procedure is used by programs to allow for mouse input in a program
that uses widgets.
Example
This is a program that allows the user to place stars on the screen. The
menu bar allows the user to quit the program at any time. The user can also
toggle the appearance of the menu bar by pressing any key.
import GUI in "%oot/lib/GUI"
var starX, starY, starColor : array 1 .. 100 of int
var numStars : int := 0
var menuVisible : boolean := true
procedure DrawStar (i : int)
if menuVisible then
View.ClipSet (0, 0, maxx,
maxy – GUI.GetMenuBarHeight)
end if
Draw.FillStar (starX (i) – 20, starY (i) – 20, starX (i) + 20,
starY (i) + 20, starColor (i))
View.ClipOff
end DrawStar
procedure Redraw
376 Object Oriented Turing Reference Manual
for i : 1 .. numStars
DrawStar (i)
end for
Text.Locate (maxrow, 1)
put "Press any key to toggle menu bar" ..
end Redraw
procedure KeyHandler (ch : char)
if menuVisible then
GUI.HideMenuBar
else
GUI.ShowMenuBar
end if
menuVisible := not menuVisible
Redraw
end KeyHandler
procedure MouseHandler (x, y : int)
if numStars = 100 then
Text.Locate (maxrow, 1)
put "Maximum number of stars exceeded!" ..
return
end if
numStars += 1
starX (numStars) := x
starY (numStars) := y
starColor (numStars) := Rand.Int (9, 15)
DrawStar (numStars)
end MouseHandler
var menu : int := GUI.CreateMenu ("File")
var menuItem : int := GUI.CreateMenuItemFull ("Quit",
GUI.Quit, '^Q', false)
GUI.SetKeyEventHandler (KeyHandler)
GUI.SetMouseEventHandler (MouseHandler)
Redraw
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetMouseEventHandler, not by calling SetMouseEventHandler.
See also
GUI.ProcessEvent.
Chapter 7 : Language Features 377
Win DOS Mac
GUI.SetNullEventHandler
X TOOT Term
Syntax
GUI.SetNullEventHandler (
nullHandler : procedure x ())
Description
Sets the new null event handler. The nullHandler parameter is the name of a
procedure that is called every time GUI.ProcessEvent is called and there
are no mouse button presses or keystrokes to be processed.
This is used by programs that need to call subprograms often, but do not
wish to interrupt the action of user widgets.
Example
The following program has a Quit button. When no widgets are being
processed, a clock in the corner is updated.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:220;50")
var oldTime : string := ""
var button : int
procedure NullHandler
var newTime : string := Time.Date
newTime := newTime (11 .. *)
if newTime not= oldTime then
Text.Locate (maxrow, maxcol – 9)
put newTime ..
oldTime := newTime
end if
end NullHandler
GUI.SetNullEventHandler (NullHandler)
button := GUI.CreateButton (10, 5, 100, "Quit", GUI.Quit)
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetNullEventHandler, not by calling SetNullEventHandler.
See also
GUI.ProcessEvent.
378 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.SetPosition
X TOOT Term
Syntax
GUI.SetPosition ( widgetID, x, y : int )
Description
Moves a widget specified by widgetID to the location (x, y). If the widget is
visible, it is moved immediately to the new location. If the widget is
hidden, it will appear at the new location when the Show procedure is
called. Note that the x and y specified here are the same as in the Create
method. For example, if you had specified a check box to be right justified
in the CreateCheckBoxFull function, then (x, y) in a call to SetPosition would
specify the lower-right corner as opposed to the lower-left corner.
Example
The following program moves the button every time the button is pressed.
import GUI in "%oot/lib/GUI"
var button : int
procedure MoveButton
var newX, newY : int
newX := Rand.Int (0, maxx – GUI.GetWidth (button))
newY := Rand.Int (0, maxy – GUI.GetHeight (button))
GUI.SetPosition (button, newX, newY)
end MoveButton
button := GUI.CreateButton (100, 100, 0, "Move Button",
MoveButton)
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling GUI.SetPosition,
not by calling SetPosition.
Win DOS Mac
GUI.SetPositionAndSize
Syntax
X TOOT Term
GUI.SetPositionAndSize ( widgetID, x, y : int,
Chapter 7 : Language Features 379
width, height : int )
Description
Changes the position and size of the widget specified by widgetID
simultaneously. The x, y, width and height parameters have the same
meaning as in the GUI.Create function for that widget. Any widget except
a menu or a menu item can be resized, although for some widgets, the
width or height parameter may be ignored.
GUI.SetPositionAndSize works the same way as the GUI.SetPosition and
GUI.SetSize procedures.
Example
The following program moves and resizes the button every time the button
is pressed.
import GUI in "%oot/lib/GUI"
var button, minWidth, minHeight : int
procedure MoveButton
var newX, newY, newWidth, newHeight : int
newWidth := max (minWidth, Rand.Int (0, 200))
newHeight := max (minHeight, Rand.Int (0, 100))
newX := Rand.Int (0, maxx – newWidth)
newY := Rand.Int (0, maxy – newHeight)
GUI.SetPositionAndSize (button, newX, newY,
newWidth, newHeight)
end MoveButton
button := GUI.CreateButton (100, 100, 0, "Move Button",
MoveButton)
minHeight := GUI.GetHeight (button)
minWidth := GUI.GetWidth (button)
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetPositionAndSize, not by calling SetPositionAndSize.
Win DOS Mac
GUI.SetScrollAmount
Syntax
X TOOT Term
GUI.SetScrollAmount (widgetID : int,
arrowInc, pageInc, thumbSize : int)
380 Object Oriented Turing Reference Manual
Description
Sets a scroll bar's arrow increment, page increment and thumb size.
Redraws the scroll bar to take into account the new thumb size. The
widgetID specifies the scroll bar to be changed. The arrowInc parameter is
the new arrow increment (the amount the scroll bar's value is changed
when the scroll arrows are pressed). A value of –1 means to use the
previously-defined arrow increment value. The pageInc parameter specifies
the new page increment (the amount the scroll bar's value is changed when
the user clicks in the page up/down section of the scroll bar). A value of –1
means to use the previously-defined page increment value. The thumbSize
parameter specifies he new thumb size. See the scroll bar explanation for
more detail on a scroll bar's thumb size. A value of –1 means to use the
previously-defined thumb size.
Example
The following program displays an image in a canvas in a window. If the
image is larger than the canvas, scroll bars to the bottom and left are used
to allow the user to see the entire image. A text field allows users to load
additional images whenever the "Load File" button is pressed.
% The "ScrollPic2" program.
import GUI in "%oot/lib/GUI"
% The maximum width/height of the canvas.
const maxSize : int := 220
const leftBorder : int := 15 % Left margin.
const bottomBorder : int := 25 % Bottom margin.
var h, v : int
% The scroll bars.
var canvas : int
% The canvas.
var pic : int
% The picture.
var fileNameField : int
% The file name text field.
var errorLabel : int
% The error message label.
var loadButton : int
% The "Load Picture" button.
procedure ScrollPic (ignore : int)
% Get the current value of the scroll bars.
var x : int := GUI.GetSliderValue (h)
var y : int := GUI.GetSliderValue (v)
GUI.PicDraw (canvas, pic, –x, –y, picCopy)
end ScrollPic
procedure LoadFile (fileName : string)
var picWidth, picHeight, canvasWidth, canvasHeight : int
var newPic : int := Pic.FileNew (fileName)
if newPic <= 0 then
GUI.SetLabel (errorLabel,
"Error loading picture: " + Error.LastMsg)
GUI.SetSelection (fileNameField, –1, –1)
return
else
GUI.SetLabel (errorLabel, "")
pic := newPic
end if
picWidth := Pic.GetWidth (pic)
picHeight := Pic.GetHeight (pic)
canvasWidth := min (picWidth, maxSize)
Chapter 7 : Language Features 381
canvasHeight := min (picHeight, maxSize)
% Hide the canvas and the three items, readjust them
% and then show them.
GUI.Hide (canvas)
GUI.Hide (h)
GUI.Hide (v)
GUI.SetSize (canvas, canvasWidth, canvasHeight)
GUI.SetSliderSize (h, canvasWidth + 1)
GUI.SetPosition (v, 15 + canvasWidth,
bottomBorder + GUI.GetScrollBarWidth - 1)
GUI.SetSliderSize (v, canvasHeight + 1)
GUI.SetSliderMinMax (h, 0, picWidth – 1)
GUI.SetSliderMinMax (v, 0, picHeight – 1)
GUI.SetScrollAmount (h, 3, 100, canvasWidth)
GUI.SetScrollAmount (v, 3, 100, canvasHeight)
GUI.SetSliderValue (h, 0)
GUI.SetSliderValue (v, picHeight)
GUI.Show (canvas)
GUI.Show (h)
GUI.Show (v)
ScrollPic (0)
end LoadFile
procedure LoadFileButton
var fileName : string := GUI.GetText (fileNameField)
LoadFile (fileName)
end LoadFileButton
View.Set ("graphics:265;295")
% We place the canvas first and everything else is placed
% relative to the canvas.
canvas := GUI.CreateCanvas (leftBorder,
bottomBorder + GUI.GetScrollBarWidth, maxSize, maxSize)
h := GUI.CreateHorizontalScrollBarFull (GUI.GetX (canvas),
GUI.GetY (canvas) – GUI.GetScrollBarWidth,
GUI.GetWidth (canvas), 0, 100, 0, ScrollPic, 3, 100, maxSize)
v := GUI.CreateVerticalScrollBarFull (
GUI.GetX (canvas) + GUI.GetWidth (canvas),
GUI.GetY (canvas), GUI.GetHeight (canvas), 0, 100,
100, ScrollPic, 3, 100, maxSize)
fileNameField := GUI.CreateTextField (GUI.GetX (canvas),
GUI.GetY (canvas) + GUI.GetHeight (canvas) + 10, 150, "",
LoadFile)
loadButton := GUI.CreateButton (GUI.GetX (fileNameField) +
GUI.GetWidth (fileNameField) + 20,
GUI.GetY (fileNameField), 0, "Load File", LoadFileButton)
errorLabel := GUI.CreateLabel (GUI.GetX (canvas), 5, "")
% Set the initial picture and return if it is not found.
GUI.SetText (fileNameField, "Forest.bmp")
LoadFileButton
if pic = 0 then
return
end if
382 Object Oriented Turing Reference Manual
loop
exit when GUI.ProcessEvent
end loop
Exported qualified.
Status
This means that you can only call the function by calling
GUI.SetScrollAmount, not by calling SetScrollAmount.
GUI.CreateHorizontalScrollBar and GUI.CreateVerticalScrollBar
See also
Win DOS Mac
GUI.SetSelection
X TOOT Term
Syntax
GUI.SetSelection ( widgetID, fromSel, toSel : int )
Description
Sets the selected text in the text field specified by widgetID. The value of the
fromSel and toSel parameters indicate the characters where the selection
will begin and end. For example, if the text was "Hello there", setting
fromSel to 2 and toSel to 5 would select "ell". Setting fromSel and toSel to –1
automatically selects the entire text.
The fromSel parameter specifies the start of the selection. This ranges from 1
(before the first character) to the number of characters in the text + 1 (after
the last character). A value of –1 for both fromSel and toSel selects the entire
text.
The toSel parameter specifies the end of the selection. This ranges from 1
(before the first character) to the number of characters in the text + 1 (after
the last character). A value of –1 for both fromSel and toSel selects the entire
text.
Example
The following program allows the user to type into a text field. When the
user presses ENTER, it searches for any non-lowercase text and if it finds
any, selects it to make it easy for the user to correct it. If all the input is
lower-case text, the program terminates.
import GUI in "%oot/lib/GUI"
var textField, lbl : int
procedure CheckInput (s : string)
for i : 1 .. length (s)
if (s (i) < 'a' or 'z' < s (i)) and s (i) not= ' ' then
GUI.SetSelection (textField, i, i + 1)
return
end if
Chapter 7 : Language Features 383
end for
GUI.Quit
end CheckInput
textField := GUI.CreateTextField (100, 100, 200, "", CheckInput)
lbl := GUI.CreateLabelFull (100 + GUI.GetWidth (textField) div 2,
100 + GUI.GetHeight (textField),
"Only Allows Lower Case Letters", 0, 0,
GUI.CENTER + GUI.BOTTOM, 0)
loop
exit when GUI.ProcessEvent
end loop
GUI.SetLabel (lbl, "Program Finished!")
Exported qualified.
Status
This means that you can only call the function by calling GUI.SetSelection,
not by calling SetSelection.
GUI.CreateTextField.
See also
Win DOS Mac
GUI.SetSize
X TOOT Term
Syntax
GUI.SetSize ( widgetID, width, height : int )
Description
Changes the size of the widget specified by widgetID. If the widget is
visible, its size is changed immediately, otherwise the widget will appear
in its new size when the widget is next made visible. Note that the width
and height parameters are no necessarily the actual width and height of the
widget. For example, the TextField widget ignores the height parameter,
calculating the widget's actual height from the height of the text in the
TextField.
Example
The following program resizes the button every time the button is pressed.
import GUI in "%oot/lib/GUI"
var button : int
procedure ResizeButton
var newWidth, newHeight : int
newWidth := Rand.Int (0, 200)
newHeight := Rand.Int (0, 200)
GUI.SetSize (button, newWidth, newHeight)
end ResizeButton
384 Object Oriented Turing Reference Manual
button := GUI.CreateButton (100, 100, 0, "Resize Button",
ResizeButton)
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling GUI.SetSize, not
by calling SetSize.
Win DOS Mac
GUI.SetSliderMinMax
X TOOT Term
Syntax
GUI.SetSliderMinMax ( widgetID, min, max : int )
Description
Sets the minimum and maximum values of the slider or scroll bar specified
by widgetID. The min parameter specifies the new minimum value of the
slider or scroll bar. The max parameter specifies the new maximum value
of the slider or scroll bar. The max parameter must be greater than the min
parameter.
GUI.SetSliderMinMax redraws the thumb to take into account the new
minimum and maximum. If the current value of the slider is outside the
new minimum/maximum, then the value is adjusted appropriately.
Example
See GUI.SetScrollAmount for an example of GUI.SetSliderMinMax.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetSliderMinMax, not by calling SetSliderMinMax.
See also
GUI.CreateHorizontalScrollBar, GUI.CreateVerticalScrollBar,
GUI.CreateHorizontalSlider, and GUI.CreateVerticalSlider.
Chapter 7 : Language Features 385
Win DOS Mac
GUI.SetSliderReverse
X TOOT Term
Syntax
GUI.SetSliderReverse ( widgetID : int )
Description
Sets a slider or scroll bar specified by widgetID into (or out of, if already
into) "reverse mode". Normally, a slider or scroll bar is at its minimum
value when the thumb is on the left hand side (bottom for a vertical slider).
This reverses it, so the minimum value is when the thumb is at the right
hand side (top for vertical sliders) of the track. Calling this routine a second
time reverses it back to normal. This procedure redraws the slider to move
the thumb to its new location.
Example
The following program creates two sliders, one of which is reversed.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:300;70")
var sBar, sBarLabel, reverseSBar, reverseSBarLabel : int
procedure SBarMoved (value : int)
GUI.SetLabel (sBarLabel, intstr (value))
end SBarMoved
procedure ReverseSBarMoved (value : int)
GUI.SetLabel (reverseSBarLabel, intstr (value))
end ReverseSBarMoved
sBar := GUI.CreateHorizontalScrollBar (10, 10, 250,
50, 150, 50, SBarMoved)
sBarLabel := GUI.CreateLabel (
GUI.GetX (sBar) + GUI.GetWidth (sBar) + 10, 10, "50")
reverseSBar := GUI.CreateHorizontalScrollBar (10, 40, 250,
50, 150, 50, ReverseSBarMoved)
GUI.SetSliderReverse (reverseSBar)
reverseSBarLabel := GUI.CreateLabel ( GUI.GetX (reverseSBar) +
GUI.GetWidth (reverseSBar) + 10, 40, "50")
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetSliderReverse, not by calling SetSliderReverse.
See also
GUI.CreateHorizontalScrollBar, GUI.CreateVerticalScrollBar,
GUI.CreateHorizontalSlider, and GUI.CreateVerticalSlider.
386 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.SetSliderSize
X TOOT Term
Syntax
GUI.SetSliderSize ( widgetID, length : int )
Description
Changes the length of a slider or scroll bar specified by widgetID to the
value specified by the length parameter. Redraws the slider or scroll bar
and changes the position of the thumb to take into account the new size of
the slider or scroll bar.
Example
See GUI.SetScrollAmount for an example of GUI.SetSliderSize.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetSliderSize, not by calling SetSliderSize.
See also
GUI.CreateHorizontalScrollBar, GUI.CreateVerticalScrollBar,
GUI.CreateHorizontalSlider, and GUI.CreateVerticalSlider.
Win DOS Mac
GUI.SetSliderValue
X TOOT Term
Syntax
GUI.SetSliderValue ( widgetID, value : int )
Description
Sets the value of a slider or scroll bar specified by widgetID to value. It
moves the thumb on the slider or scroll bar to the appropriate location and
calls the slider's action procedure with the new value.
Example
See GUI.SetScrollAmount for an example of GUI.SetSliderValue.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.SetSliderValue, not by calling SetSliderValue.
See also
GUI.GetSliderValue for reading a slider or scroll bar's value. See also
GUI.CreateHorizontalScrollBar, GUI.CreateVerticalScrollBar,
GUI.CreateHorizontalSlider, and GUI.CreateVerticalSlider.
Chapter 7 : Language Features 387
Win DOS Mac
GUI.SetText
X TOOT Term
Syntax
GUI.SetText ( widgetID : int, text : string )
Description
Sets the text of a text field specified by widgetID to text. The selection is set
to 1, 1 (i.e. the cursor is at the beginning of the text).
Status
Exported qualified.
This means that you can only call the function by calling GUI.SetText, not
by calling SetText.
Example
The following program converts all lower case input in the text field to
upper case when the user presses ENTER.
import GUI in "%oot/lib/GUI"
var textField, lbl : int
procedure CheckInput (s : string)
var newString : string := ""
for i : 1 .. length (s)
if 'a' <= s (i) and s (i) <= 'z' then
newString += chr (ord (s (i)) – 32)
else
newString += s (i)
end if
end for
GUI.SetText (textField, newString)
GUI.SetSelection (textField, –1, –1)
end CheckInput
textField := GUI.CreateTextField (100, 100, 200, "", CheckInput)
lbl := GUI.CreateLabelFull (100 + GUI.GetWidth (textField) div 2,
100 + GUI.GetHeight (textField), "Converts to Upper Case",
0, 0, GUI.CENTER + GUI.BOTTOM, 0)
loop
exit when GUI.ProcessEvent
end loop
Status
Exported qualified.
This means that you can only call the function by calling GUI.SetText, not
by calling SetText.
See also
GUI.CreateTextField.
388 Object Oriented Turing Reference Manual
Win DOS Mac
GUI.SetXOR
X TOOT Term
Syntax
GUI.SetXOR ( widgetID : int, xor : boolean )
Description
Sets the "xor mode" of the canvas specified by widgetID. If the xor parmeter
is set to true, the canvas is set to xor mode. When in xor mode, all the Draw...
procedures of a canvas are treated as if the View.Set ("xor") statement had
been executed before the Draw procedure.
Example
See GUI.SetDisplayWhenCreated for an example of GUI.Show.
import GUI in "%oot/lib/GUI"
View.Set ("graphics:400;300")
var canvas1, label1, canvas2, label2 : int
canvas1 := GUI.CreateCanvas (10, 20, maxx div 2 – 20, maxy – 30)
label1 := GUI.CreateLabelFull (10, 2, "XOR", maxx div 2 – 20, 0,
GUI.CENTER, 0)
canvas2 := GUI.CreateCanvas (maxx div 2 + 10, 20,
maxx div 2 – 20, maxy – 30)
label2 := GUI.CreateLabelFull (maxx div 2 + 10, 2, "Normal",
maxx div 2 – 20, 0, GUI.CENTER, 0)
GUI.SetXOR (canvas1, true)
for i : 1 .. 20
var x : int := Rand.Int (0, maxx div 2 – 20)
var y : int := Rand.Int (0, maxy – 20)
var c : int := Rand.Int (1, 15)
GUI.DrawFillStar (canvas1, x – 20, y – 20, x + 20, y + 20, c)
end for
GUI.SetXOR (canvas2, false)
for i : 1 .. 20
var x : int := Rand.Int (0, maxx div 2 – 20)
var y : int := Rand.Int (0, maxy – 20)
var c : int := Rand.Int (1, 15)
GUI.DrawFillStar (canvas2, x – 20, y – 20, x + 20, y + 20, c)
end for
Status
Exported qualified.
This means that you can only call the function by calling GUI.SetXOR, not
by calling SetXOR.
See also
GUI.CreateCanvas.
Chapter 7 : Language Features 389
Win DOS Mac
GUI.Show
X TOOT Term
Syntax
GUI.Show ( widgetID : int )
Description
Shows a widget specified by widgetID. Used in conjunction with GUI.Hide
to show and hide widgets. Hidden widgets cannot get events (i.e. respond
to keystrokes or mouse clicks). If an active text field (see text field) is
hidden, then any keystrokes in the window will be ignored.
In most cases where a widget is to appear, then disappear, then appear
again, it is advised to create the widget once and hide it until it is to
appear, whereupon GUI.Show is called. When the user is finished with the
widget, the widget is hidden using GUI.Hide. This saves the overhead of
creating and disposing of the same widget several times.
Example
See GUI.SetDisplayWhenCreated for an example of GUI.Show.
Status
Exported qualified.
This means that you can only call the function by calling GUI.Show, not by
calling Show.
See also
GUI.Hide.
Win DOS Mac
GUI.ShowMenuBar
X TOOT Term
Syntax
GUI.ShowMenuBar
Description
Shows the menu bar in the selected window.
Example
See GUI.SetMouseEventHandler for an example of GUI.HideMenuBar.
Status
Exported qualified.
This means that you can only call the function by calling
GUI.ShowMenuBar, not by calling ShowMenuBar.
See also
GUI.HideMenuBar. See also GUI.CreateMenu.
390 Object Oriented Turing Reference Manual
handler
exception handler
Dirty
A exceptionHandler is:
Syntax
handler ( id )
statementsAndDeclarations
end handler
Description
An exception handler is an optional block of statements and declarations
in a subprogram (or process). It is activated when the program (or process)
fails. This occurs, for example when dividing by zero.
Example
This program parses the input stream using a stack. If the stack overflows
(its top exceeds its maximum), a quit statement in the push procedure
aborts the parsing and gives control to the exception handler in the parse
procedure. The parse procedure calls parseExpn which calls push. If push
overflows the stack, it executes a quit and control is passed to the
exception handler in the parse procedure. The interrupted procedures
(parseExpn and push) are terminated and their local variables are deleted.
const stackOverflow := 500
const maxTop := 100
var top : 0 .. maxTop := 0
var stack : array 1 .. maxTop of int
procedure push ( i : int )
if top = maxTop then
quit : stackOverflow
end if
top := top + 1
stack ( top ) := i
end push
…
procedure parse
handler ( exceptionNumber )
put "Failure number ", exceptionNumber
case exceptionNumber of
label stackOverflow :
put "Stack has overflowed!!"
… other exceptions handled here …
% Unexpected failures
label :
quit >
% Pass exception further
end case
end handler
parseExpn
% Eventually push is called
end parse
Chapter 7 : Language Features 391
Details
See the quit statement for an explanation of its quitReason (stackOverflow in
the first quit statement above) and its guiltyParty (> in the second quit
statement, meaning the exception is due to causes outside of this handler).
An exception handler can appear only in the body of a subprogram (or
process), just preceding the declarations and statements. The form of a
procedure is:
procedure [ pervasive ] id
[ ( [ paramDeclaration {,paramDeclaration } ] ) ]
[ import [ [var] id {, [var] id } ] ]
[ pre trueFalseExpn ]
[ init id := expn {, id := expn } ]
[ post trueFalseExpn ]
[ exceptionHandler ]
statementsAndDeclarations
end id
Exactly the same declarations and statements can appear in a handler as
can appear in the subprogram body following the handler. In the absence
of exceptions, handlers have no observable effect. A particular handler is
activated (it becomes ready to handle an exception) when it is encountered
during execution. It remains active until the subprogram (or process)
containing it has completed, or the handler is given control. Activation of a
handler when a previous handler is already active will cause exceptions to
be passed to the newly-activated handler. In other words, handlers have a
dynamic scope that begins when the exception handler is encountered and
ends when the subprogram (or process) containing the handler has
terminated or the handler is given control.
When a handler is given control, it becomes, in effect, a replacement for the
declarations and statements following it. If the handler is in a function, it
must terminate with a result statement or with a quit. If the handler is in a
procedure (or process), the handler must terminate with a return, a quit, or
by encountering the end of the handler (which is equivalent to a return).
When a handler terminates with a result or return statement (or by
reaching the end of a procedure's handler), the subprogram's post
condition (if any) must be true. A quit statement does not need to establish
the post condition.
Programming with exception handlers easily leads to incomprehensible
software, due to the difficulty of keeping track of the flow of control. One
of the most insidious situations is when an exception occurs in a module,
class or monitor and is propagated outside of the unit. This can leave the
contained data in an inconsistent state; in the case of a monitor, it is left
locked forever. To avoid this possibility, you can use a handler in each
exported subprogram. If an exception in a process is not handled, the entire
program is aborted. If an implementation allocates dynamic arrays on the
heap, an exception may prevent the deallocation of such an array.
392 Object Oriented Turing Reference Manual
Without exception handling, a program executes according to the language definition or
else is aborted. If an exception handler is active, instead of aborting, control
is given to the handler. The quitNumber for a system-detected failure is
implementation-dependent. There is a file "%exceptions" which lists these
numbers. The user program can simulate a system exception by doing a
quit with the corresponding number.
If the user turns off checking explicitly, the system may not detect failures.
In some cases the failure may yield incorrect data or arbitrary behavior.
Some exceptions are unpredictable or implementation-dependent. For
example, in x := 24 div i + 24 / i, if i is zero, the exception could be either an
integer or a real division by zero, because the order or evaluation is
implementation-dependent.
hasch
has character function
Syntax
hasch : boolean
Description
The hasch procedure is used to determine if there is a character that has
been typed but not yet been read.
Example
The flush procedure gets rid of any characters that have been typed but not
yet read.
procedure flush
var ch : string ( 1 )
loop
exit when not hasch
getch ( ch ) % Discard this character
end loop
end flush
Details
The screen should be in a "screen" or "graphics" mode. See the setscreen
procedure for details. If the screen is not in one of these modes, it will
automatically be set to "screen" mode.
See also
getch and getchar.
See also predefined unit Input.
Chapter 7 : Language Features 393
id
Description
(identifier) name of an item in a program
Variables, constants, types, procedures, etc. in Turing programs are given
names such as incomeTax, x, and height. These names are called identifiers
(ids).
An identifier must start with a letter (large or small) or an underscore ( _ )
and can contain up to 50 characters, each of which must be a letter, a digit
(0 to 9) or an underscore ( _ ). Large and small letters are considered
distinct, so that A and a are different names. This differs from Pascal
where large and small letters in names are equivalent.
Every character in a name is significant in distinguishing one name from
another.
By convention, words that make up an identifier are capitalized (except the
first one), as in incomeTax and justInTime.
An item in a Turing program cannot be given the same name as a keyword
such as get or as a reserved word such as index. See Appendix A for a list
of keywords and reserved words. As well, there are some identifiers that
are used by the Turing error recovery procedures and are thus unavailable
for use as identifiers. Specifically, they are: endif, elseif, endloop and endfor.
if
Syntax
statement
An ifStatement is:
if trueFalseExpn then
statementsAndDeclarations
{ elsif trueFalseExpn then
statementsAndDeclarations }
[ else
statementsAndDeclarations ]
end if
394 Object Oriented Turing Reference Manual
Description
An if statement is used to choose among a set of statements (and
declarations). One set (at most) is chosen and executed and then execution
continues just beyond end if.
The expressions (the trueFalseExpressions) following the keyword if and
each elsif are checked one after the other until one of them is found to be
true, in which case the statements (and declarations) following the
corresponding then are executed. If none of these expressions evaluates to
true, the statements following else are executed. If no else is present and
none of the expressions are true, no statements are executed and execution
continues following the end if.
Example
Output a message based on value of mark.
if mark >= 50 then
put "You pass"
else
put "You fail"
end if
Example
Output A, B, C, D or F depending on mark.
if mark >= 80 then
put "A"
elsif mark >= 70 then
put "B"
elsif mark >= 60 then
put "C"
elsif mark >= 50 then
put "D"
else
put "F"
end if
Example
If x is negative, change its sign.
if x < 0 then
x := - x
Example
end if
If x is less than zero or greater than maxx, put a message.
if x < 0 or x > maxx then
put "Out of bounds!"
end if
Example
If the boolean flag is true and name is “stop”, put a message and return.
if flag and name = "stop" then
put "Exiting routine"
return
end if
Details
Several statements and declarations can appear after a particular then.
See also
case statements for another way to select among statements.
Chapter 7 : Language Features 395
#if
used for conditional compilation
A conditional compilation #if has the form:
Syntax
#if expn then
… any source text …
{ #elsif expn then
… any source text … }
[ #else
… any source text … ]
#end if
Description
An #if construct supports compile time selection of sections of source text
to make up a program (or unit of a program), in other words conditional
compilation. Any arbitrary source text (characters) can be selected.
Each of the selecting expressions (expns) have the form of a boolean
expression, with the use of the operators and, or and not (but not =>) and
parentheses. The short forms & and ~ are supported. The operands of the
expressions must be preprocessor flags, which are set by a systemdependent mechanism not described here. A flag is considered to be true if
it is explicitly set. If it is not explicitly set, it is considered false.
Unlike other parts of the language, the #if, #elsif, #else and #end if
constructs are not free format. Specifically, they must be placed by
themselves on a single line.
Example
A pair of declarations is chosen if both stats and debug are set, otherwise the
put statement is selected. The selected part becomes part of the program
and the other parts are ignored.
#if stats and debug then
var count : array 1 .. 5 of real
var message : string
#else
put "Debugging message"
#end if
396 Object Oriented Turing Reference Manual
implement
clause
An implementClause is:
Syntax
implement implementItem
Description
An implement clause is used to specify that the module, monitor or class
containing the clause is to be the implementation of another module,
monitor or class. This implementation is a special kind of expansion. The
module, monitor or class containing the clause gains access to (inherits) all
the declarations inside the target item. See inherit clause for rules about
expansions, which are also rules for implementations.
The implement clause can only be used in a unit. See unit for the
definition of a unit.
Example
Here is a stack module which defers all of its exported subprograms. This
module is an interface but not an implementation. Following stack is the
stackBody module that implements the stack module, giving the bodies for
stack's subprograms. Any call to stack's push or pop procedures, such as
stack.push("Ed"), will actually call the procedures given in stackBody.
module stack
% Interface
implement by stackBody
% stackBody has implementation
export push, pop
deferred procedure push ( s : string )
deferred procedure pop ( var s : string )
end stack
Next comes the expansion which gives the bodies for the deferred
procedures push and pop. The stackBody body also adds declarations for the
top and contents variables.
module stackBody
% Implementation
implement stack
% stack has interface
var top : int := 0
var contents : array 1 .. 100 of string
body procedure push % ( s : string )
top := top + 1
contents ( top ) := s
end push
body procedure pop
s := contents ( top )
top := top - 1
end pop
% ( var s : string )
end stackBody
Chapter 7 : Language Features 397
Details
Module, monitor or class D can be in C's implement-by clause if, and only
if, C is in D's implement clause. In other words, an interface must apply to
exactly one implementation and vice versa. A module can implement only
a module, a monitor only a monitor, and a class only a class. Classes (but
not modules and monitors) can contain inherit clauses. A class cannot
contain both an inherit and an implement clause.
An implementItem is one of:
(a) id
(b) id in fileName
The second form is used when the implement clause is for a separate unit
and the imported item is in a file whose name is different from the item's
name, as in:
implement ledger in "ledg.t"
The fileName must be an explicit character string, e.g., "ledg.t". See also
unit. Parentheses are allowed around the items in implement clauses, as
in:
implement ( ledger in "ledg.t" )
There is no restriction on the declarations that an interface may contain. In
particular, an interface (any module, monitor or class containing an
implement-by clause), can contain subprogram bodies and variable
declarations, exactly as is the case in expansions. This is different from
languages such as C++ in which there are strict rules limiting what you can
put in an interface.
Even though D contains an implement clause, D can also contain an
implement-by clause, which implies further implementation by further
automatic expansion.
Suppose class D is in class C's implement-by clause and that p is a pointer
to class C:
var p : ^ C
Even though C is implemented by D, p remains a pointer to class C. Each
creation of an object of class C actually creates an object of type D, for
example:
new p
% Creates object of class D
Class D, which implements C, could also have an implement-by clause,
which causes its implementation to be automatically created and so on. If
another class E inherits C, this expansion does not include D.
If the new statement contains an explicit class name E that is a descendant
of C (but not actually C), as in
new E, p
the object of the explicit class is created. If E has an implement-by clause,
the expansion is created.
See also
unit, module, monitor and class. See also implement by clause, inherit
clause, export list, and import list. See also deferred subprograms.
398 Object Oriented Turing Reference Manual
implement by
Syntax
clause
An implementByClause is:
implement by implementByItem
Description
An implement-by clause is used to specify that a module, monitor or class
C is to be automatically implemented by the implementByItem. C is called
the interface and the implementByItem, which must contain an implement
clause, is called the implementation. See implement clause for details and an
example.
The implement-by clause can only be used in a unit. See unit for the
definition of a unit.
An implementByItem is one of:
(a)
id
(b)
id in fileName
The second form is used when the implement-by clause is for a separate
unit and the imported item is in a file whose name is different from the
item's name, as in:
implement by ledgerBody in "ledgbod.t"
The fileName must be an explicit character string, e.g., "ledgbod.t". See also
unit. Parentheses are allowed around the items in an implement-by
clauses, as in:
implement by ( ledgerBody in "ledgbod.t" )
import
Syntax
list
An importList is:
import [ howImport ] importItem
{, [howImport] importItem }
Chapter 7 : Language Features 399
Description
An import list is used to specify those items that a procedure, function,
module, monitor, or a class uses from outside of itself. Note that a function
or procedure is not allowed to have an import list and thus automatically
imports whichever functions or procedures are used by the function or
procedure. The compiler determines the list automatically by looking to see
what items are actually used.
Example
In this example, the type T is imported into the stack module and used as
the type that can be pushed onto or popped off the stack. Since no other
items are imported, the only identifiers from outside of stack that can be
used in it must be predefined, such as sqrt, or declared to be pervasive.
type T : string
…
module stack
import T
export push, pop
var top : int := 0
var contents : array 1..100 of T
procedure push … end push
procedure pop … end pop
end stack
Details
The importItem is one of:
(a) id
(b) id in fileName
The second form is used in OOT when the list is the import list for a
separate unit (or the main program), and the imported item is in a file
whose name is different from the item's name, for example:
import ledger in "newledg.t"
The fileName must be an explicit character string. See also unit.
Parentheses are allowed around the items in an import lists, as in:
import ( ledger in "newledg.t" )
There are various ways to import items, as determined by howImport. The form of
howImport is one of:
(a) var
(b) const
(c) forward
Commonly the howImport is omitted, which means the default access for
the item is the same access as the item has. In other words, a read-write
item that is imported without a howImport is imported read-write. A readonly symbol that is imported without a howImport is imported read-only.
If the importItem is forward, the import list is part of a forward procedure
or function declaration and the imported item is itself necessarily a
procedure or function. See forward declarations for details and an
example.
400 Object Oriented Turing Reference Manual
If the import list of a module, monitor or class is omitted, the
implementation assumes that the list is import( ), meaning that no items
are imported. For example, a module must explicitly import any global
identifiers that are not predefined or pervasive.
Circular (recursive) imports are not allowed. For example, if unit A imports
B then B cannot import A. However, circular usage of separately compiled
units is possible by separating the units into interfaces and bodies and
having the bodies import the interfaces. For example, if C is the parent
class of D, D can import C, but not vice versa.
In an expansion (or implementation), the import list of the expansion
augments the import list of the parent.
An overriding subprogram (in an expansion) ignores the import list of the
target subprogram and uses its own import list.
Turing initializes modules and monitors in order of importation.
Initialization begins with the main program, which first initializes its
imports in the order given in its import list, and then initializes itself.
unit, module, monitor and class. See also export list, inherit clause,
implement clause and implement by clause.
See also
in
member of a set
Syntax
in
Description
The in operator determines if an element is in a set.
Example
type rankSet : set of 0 .. 10
var rankings : rankSet := rankSet ( 0 ) % The set {0}
…
if 5 in rankings then …
Description
% Is 5 in the rankings set?
The not in operator is exactly the opposite of in. For example, 7 not in
rankings means the same as not (7 in rankings).
The element is required to be in the set's index type. In the above example
this is satisfied because element 5 is in the index type 0 .. 10.
The keyword in is also used in lists such as import lists. See import list.
See also
the set type, infix operators, and precedence of operators.
Chapter 7 : Language Features 401
include
source files
An includeConstruct is:
Syntax
include fileName
Description
An include is used to copy parts of files so that they become part of the
Turing program. This copying is temporary, that is, no files are changed.
The file name must be an explicit string constant such as "stdstuff".
Example
On IBM PC compatible computers, there are arrow keys that produce
character values such as 200 and 208. Let us suppose that a file called
arrows contains definitions of these values:
const upArrow := 200
const downArrow := 208
const rightArrow := 205
const leftArrow := 203
These definitions can be included in any program in the following manner:
include "arrows"
…
var ch : string ( 1 )
getch ( ch )
% Read one character
case ord ( ch ) of
label upArrow :
…handle up arrow…
label downArrow :
…handle down arrow…
label rightArrow :
…handle right arrow…
label leftArrow :
…handle left arrow…
label :
…handle any other key…
end case
Details
An include file can itself contain include constructs. This can continue to
any level, although a circular pattern of includes would be a mistake, as it
would lead to an infinitely long program.
It is common to save procedures, functions and modules in separate files.
The files are collected together using include.
Details
If the filename in the include statement starts with a "%", then Turing
searches the system directory for the file. See the editor reference for the
environment to see how to set the system directory. This method can be
used to allow the system administrator to easily supply a set of routines in
a file to a large number of users by placing it in one easy-to-find location.
402 Object Oriented Turing Reference Manual
If the system directory is set to "C:\TURING", then the line
Example
include "%sorting.t"
will include the file "C:\TURING\SORTING.T" in the program.
Details
Under OOT, there are several system directories available. The "%oot"
directory is the directory where all the OOT system files are located. The
"%home" directory is the user’s home directory.
Example
If the oot directory is set to "/usr/local/lib/oot" then the line
include "%oot/teacher/sorting.t"
will include the file "/usr/local/lib/oot/teacher/sorting.t" in the program.
index
find pattern in string function
Syntax
index (s , patt : string ) : int
Description
The index function is used to find the position of patt within string s. For
example, index ( "chair", "air" ) is 3.
Example
This program outputs 2, because "ill" is a substring of "willing", starting at
the second character of "willing".
var word : string := "willing"
put index ( word, "ill" )
Details
If the pattern (patt) does not appear in the string (s), index returns 0 (zero).
For example, here is an if statement that checks to see if string s contains a
blank:
if index ( s, " " ) not= 0 then …
The index is sometimes used to efficiently determine if a character is one of
a given set of characters. For example, here is an if statement that checks to
see if ch, which is declared using var ch : string (1), is a digit:
if index ( "0123456789", ch) not= 0 then …
If a string contains more that one occurrence of the pattern, the leftmost
location is returned. For example, index ("pingpong", "ng") returns 3.
If patt is the null string, the result is 1.
Chapter 7 : Language Features 403
indexType
An indexType is one of:
Syntax
(a)
(b)
(c)
(d)
(e)
Description
subrangeType
enumeratedType
namedType % Which is a subrange or enumerated type
char
boolean
An index type defines a range of values that can be used as an array
subscript, as a case selector, as a selector (tag) for a union type, or as the
base type of a set type.
Example
var z : array 1 .. 9 of real % 0..9 is an index type
type smallSet : set of 0 .. 2 % 0..2 is an index type
indirection
operator (@)
Dangerous
Syntax
targetType @ ( expn )
Description
The indirection operator @ is used to access values that lie at absolute
machine addresses in the computer's memory. This is dangerous and
implementation-dependent and can cause arbitrary corruption of data and
programs.
Example
Copy the byte value at memory location 246 into b and then set that
memory byte to zero.
var b : nat1
b := nat1 @ (246)
Details
% One byte natural number
nat1 @ (246) := 0
The form of targetType must be one of:
(a) [ id . ] typeId
(b) int, int1, int2 or int4
404 Object Oriented Turing Reference Manual
(c)
(d)
(e)
(f)
nat, nat1, nat2 or nat4
boolean
char [ ( numberOfCharacters ) ]
string [ ( maximumLength ) ]
(g)
addressint
In form (a) the beginning identifier id must be the name of a module,
monitor or class that exports the typeId. Each of numberOfCharacters and
maximumLength must be compile time integer expressions. These are the
same target types as in type cheats.
The indirection operator @ takes an integer as an address. This value must
fit in the range of addressint. See addressint. See also pointer types and
the ^ operator (which accesses objects located by pointers).
See also
cheat. See also explicitIntegerConstant (which explains how to write
hexadecimal constants, which are often used for addresses).
infix
Syntax
operator
An infixOperator is one of:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
(l)
(m)
(n)
(o)
+
*
/
div
mod
rem
**
<
>
=
<=
>=
not=
and
% Integer and real addition; set union;
% string catenation
% Integer and real subtraction; set difference
% Integer and real multiplication; set intersection
% Real division
% Truncating integer division
% Modulo
% Remainder
% Integer and real exponentiation
% Less than
% Greater than
% Equal
% Less than or equal; subset
% Greater than or equal; superset
% Not equal
% And (boolean conjunction)
Chapter 7 : Language Features 405
(p)
(q)
(r)
(s)
(t)
(u)
(v)
Description
or
% Or (boolean disjunction)
=> % Boolean implication
in
% Member of set
not in
% Not member of set
shr % Shift right
shl % Shift left
xor % Exclusive OR
An infix operator is placed between two values or operands to produce a
third value. For example, the result of 5 + 7 is 12. In some cases the
meaning of the operator is determined by its operands. For example, in
"pine" + "apple", the + operator means string catenation while in 5 + 7 it
means integer addition. There are also prefix operators (-, + and not), which
are placed in front of a single value. See prefix operator.
In expressions with several operators, such as 3 + 4 * 5, the precedence rules determine the
order in which the operation is done (see precedence for a listing of these
rules). In this example, the multiplication is done before the addition, so
the expression is equivalent to 3 + (4 * 5).
The numerical (integer or real) operators are +, -, *, /, div, mod, and **. All
of these except div produce a real result when at least one of their
operands is real. If both operands are integers, the result is an integer
except in the case of real division (/) which always produces a real result
regardless of the operands.
The div operator is like real division (/), except that it always produces an
integer result, truncating any fraction to produce the nearest integer in the
direction of zero.
The mod operator is the modulo and the rem operator is the remainder. The
sign of the result of mod operator is the same as the sign of the second
operand. The rem operator operates like the mod operator in Turing (and
in most other languages). It produces the remainder, which is the
difference between real division (/) and integer division (div). When both
operands are positive, this is the modulo. For example, 14 mod 10 is 4. If one
of the operands is negative, a negative answer may result, for example, –7
mod 2 is –1. See also the int and real types.
The comparison operators (<, >, =, <=, >=, not=) can be applied to numbers
as well as to enumerated types. They can also be applied to strings to
determine the ordering between strings (see string type for details). Arrays,
records, unions and collections cannot be compared. Boolean values (true
and false) can be compared only for equality (= and not=); the same
applies to pointer values. Set values can be compared using <= and >=,
which are the subset and superset operators. The not= operator can be
written as ~=.
Strings are manipulated using catenation (+) as well as substring
expressions (see substring) and the index function (see index). See also the
string type.
406 Object Oriented Turing Reference Manual
The operators to combine true/false values are and, or, and =>
(implication), as well as equality (= and not=). See also the boolean type.
The set operators are union (+), intersection (*), set difference (-), subset
(<=), superset (>=), and membership (in and not in). See also the set type.
The shr (shift right), shl (shift left) and xor (exclusive OR) operators accept
and produce natural numbers. See shr, shl, and xor.
inherit
inheritance clause
An inheritClause is:
Syntax
inherit inheritItem
Description
An inherit clause specifies that the class containing the clause is to be an
expansion of another class. This expansion is called inheritance. The class
containing the clause gains access to (inherits) all the declarations inside
the target item. Expansions are used to add new declarations and exports
and to support polymorphism (overriding subprograms).
Example
Here is an example of a stack class. Following it, we show another class,
called stackWithDepth, that inherits stack by adding a function called depth.
class stack
export push, pop
var top : int := 0
var contents : array 1 .. 100 of string
procedure push ( s : string )
top := top + 1
contents (top ) := s
end push
procedure pop ( var s : string )
s := contents ( top )
top := top - 1
end pop
end stack
Next comes an expansion, which inherits the internal declarations of the
stack class and adds the depth function.
class stackWithDepth
inherit stack
export depth
function depth : int
result top
Chapter 7 : Language Features 407
end push
Details
end stackWithDepth
Objects of the inherited class stackWithDepth are like objects of the parent
class stack, except there is an additional exported function named depth.
An inheritItem is one of:
(a) id
(b) id in fileName
The second form is used when the inherit clause is for a separate unit and
the imported item is in a file whose name is different from the item's name,
for example:
inherit ledger in "newledg.t"
The fileName must be an explicit character string, e.g., "newledg.t".
Parentheses are allowed around the item in an inherit clause, as in:
inherit ( ledger in "newledg.t" )
There is a special form of inherit clause, called an implement clause, that
is used to separate an interface from an implementation. Modules and
monitors, as well as classes, use these clauses. See implement clause and
implement by clause.
If class D inherits class C, we say that C is the parent and D is the child.
Class B is said to be an ancestor of class D (and D is the descendant of B) if B
and D are the same class, or if B is the parent of D, or if B is the parent of
the parent of D, etc. We write this as follows:
B <= D
% B is an ancestor of D
If B is an ancestor of D but not the same as D, we say B is a strict ancestor of
D. We write this as:
B<D
% B is a strict ancestor of D
We also use the notations D >= B, D > B and D = B with the obvious
meanings. All of these notations can be used in a program. Their main use
is in conjunction with objectclass, which determines the class of an object
located by a pointer. For example, if p is declared to be a pointer to a stack,
we can write the following to see if p currently locates an object with the
depth operation:
% Does the object located by p have the depth operation
if stackWithDepth <= objectclass(p) then
A pointer that locates an object created as class E can be assigned to a
pointer to class B, only if B is an ancestor of E. For example, a pointer to an
object that is a stackWithDepth can be assigned to a pointer to stack, but not
vice versa. The pointer nil can be assigned to any pointer variable, but the
value nil(C) can only be assigned to a pointer to an ancestor of C.
An object (located by a pointer) can be assigned to another object only if
they were created as objects of the same class. However, assignment of
objects that are monitors or that contain dynamic arrays or collections is
not allowed.
408 Object Oriented Turing Reference Manual
Circular (recursive) inherits are not allowed. For example, if unit B inherits A then A cannot
inherit B. Only one item is allowed in an inherit clause; in other words,
Turing supports single inheritance but not multiple inheritance.
See implement clause for a special kind of expansion that separates a
module, monitor or class' interface from its implementation. See class for
an example of polymorphism, in which an inheriting class overrides
subprograms of its parent class.
The initialization of a module, a monitor or an object is immediately
preceded by the initialization of the item that it inherits or implements (if
any). Correspondingly, if the item has an implement by clause, the
implementation is initialized immediately after the initialization of the
current item.
Within a class C, with ancestor B, you can force a call to exported
subprogram p using the form C.p (or B.p). This calls the subprogram
declared in C (or in B in the case of B.p), regardless of the actual class of the
object and any overriding of p. This is similar to the notation C::p of the
C++ language. This notation can only be used inside class C.
unit, module, monitor and class. See also export list, import list,
implement clause, implement by clause and deferred subprogram. See
also objectclass.
See also
init
array initialization
Syntax
init
Description
The init (initialization) keyword is used for two different purposes in
Turing. The most common is for initializing arrays, records and unions.
The less common is for recording parameter values in subprograms for
later use in post conditions.
Example
var mensNames : array 1 .. 3 of string :=
init ( "Tom", "Dick", "Harry" )
put mensNames ( 2 )
% This outputs Dick
var names : array 1 .. 2, 1 .. 3 of string :=
init ( "Tom", "Dick", "Harry",
"Alice", "Barbara", "Cathy")
put names ( 2, 1 ) % This outputs Alice
Chapter 7 : Language Features 409
Details
The order of initializing values for multi-dimensional arrays is based on
varying the right subscripts (indexes) most rapidly. This is called row major
order. Initialization of records and unions is analogous to initializing arrays.
Values are listed in the order in which they appear in the type. See array,
record, and union types.
Example
This procedure is supposed to set integer variable i to an integer
approximation of its square root. The init clause records the initial value of
i as j so it can be used in the post condition to make sure that the
approximation is sufficiently accurate. The name j can be used only in the
post condition and nowhere else in the procedure.
procedure intSqrt ( var i : int )
pre i >= 0
init j := i
post abs ( i - sqrt ( j ) ) <= 1
… statements to approximate square root…
end intSqrt
See also
pre and post assertions and procedure and process declarations.
Input
Description
All
This unit contains the predefined procedures that deal with handling input
on a character-by-character basis.
All routines in the Input module are exported unqualified. (This means
you can call the entry points directly.)
Entry Points
getch
Gets the next character in the keyboard buffer (procedure
with a string (1) argument).
hasch
Returns true if there are characters waiting in the keyboard
buffer.
getchar
Gets the next character in the keyboard buffer (function
returning a char).
Pause
Waits for a key to be pressed.
410 Object Oriented Turing Reference Manual
Input.getch
All
Syntax
getch ( var ch : string ( 1 ) )
Description
The getch procedure is used to input a single character without waiting for
the end of a line. The parameter ch is set to the next character in the
keyboard buffer (the oldest not-yet-read character).
Example
This program contains a procedure called getKey which causes the program
to wait until a key is pressed.
View.Set ("graphics")
procedure getKey
var ch : string (1)
getch (ch)
end getKey
for i : 1 .. 1000
put i : 4, " Pause till a key is pressed"
getKey
end for
Details
The screen should be in a "screen" or "graphics" mode. See the View.Set
procedure for details. If the screen is not in one of these modes, it will
automatically be set to "screen" mode.
Some keys, such as the left arrow key, insert key, delete key, and function
keys do not produce ordinary character values. These keystrokes are
returned by getch as special values. See Appendix D for these values.
Status
Exported unqualified.
This means that you can call the function by calling getch or by calling
Input.getch.
See also
hasch (has character) procedure which is used to see if a character has been
typed but not yet read.
Chapter 7 : Language Features 411
Input.getchar
All
Syntax
getchar : char
Description
The getchar function is used to input a single character without waiting for
the end of a line. The next character in the keyboard buffer (the oldest notyet-read character) is returned.
Example
This program contains a procedure called getKey which causes the program
to wait until a key is pressed.
View.Set ("graphics")
procedure getKey
var ch : char
ch := getchar
end getKey
for i : 1 .. 1000
put i : 4, " Pause till a key is pressed"
getKey
end for
Details
The screen should be in a "screen" or "graphics" mode. See the View.Set
procedure for details. If the screen is not in one of these modes, it will
automatically be set to "screen" mode.
Some keys, such as the left arrow key, insert key, delete key, and function
keys do not produce ordinary character values. These keystrokes are
returned by getch as special values. See Appendix D for these values.
Status
Exported unqualified.
This means that you can call the function by calling getchar or by calling
Input.getchar.
See also
hasch (has character) procedure which is used to see if a character has been
typed but not yet read.
412 Object Oriented Turing Reference Manual
Input.hasch
All
Syntax
hasch : boolean
Description
The hasch procedure is used to determine if there is a character that has
been typed but not yet been read.
Example
The flush procedure gets rid of any characters that have been typed but not
yet read.
procedure flush
var ch : string ( 1 )
loop
exit when not hasch
getch ( ch ) % Discard this character
end loop
end flush
Details
The screen should be in a "screen" or "graphics" mode. See the View.Set
procedure for details. If the screen is not in one of these modes, it will
automatically be set to "screen" mode.
Status
Exported unqualified.
This means that you can call the function by calling hasch or by calling
Input.hasch.
See also
getch and getchar.
Input.Pause
All
Syntax
Input.Pause
Description
The Input.Pause procedure simply waits for a key to be pressed and then
returns. It echoes the key pressed if echo mode is set. (See View.Set for
setting echo mode)
This subprogram helps avoid having to declare a variable declaration and
then make a call to getch or getchar.
Chapter 7 : Language Features 413
This program pauses after every name read from the file.
Example
var f : int
open : f, "data", get
loop
exit when eof (f)
get : f, name : *
put name
Input.Pause
end loop
Exported qualified.
Status
This means that you can only call the function by calling Input.Pause, not
by calling Pause.
int
integer type
Syntax
int
Description
The int (integer) type has the values … –2, –1, 0, 1, 2 … Integers can be
combined by various operators such as addition (+) and multiplication (*).
Integers can also be combined with real numbers, in which case the result
is generally a real number. An integer can always be assigned to a real
variable, with implicit conversion to real.
Example
var counter, i : int
var j : int := 9
Details
var tax := 0
% The type is implicitly int because 0 is an integer
See also explicitIntegerConstant. The real type is used instead of int when
values have fractional parts as in 16.837. See the real type for details.
The operators on integers are +, -, * (multiply), div (truncating integer
division), mod (integer remainder), ** (exponentiation), as well as
comparisons (+, not=, >, >=, <, <=). The operators and, or and xor can be
applied to non-negative integer values. The bit-wise boolean result is
produced as an integer (actually, as a natural number). The shr (shift right)
and shl (shift left) operators are also introduced.
Real numbers can be converted to integers using ceil (ceiling), floor, and
round (see descriptions of these functions). Integers can be converted to
real numbers using intreal, but in practice this is rarely used, because an
integer value used in place of a real value will be automatically converted
to real.
414 Object Oriented Turing Reference Manual
Integers can be converted to strings and back using intstr and strint.
Integers can be converted to corresponding ASCII (or EBCDIC) characters
using chr and ord. See the descriptions of these functions.
Pseudo-random sequences of integers can be generated using randint. See
randint.
In current implementations of Turing, the range of integers is from 2147483647 to 2147483647. In other words, the maximum size of integer is
2**31 - 1. See maxint. This range exists because integers are stored in 4
bytes. The remaining negative value, -2147483648 records uninitialization.
The types int1, int2 and int4 specify integers that fit into 1, 2 or 4 bytes.
The intn types (int1, int2 and int4) are not checked for initialization and
allow all their bit patterns as numbers.
The natural number type nat allows only the non-negative values: 0, 1, 2, 3, … Natural
number values can be used whenever integer values are expected and vice
versa, given that the value does not exceed the range of the expected type.
nat and intn.
See also
intn
n-byte integer type
Dirty
Syntax
(a)
(b)
(c)
Description
int1
int2
int4
% 1-byte integer
% 2-byte integer
% 4-byte integer
The intn (n-byte integer) types are machine-dependent types that occupy a
specified number of bytes. By contrast, the int type is in principle a
machine-independent and mathematical type (it overflows, however, when
the value is too large or small, that is, when the value does not fit into 4
bytes).
Example
var counter1 : int1 % Range is -128 .. 127
var counter2 : int2 % Range is -32768 .. 32767
var counter4 : int4 % Range is -2147483648 .. 2147483647
Details
In current implementations of Turing, the range of the int is -2147483647 to
2147483647, which means that the int4 type allows one more value, 2147483648. This extra value is used in int to represent the state of being
initialized. The intn types allow use of all possible values that fit into n
bytes and thereby cannot check for initialization.
Chapter 7 : Language Features 415
The intn types are like the C language types short int, int, and long int,
except that the number of bytes occupied by the C types depends on the
particular C compiler.
See also
the natn types which are n byte natural (non-negative) values. See also int
and nat.
intreal
integer-to-real function
Syntax
intreal ( i : int ) : real
Description
The intreal function is used to convert an integer to a real number. This
function is rarely used, because in Turing, an integer value can be used
where ever a real value is required. When the integer value is used where
a real value is required, the intreal function is implicitly called to do the
conversion from int to real.
See also
floor, ceil and round functions.
intstr
integer-to-string function
Syntax
intstr ( i : int [ , width : int [ , base : int ] ] ) : string
Description
The intstr function is used to convert an integer to a string. The string is
equivalent to i, padded on the left with blanks as necessary to a length of
width, written in the given number base. For example, intstr (14, 4,
10)="bb14" where b represents a blank. The width and base parameters are
both optional. If they are omitted, the string is made just long enough to
hold the value, and the number base is 10. For example, intstr (14, 4) =
"bb14" and intstr (–23 ) = "–23".
The width parameter must be non-negative. If width is not large enough to
represent the value of i, the length is automatically increased as needed.
The string returned by intstr is of the form:
{blank}[-]digit{digits}
416 Object Oriented Turing Reference Manual
where {blank} means zero or more blanks, [-] means an optional minus
sign, and digit{digit} means one or more digits. The leftmost digit is either
non-zero or else a single zero digit. In other words, leading zeros are
suppressed.
The letters A, B, C … are used to represent the digit values 10, 11, 12, …
The base must be in the range 2 to 36 (36 because there are ten digits and 26
letters). For example, intstr (255, 0, 16) = "FF".
The intstr function is the inverse of strint, so for any integer i,
strint ( intstr (i)) = i.
chr, ord and strint functions. See also the natstr and strnat functions. See
also explicitIntegerConstants for the way to write non base 10 values in a
program.
See also
invariant
assertion
An invariantAssertion is:
Syntax
invariant trueFalseExpn
Description
An invariant assertion is a special form of an assert statement that is used
only in loop and for statements and in modules, monitors, and classes. It is
used to make sure that a specific requirement is met. This requirement is
given by the trueFalseExpn. The trueFalseExpn is evaluated. If it is true, all is
well and execution continues. If it is false, execution is terminated with an
appropriate message. See assert, loop and for statements and the module
declarations for more details.
Example
This program uses an invariant in a for loop. The invariant uses the
function nameInList to specify that a key has not yet been found in an array
of names.
var name : array 1 .. 100 of string
var key : string
… input name and key …
function nameInList ( n : int) : boolean
for i : 1 .. n
if key = name ( i ) then
result true
end if
end for
result false
end nameInList
Chapter 7 : Language Features 417
for j : 1 .. 100
invariant not nameInList ( j - 1)
if key = name ( j) then
put "Found name at ", j
exit
end if
end loop
Win DOS Mac
Joystick
Description
X TOOT Term
This unit contains the predefined subprogram that deal with reading the
joystick in a Turing program. The routines allow you to get the current
joystick position and whether either one of the two buttons are pressed.
All routines in the Joystick module are exported qualified (and thus must
be prefaced with "Joystick."). All the constants are exported unqualified
and thus do not need the Joystick prefix.
Entry Points
joystick1, joystick2
GetInfo
joystick name contants (exported unqualified)
Reads the current value of a joystick and status of the
joystick buttons.
Win DOS Mac
Joystick.GetInfo
X TOOT Term
Syntax
Joystick.GetInfo ( joystick : int, var xPos, yPos : int,
btn1Pressed, btn2Pressed : boolean )
Description
Reads the position and button status of the joystick specified by the joystack
parameter. The x and y parameter are returned in the xPos and yPos
parameters. If button 1 or button 2 on the joystick are currently pressed,
btn1Pressed and btn2Pressed will be set to true. The joystick parameter can be
either joystick1 or joystick2.
The x and y positions vary from joyMin to joyMax. To use them with
respect to a screen, the coordinates returned from Joystick.GetInfo must
be translated into screen coordinates. The following formula can be used:
screenX = round (maxx * (xPos – joyMin) / (joyMax – joyMin))
418 Object Oriented Turing Reference Manual
screenY = round (maxy * (yPos – joyMin) / (joyMax – joyMin))
Details
The Joystick module contains undocumented subprograms for those who
need to access more than two buttons or axes on a joystick. Contact Holt
Software if you need more information.
Example
The following program outputs the current location of joystick #1 and
draws a cursor on the screen to point out where it is showing.
var jx, jy, x, y, ox, oy : int := –1
var b1, b2, oB1, oB2 : boolean := false
loop
Joystick.GetInfo (joystick1, jx, jy, b1, b2)
% Convert joystick coordinates into screen coordinates.
x = round (maxx * (jx – joyMin) / (joyMax – joyMin))
y = round (maxy * (jy – joyMin) / (joyMax – joyMin))
if x not= ox or y not= oy or b1 not= oB1 or b2 not= oB2 then
Text.Locate (1, 1)
put "x = ", x, " y = ", y, " b1 = ", b1, " b2 = ", b2
View.Set ("xor")
Draw.Line (ox – 10, oy, ox + 10, oy, brightred)
Draw.Line (ox , oy – 10, ox , oy + 10, brightred)
Draw.Line (x – 10, y, x + 10, y, brightred)
Draw.Line (x, y – 10, x, y + 10, brightred)
ox := x
oy := y
oB1 := b1
oB2 := b2
end if
end loop
Exported qualified.
Status
This means that you can only call the function by calling Joystick.GetInfo,
not by calling GetInfo.
length
of a string function
Syntax
length ( s : string ) : int
Description
The length function returns the number of characters in the string. The
string must be initialized. For example, length("table") is 5.
Example
This program inputs three words and outputs their lengths.
var word : string
for i : 1 .. 3
Chapter 7 : Language Features 419
get word
put length ( word )
end for
If the words are "cat", "robin" and "crow", the program will output 3, 5 and
4.
Details
The length function gives the current length of the string. To find the
maximum length of a string, use upper. For example, given the declaration
var s : string (10), upper (s) returns 10.
See also
upper.
Limits
Description
All
This unit contains constants and functions used in determining the
mathematical accuracy of the language.
All routines in the Limits module are exported qualified (and thus must be
prefaced with "Limits.") except maxint, maxnat, minint and minnat,
which are exported unqualified (this means you can call those entry points
directly).
Entry Points
DefaultFW
Default fraction width used in printing using the "put"
statement.
DefaultEW
Default exponent width used in printing using the "put"
statement.
minint
The minimum integer in Turing (exported unqualified).
maxint
The maximum integer in Turing (exported unqualified).
minnat
The minimum natural number in Turing (exported
unqualified).
maxnat
The maximum natural number in Turing (exported
unqualified).
Real numbers are represented in Turing as:
f * (radix ** e) or 0
where for non-zero f:
(1 / radix) <= abs (f) and abs (f) < 1.0
minexp <= e and e <= maxexp.
Radix
The "radix" (usually 2).
420 Object Oriented Turing Reference Manual
ln
NumDigits
The number of radix digits in f.
MinExp
"minexp" (the smallest exponent allowed).
MaxExp
"maxexp" (the largest exponent allowed).
GetExp
Function that returns the value of "e".
SetExp
Procedure that sets the value of "e".
Rreb
The relative round-off error bound.
natural logarithm function
Syntax
ln ( r : real ) : real
Description
The ln function is used to find the natural logarithm (base e) of a number.
For example, ln ( 1) is 0.
Example
This program prints out the logarithms of 1, 2, 3, … up to 100.
for i : 1 .. 100
put "Logarithm of ", i, " is ", ln ( i )
end for
Details
See also the exp (exponential) function. You cannot take the logarithm of
zero or a negative number.
Note
logn (i ) = ln (i ) / ln (n )
See also
exp (the exponentiation function).
See also predefined unit Math.
locate
Syntax
procedure
locate ( row, column : int )
Chapter 7 : Language Features 421
Description
The locate procedure is used to move the cursor so that the next output
from put will be at the given row and column. Row 1 is the top of the
screen and column 1 is the left side of the screen.
Example
This program outputs stars of random colors to random locations on the
screen. The variable colr is purposely spelled differently from the word
color to avoid the procedure of that name (used to set the color of output).
The row number is purposely chosen so that it is one less than maxrow.
This avoids the scrolling of the screen which occurs when a character is
placed in the last column of the last row.
setscreen ("screen")
var row, column, colr : int
loop
randint ( row, 1, maxrow - 1)
randint ( column, 1, maxcol)
randint (colr, 0, maxcolor)
color (colr)
locate (row, column )
put "*" .. % Use dot-dot to avoid clearing end of line
end loop
Details
The locate procedure is used to locate the next output based on row and
column positions. See also the locatexy procedure which is used to locate
the output based x and y positions, where x=0, y=0 is the left bottom of the
screen.
The screen should be in a "screen" or "graphics" mode. See the setscreen
procedure for details. If the screen is not in one of these modes, it will
automatically be set to"screen" mode.
See also
setscreen and drawdot.
See also predefined unit Text.
locatexy
graphics procedure
Pixel graphics only
Syntax
locatexy ( x , y : int )
Description
The locatexy procedure is used to move the cursor so that the next output
from put will be at approximately (x, y). The exact location may be
somewhat to the left of x and below y to force alignment to a character
boundary.
Example
This program outputs "Hello" starting at approximately (100, 50) on the
screen.
422 Object Oriented Turing Reference Manual
setscreen ("graphics")
locatexy ( 100, 50 )
put "Hello"
The locatexy procedure is used to locate the next output based on x and y
positions, where the position x=0, y=0 is the left bottom of the screen. See
also the locate procedure which is used to locate the output-based row and
column positions, where row 1 is the top row and column 1 is the left
column.
Details
The screen should be in a "graphics" mode. See the setscreen procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
setscreen and drawdot.
See also
See also predefined unit Text.
loop
statement
A loopStatement is:
Syntax
loop
statementsAndDeclarations
end loop
Description
A loop statement causes the statements (and declarations) in it to be
repeatedly executed. This continues until terminated by one of its enclosed
exit statements (or by an enclosed return or result statement).
Example
Output on separate lines: Happy, Happy, Happy, etc.
loop
put "Happy"
Example
end loop
Read words up to the word Stop.
var word : string
loop
get word
exit when word = "Stop"
end loop
Chapter 7 : Language Features 423
Details
A loop statement can contain more than one exit, or none at all (in which
case it is an infinite loop). When the exit when is at the beginning of the
loop, the loop works like Pascal's do while; when at the end, the loop
works like Pascal's repeat until.
Just preceding the statements and declarations, you are allowed to write an
"invariant clause" of the form:
invariant trueFalseExpn
This clause is equivalent to: assert trueFalseExpn.
lower
bound
Syntax
lower (reference [ , dimension ] ) : int
Description
The lower attribute is used to find the lower bound of an array, string,
char(n) or non-opaque subrange type. Since the lower bound is necessarily
known at compile time, lower is rarely used.
See also
upper which finds the upper bound.
Math
Description
All
This unit contains all the mathematical routines. There are three routines
that are part of the language, but are conceptually part of the Math unit.
All routines in the Math unit are exported unqualified. (This means you
can call the entry points directly.)
Descriptions of all the subprograms iin the Math module can be found in
this chapter.
Entry Points
abs*
The absolute value function.
arctan
The arctan function (radians).
arctand
The arctan function (degrees).
424 Object Oriented Turing Reference Manual
cos
The cosine function (radians).
cosd
The cosine function (degrees).
exp
The exponentiation function.
ln
The natural logarithm function.
max*
The maximum value function.
min*
The minimum value function.
sign
The sign function.
sin
The sine function (radians).
sind
The sine function (degrees).
sqrt
The square root function.
* Part of the language, conceptually part of the Math unit.
max
maximum function
Syntax
max ( expn , expn )
Description
The max function is used to find the maximum of two numbers (the two
expn's). For example, max ( 5, 7 ) is 7. If both numbers are int, the result is
int. If both numbers are nat (natural numbers), the result is nat. But if one
or both of the numbers are real, the result is real. See also the min function.
Example
This program outputs 85.72.
var x : real := 74.61
var y : real := 85.72
put max ( x, y )
Example
% Outputs 85.72
This program inputs 10 numbers and outputs their maximum.
var m, t : real
% Input first number
get m
for i : 2 .. 10
% Handle remaining 9 numbers
get t
m := max ( m, t )
end for
put "The maximum is ", m
Chapter 7 : Language Features 425
See also predefined unit Math.
See also
maxcol
maximum column function
Syntax
maxcol : int
Description
The maxcol function is used to determine the number of columns on the
screen.
Example
This program outputs the maximum column number.
put "Number of columns on the screen is ", maxrow
Details
For IBM PC compatibles as well as most UNIX dumb terminals, in "text" or
"screen" mode, maxcol = 80. For the default IBM PC compatible "graphics"
mode (CGA), maxcol = 40.
See also
locate procedure for an example of the use of maxcol.
maxcolor
graphics function
Syntax
maxcolor : int
Description
The maxcolor function is used to determine the maximum color number
for the current mode of the screen. The alternate spelling is maxcolour.
Example
This program outputs the maximum color number.
setscreen ("graphics")
…
put "The maximum color number is ", maxcolor
Details
The screen should be in a "screen" or "graphics" mode. If it is not, it will
automatically be set to "screen" mode. See setscreen for details.
For IBM PC compatibles in "screen" mode, maxcolor = 15. For the default
IBM PC compatible "graphics" mode (CGA), maxcolor = 3.
426 Object Oriented Turing Reference Manual
See also
drawdot and palette for examples of the use of maxcolor. See the color
procedure which is used for setting the currently active color.
maxint
maximum integer function
Syntax
maxint : int
Description
The maxint function is used to determine the largest integer (int) that can
be used in a program.
Example
This program outputs the maximum integer.
put "The largest integer that can be used is ", maxint
In current Turing and OOT implementations, int values are stored in 4
bytes, i.e., 32 bits. This determines the maximum int value, which is 2**31 –
1, equaling 2147483647.
Details
There is an anomaly in computer arithmetic in that the absolute value of
the largest negative integer is one larger than maxint. Turing reserves this
extra value to represent the uninitialized integer. This value can be
computed but any attempt to assign it to an int variable is detected as an
overflow. You can use this extra value by using the int4 type instead of int,
but this type has no initialization checking.
See also
maxnat and minint.
See also OOT predefined unit Math.
maxnat
maximum natural number function
Syntax
maxnat : nat
Description
The maxnat function is used to determine the largest natural number that
can be used in a program.
Chapter 7 : Language Features 427
Example
This program outputs the maximum natural number.
put "The largest natural number that can be used is ", maxnat
In current implementations, natural numbers are stored in 4 bytes, i.e., 32
bits. This determines the maximum natural number, which is 2**32 - 2,
equaling 4294967294.
Details
In four bytes it is possible to represent one more value, namely, 2**32 - 1 =
4294967295. This extra value is used in Turing to represent the
uninitialized natural number. Although it can be computed, any attempt to
assign it to a nat variable is detected as an overflow. You can use this extra
value by using the nat4 type instead of nat, but this type has no
initialization checking.
See also
maxint and minnat.
See also predefined unit Limits.
maxrow
maximum row function
Syntax
maxrow : int
Description
The maxrow function is used to determine the number of rows on the
screen.
Example
This program outputs the maximum row number.
put "Number of rows on the screen is ", maxrow
Details
For IBM PC compatibles, maxrow = 25. For many UNIX dumb terminals,
maxrow = 24.
See also
locate procedure for an example of the use of maxrow.
maxx
Syntax
graphics function
maxx : int
428 Object Oriented Turing Reference Manual
Pixel graphics only
Description
The maxx function is used to determine the maximum value of x for the
current graphics mode.
Example
This program outputs the maximum x value.
setscreen ("graphics")
…
put "The maximum x value is ", maxx
The screen should be in a "graphics" mode. If it is not, it will automatically
be set to "graphics" mode. See setscreen for details.
Details
For the default IBM PC compatible graphics mode (CGA), maxx = 319.
drawdot for an example of the use of maxx and for a diagram illustrating x
and y positions.
See also
maxy
graphics function
Pixel graphics only
Syntax
maxy : int
Description
The maxy function is used to determine the maximum value of y for the
current graphics mode.
Example
This program outputs the maximum y value.
setscreen ("graphics")
…
put "The maximum y value is ", maxy
Details
The screen should be in a "graphics" mode;. If it is not, it will automatically
be set to "graphics" mode. See setscreen for details.
For the default IBM PC compatible graphics mode (CGA), maxy = 199.
See also
drawdot for an example of the use of maxy and for a diagram illustrating x
and y positions.
Chapter 7 : Language Features 429
min
minimum function
Syntax
min ( expn , expn )
Description
The min function is used to find the minimum of two numbers (the two
expn's). For example, min ( 5, 7 ) is 5. If both numbers are int, the result is
int. If both numbers are nat (natural numbers), the result is nat. But if one
or both of the numbers are real, the result is real. See also the max function.
Example
This program outputs 74.61.
var x : real := 74.61
var y : real := 85.72
put min ( x, y )
% Outputs 74.61
This program inputs 10 numbers and outputs their minimum.
Example
var m, t : real
% Input first number
get m
for i : 2 .. 10
% Handle remaining 9 numbers
get t
m := min ( m, t )
end for
put "The minimum is ", m
See also
See also predefined unit Math.
minint
minimum integer function
Syntax
minint : int
Description
The minint function is used to determine the smallest integer (int) that can
be used in a program.
Example
This program outputs the maximum integer.
put "The smallest integer that can be used is ", minint
430 Object Oriented Turing Reference Manual
In current implementations, int values are stored in 4 bytes, i.e., 32 bits.
This determines the minimum int value, which is -2**31 -1, equaling 2147483647.
Details
There is an anomaly in computer arithmetic in that the absolute value of
the largest negative integer is one larger than maxint. Turing reserves this
extra value to represent the uninitialized integer. This value can be
computed but any attempt to assign it to an int variable is detected as an
overflow. You can use this extra value by using the int4 type instead of int,
but this type has no initialization checking.
See also
minnat and maxint.
See also predefined unit Limits.
minnat
minimum natural number function
Syntax
minnat : nat
Description
The minnat function is used to determine the smallest natural number that
can be used in a program.
Example
This program outputs the smallest natural number.
put "The smallest natural number that can be used is ", minnat
Details
In current Turing and OOT implementations, natural numbers are stored
in 4 bytes, i.e., 32 bits. However, the minimum natural number in all
implementations is 0. minnat is provided for purposes of symmetry with
minint, maxint and maxnat.
In four bytes it is possible to represent one more value, namely, 2**32 - 1 =
4294967295. This extra value is used in Turing to represent the
uninitialized natural number. Although it can be computed, any attempt to
assign it to a nat variable is detected as an overflow. You can use this extra
value by using the nat4 type instead of nat, but this type has no
initialization checking.
See also
minint and maxnat.
See also predefined unit Limits.
Chapter 7 : Language Features 431
mod
modulo operator
Syntax
mod
Description
The mod (modulo) operator produces the modulo of one number with
another. In other words, the result is always a number between 0 and the
second operand. If both operands are positive, the result is identical to the
remainder operator. For example, 7 mod 2 produces 1 and –12 mod 5
produces 3.
Example
In this example, hours is the current time. It is moved back and forth by a
random amount, but the final result must always be between 1 and 12 (the
mod operation produces a number between 0 and 11 and then 0 becomes
12).
var hours : int := 12
var hoursPassed : int
put "The time is now ", hours, " o'clock"
loop
randint (hoursPassed, -12, 12)
exit when hoursPassed = 0
if hoursPassed < 0 then
put hoursPassed, " hours before " ..
else
put hoursPassed, " hours later " ..
end if
put hours, " o'clock" ..
hours := (hours + hoursPassed) mod 12
if hours = 0 then
hours = 12
end if
put " it was ", hours, " o'clock"
end loop
Details
If the second operand is positive, then the result is always non-negative.
Likewise, if the second operand is negative, then the result is always nonpositive. If both operands are negative, the result is the same as the
remainder operator.
See also
infix operators, precedence of operators and the rem and div operators.
432 Object Oriented Turing Reference Manual
module
declaration
A moduleDeclaration is:
Syntax
module id
[ implement implementItem ]
[ implement by implementByItem ]
[ import [ var ] importItem
{, [ var ] importItem } ]
[ export [ howExport ] id {,[howExport ] id }]
statementsAndDeclarations
end id
Description
A module declaration creates a package of variables, constants, types,
subprograms, etc. The name of the module (id) is given in two places, just
after module and just after end. Items declared inside the module can be
accessed outside of the module only if they are exported. Items from
outside the module that are to be used in the module need to be imported
(unless they are predefined or pervasive).
Example
This module implements a stack of strings.
module stack
% Implements a LIFO list of strings
export push, pop
var top : int := 0
var contents : array 1 .. 100 of string
procedure push ( s : string )
top := top + 1
contents ( top ) := s
end push
procedure pop ( var s : string )
s := contents ( top )
top := top - 1
end pop
end stack
stack . push ( "Harvey" )
var name : string
stack . pop ( name )
% This sets name to Harvey
Chapter 7 : Language Features 433
Outside of the stack module, the procedures push and pop can be called using the notation
stack.push and stack.pop. This access is allowed because push and pop are
exported from the module. Other items declared in the module (top and
contents) cannot be accessed from outside because they are not exported.
Details
In some other programming languages, a module is called a package, cluster
or object.
A module declaration is executed (it is initialized) by executing its
declarations and statements. For example, the stack module is initialized by
setting the top variable to 0. This initialization executes all the statements
and declarations in the module that are not contained in procedures or
functions. The initialization is completed before any procedure or function
of the module can be called from outside the module. An exported
subprogram must not be called until initialization of the module is
complete.
A call to an exported procedure or function from outside the module
executes the body of that procedure or function (the module is not
initialized with each such call). See also monitor and class declarations.
The import list gives the names of items declared outside the module that
can be accessed inside the module. Since stack has no import list, it is not
allowed to access any names declared outside of it. See also import lists.
Separately-compiled units that are imported are initialized before the
importing unit.
The export list is used to implement information hiding, which isolates
implementation details inside the module. The export list gives the names
of items declared inside the module that can be used outside the module.
For example, push and pop are exported from stack. Each such use of an
exported item must be preceded by the module name and a dot, for
example, stack.push. (See unqualified for advice on how to avoid using the
prefix "stack."). Names that are not exported, such as top and contents,
cannot be accessed outside the module.
Procedures, functions, variables, constants and types can be exported;
modules, monitors or classes cannot be exported.
A class is essentially a template for creating individual modules (objects).
See class for details. A monitor is essentially a module in which only one
process can be active at a time. See monitor and process for details.
The opaque keyword is used (only) in export lists to precede exported type
names that have declarations in the module. Outside of the module, the
type will be distinct from all others types. This means, for example, that if
the opaque type is a record, its fields cannot be accessed outside of the
module. Opaque types are used to guarantee that certain items are
inspected and manipulated in only one place, namely, inside the module.
These types are sometimes called abstract data types. See also export lists,
which also describes unqualified and pervasive exports.
Implement and implement-by lists are used to separate a module's interface from its body.
This allows only a part of a module (its interface) to be visible to its users
(its importers), while hiding its implementation. See implement and
implement by lists.
434 Object Oriented Turing Reference Manual
Example
Use an opaque type to implement complex arithmetic.
module complex
export opaque value, constant, add,
… other operations …
type value :
record
realPt, imagPt : real
end record
function constant (realPt, imagPt: real ) : value
var answer : value
answer . realPt := realPt
answer . imagPt := imagPt
result answer
end constant
function add (L, R : value ) : value
var answer : value
answer . realPt := L . realPt + R . realPt
answer . imagPt := L . imagPt + R . imagPt
result answer
end add
… other operations for complex arithmetic go here …
end complex
var c,d : complex .value :=complex.constant ( 1, 5 )
% c and d become the complex number (1,5)
var e : complex .value := complex.add (c, d )
% e becomes the complex number (2,10)
Details
Module declarations can be nested inside other modules but cannot be
nested inside procedures or functions. A module must not contain a bind
as one of its (outermost) declarations. A return statement cannot be used as
one of the (outermost) statements in a module.
The syntax of a moduleDeclaration presented above has been simplified by leaving out pre,
invariant and post clauses; the full syntax is:
module id
[ implement implementItem ]
[ implement by implementByItem ]
[ import [ var ] importItem {, [ var ] importItem } ]
[ export [ howExport ] id {, [ howExport ] id } ]
[ pre trueFalseExpn ]
statementsAndDeclarations
[ invariant trueFalseExpn ]
statementsAndDeclarations
[ post trueFalseExpn ]
end id
Chapter 7 : Language Features 435
The true/false expression in the pre and post clauses must be true when
initialization reaches each of them. After that, these have no effect. The
true/false expression in the invariant must be true any time the module is
exited (when finishing initialization or when returning from an external
call to an exported subprogram) or called (via an exported subprogram).
These clauses (pre,post and invariant) are not inherited by expansions. For
example, if module B inherits A, the subprograms of B are bound by B's
clauses and not by A's.
unit, monitor and class. See also export list, import list, implement list,
implement by list, inherit list and deferred subprogram.
See also
monitor
declaration
A monitorDeclaration is:
Syntax
monitor id
[ implement implementItem ]
[ implement by implementByItem ]
[ import [ var ] importItem
{, [ var ] importItem } ]
[ export [ howExport ] id {,[howExport ] id }]
statementsAndDeclarations
end id
Description
A monitor is a special purpose module (see module) that is used with
concurrent processes (see process). At most, one concurrent process (see
process) can be active in a monitor at a time. This means that a process will
be blocked if it calls a monitor that is already active. The process will not be
allowed to proceed until the monitor is inactive. The monitor provides
mutually exclusive access to the monitor's internal data.
Example
This monitor controls access to the count variable so it can be updated by
two processes (the observer and the reporter) without being corrupted by
this concurrent access. Generally, it is not safe to have one process update a
variable that other processes are simultaneously accessing. The observer
process repeatedly increments the counter when it observes an event. The
reporter process repeatedly writes out the number of events that have
occurred since the last report, resetting the counter to zero.
monitor controller
export observe, report
436 Object Oriented Turing Reference Manual
var counter : int := 0
procedure observe
counter := counter + 1
end observe
procedure report (var n : int )
n := counter
counter := 0
end report
end controller
process observer
loop
… observe one event …
controller . observe
end loop
end observer
process reporter
var n : int
loop
controller.report ( n )
… report n events …
end loop
end reporter
fork observer
Details
% Activate the observer
fork reporter
% Activate the reporter
A monitor is essentially a module in which only one process can be active
at a time. See module declarations for details about initialization.
Initialization is the same for modules and monitors.
A monitor can contain wait statements (that put processes to sleep) and
signal statements (that wake them up again). These statements operate on
condition variables, which are essentially queues of sleeping processes.
A class is essentially a template for creating individual modules (objects).
See class for details. If the class declaration is preceded by the keyword
monitor, the created modules are actually monitors. Monitor classes can
only inherit (inherit from) other monitor classes.
The body of a monitor has the same form as that of a module, except that
modules, monitors and processes cannot be declared inside monitors, and
certain statements (wait and signal) are allowed in monitors.
Details
The syntax of a monitorDeclaration presented above has been simplified by
leaving out pre, invariant and post clauses. See module for an explanation
of these extra features. There is also an optional
compilerTimeIntegerExpression in the first line, which is explained below.
The full syntax is:
monitor id [ : compileTimeIntegerExpn ]
[ implement implementItem ]
[ implement by implementByItem ]
Chapter 7 : Language Features 437
[ import [ var ] importItem {, [ var ] importItem } ]
[ export [ howExport ] id {, [ howExport ] id } ]
[ pre trueFalseExpn ]
statementsAndDeclarations
[ invariant trueFalseExpn ]
statementsAndDeclarations
[ post trueFalseExpn ]
end id
If the optional compileTimeIntegerExpression is present, this is a device monitor. Its exclusive
access is enforced by an implementation-dependent trick, such as
executing it at a hardware priority level given by the expression. A device
monitor is restricted from calling monitors (directly or indirectly). This
restriction is imposed to eliminate the possibility of blocking a process with
a non-zero hardware priority (as this would inadvertently allow multiple
entry into a device monitor). It is the programmer's responsibility to meet
this restriction; the compiler will not in general enforce the restriction. The
current (1999) implementation ignores this compileTimeIntegerExpression.
Details
An unexported parameterless procedure in a monitor can be specified to be
an interrupt handling procedure by specifying a device in its header, using
the form:
procedure id [ : deviceSpecification ]
The deviceSpecification is a compile time natural number that designates, to
the implementation, the class of interrupts that effectively call this
procedure. Interrupt handling procedures cannot be called explicitly
within the program.
There are two restrictions that the programmer must follow when using
interrupt handling procedures; these restrictions will not necessarily be
enforced by the software. The first is that an interrupt handling procedure
must not execute a wait, either directly or indirectly, by calling another
procedure. The second is that the interrupt handling procedure must not
directly or indirectly cause an exception, unless the exception will be
caught by an exception handler that is activated directly or indirectly by
the interrupt handling procedure.
Details
Declarations of monitors within monitors are disallowed. This would be
redundant anyway, as only one process can be inside the outer monitor, so
the inner monitor is guaranteed to be successful.
Declarations of classes within monitors are also disallowed.
Any subprogram declared within a subprogram is now allowed to be
assigned to a subprogram variable, nor passed as a parametric
subprogram.
See also
unit, module and class. See also export list, import list, implement list,
implement by list and deferred subprogram.
438 Object Oriented Turing Reference Manual
Win DOS Mac
Mouse
Description
X TOOT Term
This unit contains the predefined subprograms that deal with using the
mouse in a Turing program. The routines allow you to get the current
mouse cursor position, check if a button has been pressed and get the
information if it has. There are also routines to hide and show the mouse
on systems where it makes sense. (On GUI based systems like the
Macintosh, the mouse can't be hidden as it may be needed by other
applications running at the same time.)
All routines in the Mouse module are exported qualified (and thus must be
prefaced with "Mouse.").
Entry Points
Where
Gets the current location of the mouse cursor and status of
the mouse buttons.
Hide
Makes the mouse cursor disappear.
Show
Displays the mouse cursor.
ButtonMoved Checks to see if a mouse button has been pressed.
ButtonWait
Gets information about a mouse button being pressed such
as where it was pressed, which button was pressed, etc.
ButtonChoose Selects the mode for the mouse (either single button mode
or multi-button mode).
Win DOS Mac
Mouse.ButtonChoose
X TOOT Term
Syntax
Mouse.ButtonChoose (choice : string)
Description
The Mouse.ButtonChoose procedure is used to change the mode of the
mouse. In Turing, the mouse can either be in "single-button mode" or in
"multi-button mode". In "single-button mode" the mouse is treated as a one
button mouse. A button is considered pressed when any button is pressed
and released only when all buttons have been released.
In Turing, the mouse starts in "single-button mode".
Chapter 7 : Language Features 439
The parameter choice can be one of "singlebutton", "onebutton" (which
switch the mouse into "single-button mode") or "multibutton" (which
switches the mouse into "multi-button mode").
Example
A program that displays the status of the mouse at the top left corner of the
screen.
Mouse.ButtonChoose ("multibutton")
var x, y, button, left, middle, right : int
Mouse.Where (x, y, button)
left := button mod 10
% left = 0 or 1
middle := (button - left) mod 100 % middle = 0 or 10
% right = 0 or 100
right := button - middle - left
if left = 1 then
put "left button down"
end if
if middle = 10 then
put "middle button down"
end if
if right = 100 then
put "right button down"
end if
Details
Under DOS, the mouse does not start initialized. When the first mouse
command is received (Mouse.Hide, Mouse.Show, Mouse.Where,
Mouse.ButtonMoved, Mouse.ButtonWait, Mouse.ButtonChoose) Turing
attempts to initialize the mouse. If successful, the mouse cursor will appear
at that time.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
Status
Exported qualified.
This means that you can only call the function by calling
Mouse.ButtonChoose, not by calling ButtonChoose.
See also
Mouse.ButtonMoved and Mouse.ButtonWait to get mouse events saved
in a queue. See also Mouse.Where to get the current status of mouse
button(s).
Win DOS Mac
Mouse.ButtonMoved
Syntax
X TOOT Term
Mouse.ButtonMoved (motion : string) : boolean
440 Object Oriented Turing Reference Manual
Description
The Mouse.ButtonMoved function indicates whether there is a mouse
event of the appropriate type on the mouse queue. Events are either "up",
"down", "updown" or "downup" events (although the "downup" and
"updown" are the same event).
The parameter motion must be one of "up", "down", "updown" or
"downup". If an event of the type requested is in the queue,
Mouse.ButtonMoved returns true. If the event is not in the queue, then
Mouse.ButtonMoved returns false.
In "single-button mode" (where the mouse is treated like a one-button
mouse), a "down" event occurs whenever all the buttons are up and a
button is pressed. An "up" event takes place when the last button is
released so that no buttons remain pressed.
In "multi-button mode", a "down" event occurs whenever any button is
pressed, and an "up" event occurs whenever any button is released.
Example
This program draws random circles on the screen until the user clicks the
mouse button, whereupon is starts drawing random boxes. Clicking the
mouse button switches between the two.
var circles: boolean := true
loop
var x, y, radius, clr: int
if Mouse.ButtonMoved ("down") then
var buttonnumber, buttonupdown : int
Mouse.ButtonWait ("down", x, y, buttonnumber,
buttonupdown)
circles := not circles
end if
x := Rand.Int (0, maxx)
y := Rand.Int (0, maxy)
radius := Rand.Int (0, 100)
clr := Rand.Int (0, maxcolor)
if circles then
Draw.FillOval (x, y, radius, radius, clr)
else
Draw.FillBox (x, y, x + radius, y + radius, clr)
end if
end loop
Example
This is an example demonstrating how to check for both character and
mouse input at the same time.
var ch : string (1)
var x, y, btnnum, btnupdown : int
loop
if hasch then
getch (ch)
Text.Locate (1, 1)
put "The character entered is a: ", ch
end if
if Mouse.ButtonMoved ("down") then
Mouse.ButtonWait ("down", x, y, btnnum, btnupdown)
Text.Locate (1, 1)
put "The button was clicked at position: ", x, ", ",y
Chapter 7 : Language Features 441
end if
Details
end loop
Mouse.ButtonMoved can be thought of as the mouse equivalent of hasch
in that they both check for something in a queue and both return
immediately.
Under DOS, the mouse does not start initialized. When the first mouse
command is received (Mouse.Hide, Mouse.Show, Mouse.Where,
Mouse.ButtonMoved, Mouse.ButtonWait, Mouse.ButtonChoose), Turing
attempts to initialize the mouse. If the initialization is successful, the mouse
cursor will appear.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
Status
Exported qualified.
This means that you can only call the function by calling
Mouse.ButtonMoved, not by calling ButtonMoved.
See also
Mouse.ButtonMoved to get mouse events saved in the queue. See also
Mouse.ButtonChoose to switch between "single-button mode" and "multibutton mode".
Win DOS Mac
Mouse.ButtonWait
X TOOT Term
Syntax
Mouse.ButtonWait (motion : string,
var x, y, buttonNumber, buttonUpDown : int)
Description
The Mouse.ButtonWait procedure gets information about a mouse event
and removes it from the queue.
The parameter motion must be one of "up", "down", "updown" or
"downup". If an event of the type requested is in the queue,
Mouse.ButtonWait returns instantly. If there isn't such an event,
Mouse.ButtonWait waits until there is one and then returns (much like
getch handles keystrokes).
In "single-button mode" (where the mouse is treated like a one-button
mouse), a "down" event occurs whenever all the buttons are up and a
button is pressed. An "up" event takes place when the last button is
released so that no buttons remain pressed.
In "multi-button mode", a "down" event occurs whenever any button is
pressed, and an "up" event occurs whenever any button is released.
442 Object Oriented Turing Reference Manual
The parameters x and y are set to the position of the mouse cursor when
the button was pressed. The parameter buttonnumber is set to 1 when in
"single-button mode". In "multi-button mode", it is set to 1 if the left button
was pressed, 2 if the middle button was pressed, and 3 if the right button
was pressed. The parameter buttonupdown is set to 1, if a button was
pressed and 0 if a button was released.
Example
This program draws lines. It starts a line where the user presses down and
continues to update the line while the mouse button is held down. When
the button is released, the line is permanently draw and the user can draw
another line.
var x, y, btnNumber, btnUpDown, buttons : int
var nx, ny : int
loop
Mouse.ButtonWait ("down", x, y, btnNumber, btnUpDown)
nx := x
ny := y
loop
Draw.Line (x, y, nx, ny, 0) % Erase previous line
exit when Mouse.ButtonMoved ("up")
Mouse.Where (nx, ny, buttons)
Draw.Line (x, y, nx, ny, 1) % Draw line to position
end loop
Mouse.ButtonWait ("up", nx, ny, btnNumber, btnUpDown)
Draw.Line (x, y, nx, ny, 2) % Draw line to final position
end loop
Example
This is an example demonstrating how to check for both character and
mouse input at the same time.
var ch : string (1)
var x, y, btnNum, btnUpDown : int
loop
if hasch then
getch (ch)
Text.Locate (1, 1)
put "The character entered is a: ", ch
end if
if Mouse.ButtonMoved ("down") then
Mouse.ButtonWait ("down", x, y, btnNum, btnUpDown)
Text.Locate (1, 1)
put "The button was clicked at position: ", x, ", ",y
end if
end loop
Details
Mouse.ButtonWait can be thought of as the mouse equivalent of getch in
that they both read something in a queue and both wait until they get the
thing they're looking for.
Under DOS, the mouse does not start initialized. When the first mouse
command is received (Mouse.Hide, Mouse.Show, Mouse.Where,
Mouse.ButtonMoved, MouseButtonWait, Mouse.ButtonChoose), Turing
attempts to initialize the mouse. If the initialization is successful, the mouse
cursor will appear.
Chapter 7 : Language Features 443
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
Exported qualified.
Status
This means that you can only call the function by calling
Mouse.ButtonWait, not by calling ButtonWait.
Mouse.ButtonWait to see if an appropriate event is in the queue. See also
Mouse.ButtonChoose to switch between "single-button mode" and "multibutton mode".
See also
Win DOS Mac
Mouse.Hide
X TOOT Term
Syntax
Mouse.Hide
Description
This function causes the mouse cursor to disappear. Mouse.Show makes
the mouse cursor re-appear. While the cursor is hidden, the program still
gets button events and Mouse.Where still functions normally. Mouse.Hide
has no effect on systems with Windows (Macintosh, X Windows, MS
Windows).
Example
A program that hides the mouse cursor whenever it is outside the box
(100, 100) - (200, 200).
var x, y, button : int
loop
Mouse.Where (x, y, button)
if x < 100 or x > 200 or y< 100 or y> 200 then
Mouse.Hide
else
Mouse.Show
end if
end loop
Details
Under DOS, the mouse does not start initialized. When the first mouse
command is received (Mouse.Hide, Mouse.Show, Mouse.Where,
Mouse.ButtonMoved, Mouse.ButtonWait, Mouse.ButtonChoose), Turing
attempts to initialize the mouse. If the initialization is successful, the mouse
cursor will appear.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
444 Object Oriented Turing Reference Manual
Exported qualified.
Status
This means that you can only call the function by calling Mouse.Hide, not
by calling Hide.
Mouse.Show to make the mouse cursor re-appear.
See also
Win DOS Mac
Mouse.Show
X TOOT Term
Syntax
Mouse.Show
Description
This function causes the mouse cursor to appear. Mouse.Hide makes the
mouse cursor disappear. Mouse.Show has no effect on systems with
Windows (Macintosh, X Windows, MS Windows).
Example
A program that hides the mouse cursor whenever it is outside the box
(100, 100) - (200, 200).
var x, y, button : int
loop
Mouse.Where (x, y, button)
if x < 100 or x > 200 or y < 100 or y > 200 then
Mouse.Hide
else
Mouse.Show
end if
end loop
Details
Under DOS, the mouse does not start initialized. When the first mouse
command is received (Mouse.Hide, Mouse.Show, Mouse.Where,
Mouse.ButtonMoved, Mouse.ButtonWait, Mouse.ButtonChoose), Turing
attempts to initialize the mouse. If the initialization is successful, the mouse
cursor will appear.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
Status
Exported qualified.
This means that you can only call the function by calling Mouse.Show, not
by calling Show.
See also
Mouse.Hide to make the mouse cursor disappear.
Chapter 7 : Language Features 445
Win DOS Mac
Mouse.Where
X TOOT Term
Syntax
Mouse.Where (var x, y, button : int)
Description
This Mouse.Where procedure is used to get current information about the
status of the mouse. The parameters x and y are set to the current location
of the mouse cursor. If the program is running on a system using windows,
the cursor may be outside the window. This means that x and y may be set
to values outside of the bounds of 0 to maxx and 0 to maxy. When running
under DOS, x, y and button are set to –1 if there is no mouse available.
The parameter button is set depending on the current mode. In "singlebutton mode" (where the mouse is treated like a one-button mouse), button is
set to 0 if all the mouse buttons are up, and 1 if any of the mouse buttons
are down. In "multi-button mode", button is assigned the sum of 1 if the left
button is down, 10 if the middle button is down, and 100 if the right button
is down. Thus if button has the value of 101, then it means that the left and
right mouse buttons were depressed.
Example
A program that displays the status of the mouse at the top left corner of the
screen.
var x, y, button : int
loop
Mouse.Where (x, y, button)
Text.Locate (1, 1)
if button = 0 then
put x : 4, " ", y : 4, " button up"
else
put x : 4, " ", y : 4, " button down"
end if
end loop
Details
Under DOS, the mouse does not start initialized. When the first mouse
command is received (Mouse.Hide, Mouse.Show, Mouse.Where,
Mouse.ButtonMoved, Mouse.ButtonWait, Mouse.ButtonChoose) Turing
attempts to initialize the mouse. If the initialization is successful, the mouse
cursor will appear.
Note that under DOS, there must be both a mouse attached and a mouse
driver (usually found in the autoexec.bat or config.sys files).
Status
Exported qualified.
This means that you can only call the function by calling Mouse.Where,
not by calling Where.
446 Object Oriented Turing Reference Manual
See also
Mouse.ButtonMoved and Mouse.ButtonWait to get mouse events saved
in a queue. See also Mouse.ButtonChoose to switch between "single-button
mode" and "multi-button mode".
Music
Description
This unit contains the predefined subprograms that deal with sound and
music. Some of these routines have not been implemented at the time of
the writing of this manual and will be implemented in future releases.
All routines in the Music module are exported qualified (and thus must be
prefaced with "Music.").
Entry Points
Play
Plays a series of notes.
PlayFile
Plays music from a file. File must be in an allowable
format.
Sound
Plays a specified frequency for a specified duration.
SoundOff
Immediately terminates any sound playing.
Win DOS Mac
Music.Play
X TOOT Term
Syntax
Music.Play ( music : string )
Description
The Music.Play procedure is used to sound musical notes on the computer.
Sounds are produced synchronously on a per process basis. This means
that when a process executes a Music.Sound or Music.Play command, it
stops until the command is finished. However, other processes will
continue to executing.
Example
This program sounds the first three notes of the C scale.
Music.Play ( "cde" )
Example
This program plays from middle C to one octave above middle C and
down again in 8th notes.
Chapter 7 : Language Features 447
Music.Play ( "8cdefgab>c" )
Music.Play ( "<bagfedc" )
Details
The syntax of the play string may be enhanced in the future.
The Music.Play procedure takes strings containing characters that specify
notes, rests, sharps, flats and duration. The notes are the letters a to g (or A
to G). A rest is p (for pause). A sharp is + and a flat is -. The durations are 1
(whole note), 2 (half note), 4 (quarter note), 8 (eight note) and 6 (sixteenth
note). The character > raises to the next octave and < lowers. For example,
this is the way to play C and then C sharp one octave above middle C with
a rest between them, all in sixteenth notes: Music.Play(">6cpc+"). Blanks
can be used for readability and are ignored by Music.Play.
Under some systems such as UNIX, the Music.Play procedure may have
no effect.
Status
Exported qualified.
This means that you can only call the function by calling Music.Play, not
by calling Play.
See also
the Music.Sound procedure, which makes a sound of a given frequency
(Hertz) and duration (milliseconds).
Win DOS Mac
Music.PlayFile
X TOOT Term
Syntax
Music.PlayFile ( fileName : string )
Description
The Music.PlayFile procedure is used to play a file of music. The file must
be in one of the acceptable formats and the machine, must have the
appropriate hardware.
The fileName parameter must give the format of the file:
WAV files
"WAV:filename" or "filename.WAV"
SND files
"SND:filename" or "filename.SND"
MOD files
"MOD:filename" or "filename.MOD"
Sounds are produced synchronously on a per process basis. This means
that when a process executes a Music.Sound, Music.Play or
Music.PlayFile command, it stops until the command is finished.
However, other processes will continue executing.
Example
This program plays the music in the file "branden3.wav" while drawing
ovals on the screen.
448 Object Oriented Turing Reference Manual
process DoMusic
loop
Music.PlayFile ( "branden3.wav" )
end loop
end DoMusic
fork DoMusic
var x, y, clr : int
loop
x := Rand.Int (0, maxx)
y := Rand.Int (0, maxy)
clr := Rand.Int (0, maxcolor)
Draw.FillOval (x, y, 30, 30, clr)
end loop
The file formats that each system can play will vary. As well, the correct
hardware must be available. The release notes document which format
each system can handle.
Details
At the time of writing, this subprogram was not supported on any
platform. Check the release notes to see which file types are supported in
the current version.
Exported qualified.
Status
This means that you can only call the function by calling Music.PlayFile,
not by calling PlayFile.
Win DOS Mac
Music.Sound
X TOOT Term
Syntax
Music.Sound ( frequency, duration : int )
Description
The Music.Sound statement is used to cause the computer to sound a note
of a given frequency for a given time. The frequency is in cycles per second
(Hertz). The time duration is in milliseconds. For example, middle A on a
piano is 440 Hertz, so Music.Sound(440, 1000) plays middle A for one
second.
Sounds are produced synchronously on a per process basis. This means
that when a process executes a Music.Sound or Music.Play command, it
stops until the command is finished. However, other processes will
continue executing.
Example
This program plays a siren sound in the background.
process siren
loop
Chapter 7 : Language Features 449
for i : 100 .. 3000 by 100
Music.Sound ( i, 50 ) % Sound note
end for
for decreasing i : 2900 .. 200 by 100
Music.Sound ( i, 50 ) % Sound note
end for
end loop
end siren
fork siren
… the rest of the program goes here while the siren continues …
Details
On IBM PC compatibles, the hardware resolution of duration is in units of
55 milliseconds. For example, Music.Sound(440, 500) will delay the
program by about half a second, but may be off by as much as 55
milliseconds.
Status
Exported qualified.
This means that you can only call the function by calling Music.Sound, not
by calling Sound.
See also
Music.Play statement, which plays notes based on musical notation. For
example, Music.Play("8C") plays an eighth note of middle C.
Win DOS Mac
Music.SoundOff
X TOOT Term
Syntax
Music.SoundOff
Description
The Music.SoundOff procedure stops any sound or music that is currently
playing or is waiting to play.
Status
Exported qualified.
This means that you can only call the function by calling Music.SoundOff,
not by calling SoundOff.
See also
Music.Play, Music.PlayFile, and Music.Sound procedures, which make
sounds that can be turned off with Music.SoundOff.
450 Object Oriented Turing Reference Manual
named
type
A namedType is one of:
Syntax
(a)
(b)
typeId
moduleId . typeId
Description
A type can be given a name (typeId) and later this name can be used
instead of writing out the type.
Example
In this example, phoneRecord is a named type.
type phoneRecord :
record
name : string ( 20 )
phoneNumber : int
address : string ( 50 )
end record
…
var oneEntry : phoneRecord
var phoneBook : array 1 .. 100 of phoneRecord
Details
Form (a) is the most common kind of named type. Form (b) is used when
the type name has been exported from a module.
Arrays whose bounds are not known at compile time cannot be named.
nargs
number of arguments
Syntax
nargs : int
Description
The nargs function is used to determine the number of arguments that
have been passed to a program from the command line. For example, if the
program is run from the Turing environment using
:r file1 file2
then nargs will return 2. If a program called prog.x is run under UNIX
using this command:
prog.x file1 file2
Chapter 7 : Language Features 451
the value of nargs will similarly be 2.
The nargs function is usually used together with the fetcharg function to
access the arguments that have been passed to the program.
fetcharg for an example of the use of nargs.
See also
nat
natural number type
Syntax
nat
Description
The nat (natural number) type has the values 0, 1, 2, 3 … Natural numbers
can be combined by various operators, such as addition (+) and
multiplication (*). Natural numbers can be combined with integers (type
int), in which case the result is an integer. Natural numbers can also be
combined with real numbers, in which case the result is generally a real
number. Natural numbers can always be assigned to real variables, with
implicit conversion to real.
Example
var counter : nat
Details
var j : nat := 9
See also explicitIntegerConstant. The nat type is used instead of int when the
values are known to be non-negative.
The Turing operators on natural numbers are the same as those for
integers: +, -, * (multiply), div (truncating integer division), mod (integer
remainder), ** (exponentiation), as well as comparisons (+, not=, >, >=, <,
<=). The operators and, or and xor to be applied to natural number values.
The bit-wise boolean result is produced as a natural number. The shr (shift
right) and shl (shift left) operators are also introduced.
In the current implementation, the range of natural numbers is from 0 to
4294967294. In other words, the maximum value of a natural number is
2**32 - 2. This range exists because natural numbers are stored in 4 bytes.
The types nat1, nat2 and nat4 specify natural numbers that fit into 1, 2 or 4
bytes.
Explicit constants such as 213 and 0 are considered to be integers. As a
result the type of tax in this declaration is int:
var tax := 0 % The type is int
Natural number values can be used whenever integer values are expected
and vice versa, given that the value does not exceed the range of the
expected type.
452 Object Oriented Turing Reference Manual
When integer and natural numbers are combined using a binary operator such as +, the
result is an integer. This means, for example, that if counter is a natural
number, counter + 1 is considered to be an integer. As long as the result fits
into the range that is the intersection of the ranges of int and nat, the result
will be as expected. Anomalies occur when the result is (or would be)
greater than the largest integer (maxint=2147483647). For example, if
natural number n is greater than maxint, the expression n + 1 will
overflow, because its result is an int (because 1 is an int). To avoid this
problem, you must be careful that both operands are natural numbers.
Suppose we have this declaration:
const natOne : nat := 1
We can safely compute n + natOne because both operands have type nat.
Natural numbers can be converted to real numbers using natreal, but in
practice this is rarely used, because a natural value used in place of a real
value will be automatically converted to real.
Natural numbers can be converted to strings and back using natstr and
strnat.
In the C language, a natural number is said to be "unsigned".
maxnat, int, natn, intn, natstr, strnat and natreal.
See also
natn
n-byte natural number type
Dirty
Syntax
(a)
(b)
(c)
Description
nat1
nat2
nat4
% 1-byte natural number
% 2-byte natural number
% 4-byte natural number
The natn (n-byte natural number) types are machine-dependent types that
occupy a specified number of bytes. By contrast, the nat type is in principle
a machine-independent and mathematical type (it overflows, however,
when the value is too large or small, that is, when the value does not fit
into 4 bytes).
Example
var counter1 : nat1
var counter2 : nat2
% Range is 0 .. 255
% Range is 0 .. 65536
var counter4 : nat4
% Range is 0 .. 4294967295
Chapter 7 : Language Features 453
Details
In Turing, the range of the nat is 0 to 4294967294, which means that the
nat4 type allows one more value, 4294967295. This extra value is used in
nat to represent the state of being uninitialized. The natn types allow use of
all possible values that fit into n bytes and thereby eliminates checking for
initialization.
The natn types are like the C language types short unsigned, unsigned, and
long unsigned, except that the number of bytes occupied by the C types
depends on the particular C compiler.
See also
the intn types which are n byte integer values. See also nat and int.
natreal
natural number to real function
Syntax
natreal ( n : nat ) : real
Description
The natreal function is used to convert a natural number to a real number.
This function is rarely used, because in Turing, a natural number can be
used anyplace a real value is required. When this is done, the natreal
function is implicitly called to do the conversion from nat to real. The
natreal function is similar to intreal, except that natreal handles values
that are larger than int values and does not handle negative values.
See also
nat. See also the intreal, floor, ceil and round functions.
natstr
natural-number-to-string function
Syntax
natstr ( n : nat [ , width : int [, base : int ] ] ) : string
Description
The natstr function is used to convert a natural number to a string. The
string is equivalent to n, padded on the left with blanks as necessary to a
length of width, written in the given number base. For example, natstr (14,
4, 10)="bb14" where b represents a blank. The width and base parameters are
both optional. It they are omitted, the string is made just long enough to
hold the value and the number base is 10. For example, natstr (23) = "23".
454 Object Oriented Turing Reference Manual
The width parameter must be non-negative. If width is not large enough to
represent the value of i, the length is automatically increased as needed.
The string returned by natstr is of the form:
{blank}digit{digits}
where {blank} means zero or more blanks and digit{digit} means one or
more digits. The leftmost digit is either non-zero, or a single zero digit; in
other words, leading zeros are suppressed.
The letters A, B, C … are used to represent the digit values 10, 11, 12, …
The base must be in the range 2 to 36 (36 because there are ten digits and 26
letters). For example, natstr (255, 0, 16) = "FF".
The natstr function is the inverse of strnat, so for any natural number n,
strnat ( natstr(n) ) = n.
See also
chr, ord and strnat functions. See also the intstr and strint functions. See
also explicitIntegerConstant for the way to write values in base 2 and base 16
in a program.
Win DOS Mac
Net
Description
X TOOT Term
The Net module allows TCP/IP equipped machines to communicate. In the
current implementation (WinOOT 3.0), this is available only under Win32
(Windows 95, 98, NT and later).
To allow two machines to communicate, there must be a server (which calls
Net,WaitForConnection) and a client (which calls Net.OpenConnection).
The server waits until a client connects and then starts communication
between the two. When a connection is established, a net stream is
returned that can be used in the same fashion as a file stream (i.e. using
puts and gets). Once the connection is finished, the programs call
Net.CloseConnection.
For ease of reading web pages, the Net.OpenURLConnection opens up a
URL for reading with the get statement. It is up to the user program to
interpret the HTML or file located at the URL.
All subprograms in the Net unit are exported qualified (and thus must be
prefaced with "Net.").
Entry Points
WaitForConnection
Waits until a client connects to a specified port.
OpenConnection
Opens a connection to a specified machine.
OpenURLConnection Opens a connection to a file specified by a URL.
Chapter 7 : Language Features 455
CloseConnection
Closes a specified connection.
BytesAvailable Returns the number of bytes available to be read from a
net stream.
CharAvailable Returns true if there is a character available to be read
from a net stream.
LineAvailable Returns true if there is a line of text available to be read
from a net stream.
TokenAvailable
Returns true if there is a token available to be read
from a net stream.
HostAddressFromName
Returns a host's address given its host name.
HostNameFromAddress
Returns a host's name given its address.
LocalAddress Returns the host name of the local machine.
LocalName
Returns the TCP/IP address of the local machine.
Win DOS Mac
Net.BytesAvailable
X TOOT Term
Syntax
Net.BytesAvailable ( netStream : int ) : int
Description
Returns the number of bytes available for reading from the net stream
specified by the netStream parameter.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Status
Exported qualified.
This means that you can only call the function by calling
Net.BytesAvailable, not by calling BytesAvailable.
See also
Net.CharAvailable, Net.LineAvailable, and Net.TokenAvailable.
456 Object Oriented Turing Reference Manual
Win DOS Mac
Net.CharAvailable
X TOOT Term
Syntax
Net.CharAvailable ( netStream : int ) : boolean
Description
Returns true if a character is waiting to be read from the net stream
specified by the netStream parameter. If Net.CharAvailable returns true,
then a single character can be read from the stream without blocking.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
The following program fragment reads a character from netStream only if
there is one waiting to be read.
Example
if Net.CharAvailable (netStream) then
var ch : char
get : netStream, ch
put ch ..
end if
Status
Exported qualified.
This means that you can only call the function by calling
Net.CharAvailable, not by calling CharAvailable.
See also
Net.BytesAvailable, Net.LineAvailable, and Net.TokenAvailable.
Win DOS Mac
Net.CloseConnection
X TOOT Term
Syntax
Net.CloseConnection ( netStream : int )
Description
Closes a network connection made with Net.OpenConnection or
Net.WaitForConnection. After the connection is closed, the net stream
cannot be used for any purpose on either side of the connection.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
Chapter 7 : Language Features 457
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
The following program fragment connects to port 5300 on the machine
specified by netAddress, sends the work OK to it and closes the connection.
Example
netStream := Net.OpenConnection (netAddress, chatPort)
if netStream <= 0 then
put "Unable to connect to ", netAddress
return
end if
put : netStream, "OK"
Net.CloseConnection (netStream)
Exported qualified.
Status
This means that you can only call the function by calling
Net.CloseConnection, not by calling CloseConnection.
Net.OpenConnection and Net.WaitForConnection.
See also
Win DOS Mac
Net.HostAddressFromName
X TOOT Term
Syntax
Net.HostAddressFromName (
hostName : string ) : string
Description
Returns the numeric TCP/IP address of the machine whose hostname is
specified by the hostName parameter.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Example
The following program prints out the hostname of the current machine.
var hostName : string := "www.holtsoft.com"
put "The machine address of ", hostName , " is ",
Net.HostAddressFromName (hostName)
Status
Exported qualified.
This means that you can only call the function by calling
Net.HostAddressFromName, not by calling HostAddressFromName.
See also
Net.HostNameFromAddress.
458 Object Oriented Turing Reference Manual
Win DOS Mac
Net.HostNameFromAddress
X TOOT Term
Syntax
Net.HostNameFromAddress (
hostAddr : string ) : string
Description
Returns the TCP/IP hostname of the machine whose numeric address is
specified by the hostAddr parameter.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
The following program prints out the hostname of the machine whose
TCP/IP numeric address is "128.100.5.1".
Example
var hostAddr : string := "128.100.5.1"
put "The machine name of ", hostAddr, " is ",
Net.HostNameFromAddress (hostAddr)
Status
Exported qualified.
This means that you can only call the function by calling
Net.HostNameFromAddress, not by calling LocalName.
See also
Net.HostAddressFromName.
Win DOS Mac
Net.LineAvailable
X TOOT Term
Syntax
Net.LineAvailable ( netStream : int ) : boolean
Description
Returns true if a line of input is waiting to be read from the net stream
specified by the netStream parameter. If Net.LineAvailable returns true,
then a line of input can be read from the stream without blocking.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
Chapter 7 : Language Features 459
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
The following program fragment reads a character from netStream only if
there is one waiting to be read.
Example
if Net.LineAvailable (netStream) then
var line : string
get : netStream, line : *
put line
end if
Status
Exported qualified.
This means that you can only call the function by calling
Net.LineAvailable, not by calling LineAvailable.
See also
Net.BytesAvailable, Net.CharAvailable, and Net.TokenAvailable.
Win DOS Mac
Net.LocalAddress
X TOOT Term
Syntax
Net.LocalAddress : string
Description
Returns the TCP/IP numeric address of the machine the program is
running on. The numeric address is of the form xxx.yyy.zzz.www where
each segment is a number from 0 to 255.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Example
The following program prints out the TCP/IP numeric address of the
current machine.
put "Your machine address is ", Net.LocalAddress
Status
Exported qualified.
This means that you can only call the function by calling
Net.LocalAddress, not by calling LocalAddress.
See also
Net.LocalName.
460 Object Oriented Turing Reference Manual
Win DOS Mac
Net.LocalName
X TOOT Term
Syntax
Net.LocalName : string
Description
Returns the TCP/IP hostname of the machine the program is running on.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Example
The following program prints out the hostname of the current machine.
put "Your machine name is ", Net.LocalName
Status
Exported qualified.
This means that you can only call the function by calling Net.LocalName,
not by calling LocalName.
See also
Net.LocalAddress.
Win DOS Mac
Net.OpenConnection
X TOOT Term
Syntax
Net.OpenConnection ( netAddr : string,
port : int ) : int
Description
Attempts to open a connection to port specified by the port parameter on
the machine specified by netAddr parameter. There must be a program
listening to that port for the connection to be made. In OOT, this is done
using the Net.WaitForConnection function.
If successful, Net.OpenConnection returns a network stream descriptor
which can be used with the put, get, read, and write statements and eof
function to send and receive data to the listening program. It is also the
parameter used for the Net.CloseConnection, Net.BytesAvailable,
Net.CharAvailable, Net.LineAvailable, and Net.TokenAvailable
functions.
Chapter 7 : Language Features 461
The netAddr parameter is a string specifying the net address of the machine
to be connected to. This can either be the full hostname or the numerical
address.
In general, system program listen in on ports with numbers below 1024.
Port numbers above 1024 are generally available for use by user created
programs.
The program will wait for an indeterminate amount of time to make the
connection. If it fails, it will return a non-positive value.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Example
The following program implements a "Chat" program. One user runs the
program on their machine as a server, which waits for another machine to
connect to it. The second user specifies the machine to connect to and then
connects. The two can then type at each other.
% The "Chat" program
const chatPort : int := 5055
var choice : int
loop
put "Enter 1 to run chat server"
put "Enter 2 to run chat session"
put "Choice: " ..
get choice
exit when choice = 1 or choice = 2
end loop
var netStream : int
var netAddress : string
if choice = 1 then
netStream := Net.WaitForConnection (chatPort, netAddress)
else
put "Enter the address to connect to: " ..
get netAddress
netStream := Net.OpenConnection (netAddress, chatPort)
if netStream <= 0 then
put "Unable to connect to ", netAddress
return
end if
end if
Draw.Cls
put "Connected to ", netAddress
var localRow : int := 2
var localCol : int := 1
var remoteRow := maxrow div 2
var remoteCol : int := 1
var ch : char
View.Set ("noecho")
462 Object Oriented Turing Reference Manual
loop
if hasch then
ch := getchar
put : netStream, ch
if ch = '\n' then
localRow := localRow mod (maxrow div 2) + 1
localCol := 1
Text.Locate (localRow, localCol)
put "" % Clear to end of line
Text.Locate (localRow, localCol)
else
Text.Locate (localRow, localCol)
put ch ..
localCol += 1
end if
end if
if Net.CharAvailable (netStream) then
get : netStream, ch
if ch = '\n' then
remoteRow := remoteRow mod (maxrow div 2) +
1 + (maxrow div 2)
remoteCol := 1
Text.Locate (remoteRow, remoteCol)
put "" % Clear to end of line
Text.Locate (remoteRow, remoteCol)
else
Text.Locate (remoteRow, remoteCol)
put ch ..
remoteCol += 1
end if
end if
end loop
Status
Exported qualified.
This means that you can only call the function by calling
Net.OpenConnection, not by calling OpenConnection.
See also
Net.WaitForConnection and Net.CloseConnection.
Win DOS Mac
Net.OpenURLConnection
X TOOT Term
Syntax
Net.OpenURLConnection ( urlAddr : string ) : int
Description
Attempts to open a http connection to pthe URL (Universal Resource
Locator) specified by the urlAddr.
Chapter 7 : Language Features 463
If successful, Net.OpenURLConnection returns a network stream
descriptor which can be used with the get statement and eof function to
read the web page located at the URL.
The program will wait for an indeterminate amount of time to make the
connection. If it fails, it will return a non-positive value.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Example
The following program prints out the contents of the file specified by the
user.
var url : string
put "Enter the URL to load: " ..
get url
var netStream : int
var line : string
netStream := Net.OpenURLConnection (url)
if netStream <= 0 then
put "Unable to connect to ", url
return
end if
loop
exit when eof (netStream)
get : netStream, line
put line
end loop
Net.CloseConnection (netStream)
Status
Exported qualified.
This means that you can only call the function by calling
Net.OpenURLConnection, not by calling OpenURLConnection.
See also
Net.CloseConnection.
Win DOS Mac
Net.TokenAvailable
Syntax
X TOOT Term
Net.TokenAvailable ( netStream : int ) : boolean
464 Object Oriented Turing Reference Manual
Description
Returns true if a line of input is waiting to be read from the net stream
specified by the netStream parameter. If Net.TokenAvailable returns true,
then a single token (character surrounded by whitespace) can be read from
the stream without blocking.
Details
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
The following program fragment reads a character from netStream only if
there is one waiting to be read.
Example
if Net.TokenAvailable (netStream) then
var token : string
get : netStream, token
put token
end if
Status
Exported qualified.
This means that you can only call the function by calling
Net.TokenAvailable, not by calling TokenAvailable.
See also
Net.BytesAvailable, Net.CharAvailable, and Net.LineAvailable.
Win DOS Mac
Net.WaitForConnection
X TOOT Term
Syntax
Net.WaitForConnection ( port : int,
var netAddr : string ) : int
Description
Listens for a connection at the port specified by the port parameter. When
another program connects to the port, then the function returns. The
address of the connecting machine is specified in the netAddr parameter
and the Net.WaitForConnection returns a network stream descriptor
which can be used with the put, get, read, and write statements and eof
function to send and receive data to the connecting program. It is also the
parameter used for the Net.CloseConnection, Net.BytesAvailable,
Net.CharAvailable, Net.LineAvailable, and Net.TokenAvailable
functions.
In OOT, the connection to a port is made with the Net.OpenConnection
function.
The netAddr parameter is a string specifying the net address of the machine
that connected to the port. It is the machines numerical address.
Chapter 7 : Language Features 465
In general, system program listen in on ports with numbers below 1024.
Port numbers above 1024 are generally available for use by user created
programs.
The program will wait for indefinitely for a connection to made to the port.
The Net module requires a TCP/IP stack to be installed and operating in
order to function. It does not communicate using any other protocols
Details
It is possible for Firewalls to interfere with the actions of the Net module,
preventing connections from taking place.
Example
See Net.OpenConnection for an example of Net.WaitForConnection.
Status
Exported qualified.
This means that you can only call the function by calling
Net.WaitForConnection, not by calling WaitForConnection.
Net.OpenConnection and Net.CloseConnection.
See also
new
statement
A newStatement is:
Syntax
new [ collectionOrClassId , ] pointerVariableReference
Description
A new statement creates (allocates) a new element and assigns its location
to the pointer variable. This element can be an object of a collection or class
or a value of a type. If the collectionOrClassId is omitted, the choice of
element is determined by the type of the pointer. For example, if the
pointer is to class C, an object of class C will be allocated.
Example
Using a collection, declare a list of records and allocate one record.
var list : collection of
record
contents : string ( 10 )
next : pointer to list % Short form: next : ^ list
end record
% Short form: var first : ^ list
var first : pointer to list
new list, first
Example
% Short form: new first
Using a class, create an object of that class. The object is located by the start
pointer.
class node
export var next, var name
name : string (25)
466 Object Oriented Turing Reference Manual
next : pointer to node % Short form: next : ^ node
end node
var start : pointer to node % Short form: var start : ^ node
new node, start
Example
% Short form: new start
Using a record type, declare a list of records and allocate one record.
type item:
record
contents : string ( 10 )
next : pointer to item % Short form: next : ^ item
end record
% Short form: var first : ^ item
var first : pointer to item
new first
Details
As the examples in this section show, a pointer can locate one of three
things: an object of a collection, an object of a class or a value of a type.
In the new statement, the collectionOrClassId can be omitted. If the pointer
locates a type, it must be omitted. The free statement is used to deallocate
an element.
An imported class can have one of its objects created (by the new statement) only if the
class is imported var.
If there is no more space to allocate an element, new will set the pointer to
be the nil value, and the program will continue executing.
If the pointer locates class C and C contains an implement by list, the
object created by new is the inherited object (through any number of levels
of implement by). The pointer, however, remains a pointer to C.
The form new p is a short form for new C, p when C is the class or
collection given in p's type.
If p is a pointer to class C and C has a descendant (expansion) class D, a
new statement can be used to allocate an object of type D, as in:
new D, p
% Allocates an object of class D
If D has an implement by clause, the expansion is created.
Details
The new statement can also be used to resize a flexible array. If an array
has been declared flexible using the syntax .
var name : flexible array indexType { , indexType } of typeSpec
The indices may have compile-time or run-time upper bounds (the lower
bound must be compile-time). The upper bounds can be changed by using:
new name , newUpper1 {,newUpper2}
The existing array entries will retain their values, except that any index
made smaller will have the corresponding array entries lost. Any index
made larger will have the new array entries uninitialized (if applicable).
Additionally, the upper bound (both in the declaration and the new
statement) can be made one less than the lower bound. This effectively
makes an array that contains 0 elements. It can later be increased in size
with another new.
Chapter 7 : Language Features 467
In the current implementation (1999), with a multi-dimensional array with
a non-zero number of total elements, it is a run-time error to change any
but the first dimension (unless one of the new upper bounds is one less
than the corresponding lower bound, giving 0 elements in the array) as the
algorithm to rearrange the element memory locations has not yet been
implemented.
Currently, only variables can be declared in this form. There is no flexible
array parameter type, although a flexible array can be passed to an array
parameter with "*" as the upper bound.
Example
See array for an example of flexible arrays.
See also
class and collection declarations, pointer type, free statement, nil value
and implement by list.
For flexible arrays, see also array and flexible.
nil
pointer to a collection
Syntax
nil [ (collectionOrClassId ) ]
Description
The nil pointer does not locate any element (object). Pointers locate items
in collections, classes and types. The collectionOrClassId is optional.
This nil pointer is distinct from pointers to actual elements, and it can be
compared to these pointers. It is also distinct from the uninitialized pointer
value.
Example
In this example, the pointer called first is set to the nil pointer of collection
c, that is, to nil(c).
var c : collection of
record
name : string ( 50 )
next : pointer to c
end record
var first : pointer to c := nil ( c )
Details
See also collection, class and pointer. When nil is written without the
collectionOrClassId, it can be assigned to a pointer to any collection, class or
type.
The type of nil without the collectionOrClassId is effectively a pointer to
everyClass, an imaginary class that has no objects and is the descendant of
all classes. This implies that it can be assigned to any other class pointer,
because it is a descendant of all classes.
468 Object Oriented Turing Reference Manual
Turing allows you to write nil (id) after a forward declaration of id (the
name of a collection, class or type) before (and after) the resolution of the
id.
not
true/false (boolean) operator
Syntax
not
Description
The not (boolean negation) operator produces the opposite of a true/false
value. For example, not (x >y) is equivalent to x <= y.
Example
var error : boolean := false
var success : boolean
…
success := not error% success becomes the opposite of error
Details
The not operator takes true and produces false and takes false and
produces true. The not operator can be written as ~. See also the boolean
type, prefix operators, and precedence of operators.
The not operator can be applied to sets.
objectclass
of a pointer
Syntax
objectclass ( pointerExpn )
Description
The objectclass attribute is used to find the class of an object located by a
pointer. The pointerExpn must be an expression that is a pointer to a class.
Chapter 7 : Language Features 469
See class for an example of classes and inheritance, in which a class called
TextFile is inherited by a class called Device. The Device class adds a new
exported procedure called ioCtl. In the present example, objectclass is used
to test to make sure that the textFilePtr currently locates an object that was
created as a Device (or as a descendant of Device). The notation
Device(textFilePtr) converts the pointer to be a pointer to a Device so that
ioCtl can be called.
Example
var textFilePtr : ^ TextFile
…
if objectclass ( textFilePtr ) >= Device then
% Can safely treat object as a Device
Device ( textFilePtr ) . ioCtl
…
end if
Details
This example uses the class comparison operator >= which means "is a
descendant of". See class.
You can only use objectclass in class comparisons. In particular,
objectclass cannot be used to declare pointers. For example, this:
var p : ^objectclass (q)
is not allowed.
opaque
type
Description
When a type T is exported from module, monitor or class M using the
keyword opaque, the type M.T is distinct from all other types. Opaque
types are used to guarantee that updates to values of the type are done
within M.
See also
module declarations for an example of an opaque type used to implement
complex arithmetic. See also equivalence of types for the definition of the
type matching rules for opaque types.
open
Syntax
file statement
An openStatement is one of:
470 Object Oriented Turing Reference Manual
(a) open : fileNumberVar, fileName,
ioCapability
{ , ioCapability }
(b) open : fileNumberVar, argNum, ioCapability
{ , ioCapability }
Description
The open statement connects the program to a file so the program can
perform operations such as read on the file. In form (a), the open statement
translates a fileName, such as "Master", to a file number such as 5. Form (b),
which is less-commonly used, opens a file whose name is given by a
program argument. This is described below.
The read statement uses the file number, not the file name, to access the
file. When the program is finished using the file, it disconnects from the file
using the close statement. Each ioCapability is the name of an operation,
such as read, that is to be performed on the file.
Example
This programs illustrates how to open, read and then close a file.
var fileName : string := "Master"% Name of file
var fileNo : int
% Number of file
var inputVariable : string ( 100 )
open : fileNo, fileName, read
…
read : fileNo, inputVariable
…
close : fileNo
Details
The open statement always sets the fileNumber to a positive number. If the
open fails (generally because the file does not exist), the fileNumber is set to
zero.
An ioCapability is one of:
get, put, read, write, seek, mod
A file can be accessed using only the statements corresponding to the
input/output capabilities with which it was opened. Note: tell is allowed
only if the open is for seek.
The open statement truncates the file to length zero if the ioCapabilities
includes put or write but not mod (which stands for modify). In all other
cases, open leaves the existing file intact. The mod ioCapability specifies
that the file is to be modified without being truncated. Each open positions
to the beginning of a file. There is no mechanism to delete a file.
To open for appending to the end of the file, one has to open for seek, mod and for write or
put and then seek to the end of the file. See the seek statement.
Mixed mode files, which combine get and read (or put and write), are
supported by some operating systems, such as UNIX, but not by others,
such as MS-DOS.
Chapter 7 : Language Features 471
On the IBM PC, one should note that opening files in other directories uses
the backslash character. This is because the backslash is a special character
in Turing (as in \t for tab and \n for a newline). To get a single backslash,
use \\.
e.g. open : f, "C:\\STUDENTS\\SMITH\\ACCT.DAT", put
Form (b) of the syntax allows you to open a file whose name is given as a
program argument on the command line. For example, under UNIX, the
command line:
prog.x infile outfile
specifies to execute prog.x with program arguments infile and outfile.
Similarly, in the Turing programming environment, the run command can
accept program arguments. The argNumber is the position of the argument
on the command line. (The first argument is number 1.) The name of the
file to be opened is the corresponding program argument. If there is no
such argument, or if the file cannot be opened successfully,
fileNumberVariable is set to zero. See also nargs, which gives the number of
arguments, and fetcharg, which gives the n-th argument string.
Program argument files referenced by argument number and used in put,
get, read or write statements need not be explicitly opened, but are
implicitly opened with the capability corresponding to the input/output
statement in which they are first used. (The fileNumber gives the number of
the argument.)
The operating system standard files (error, output and input) are accessed
using file numbers 0, -1, and -2, respectively (although this may be subject
to change). These files are not opened explicitly, but are used simply by
using form (b) with the number. Beware of the anomalous case of a failed
open that gives you file number 0. A subsequent use of this number in a
put will produce output that goes to the standard error stream, with no
warning that the file you attempted to open is not actually being used.
To append to a file, the file must be opened with the mod and seek
capability and then there must be a seek to the end of file. For example:
var streamnumber : int
open : streamnumber, "myfile", put, mod, seek
seek : streamnumber, *
put : streamnumber, "This appears at the end of the file"
There is an older and still acceptable version of open that has this syntax:
open ( var fileNumber : int, fileName : string, mode : string)
The mode must be "r" (for get) or "w " (for put).
Details
The path name specified in the open statement and elsewhere can always
be in UNIX format (i.e. with forward slashes, an initial forward slash
indicating an absolute directory). On the PC, absolute paths would have
the form:
a:/dir1/dir2/filename
On the Macintosh, they would have the form:
/volume name/directory1/directory2/file name
472 Object Oriented Turing Reference Manual
Note that in addition to the UNIX path format, on the PC, you can always
use standard PC path notation and on the Macintosh, you can use standard
Macintosh path notation. On the Macintosh volume, directory and file
names can have spaces in them.
All routines (such as the File and Dir module routines) will return files
names in UNIX format, regardless of the machine the program is run on.
close, get, put, read, write, seek and tell statements.
See also
or
operator
Syntax
A or B
Description
The or (boolean) operator yields a result of true if at least one (or both) of
the operands is true. or is a short circuit operator. For example, if A is true
in A or B then B is not evaluated.
Example
var success : boolean := false
var continuing := true % the type is boolean
…
continuing := continuing or success
Details
continuing is set to false, if and only if, both continuing and success are false.
Since Turing uses short circuit operators, once continuing is true, success
will not be looked at.
The or operator can be applied to natural numbers. The result is the natural
number that is the bit-wise or of the operands. See nat (natural number).
See also
boolean (which discusses true/false values), explicitTrueFalseConstant
(which discusses the values true and false), precedence and expn
(expression).
Chapter 7 : Language Features 473
ord
character-to-integer function
Syntax
ord ( ch : char ) : int
Description
The ord function accepts an enumerated value, char, or a string of length 1,
and returns the position of the value in the enumeration, or of the character
in the ASCII (or EBCDIC for IBM mainframes) sequence. Values of an
enumerated type are numbered left to right starting at zero. For example,
ord ( "A" ) is 65. The ord function is the inverse of chr, so for any character
c, chr ( ord (c)) = c.
See also
chr, intstr and strint functions.
Win DOS Mac
palette
graphics procedure
X TOOT Term
Syntax
palette ( p : int )
Description
The palette procedure is used to change the palette number to p.
Example
This program shows all the colors of palette number 3 for an IBM PC
compatible using CGA graphics. The first line of output, for color 0, will
not be visible, because the background is also color 0.
setscreen ("graphics")
palette ( 3 )
for colorNumber : 0 .. maxcolor
color (colorNumber)
put "Color number ", colorNumber
end for
Details
The palette depends on the display hardware on the computer. On IBM PC
compatibles under CGA (the default graphics mode), there are palettes
numbered 0 to 3. The main palettes are numbers 2 and 3. Here is the
meaning of the color numbers under these CGA palette numbers.
Palette 2: 1 = cyan (blue), 2 = magenta (red), and 3 = white.
Palette 3: 1 = green, 2 = red, 3 = brown.
474 Object Oriented Turing Reference Manual
Palette number 0 is like 2 but not as bright. Palette 1 is like 3 but not as
bright.
The palette procedure is meaningful only in a "graphics" mode. See
setscreen for a description of the graphics modes.
whatpalette, which is used to determine the current palette number. See
also drawdot and maxcolor.
See also
See also predefined unit View.
Win DOS Mac
parallelget
parallel port function
X TOOT Term
Syntax
parallelget : int
Description
The parallelget procedure is used on a PC to read the value of certain pins
on the parallel port. This port corresponds to the MS-DOS device "LPT1".
This procedure can be used to control robots and peripherals.
Example
This program reads and prints the values of the five data pins of the PC’s
parallel port.
const val : int := parallelget
% Read in the set of pin values
put "Pin 10 is: ", (val div 64) mod 2
put "Pin 11 is: ", (val div 128) mod 2
put "Pin 12 is: ", (val div 32) mod 2
put "Pin 13 is: ", (val div 16) mod 2
put "Pin 15 is: ", (val div 8) mod 2
Details
The five pins that are used for parallel input are pins 10-15. The parallelget
procedure returns the sum of
64
Pin 10 high
128
Pin 11 high
32
Pin 12 high
16
Pin 13 high
8
Pin 15 high
The mod and div operators can be used to determine which pins are high
or low.
See also
the parallelput procedure for a diagram of the pins. That procedure is
used to set the values on the parallel port.
Chapter 7 : Language Features 475
Win DOS Mac
parallelput
parallel port procedure
X TOOT Term
Syntax
parallelput ( p : int )
Description
The parallelput procedure is used on a PC to set the values on the data
pins on the parallel port. This port corresponds to the MS-DOS device
"LPT1". This procedure can be used to control robots and peripherals.
Example
This program sets data bit 0, data bit 1 and so on to data bit 7.
for i : 0 .. 7
parallelput (2 ** i)
%Set data bit i on the parallel port
put "Data bit ", i, " or Pin ", i + 2, "has just been set"
end for
Details
The parallelput procedure is used to set the eight data bits on the PC’s
parallel port. These data bits 0 - 7 correspond to pins 2 - 9 on the parallel
port.
Data pins 7,6...,0
Parallel Pin Out - From the IBM PC
The value sent to parallelput is the sum of the following:
1
2
4
8
Data bit 0
Data bit 1
Data bit 2
Data bit 3
16
32
64
128
Data bit 4
Data bit 5
Data bit 6
Data bit 7
For example, the command parallelput (97) sets data bits 0, 5 and 6 high
(97 = 1 + 32 + 64) and sets the other data pins low. Because there are only 8
data pins in the parallel port, the value passed to parallelput must be from
0 to 255.
The parallelput procedure is not meant for sending a stream of characters
to the parallel port (for example, if you want to send the string "Hello" to
the printer). If you want to do this, open the file "LPT1" using the open
statement and put to the file.
476 Object Oriented Turing Reference Manual
the parallelget function, which is used to read the values of pins on the
parallel port.
See also
paramDeclaration
parameter declaration
A paramDeclaration is one of:
Syntax
(a)
(b)
Description
[ var ] id {, id } : typeSpec
subprogramHeader
A parameter declaration, which is part of the header of a procedure or
function, specifies a formal parameter (see also procedure and function
declarations). Form (a) above is the most common case. Form (b) specifies
procedures and functions that are themselves passed as parameters.
Example
procedure putTitle ( title : string )
% The parameter declaration is: title : string
put title
end putTitle
procedure x (var s : array 1 .. * of string (*))
% Set each element of s to the null string
for i : 1 .. upper ( s )
s ( i ) := ""
end for
end x
Details
Parameters to a procedure may be declared using var, which means that
the parameter can be changed inside the procedure. For example, s is
changed in the x procedure. If a parameter is declared without var, it
cannot be changed. (This differs from Pascal, where non-var parameters
can be changed.) Parameters to functions cannot be declared to be var.
Parameters declared var are passed by reference, which means that a
pointer to the value is passed to the procedure, rather than passing the
actual value. This implies that in the call p ( a (i)), in which array element
a(i) is passed to procedure p, a change to i in p does not change the element
referred to by p's actual parameter. Every non-scalar (not integer,
subrange, real, boolean, enumerated, pointer or the char type) parameter is
passed by reference whether or not it is declared var. In all other cases
(scalar non-var parameters) the parameter is passed by value (the actual
value is copied to the procedure).
Chapter 7 : Language Features 477
The upper bound of an array or string that is a formal parameter may be
specified as an asterisk (*), as is done above for parameter s in procedure x.
This specifies that the size of the upper bound is inherited from the
corresponding actual parameter. Parameters declared using star are called
dynamic parameters.
The names of the formal parameters must be distinct from each other, from the procedure
or function name, and from pervasive identifiers. However, they need not
be distinct from names outside of the procedure or function.
Example
Find the zero of function f. This example illustrates form (b), which is a
parameter that is a function. See also subprogramHeader.
function findZero ( function f ( x : real) : real,
left, right, accuracy : real ) : real
pre sign ( f ( left ) ) not= sign ( f ( right)) )
and accuracy > 0
var L : real := left
var R : real := right
var M : real
const signLeft := sign ( f ( left ) )
loop
M := ( R + L) / 2
exit when abs ( R - L ) <= accuracy
if signLeft =sign ( f ( M ) ) then
L := M
else
R := M
end if
end loop
result M
end findZero
Details
Form (b) of paramDeclaration is used to specify formal parameters that are
themselves procedures or functions. For example, in the findZero function, f
is a formal parameter that is itself a function. The subprogram type can be
used to replace form (b). In particular, the header to the findZero function
can be replaced by the following with no change in the action. The names g
and x serve no purpose, except as place holders in the declaration of f.
function findZero ( f : function g ( x : real) : real,
left, right, accuracy : real ) : real
Details
Parameters that are declared non var should, in principle, be constant.
Unfortunately, there is an anomalous situation in which these can change.
This occurs when the parameter is passed by reference, because it is a non
scalar such as a string. If the actual parameter is changed while the
subprogram is executing, the formal parameter will change as well.
You can also optionally use the register keyword to request that the
variable be placed in a machine register. This changes form (a) to allow the
optional use of the register keyword. The syntax for form (a) is actually:
[ var ] [ register ] id {, id } : [ cheat ] typeSpec
478 Object Oriented Turing Reference Manual
In the current (1999) implementation, programs are run interpretively using pseudo-code,
which has no machine registers, and the register keyword is ignored. See
register for restrictions on the use of register parameters.
The optional keyword cheat means that the parameter has a type cheat. See
cheat. Any variable or constant non scalar (in other words, items passed by
reference) can be passed to a type cheat parameter. The internal
representation will be interpreted as a value of the specified type. This is
dangerous as it provides unconstrained access to the underlying computer
memory.
This procedure outputs the values of n bytes starting at the address of
formal parameter a, using a parameter type cheat.
Example
procedure dump (a : cheat array 0 .. 10000 of nat1, n : int )
for i : 0 .. n - 1
put i, a ( i ) : 4
end for
end dump
var s : string := "abc"
dump ( s, 5 )
% Dumps 5 bytes, starting with "abc"
pause
Syntax
statement
A pauseStatement is:
pause expn
Description
The pause statement blocks the program (or just the process in the case of a
concurrent program) for a given number of simulated time units. The expn
must be a non-negative int value giving the number of time units. This is
analogous to the delay statement, which causes blocking for a given
amount of real time (actual physical time).
The interpreter maintains a counter which it considers to be simulated
time. The only execution that causes this counter to increase is the pause
statement. The process executing the pause is blocked until the counter has
counted forward the number of units given by expn. All other statements
(except wait) are considered to be infinitely fast. Several processes can be
executing pause statements simultaneously.
The use of simulated time allows Turing to be used as a simulation
language in which the pause statement simulates the passage of time in the
simulated system.
Chapter 7 : Language Features 479
Win DOS Mac
PC
Description
X TOOT Term
This unit contains the predefined subprograms that deal with direct access
to the hardware under the IBM PC architecture (available only under DOS
OOT).
All routines in the PC unit are exported qualified (and thus must be
prefaced with "PC.").
Entry Points
InPort
Returns a nat1 (byte) value from a specified PC port.
InPortWord
Returns a nat2 (word) value from a specified PC port.
OutPort
Sends a nat1 (byte) value to a specified PC port.
OutPortWord
Sends a nat2 (word) value to a specified PC port.
Interrupt
Calls a PC interrupt specifying the registers to be passed to
the interrupt handler.
InterruptSegs Calls a PC interrupt specifying the segments and registers
to be passed to the interrupt handler.
ParallelGet
Returns the value of the pins set on the parallel port.
ParallelPut
Sets the values of the pins on the parallel port.
SerialGet
Reads a character from the serial port.
SerialPut
Sends a character to the serial port.
SerialHasch
Returns whether there’s a character waiting on the serial
port.
Win DOS Mac
PC.InPort
Syntax
X TOOT Term
PC.InPort (portAddress : nat2) : nat1
480 Object Oriented Turing Reference Manual
Description
PC.InPort returns a nat1 (byte) value from a specified PC port. Ports on the
IBM PC control a variety of hardware related activities such as video color
maps and refresh rates, disk I/O activity, etc. Note that it’s very easy to
crash the machine with an incorrect portAddress. Read the hardware
manuals to determine the correct portAddress .
Example
The program prints out the value of the current overscan color on VGA
displays (this will usually be 0) and then sets the value to colour 4.
% This only works on VGA systems
% Determine the address of the Attribute Controller Port
var crtStatus : int := nat2@(16#463) + 6
var prevColor : int
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#11)
% Set the register number
% Read the register contents
prevColor := PC.InPort (16#3C1)
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#20)
% Restore the palette
put "The overscan color was ", prevColor
Input.Pause
% Wait for a keystroke
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#11)
% Set the register number
PC.OutPort (16#3C0, 4)
% Write the register contents
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#20)
% Restore the palette
put "The overscan color is now 4 "
Input.Pause
% Wait for a keystroke
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#11)
% Set the register number
PC.OutPort (16#3C0, prevColor)
% Write the register contents
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#20)
% Restore the palette
put "The overscan color is now ", prevColor
Status
Exported qualified.
This means that you can only call the function by calling PC.InPort, not by
calling InPort.
Win DOS Mac
PC.InPortWord
Syntax
X TOOT Term
PC.InPortWord (portAddress : nat2) : nat2
Chapter 7 : Language Features 481
Description
PC.InPortWord returns a nat2 (word) value from a specified PC port. Ports
on the IBM PC control a variety of hardware related activities such as
video color maps and refresh rates, disk I/O activity, etc. Note that it’s
very easy to crash the machine with an incorrect portAddress. Read the
hardware manuals to determine the correct portAddress .
Status
Exported qualified.
This means that you can only call the function by calling PC.InPortWord,
not by calling InPortWord.
Win DOS Mac
PC.Interrupt
X TOOT Term
Syntax
PC.Interrupt (interrupt : nat2, var eax, ebx, ecx, edx,
esi, edi, cflag : nat)
Description
PC.Interrupt calls the specified interrupt after loading the 8x86 registers
with the values passed as parameters. After calling the interrupt, the
values of the parameters are assigned from the registers.
To determine the interrupts and register arguments, consult a DOS, BIOS
or PC hardware reference manual.
Note that it’s very easy to crash the machine with an incorrect call to
PC.Interrupt. Before running any program that uses the PC.Interrupt
subprogram, save the program.
Example
In “screen” mode, foreground colors 16-31 normally cause characters to
blink. This reprograms the video controller to eliminate blinking. Any
character that was blinking will now have a high intensity background
color. This program flips between blinking and high intensity background.
% This only works on EGA/VGA systems
for back : 0 .. 7
for fore : 0 .. 31
Text.Locate (back + 1, fore + 1)
Text.Color (fore)
Text.ColorBack (back)
put "X" ..
end for
end for
Input.Pause
% Wait for a keystroke
var ax, bx, cx, dx, si, di, flag : nat
% Turn off blinking, use high intensity background instead
ax := 16#1003
482 Object Oriented Turing Reference Manual
bx := 0
PC.Interrupt (16#10, ax, bx, cx, dx, si, di, flag)
Input.Pause
% Wait for a keystroke
% Turn on blinking, no high intensity background colors
bx := 1
PC.Interrupt (16#10, ax, bx, cx, dx, si, di, flag)
Details
Under DOS OOT, the processor is running in 386 or 32 bit mode. This
means that all addresses are 32 bits and must be set as such for those
routines that involve passing an address to an interrupt. In general, the
segment register should be set to 0 and the register should be set to the flat
address of the object. The flat address is calculated by multiplying the
segment by 16 and adding the offset.
The DOS extender will translate the addresses into standard 16 bit
addresses and back again. While most of the available interrupts are
supported by the DOS extender, check with Holt Software before trying to
use an unusual interrupt.
To set various registers such as al, ah, etc., use the following guide:
To set al in eax, use bits (eax, 0 .. 7) := value
To set ah in eax, use bits (eax, 8 .. 15) := value
To set ax in eax, use bits (eax, 0 .. 15) := value
See also
PC.InterruptSegs for calling interrupts where the segment registers need
to be set.
Status
Exported qualified.
This means that you can only call the function by calling PC.Interrupt, not
by calling Interrupt.
Win DOS Mac
PC.InterruptSegs
Syntax
X TOOT Term
PC.InterruptSegs (interrupt : nat2,
var eax, ebx, ecx, edx, esi, edi, cflag : nat,
var ds, es, ss, cs : nat)
Chapter 7 : Language Features 483
Description
PC.InterruptSegs calls the specified interrupt after loading the 8x86
registers and segment registers with the values passed as parameters. After
calling the interrupt, the values of the parameters are then assigned from
the registers and segment registers.
To determine the interrupts and register arguments, consult a DOS, BIOS
or PC hardware reference manual.
Note that it’s very easy to crash the machine with an incorrect call to
PC.InterruptSegs. Before running any program that uses the
PC.InterruptSegs subprogram, save the program.
Details
Under DOS OOT, the processor is running in 386 or 32 bit mode. This
means that all addresses are 32 bits and must be set as such for those
routines that involve passing an address to an interrupt. In general, the
segment register should be set to 0 and the register should be set to the flat
address of the object. The flat address is calculated by multiplying the
segment by 16 and adding the offset.
The DOS extender will translate the addresses into standard 16 bit
addresses and back again. While most of the available interrupts are
supported by the DOS extender, check with Holt Software before trying to
use an unusual interrupt.
To set various registers such as al, ah, etc., use the following guide:
To set al in eax, use bits (eax, 0 .. 7) := value
To set ah in eax, use bits (eax, 8 .. 15) := value
To set ax in eax, use bits (eax, 0 .. 15) := value
Status
Exported qualified.
This means that you can only call the function by calling PC.InterruptSegs,
not by calling InterruptSegs.
Win DOS Mac
PC.OutPort
X TOOT Term
Syntax
PC.OutPort (portAddress : nat2, value : nat1)
Description
PC.OutPort writes a nat1 (byte) value to a specified PC port. Ports on the
IBM PC control a variety of hardware related activities such as video color
maps and refresh rates, disk I/O activity, etc. Note that it’s very easy to
crash the machine with an incorrect portAddress. Read the hardware
manuals to determine the correct portAddress .
484 Object Oriented Turing Reference Manual
The program prints out the value of the current overscan color on VGA
displays (this will usually be 0) and then sets the value to colour 4.
Example
% This only works on VGA systems
% Determine the address of the Attribute Controller Port
var crtStatus : int := nat2@(16#463) + 6
var prevColor : int
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#11)
% Set the register number
% Read the register contents
prevColor := PC.InPort (16#3C1)
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#20)
% Restore the palette
put "The overscan color was ", prevColor
Input.Pause
% Wait for a keystroke
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#11)
% Set the register number
PC.OutPort (16#3C0, 4)
% Write the register contents
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#20)
% Restore the palette
put "The overscan color is now 4 "
Input.Pause
% Wait for a keystroke
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#11)
% Set the register number
PC.OutPort (16#3C0, prevColor)
% Write the register contents
dummy := PC.InPort (crtStatus ) % Toggle the controller
PC.OutPort (16#3C0, 16#20)
% Restore the palette
put "The overscan color is now ", prevColor
Status
Exported qualified.
This means that you can only call the function by calling PC.OutPort, not
by calling OutPort.
Win DOS Mac
PC.OutPortWord
X TOOT Term
Syntax
PC.OutPortWord (portAddress : nat2, value : nat2)
Description
PC.OutPortWord writes a nat2 (word) value to a specified PC port. Ports
on the IBM PC control a variety of hardware related activities such as
video color maps and refresh rates, disk I/O activity, etc. Note that it’s
very easy to crash the machine with an incorrect portAddress. Read the
hardware manuals to determine the correct portAddress .
Chapter 7 : Language Features 485
Exported qualified.
Status
This means that you can only call the function by calling PC.OutPortWord,
not by calling OutPortWord.
Win DOS Mac
PC.ParallelGet
X TOOT Term
Syntax
PC.ParallelGet (port : int) : nat1
Description
The PC.ParallelGet function is used to read the value of certain pins on a
parallel port. The port is specified with the port parameter which can have
the value 1, 2 or 3 corresponding to "LPT1", "LPT2" and "LPT3". This
procedure can be used to control robots and peripherals.
Example
This program reads and prints the values of the five data pins of the PC’s
parallel port.
% Read in the set of pin values from LPT1
const val : int := PC.ParallelGet (1)
put "Pin 10 is: ", (val div 64) mod 2
put "Pin 11 is: ", (val div 128) mod 2
put "Pin 12 is: ", (val div 32) mod 2
put "Pin 13 is: ", (val div 16) mod 2
put "Pin 15 is: ", (val div 8) mod 2
Details
The five pins that are used for parallel input are pins 10-15. The
PC.ParallelGet procedure returns the sum of
64
Pin 10 high
128
Pin 11 high
32
Pin 12 high
16
Pin 13 high
8
Pin 15 high
The mod and div operators can be used to determine which pins are high
or low.
Status
Exported qualified.
This means that you can only call the function by calling PC.ParallelGet,
not by calling ParallelGet.
See also
PC.ParallelPut procedure for a diagram of the pins. That procedure is used
to set the values on the parallel port.
486 Object Oriented Turing Reference Manual
Win DOS Mac
PC.ParallelPut
X TOOT Term
Syntax
PC.ParallelPut (port : int, value : int)
Description
The PC.ParallelPut procedure is used on a PC to set the values on the data
pins on the parallel port. The port is specified with the port parameter
which can have the value 1, 2 or 3 corresponding to "LPT1", "LPT2" and
"LPT3". This procedure can be used to control robots and peripherals.
Example
This program sets data bit 0, data bit 1 and so on to data bit 7.
for i : 0 .. 7
%Set data bit i on parallel port LPT2
PC.ParallelPut (2, 2 ** i)
put "Data bit ", i, " or Pin ", i + 2, "has just been set"
end for
Details
The PC.ParallelPut procedure is used to set the eight data bits on the PC’s
parallel port. These data bits 0 - 7 correspond to pins 2 - 9 on the parallel
port.
Data pins 7,6...,0
Parallel Pin Out - From the IBM PC
The value sent to PC.ParallelPut is the sum of the following:
1
Data bit 0
16
Data bit 4
2
Data bit 1
32
Data bit 5
4
Data bit 2
64
Data bit 6
8
Data bit 3
128
Data bit 7
For example, the command PC.ParallelPut (97) sets data bits 0, 5 and 6
high (97 = 1 + 32 + 64) and sets the other data pins low. Because there are
only 8 data pins in the parallel port, the value passed to PC.ParallelPut
must be from 0 to 255.
Chapter 7 : Language Features 487
The PC.ParallelPut procedure is not meant for sending a stream of
characters to the parallel port (for example, if you want to send the string
"Hello" to the printer). If you want to do this, open the file "LPT1" using the
open statement and put to the file.
Exported qualified.
Status
This means that you can only call the function by calling PC.ParallelPut,
not by calling ParallelPut.
PC.ParallelGet function, which is used to read the values of pins on the
parallel port.
See also
Win DOS Mac
PC.SerialGet
X TOOT Term
Syntax
PC.SerialGet (port : int) : nat1
Description
The PC.SerialGet procedure is used on a PC to read a single character
from a serial port. The port is specified with the port parameter which can
have the value 1, 2 or 3 corresponding to "COM1", "COM2" and "COM3".
This function can be used to read characters from a modem or from
another computer through a null-modem cable.
Example
This program sends a message to the COM1 port and then gets a reply
ending with a newline.
% We assume that the serial port has been set up previously using
% the DOS mode command.
const message : string := "This is a test"
for i : 1.. length (message)
PC.SerialPut (1, ord (message (i))
end for
PC.SerialPut (1, 10)
var inmessage : string := ""
var ch : int
loop
ch := PC.SerialGet (1)
exit when ch = 10
inmessage := inmessage + chr (ch)
end loop
put inmessage
Details
PC.SerialPut and PC.SerialGet assume that the port’s baud rate, stop bits,
etc. have been properly set with the DOS mode command outside of
Turing. Check the DOS manual for information on how to use this
command.
488 Object Oriented Turing Reference Manual
PC.SerialPut and PC.SerialGet will work reliably for communication up to
1200 baud. These routines are not suitable for rates higher than this.
Note that these routines pass integers that represent characters.
Consequently, they should be passed (and will return) values that are
between 0 and 255. Because they pass integers, use the chr and ord
functions to convert integers to characters and vice-versa.
Exported qualified.
Status
This means that you can only call the function by calling PC.SerialGet, not
by calling SerialGet.
PC.SerialPut procedure which sends characters out the serial port. See also
chr and ord functions for converting between a character and its ASCII
value.
See also
Win DOS Mac
PC.SerialHasch
X TOOT Term
Syntax
PC.SerialHasch (port : int) : boolean
Description
The PC.SerialHasch procedure is used on a PC to determine if there is a
character waiting to be read on a serial port. The port is specified with the
port parameter which can have the value 1, 2 or 3 corresponding to
"COM1", "COM2" and "COM3". This function can be in conjunction with
PC.SerialGet and PC.SerialPut to receive and send characters to a modem
or to another computer through a null-modem cable.
Example
This program continuously send the letter “A” over COM2 until it receives
the letter “B” as acknowledgment.
% We assume that the serial port has been set up previously using
% the DOS mode command.
loop
PC.SerialPut (2, "A")
if SerialHasch (2) then
var value : int := PC.SerialGet (2)
exit when chr (value) = "B"
put "Did not receive a \"B\"."
put "Received a ", chr (value), "instead."
end if
end loop
Details
PC.SerialHasch assumes that the port’s baud rate, stop bits, etc. have been
properly set with the DOS mode command outside of Turing. Check the
DOS manual for information on how to use this command.
Chapter 7 : Language Features 489
PC.SerialHasch will work reliably for communication up to 1200 baud.
These routines are not suitable for rates higher than this.
Exported qualified.
Status
This means that you can only call the function by calling PC.SerialHasch,
not by calling SerialHasch.
PC.SerialGet and PC.SerialPut procedures which receive and send
characters out a serial port. See also chr and ord functions for converting
between a character and its ASCII value.
See also
Win DOS Mac
PC.SerialPut
X TOOT Term
Syntax
PC.SerialPut (port : int, value : int)
Description
The PC.SerialPut procedure is used on a PC to send a single character out
the serial port. The port is specified with the port parameter which can
have the value 1, 2 or 3 corresponding to "COM1", "COM2" and "COM3".
This function can be used to send characters to a modem or to another
computer through a null-modem cable.
Example
This program sends a message to the COM2 port and then gets a reply
ending with a newline.
% We assume that the serial port has been set up previously using
% the DOS mode command.
const message : string := "This is a test"
for i : 1.. length (message)
PC.SerialPut (1, ord (message (i))
end for
PC.SerialPut (1, 10)
var inmessage : string := ""
var ch : int
loop
ch := PC.SerialGet (1)
exit when ch = 10
inmessage := inmessage + chr (ch)
end loop
put inmessage
Details
PC.SerialPut and PC.SerialGet assume that the port’s baud rate, stop bits,
etc. have been properly set with the DOS mode command outside of
Turing. Check the DOS manual for information on how to use this
command.
490 Object Oriented Turing Reference Manual
PC.SerialPut and PC.SerialGet will work reliably for communication up to
1200 baud. These routines are not suitable for rates higher than this.
Note that these routines pass integers that represent characters.
Consequently, they should be passed (and will return) values that are
between 0 and 255. Because they pass integers, use the chr and ord
functions to convert integers to characters and vice-versa.
Exported qualified.
Status
This means that you can only call the function by calling PC.SerialPut, not
by calling SerialPut.
PC.SerialGet procedure which sends characters out the serial port. See also
chr and ord functions for converting between a character and its ASCII
value.
See also
pervasive
Description
declaration modifier
When a variable, constant, type or subprogram is declared, you can specify
that it is to be pervasive, which means that it does not need to be explicitly
imported into modules, monitors or classes in the current scope. The
keyword pervasive can be abbreviated as an asterisk (*).
Example
var pervasive counter : int % Short form: var * count : int
const * maxCounter : int := 100
procedure * p ( x : real )
…
end p
Details
The keyword pervasive is also used in export lists along with the keyword
unqualified. See export list for details.
See also
var declaration, const declaration, procedure declaration, function
declaration, subprogram header and export list for uses of pervasive.
Chapter 7 : Language Features 491
Pic
Description
This unit contains the predefined subprograms that deal with taking
pictures of part of the screen, displaying them and moving pictures from
file to screen and back.
All routines in the Pic unit are exported qualified (and thus must be
prefaced with "Pic.").
Entry Points
New
Creates a picture from a specified portion of the screen.
Draw
Draws a picture on the screen.
Free
Frees up the picture created by using New or FileNew.
FileNew
Creates a picture from a file in an allowed format.
Save
Saves a picture as a file in an allowed format.
ScreenLoad
Displays a file in an allowed format on the screen directly.
ScreenSave
Saves a specified portion of the screen as a file in an
allowed format.
Win DOS Mac
Pic.Draw
X TOOT Term
Syntax
Pic.Draw (picID, x, y, mode : int)
Description
Pic.Draw is used to draw a picture on the screen. The picture is drawn
with the lower left corner at (x, y).
The mode parameter has one of the following values:
picCopy
This draws the picture on top of what was
underneath, obscuring it completely.
picXor
This draws the picture XORing it with the
background. In DOS, you can use this function to
do animation. Drawing an object on top of itself
with XOR erases it and restores the background.
492 Object Oriented Turing Reference Manual
picMerge
This draws the picture like picCopy except that any
occurrence of the background color in the picture
is not drawn to the screen. This allows you to
draw an irregularly-shaped object and draw it to
the screen.
picUnderMerge
This draws the picture, but only where the
background color was displayed underneath it.
The effect of this is to make the picture appear to
be displayed behind the background.
Details
If the Pic.Draw call fails, Error.Last will return a non-zero value indicating
the reason for the failure. Error.LastMsg will return a string which contains
the textual version of the error.
Details
At the time of writing, under DOS OOT and X OOT, only picCopy and
picXor are supported.
Example
The program draws a graphic on the screen and then repeats it 50 times in
random positions.
var picID: int
var x, y : int
Draw.FillBox (50, 50, 150, 150, red)
Draw.FillStar (50, 50, 150, 150, green)
Draw.FillOval (100, 100, 50, 50, blue)
picID := Pic.New (50, 50, 150, 150)
for i : 1 .. 50
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
Pic.Draw (picID, x, y, picCopy)
end for
Pic.Free (picID)
Status
Exported qualified.
This means that you can only call the function by calling Pic.Draw, not by
calling Draw.
Win DOS Mac
Pic.FileNew
Syntax
X TOOT Term
Pic.FileNew (fileName : string) : int
Chapter 7 : Language Features 493
Description
Pic.FileNew is used to obtain a picture from a file. The Pic.FileNew
procedure allocates the memory for the picture, which can be very large for
pictures of large areas. The memory is freed up when the program calls
Pic.Free with the picture ID. The picture can be used with the Pic.Draw
and Pic.Save.
Various versions of OOT can convert different formats of files. Look at the
release notes to see which file formats can be read and written.
The fileName parameter must give the format of the file:
Details
PCX files
"PCX:filename" or "filename.PCX"
BMP files
"BMP:filename" or "filename.BMP"
TIM files
"TIM:filename" or "filename.TIM"
On a DOS system, the Pic.FileNew routine may end up remapping the
color palette.
If the Pic.FileNew call fails, then it returns 0. Also Error.Last will return a
non-zero value indicating the reason for the failure. Error.LastMsg will
return a string which contains the textual version of the error.
Details
At the time of this writing, DOS OOT supported only DOS OOT native
graphics files (i.e those files with no recognized suffix or prefix). These files
can only be used by DOS OOT. PCX support will appear in the next major
release. Consult the release notes to find out which file formats are
currently supported.
At the time of writing, WinOOT supported only BMP files. Consult the
release notes to find out which file formats are currently supported.
Example
The program reads a graphic from the file mypic.pcx and then draws it 50
times.
var picID: int
var x, y : int
picID := Pic.FileNew ("mypic.pcx")
for i : 1 .. 50
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
Pic.Draw (picID, x, y, picCopy)
end for
Pic.Free (picID)
Status
Exported qualified.
This means that you can only call the function by calling Pic.FileNew, not
by calling FileNew.
494 Object Oriented Turing Reference Manual
Win DOS Mac
Pic.Free
X TOOT Term
Syntax
Pic.Free (picID : int)
Description
Pic.Free is used to release the memory allocated by Pic.New. It frees up the
memory allocated to the parameter picID. This means that picID can not be
used in a Draw.Pic or Draw.Save procedure after Pic.Free is called.
Details
If Pic.Free is passed an invalid picture ID, a fatal error occurs. If the
Pic.Free call fails for other (non-fatal) reasons, Error.Last will return a nonzero value indicating the reason for the failure. Error.LastMsg will return a
string which contains the textual version of the error.
Example
The program draws a graphic on the screen and then draws it 50 times.
var picID: int
var x, y : int
Draw.FillBox (50, 50, 150, 150, red)
Draw.FillStar (50, 50, 150, 150, green)
Draw.FillOval (100, 100, 50, 50, blue)
picID := Pic.New (50, 50, 150, 150)
for i : 1 .. 50
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
Pic.Draw (picID, x, y, picCopy)
end for
Pic.Free (picID)
Status
Exported qualified.
This means that you can only call the function by calling Pic.Free, not by
calling Free.
Win DOS Mac
Pic.New
Syntax
X TOOT Term
Pic.New (x1, y1, x2, y2 : int) : int
Chapter 7 : Language Features 495
Description
Pic.New is used to obtain a picture of a portion of the screen. The Pic.New
procedure allocates the memory for the picture, which can be very large for
pictures of large areas. The memory is freed up when the program calls
Pic.Free with the picture ID. The picture can be used with the Pic.Draw
and Pic.Save.
The picture is of the screen area defined by the rectangle (x1, y1) - (x2, y2).
Details
If the Pic.New call fails, then it returns 0. Also Error.Last will return a nonzero value indicating the reason for the failure. Error.LastMsg will return a
string which contains the textual version of the error.
Example
The program draws a graphic on the screen and then draws it 50 times.
var picID: int
var x, y : int
Draw.FillBox (50, 50, 150, 150, red)
Draw.FillStar (50, 50, 150, 150, green)
Draw.FillOval (100, 100, 50, 50, blue)
picID := Pic.New (50, 50, 150, 150)
for i : 1 .. 50
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
Pic.Draw (picID, x, y, picCopy)
end for
Pic.Free (picID)
Status
Exported qualified.
This means that you can only call the function by calling Pic.New, not by
calling New.
Win DOS Mac
Pic.Save
X TOOT Term
Syntax
Pic.Save (picID : int, fileName : string)
Description
Pic.Save is used to save a picture on the screen to a file.
Various versions of OOT can save different formats of files. Look at the
release notes to see which file formats can be read and written.
The fileName parameter must give the format of the file:
PCX files
"PCX:filename" or "filename.PCX"
BMP files
"BMP:filename" or "filename.BMP"
TIM files
"TIM:filename" or "filename.TIM"
496 Object Oriented Turing Reference Manual
Details
At the time of this writing, DOS OOT supported only DOS OOT native
graphics files (i.e those files with no recognized suffix or prefix). These files
can only be used by DOS OOT. PCX support will appear in the next major
release. Consult the release notes to find out which file formats are
currently supported.
At the time of writing, WinOOT supported only BMP files. Consult the
release notes to find out which file formats are currently supported.
Details
Example
If Pic.Save is passed an invalid picture ID, a fatal error occurs. If the
Pic.Save call fails for other (non-fatal) reasons, Error.Last will return a nonzero value indicating the reason for the failure. Error.LastMsg will return a
string which contains the textual version of the error.
The program draws a graphic on the screen and then saves it as a BMP file.
var picID: int
var x, y : int
Draw.FillBox (50, 50, 150, 150, red)
Draw.FillStar (50, 50, 150, 150, green)
Draw.FillOval (100, 100, 50, 50, blue)
Example
picID := Pic.New (50, 50, 150, 150)
Pic.Save (picID, "BMP:mypic.dat")
Pic.Free (picID)
The following two programs save and load a file in DOS OOT native
format.
% Program to save a picture in mypic.pic
var picID: int
var x, y : int
Draw.FillBox (50, 50, 150, 150, red)
Draw.FillStar (50, 50, 150, 150, green)
Draw.FillOval (100, 100, 50, 50, blue)
picID := Pic.New (50, 50, 150, 150)
Pic.Save (picID, "mypic.pic")
Pic.Free (picID)
% Program to load the picture back again and draw 50 copies
var picID: int
var x, y : int
picID := Pic.FileNew ("mypic.pic")
for i : 1 .. 50
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
Pic.Draw (picID, x, y, picCopy)
end for
Pic.Free (picID)
Status
Exported qualified.
This means that you can only call the function by calling Pic.Save, not by
calling Save.
Chapter 7 : Language Features 497
Win DOS Mac
Pic.ScreenLoad
X TOOT Term
Syntax
Pic.ScreenLoad (fileName : string, x, y, mode : int)
Description
Pic.ScreenLoad displays a picture from a file straight to the screen.
Various versions of OOT can convert different formats of files. Look at the
release notes to see which file formats can be read and written.
The fileName parameter must give the format of the file:
PCX files
"PCX:filename" or "filename.PCX"
BMP files
"BMP:filename" or "filename.BMP"
TIM files
"TIM:filename" or "filename.TIM"
The x and y parameters set the lower left hand corner of the picture.
The mode parameter has one of the following values:
Details
picCopy
This draws the picture on top of what was
underneath, obscuring it completely.
picXOR
This draws the picture XORing it with the
background. In DOS, you can use this function to
do animation. Drawing an object on top of itself
with XOR erases it and restores the background.
picMerge
This draws the picture like picCopy except that any
occurrence of the background color in the picture
is not drawn to the screen. This allows you to
draw an irregularly-shaped object and draw it to
the screen.
picUnderMerge
This draws the picture, but only where the
background color was displayed underneath it.
The effect of this is to make the picture appear to
be displayed behind the background.
On a DOS system, the Pic.ScreenLoad routine may end up remapping the
color palette.
If the Pic.ScreenLoad fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
498 Object Oriented Turing Reference Manual
Details
At the time of writing, under DOS OOT and X OOT, only picCopy and
picXOR are supported. As well, DOS OOT supports only “TM2” files (a
file format used under Turing) and DOS OOT native graphics files (i.e
those files with no recognized suffix or prefix). DOS OOT native graphics
files can only be used by DOS OOT. PCX support will appear in the next
major release. Consult the release notes to find out which file formats are
currently supported.
At the time of writing, WinOOT supported only BMP files. Consult the
release notes to find out which file formats are currently supported.
At the time of writing, MacOOT supported only PICT files. Consult the
release notes to find out which file formats are currently supported.
Example
The program displays a picture on the screen from the PCX file mypic.PCX.
Pic.ScreenLoad ("mypic.pcx", 0, 0, picCopy)
Status
Exported qualified.
This means that you can only call the function by calling Pic.ScreenLoad,
not by calling ScreenLoad.
Win DOS Mac
Pic.ScreenSave
X TOOT Term
Syntax
Pic.ScreenSave (x1, y1, x2, y2 : int, fileName : string)
Description
Pic.ScreenSave saves a portion of the screen into a file in a format specified
by the file name.
The picture saved to the file is the portion of the screen defined by the
rectangle (x1, y1) - (x2, y2).
Various versions of OOT can convert different formats of files. Look at the
release notes to see which file formats can be read and written.
The fileName parameter must give the format of the file:
Details
PCX files
"PCX:filename" or "filename.PCX"
BMP files
"BMP:filename" or "filename.BMP"
TIM files
"TIM:filename" or "filename.TIM"
At the time of this writing, DOS OOT supported only DOS OOT native
graphics files (i.e those files with no recognized suffix or prefix). These files
can only be used by DOS OOT. PCX support will appear in the next major
release. Consult the release notes to find out which file formats are
currently supported.
Chapter 7 : Language Features 499
At the time of writing, WinOOT supported only BMP files. Consult the
release notes to find out which file formats are currently supported.
Details
If the Pic.ScreenSave fails, then Error.Last will return a non-zero value
indicating the reason for the failure. Error.LastMsg will return a string
which contains the textual version of the error.
Example
The program draws a graphic and saves it as a PICT file called draw.
Draw.FillBox (50, 50, 150, 150, red)
Draw.FillStar (50, 50, 150, 150, green)
Draw.FillOval (100, 100, 50, 50, blue)
picID := Pic.ScreenSave (50, 50, 150, 150, "PICT:draw")
Exported qualified.
Status
This means that you can only call the function by calling Pic.ScreenSave,
not by calling ScreenSave.
Win DOS Mac
play
procedure
X TOOT Term
Syntax
play ( music : string )
Description
The play procedure is used to sound musical notes on the computer.
Example
This program sounds the first three notes of the C scale.
play ( "cde" )
Example
This program plays from middle C to one octave above middle C and
down again in 8th notes.
play ( "8cdefgab>c" )
play ( "<bagfedc" )
Details
The play procedure takes strings containing characters that specify notes,
rests, sharps, flats and duration. The notes are the letters a to g (or A to G).
A rest is p (for pause). A sharp is + and a flat is -. The durations are 1
(whole note), 2 (half note), 4 (quarter note), 8 (eight note) and 6 (sixteenth
note). The character > raises to the next octave and < lowers. For example,
this is the way to play C and then C sharp one octave above middle C with
a rest between them, all in sixteenth notes: play(">6cpc+"). Blanks can be
used for readability and are ignored by play.
Under some systems such as UNIX, the play procedure may have no effect.
The current (1999) implementation does not support play.
500 Object Oriented Turing Reference Manual
the playdone function, which is used to see if a note has finished sounding.
See also the sound procedure, which makes a sound of a given frequency
(Hertz) and duration (milliseconds).
See also
See also predefined unit Music.
Win DOS Mac
playdone
function
X TOOT Term
Syntax
playdone : boolean
Description
The playdone function is used to determine when notes played by the play
procedure have finished sounding.
Example
This program sounds the first three notes of the C scale and outputs "All
done" as soon as they are finished. Without the loop, the message would
come out before the notes are finished.
play ( "cde" )
loop
exit when playdone
end loop
put "All done"
Details
Under some systems such as UNIX, the playdone procedure may be
meaningless.
See also
the play procedure. See also the sound procedure which makes a sound of
a given frequency (Hertz) and duration (milliseconds).
See also predefined unit Music.
pointer
Syntax
type
A pointerType is one of:
(a)
(b)
[unchecked] pointer to collectionId
% Short form: ^ collectionId
[unchecked] pointer to classId
Chapter 7 : Language Features 501
(c)
% Short form: ^ classId
[unchecked] pointer to typeSpec
% Short form: ^ typeSpec
Description
A variable declared as a pointer type is used to locate an element of a
collection or class or a value of a type. The new statement creates a new
element (or value) and places the element's location in a pointer variable.
The free statement destroys an element located by a pointer variable.
Example
Using a collection, declare a list or records and allocate one record.
var list : collection of
record
contents : string ( 10 )
next : pointer to list
end record
var first : pointer to list
new list, first
Example
Create a collection that will represent a binary tree.
var tree : collection of
record
name : string ( 10 )
left, right : pointer to tree
end record
var root : pointer to tree
new tree, root
tree ( root ).name := "Adam"
Example
Using a class, create an object of that class. The object is located by the start
pointer. The name field of the object is set to Ed.
class node
export var next, var name
name : string ( 25 )
next : pointer to node % Short form: next : ^ node
end node
var start : pointer to node % Short form: var start : ^ node
new node, start
% Short form: new start
node ( start ) . name := "Ed"
Details
% Short form: start->name:="Ed"
For collections and classes, a pointer is effectively a subscript (an index) for
that collection or class. Pointers can be assigned, compared for equality and
passed as parameters.
The keywords pointer to can be replaced by the short form ^, as in
var first : ^ item
Given a pointer p that locates an object of class or collection C, the object is
referenced as C(p) or as the short form ^ p. A field f of the object is
referenced as C(p).f or ^p.f or as the short form p->f. For example, in the
class given above, the name field of the object located by the start pointer
can be set to Alice by:
502 Object Oriented Turing Reference Manual
start -> name := "Alice"
Pointers to types use the same notation, except that pointers to types are
not allowed to use the form typeSpec(p). See class for an example of the use
of a class with pointers.
The carat ^ is called the dereferencing operator and has the highest
precedence. For example, in ^ p.a, the carat applies to p and not to p.a. To
apply ^ to all of p.a, use parentheses: ^(p.a). Several carats can appear in a
row, for example,
var r : ^ ^ int
declares a pointer to a pointer to an integer and ^ ^ r is the notation for
referencing the integer.
A reference cannot begin with a left parenthesis, but can be surrounded by
^(…), as in ^ (q.b). If f is a parameterless function declared without
parentheses that returns a pointer, the form ^ f calls f before dereferencing
the pointer.
By default, all pointers are checked. This means there is a run time test to make sure that
references such as C(p) actually locate elements, i.e., that p is initialized, is
not nil and is not dangling (locating an object that has been freed). This
checking requires extra space (the implementation attaches a time stamp to
each pointer and object) and time. In high-performance programs in which
this extra space and time are not acceptable, the pointer can be declared to
be unchecked. When this is done, the program becomes dangerous and it is
the programmer's responsibility to make sure that all pointer usage is
valid.
If this is not the case, the program becomes susceptible to uncontrolled
crashes.
Checked pointers cannot be assigned to unchecked pointers nor vice versa.
However, you may, at your peril, use an implementation-dependent type
cheat, to convert a checked pointer to a unchecked pointer, as in:
type checkedPtr : ^ R
type uncheckedPtr : unchecked ^ R
var c : checkedPtr
% c is an checked pointer
var u : uncheckedPtr
% u is an unchecked pointer
…
u := cheat (uncheckedPtr, d)
% This is a type cheat
Unchecked pointers are equivalent to the pointers of the C language, which
are inherently error prone and cause difficult to locate bugs. An entire
collection (but not a class) can be declared unchecked, in which case all of
its pointers are implicitly unchecked. See collection.
See also
inherit lists for a description of the assignability rules for pointers. See
classes and collections for more details about the use of pointers. See also
new and free statements. See also nil, objectclass and anyclass.
Chapter 7 : Language Features 503
post
assertion
An postAssertion is:
Syntax
post trueFalseExpn
Description
A post assertion is a special form of an assert statement that is used in a
procedure or function. It is used to give requirements that the body of the
procedure or function is supposed to satisfy. These requirements are given
by the trueFalseExpn. After the body has executed and just before the
procedure or function returns, the trueFalseExpn is evaluated. If it is true,
all is well and execution continues. If it is false, execution is terminated
with an appropriate message. See assert statements and procedure and
function declarations for more details. See also pre and invariant
assertions.
Example
This function is supposed to produce an integer approximation of the
square root of integer i. The post condition requires that this result, which
is called answer, must be within a distance of 1 from the corresponding real
number square root.
function intSqrt ( i : int) answer : int
pre i >= 0
post abs ( answer - sqrt ( i ) ) <= 1
… statements to approximate square root…
end intSqrt
Details
A post assertion can also be used in a module, monitor, class or process
declaration to make sure that the initialization satisfies its requirements.
See also
module and process.
pre
Syntax
assertion
An preAssertion is:
pre trueFalseExpn
504 Object Oriented Turing Reference Manual
Description
A pre assertion is a special form of an assert statement that is used at the
beginning of a procedure or function. It is used to give requirements that
the caller of the procedure or functions is supposed to satisfy. These
requirements are given by the trueFalseExpn. The trueFalseExpn is
evaluated. If it is true, all is well and execution continues. If it is false,
execution is terminated with an appropriate message. See assert statements
and procedure and function declarations for more details. See also post
and invariant assertions.
Example
This function computes the average of n values. Its pre condition requires
that n must be strictly positive, to avoid the possibility of dividing by zero
when computing the average.
function average ( a : array 1 .. * of real, n : int) : real
pre n > 0
var sum : real := 0
for i : 1 .. n
sum := sum + a ( i )
end for
result sum / n
end average
Details
A pre assertion can also be used in a module, monitor, class or process
declaration to make sure that requirements for initialization are met.
See also
module and process.
precedence
Description
of operators
Turing's precedence rules determine the order of applying operators in an
expression such as 3 + 4 * 5. These rules state, for example, that
multiplication is done before addition, so this expression is equivalent to
3 + (4 * 5).
Parenthesized parts of an expression are evaluated before being used. For
example, in (1 + 2) * 3, the addition is done before the multiplication.
The precedence rules are defined by this table, in which operators
appearing earlier in the table are applied first. For example, multiplication
is applied before addition:
(1)
**, ^, #
(2)
prefix + and -
(3)
*, /, div, mod, rem, shr, shl
(4)
infix +, -, xor
Chapter 7 : Language Features 505
(5)
<, >, =, <=, >=, not=, in, not in
(6)
not
(7)
and
(8)
or
(9)
=>
( boolean implication)
Operators appearing on a single line in this table are applied from left to
right. For example, a–b–c is the same is (a–b)–c.
Here are some examples illustrating precedence, in which the left and right
expressions are equivalent:
–1**2
–(1**2)
a+b*c
a+(b*c)
a*b/c
(a*b)/c
b or c and d
b or (c and d)
x < y and y < z
(x < y) and (y < z)
The final example illustrates the fact that in Turing, parentheses are not
required when combining comparisons using and and or. These would be
required in the Pascal language.
The type cheat operator # is applied after subscripting, subprogram calling, dotting, and –
>. For example, in each of the following, # applies to the entire reference to
the right.
#a(i)
#r.y
#p->x
The pointer following operator ^ is applied before subscripting,
subprogram calling, dotting, and ->. For example, in the following, ^
applies to a, r and p.
^a(i)
^r.y
^p->x
Use parentheses to force ^ to apply to more of the reference. For example,
in ^(a(i)), the ^ applies to a(i).
See also
infix and prefix operators. See the int, real, string, boolean, set, enum, char
and char(n) types.
506 Object Oriented Turing Reference Manual
pred
predecessor function
Syntax
pred ( expn )
Description
The pred function accepts an integer, character, or an enumerated value
and returns the integer minus one, the previous character, or the previous
value in the enumeration. For example, pred ( 7 ) is 6.
Example
This part of a Turing program fills up array a with the enumerated values
red, yellow, green, red, yellow, green, etc.
type colors : enum ( green, yellow, red )
var a : array 1 .. 100 of colors
var c : colors := colors . red
for i : 1 .. 100
a ( i ) := c
if c = colors . green then
c := colors . red
else
c := pred ( c )
end if
end for
Details
It is illegal to apply pred to the first value of an enumeration.
See also
succ, lower and upper functions.
prefix
Syntax
operator
A prefixOperator is one of:
(a)
+
(b)
(c)
(d)
(e)
–
not
#
^
% Integer and real identity
% (does not change value)
% Integer and real negation
% Not (Boolean negation)
% Type cheat
% Pointer following
Chapter 7 : Language Features 507
A prefix operator is placed before a value or operand to produce another
value. For example, if the value of x is seven, then -x is negative seven.
There are also infix operators such as multiplication (*) and addition (+),
which are placed between two values to produce a third value. See infix
operator.
Description
The + and – prefix operators can be applied only to numeric values
(integer, real and natural numbers). The not prefix can be applied only to
true/false (boolean) values. For example not (x > y) is equivalent to x <= y.
The not operator produces true from false and false from true.
The # operators is a type cheat (see cheat), and the ^ operator is pointer
following (see pointer).
See also
int, real and boolean types, as well as precedence (for the order of applying
operators) and infix operators.
procedure
declaration
A procedureDeclaration is:
Syntax
procedure id [(paramDeclaration {, paramDeclaration })]
statementsAndDeclarations
end id
Description
A procedure declaration creates (but does not run) a new procedure. The
name of the procedure (id) is given in two places, just after procedure and
just after end.
Example
procedure greetings
put "Hello world"
end greetings
greetings
% This outputs Hello world
procedure sayItAgain ( msg : string, n : int )
for i : 1 .. n
put msg
end for
end sayItAgain
sayItAgain ("Toot", 2 ) % Toot is output twice
procedure double ( var x : real )
x := 2 * x
508 Object Oriented Turing Reference Manual
end double
var y : real := 3.14
Details
double ( y )
% This doubles the value of y
The set of parameters declared with the procedure are called formal
parameters. In the double procedure, for example, x is a formal parameter.
A procedure is called (invoked) by a procedure call statement which
consists of the procedure's name followed by the parenthesized list of
actual parameters (if any). For example, double(y) is a call having y as an
actual parameter. If there are no parameters (see the greet procedure
above), the call does not have parentheses. The keyword procedure can be
abbreviated to proc.
Ordinarily, a procedure returns (finishes and goes back to the place where it was called) by
reaching its end. However, the return statement in a procedure causes it to
return immediately. Note that return can also be used in the main program
to cause it to halt immediately.
Only parameters declared using var may be changed in the procedure, for
example, x is changed in the double procedure. The upper bounds of arrays
and strings that are parameters may be declared to be an asterisk (*). This
means that the bound is that of the actual parameter. See paramDeclaration
for details about parameters.
Procedures and functions cannot be declared inside other procedures and
functions.
The syntax of a procedureDeclaration presented above has been simplified
by leaving out the optional import list, pre condition, init clause, post
condition and exception handler. The full syntax is
procedure [ pervasive ] id
[ ( [ paramDeclaration {,paramDeclaration } ] ) ]
[ : deviceSpecification ]
[ pre trueFalseExpn ]
[ init id := expn {, id := expn } ]
[ post trueFalseExpn ]
[ exceptionHandler ]
statementsAndDeclarations
end id
A procedure must be declared before being called. To allow for mutually
recursive procedures, there are forward declarations of procedures with
later declaration of each procedure body. See forward and body
declarations for explanations.
See also
import list, pre condition, init clause, post condition and exceptionHandler
for explanations of these features. See pervasive for information on
pervasive procedures. See exceptionHandler. The optional
deviceSpecification is used only in procedures declared in monitors and is
used to create an interrupt handling procedure. See monitor for details.
Chapter 7 : Language Features 509
procedureCall
statement
A procedureCall is:
Syntax
procedureId [ ( [ expn { , expn } ] ) ]
Description
A procedure call is a statement that calls (invokes or activates) a
procedure. If the procedure has parameters, a parenthesized list of
expressions (expns) must follow the procedure's name (procedureId).
Example
procedure greet
put "Hello"
end greet
greet
% This is a call to the greet procedure
procedure times ( var i : int, factor : int)
i := factor * i
end times
var j : int
Details
times ( j, 4 )
% Multiply j by 4
A parameter declared in the header of a procedure is a formal parameter.
For example, i and factor above are formal parameters. Each expression in
the call is an actual parameter. For example, j and 4 above are actual
parameters.
If a formal parameter is declared using var, then the expression passed to
that parameter must be a variable reference (so its value can potentially be
changed by the procedure). This means, for example, that it would be
illegal to pass j+3 as the first parameter to times. The variable reference and
the formal parameter must have equivalent types (see equivalence for
details).
Each actual parameter passed to a non-var formal parameter must be
assignable to that parameter (see assignability for details). See also
procedureDeclaration.
In this explanation of procedureCall, we have up to this point ignored the
possibility of procedures exported from modules, monitors and classes. If
the procedure is being called from outside of a module or monitor M from
which it has been exported, the syntax of the procedureCall is:
M . procedureId [ ( [ expn {, expn } ] ) ]
510 Object Oriented Turing Reference Manual
In other words, the module or monitor name and a dot must precede the procedure's name.
If the procedure is being called from outside of a class from which it has
been exported, the syntax of the procedureCall is one of:
(a) classId (p) . procedureId [ ( [ expn {, expn } ] ) ]
(b)
p -> procedureId [ ( [ expn {, expn } ] ) ]
In these, p must the a pointer value that locates an object in the class. Form
(b) is a short form for form (a).
class.
See also
process
declaration
A processDeclaration is:
Syntax
process id [ ( [ paramDeclaration {,paramDeclaration } ] )]
statementsAndDeclarations
end id
Description
A process declaration is much like a procedure declaration, but is activated
by a fork statement rather than by a call. The fork statement starts
concurrent (parallel) execution of the process while the statements
following the fork continue to execute.
Example
This program initiates (forks) two concurrent processes, one of which
repeatedly outputs Hi and the other Ho. The resulting output is an
unpredictable sequence of Hi's and Ho's as greetings executes twice
concurrently, one instance with word set to Hi and the other with word set
to Ho.
process greetings ( word : string )
loop
put word
end loop
end greetings
fork greetings ( "Hi" )
fork greetings ( "Ho" )
Details
The process declaration creates a template for a process (a concurrent
activity), which is activated by a fork statement.
A process declaration can appear wherever a module declaration is
allowed except that a process declaration is not allowed in a class. The
declarations and statements in a process declaration are the same as those
in a procedure.
Chapter 7 : Language Features 511
See paramDeclaration for details about parameters. There is an anomaly in
parameters to processes, that can lead to errors. In particular, non-var
parameters that are non-scalars (such as strings and arrays) are passed by
reference. The result is that the target of the reference may change value
while the process is executing, which in turn means that the seemingly
constant parameter is not really constant. For example, if the string variable
s were passed to the greetings process and subsequently changed, the value
of greetings' formal parameter would change.
The syntax of a processDeclaration presented above has been simplified by
leaving out the optional stack size (compileTimeExpn), import list, pre
condition, init clause, post condition and exception handler.
The full syntax is:
process [ pervasive ] id
[ ( [ paramDeclaration {,paramDeclaration } ] )]
[ : compileTimeExpn ]
[ pre trueFalseExpn ]
[ init id := expn {, id := expn } ]
[ post trueFalseExpn ]
[ exceptionHandler ]
statementsAndDeclarations
end id
See pervasive for information on pervasive processes. The optional
compileTimeExpn following the parameter list (if any) is used to specify the
number of bytes for the process' stack.
import list, pre condition, init clause, post condition and exceptionHandler
for explanations of these additional features.
See also
program
a (main) program
A program is:
Syntax
statementsAndDeclarations
Description
A Turing program consists of a list of statements and declarations.
Example
This is a complete Turing program. It outputs Alan M. Turing.
put "Alan M. Turing"
Example
This is a complete Turing program. It outputs a triangle of stars.
var stars : string := "*"
loop
put stars
512 Object Oriented Turing Reference Manual
stars := stars + "*"
end loop
This is a complete Turing program. It outputs Hello once and Goodbye
twice.
Example
procedure sayItAgain ( what : string, n : int )
for i : 1 .. n
put what
end for
end sayItAgain
sayItAgain ( "Hello", 1)
sayItAgain ( "Goodbye", 2 )
Details
In a program there can be many units (see unit), one of which is the
program (called the main program), the others of which are modules,
monitors and classes. The main program is optionally preceded by an
import list, which lists the units that it uses.
See also
import list.
put
Syntax
statement
A putStatement is:
put [ : fileNumber , ] putItem { , putItem } [ .. ]
Description
The put statement outputs each of the putItems. Usually, a new line is
started in the output after the final putItem. If the optional dot-dot (..) is
present, though, subsequent output will be continued on the current
output line. With character graphics, the omission of dot-dot causes the
remainder of the output line to be cleared to blanks.
Ordinarily, the output goes to the screen. However, if the fileNumber is
present, the output goes to the file specified by the file number (see the
open statement for details). Also, output can be redirected from the screen
to a file, in which case all put statements without a file number are sent to
the file instead of the screen.
Example
var n : int := 5
put "Alice owes me $", n
% Output is: Alice owes me $5
% Note that no extra space is
% output before an integer such as n.
Chapter 7 : Language Features 513
Example
Statement
Output
Notes
put 24
24
put 1/10
0.1
% Trailing zeros omitted
put 100/10
10
% Decimal point omitted
put 5/3
1.666667 % 6 fraction digits
put sqrt (2)
1.414214 % 6 fraction digits
put 4.86 * 10**9
4.86e9 % Exponent for ≥ 1e6
put 121 : 5
bb121 % Width 5; b is blank
put 1.37 : 6 : 3
b1.370 % Fraction width of 3
put 1.37 : 11 : 3 : 2 bb1.370e+00% Exponent width of 2
put "Say \"Hello\"" Say "Hello"
put "XX" : 4, "Y"
XXbbY % Blank shown as b
put true and false false
% Put out a boolean value
put 1 < 2
true
% Put out a boolean value
Example
A single blank line is output this way:
put "" % Output null string then new line
This put statement is sometimes used to close off a line that has been
output piece by piece using put with dot-dot.
Details
The general form of a putItem is one of:
(a)
expn [:widthExpn [:fractionWidth [:exponentWidth ] ] ]
(b)
skip
See the above examples for uses of widthExpn, fractionWidth and
exponentWidth. For the exact meaning of these three widths, see the
definitions of the functions realstr, frealstr and erealstr. The skip item is used
to end the current output line and start a new line.
Details
The put semantics allow put's of enum values. The values printed are the
element names themselves, case sensitive. For example:
type colors : enum ( red, green, blue )
var c : colors := colors . red
put c
Details
% outputs "red" (without the quotes)
The put semantics allow put's of boolean values. The values printed are
either “true” or “false” (without the quotes). For example:
var c : boolean := true or false
put c
% outputs "true" (without the quotes)
514 Object Oriented Turing Reference Manual
quit
fail statement
A quitStatement is:
Syntax
quit [ guiltyParty ] [ : quitReason ]
Description
The quit statement causes a program (or concurrent process) to fail. The
failure (called an exception) either aborts the program (or process) or causes
control to be passed to an exception handler.
Example
In the inputLines procedure, halt the program if end of file is encountered
before the string "stop" is read. Note that a return statement in the
procedure would terminate the procedure but not the entire program.
var line : array 1 .. 50 of string
procedure inputLines
var i : int := 0
loop
if eof then
put "Missing 'stop' in input"
quit
% Halt entire program
end if
i := i + 1
get line ( i )
exit when line ( i ) = "stop"
end loop
end inputLines
inputLines
Details
In the simple case, the optional guiltyParty and quitReason are omitted. The
guiltyParty option is used to specify the position of failure. See
exceptionHandler for an example of a quit statement used in conjunction
with a handler. A handler, which is located at the beginning of a
subprogram body, is given control when a quit is executed or a failure,
such as division by zero, occurs in the subprogram.
The guiltyParty option is used to designate the location of the failure, for
example, to tell the debugger what line is considered to be the location of
the failure. A guiltyParty is one of:
(a)
<
(b)
>
Chapter 7 : Language Features 515
If guiltyParty is omitted, the failure is considered to occur at the quit statement. If it is <, the
failure is considered to occur at the call to the present subprogram. For
example, if the present subprogram implements square root sqrt and is
passed a negative argument, it can use < to specify that the caller provided
a faulty argument. If guiltyParty is >, this means the failure has already
occurred and is being passed on to the next handler or to the system. To
summarize, the three possibilities for designating the location of the failure
are:
(a)
<
Caller is cause of failure
(b)
>
The exception being handled is the cause.
(c)
(omitted guiltyParty) The present quit is the cause.
The quitReason is an integer expression which is used to identify the kind of
failure. If it is omitted, a default value is chosen in the following manner. If
guiltyParty is omitted or is <, the default is 1. If guiltyParty is > and an
exception handler is active, the default is the quitReason of the exception
being handled. If no exception is being handled, the default is 1. In the case
of program abortion, the implementation may pass the quitReason to the
operating system or programming environment.
See also
exceptionHandler, return and result.
Rand
Description
All
This unit contains the predefined subprograms that deal with random
numbers.
All routines in the Rand unit are exported qualified (and thus must be
prefaced with "Rand.").
Entry Points
Real
Returns a random real number.
Int
Returns a random integer.
Reset
Sets the seed in the default sequence to a default value.
Set
Sets the seed in the default sequence to a specified value.
Next
Returns a random real number from a sequence.
Seed
Sets a seed in a sequence.
516 Object Oriented Turing Reference Manual
rand
random real number procedure
Syntax
rand ( var r : real )
Description
The rand statement is used to create a pseudo-random number in the range
zero to one. For example, if x is a real number, after rand(x), x would have
a value such as 0.729548 or 0.352879.
Example
This program repeatedly and randomly prints out Hi ho, hi ho or It's off to
work we go.
var r : real
loop
rand ( r )
if r > 0.5 then
put "Hi ho, hi ho"
else
put "It's off to work we go"
end if
end loop
Details
The rand statement sets its parameter to the next value of a sequence of
pseudo-random real numbers that approximates a uniform distribution
over the range 0<r <1.
Each time a program runs, rand uses the same pseudo-random number
sequence. To get a different sequence (actually, to start the sequence at a
different point), use the randomize procedure.
To use several sequences of repeatable pseudo-random number sequences,
use the randseed and randnext procedures.
In many languages, rand would be a function rather than a procedure. It
has been designed as a procedure in Turing to respect the mathematical
idea that every call to a function using the same arguments (or no
arguments at all) should return the same value. If rand were a function,
this would not be true.
See also
randint, randomize, randseed and randnext.
See also predefined unit Rand.
Chapter 7 : Language Features 517
Rand.Int
All
Syntax
Rand.Int ( low, high : int ) : int
Description
The Rand.Int statement is used to create a pseudo-random integer in the
range low to high, inclusive. For example, if i is an integer, after
i := Rand.Int(i, 1, 10), i would have a value such as 7 or 2 or 10.
Example
This program simulates the repeated rolling of a six sided die.
loop
put "Rolled ", Rand.Int (1, 6)
end loop
Details
The Rand.Int statement sets its parameter to the next value of a sequence
of pseudo-random integers that approximates a uniform distribution over
the range low ≤i ≤ high. It is required that low ≤ high.
Each time a program runs, Rand.Int uses a different pseudo-random
number sequence. To always get the same sequence (actually, to start the
sequence at the same point), use the Rand.Set procedure.
To use several sequences of repeatable pseudo-random number sequences,
use the Rand.Seed and Rand.Next procedures.
Status
Exported qualified.
This means that you can only call the function by calling Rand.Int, not by
calling Int.
See also
Rand.Real, Rand.Set, Rand.Seed and Rand.Next.
Rand.Next
Syntax
All
Rand.Next (seq : 1 .. 10 ) : real
518 Object Oriented Turing Reference Manual
Description
The Rand.Next procedure is used when you need several sequences of
pseudo-random numbers, and you need to be able to exactly repeat these
sequences for a number of simulations. The Rand.Next procedure is the
same as rand, except seq specifies one of ten independent and repeatable
sequences of pseudo-random real numbers.
The Rand.Seed procedure is used to start one of these sequences at a
particular point.
Exported qualified.
Status
This means that you can only call the function by calling Rand.Next, not by
calling Next.
Rand.Seed, Rand.Int, Rand.Real and Rand.Next.
See also
Rand.Real
All
Syntax
Rand.Real : real
Description
The Rand.Real function returns a pseudo-random number in the range
zero to one. For example, if x is a real number, after x := Rand.Real, x
would have a value such as 0.729548 or 0.352879.
Example
This program repeatedly and randomly prints out Hi ho, hi ho or It's off to
work we go.
loop
if Rand.Real > 0.5 then
put "Hi ho, hi ho"
else
put "It's off to work we go"
end if
end loop
Details
The Rand.Real function sets its parameter to the next value of a sequence
of pseudo-random real numbers that approximates a uniform distribution
over the range 0<r <1.
Each time a program runs, Rand.Real uses a different pseudo-random
number sequence. To always get the same sequence (actually, to start the
sequence at the same point), use the Rand.Set procedure.
To use several sequences of repeatable pseudo-random number sequences,
use the Rand.Seed and Rand.Next procedures.
Chapter 7 : Language Features 519
Exported qualified.
Status
This means that you can only call the function by calling Rand.Real, not by
calling Real.
Rand.Int, Rand.Set, Rand.Seed and Rand.Next.
See also
Rand.Reset
All
Syntax
Rand.Reset
Description
This is a procedure with no parameters that resets the sequences of
pseudo-random numbers produced by Rand.Real and Rand.Int. This
allows identical executions of the same program to produce identical
results.
Example
This program simulates the repeated rolling of a six sided die. Each time
the program runs, the same sequence of rolls occurs.
Rand.Reset
loop
put "Rolled ", Rand.Int (1, 6)
end loop
Details
If Rand.Reset and Rand.Set are not used, each time a program runs
Rand.Real and Rand.Int use a different pseudo-random number sequence.
To get the same sequence each time (actually, to start the sequence at a
different point), use Rand.Reset or Rand.Set.
The Rand.Reset procedure can be called any time. However, to make it
work, it should only be called once per program. Any call to Rand.Reset
after the first one is ignored.
To use several sequences of repeatable pseudo-random number sequences,
use the Rand.Seed and Rand.Next procedures.
Status
Exported qualified.
This means that you can only call the function by calling Rand.Reset, not
by calling Reset.
See also
Rand.Set, Rand.Int, Rand.Real, Rand.Seed and Rand.Next.
520 Object Oriented Turing Reference Manual
Rand.Seed
All
Syntax
Rand.Seed ( seed : nat4, seq : 1 .. 10 )
Description
The Rand.Seed procedure restarts one of the sequences generated by
Rand.Next. Each restart with the same seed causes Rand.Next to produce
the same sequence for the given sequence.
Status
Exported qualified.
This means that you can only call the function by calling Rand.Seed, not by
calling Seed.
Rand.Next, Rand.Int, Rand.Real, and Rand.Set.
See also
Rand.Set
All
Syntax
Rand.Set (seed : nat4)
Description
This procedure sets the seed for sequences of pseudo-random numbers
produced by Rand.Real and Rand.Int. This allows identical executions of
the same program to produce identical results.
Example
This program simulates the repeated rolling of a six sided die. Each time
the program runs, the same sequence of rolls occurs.
Rand.Set (16#1234ABCD)
loop
put "Rolled ", Rand.Int (1, 6)
end loop
Details
If Rand.Reset and Rand.Set are not used, each time a program runs
Rand.Real and Rand.Int use a different pseudo-random number sequence.
To get the same sequence each time (actually, to start the sequence at a
different point), use Rand.Reset or Rand.Set.
To use several sequences of repeatable pseudo-random number sequences,
use the Rand.Seed and Rand.Next procedures.
Chapter 7 : Language Features 521
Exported qualified.
Status
This means that you can only call the function by calling Rand.Set, not by
calling Set.
Rand.Reset, Rand.Int, Rand.Real, Rand.Seed and Rand.Next.
See also
randomize
procedure
Syntax
randomize
Description
This is a procedure with no parameters that resets the sequences of
pseudo-random numbers produced by rand and randint. This allows
different executions of the same program to produce different results.
Example
This program simulates the repeated rolling of a six sided die. Each time
the program runs, a different sequence of rolls occurs (unless there is quite
a coincidence or you run the program a lot of times!).
randomize
var roll : int
loop
rand ( i )
put "Rolled ", i
end loop
Details
If randomize is not used, each time a program runs, rand and randint use
the same pseudo-random number sequences. To get a different sequence
(actually, to start the sequence at a different point), use randomize.
randomize on the PC and Mac initialize the random sequence based on the
time of day (among other things). If randomize is called more than once in
the same second, then it will restart the sequence with the same random
number. Consequently, one should not place randomize in a loop. It
should be called once per program at or near the beginning of the program.
To use several sequences of repeatable pseudo-random number sequences,
use the randseed and randnext procedures.
See also
randint, rand, randseed and randnext.
See also predefined unit Rand.
522 Object Oriented Turing Reference Manual
read
file statement
Dangerous parts
A readStatement is:
Syntax
read : fileNumber [ : status ] , readItem { ,readItem
}
Description
The read statement inputs each of the readItems from the specified file.
These items are input directly using the binary format that they have on the
file. In other words, the items are not in source (ASCII or EBCDIC) format.
In the common case, these items have been output to the file using the
write statement.
By contrast, the get and put statements use source format, which a person
can read using an ordinary text editor.
Example
This example shows how to input a complete employee record using a read
statement.
var employeeRecord :
record
name : string ( 30 )
pay : int
dept : 0 .. 9
end record
var fileNo : int
open : fileNo, "payroll", read
…
read : fileNo, employeeRecord
Details
The fileNumber must specify a file that is open with read capability (or a
program argument file that is implicitly opened).
The optional status is an int variable that is set to implementationdependent information about the read. If status is returned as zero, the
read was successful. Otherwise status gives information about the
incomplete or failed read (which is not documented here). You commonly
use status when you are reading a record or array from a file and you are
not sure if the entire item exists on the file. If it does not exist, the read will
fail part way through, but your program can continue and diagnose the
problem by inspecting status.
A readItem is:
variableReference [ : requestedSize [ : actualSize ] ]
Chapter 7 : Language Features 523
Each readItem specifies a variable to be read in internal form. The optional requestedSize is an
integer value giving the number of bytes of data to be read. The
requestedSize should be less than or equal to the size of the item's internal
form in memory (else a warning message is issued). If no requestedSize is
given, the size of the item in memory is used. The optional actualSize is an
int variable that is set to the number of bytes actually read.
An array, record or union may be read and written as a whole.
It is dangerous to read into pointer variables, as this allows the possibility
of creating incorrect addresses in the pointers. It is also dangerous to read
more bytes than are in the readItem.
the write, open, close, seek, tell, get and put statements.
See also
real
the real number type
Syntax
real
Description
The real number type is used for numbers that have fractional parts, for
example, 3.14159. Real numbers can be combined by various operators
such as addition (+) and multiplication (*). Real numbers can also be
combined with integers (whole numbers, such as 23, 0 and -9), in which
case the result is generally a real number. An integer can always be
assigned to a real variable, with implicit conversion to real.
Example
var weight, x : real
var x : real := 9.83
var tax := 0.7
% The type is implicitly real because
% 0.7 is a real number
Details
See also explicitRealConstant. The int type is used instead of real, when
values are whole numbers. See int for details.
Real numbers can be converted to integers using ceil (ceiling), floor, or
round. Real numbers can be converted to strings using erealstr, frealstr,
and realstr. These conversion functions correspond exactly to the
formatting used for the put statement with real numbers. Strings can be
converted to real numbers using strreal. See descriptions of these
conversion functions.
The predefined functions for real numbers include min, max, sqrt, sin,
cons, arctan, sind, cosd, arcand, ln and exp. See the descriptions of these
functions.
524 Object Oriented Turing Reference Manual
Pseudo-random sequences of real numbers can be generated using rand.
See the description of this procedure.
The Turing Report gives a formal definition (not repeated here) of
implemented real numbers in terms of their required accuracy relative to
infinitely accurate (mathematical) real numbers.
Turing implements real numbers using 8 byte floating point representation.
This provides 14 to 16 decimal digits of precision and an exponent range of
at least -38 .. 38. The PC and Macintosh versions of Turing have 16 decimal
digits of accuracy because they use IEEE standard floating point
representation.
realn.
See also
realn
n-byte real number type
Syntax
(a)
(b)
Description
real4
real8
% 4-byte real number
% 8-byte real number
The realn (n-byte real number) types are machine-dependent types that
occupy a specified number of bytes. By contrast, the real type is, in
principle, a machine-independent and mathematical type (however, it
overflows when the exponent of the value is too large or small and it has
only a limited amount of precision).
Example
var width : real4
Details
var height : real8
Turing implements the type real using 8 byte floating point representation.
This provides 14 to 16 decimal digits of precision and an exponent range of
at least -38 .. 38. The PC and Macintosh versions of Turing have 16 decimal
digits of accuracy because they use IEEE standard floating point
representation.
This implies that real8 and real are essentially the same type, so in practice
there is no advantage to using real8 rather than real. However, real4 has
the advantage of occupying half as much space (with correspondingly
reduced precision).
Arithmetic for all real types (real, real4 and real8) is carried out with the
accuracy and exponent range of 8-byte reals.
Chapter 7 : Language Features 525
The type real4 is sometimes called single precision (because it occupies a
single 4-byte word) and real8 is sometimes called double precision.
realstr
real-to-string function
Syntax
realstr ( r : real, width : int ) : string
Description
The realstr function is used to convert a real number to a string. For
example, realstr (2.5e1, 4)="bb25" where b represents a blank. The string is
an approximation to r, padded on the left with blanks as necessary to a
length of width.
The width parameter must be non-negative. If the width parameter is not
large enough to represent the value of r it is implicitly increased as needed.
The displayed value is rounded to the nearest decimal equivalent with this
accuracy. In the case of a tie, the display value is rounded to the next larger
value.
The string realstr (r, width ) is the same as the string frealstr (r, width,
defaultfw ) when r =0 or when 1e-3 < abs (r ) < 1e6, otherwise the same as
erealstr (r, width, defaultfw, defaultew), with the following exceptions. With
realstr, trailing fraction zeroes are omitted, and the decimal point is omitted
if the entire fraction is zero. (These omissions take place even if the
exponent part is printed.) If an exponent is printed, any plus sign and
leading zeroes are omitted. Thus, whole number values are in general
displayed as integers.
Defaultfw is an implementation-defined number of fractional digits to be
displayed. For most implementations, defaultfw will be 6.
Defaultew is an implementation-defined number of exponent digits to be
displayed. For most implementations, defaultew will be 2.
The realstr function approximates the inverse of strreal, although roundoff errors keep these from being exact inverses.
See also
the erealstr, frealstr, strreal, intstr and strint functions.
526 Object Oriented Turing Reference Manual
record
type
A recordType is:
Syntax
record
id {, id } : typeSpec
{ id {, id } : typeSpec }
end record
Description
Each value of a record type consists of fields, one field for each name (id)
declared inside the record. In the following example, the fields are name,
phoneNumber and address.
Example
type phoneRecord :
record
name : string ( 20 )
phoneNumber : int
address : string ( 50 )
end record
…
var oneEntry : phoneRecord
var phoneBook : array 1 .. 100 of phoneRecord
var i : int
oneEntry .name := "Turing, Alan"
…
phoneBook ( i ) := oneEntry
Details
% Assign whole record
In a record, id's of fields must be distinct. However, these need not be
distinct from identifiers outside the record. Records can be assigned as a
whole (to records of an equivalent type), but they cannot be compared. A
semicolon can optionally follow each typeSpec.
Any array contained in a record must have bounds that are known at
compile time.
The notation –> can be used to access record fields. For example, if p is a
pointer to phoneRecord, p–>name locates the name field. See pointer.
Chapter 7 : Language Features 527
register
Description
use machine register
Dirty
When a variable, constant or parameter is declared, you can request that
the item be placed in a machine register. This should be done only for
programs requiring considerable efficiency.
Example
var register counter : int
const register maxCounter : int := 100
procedure p ( register x : real )
…
end p
Details
Items can be requested to be in registers only if they are local to a
subprogram (not global variables, declared in the main program, a module,
monitor or class). Items requested to be in registers cannot be bound to,
passed to reference parameters, have their address taken by addr, or have
certain type cheats applied to them (since a machine register has no
address).
The request to use a register may be ignored. For example, the current
(1999) interpretive implementation uses pseudo-code, which has no
machine registers, and so ignores the register keyword. For the syntax of
using this keyword, see var declaration, const declaration and
paramDeclaration.
rem
remainder operator
Syntax
rem
Description
The rem (remainder] operator produces the remainder of one number
divided by another. For example, 7 rem 2 produces 1 and -12 rem 5
produces -2.
Example
In this example, eggCount is the total number of eggs. The first put
statement determines how many dozen eggs there are. The second put
statement determines how many extra eggs there are beyond the last
dozen.
528 Object Oriented Turing Reference Manual
var eggCount : int
get eggCount
put "You have ", eggCount div 12, " dozen eggs"
put "You have ", eggCount rem 12, " left over"
infix operators, precedence of operators and the mod and div operators.
See also
repeat
make copies of string function
Syntax
repeat ( s : string, i : int ) : string
Description
The repeat function returns i copies of string s catenated together. For
example, repeat ("X", 4) is XXXX.
Example
This program outputs HoHoHo.
var word : string := "Ho"
put repeat ( word, 3 )
If i is less than or equal to zero, the null string "" is returned. The repeat
function is often used for spacing of output. For example, this statement
skips 20 blanks before outputting x.
Details
put repeat (" ", 20), x
result
statement
A resultStatement is:
Syntax
result expn
Description
A result statement, which must appear only in a function, is used to
provide the value of the function.
Example
This function doubles its parameter.
function double ( x : real ) : real
result 2 * x
end double
put double ( 5.3 )
% This outputs 10.6
Chapter 7 : Language Features 529
This function finds the position of a name in a list.
Example
function find ( a : array 1 .. 100 of string ) : int
for i : 1 .. 100
if a ( i ) = name then
result i
end if
end for
end find
The execution of a result statement computes the value of the expression
(expn) and terminates the function, returning the value as the value of the
function.
Details
The expression must be assignable to the result type of the function, for
example, in double, 2*x is assignable to real. (See the assignmentStatement for
the definition of assignable.)
A function must terminate by executing a result statement and not by
reaching the end of the function.
return
statement
A returnStatement is:
Syntax
return
Description
A return statement terminates the procedure (or main program) in which it
appears. Ordinarily, a procedure (or main program) terminates by
reaching its end; the return statement is used to cause early termination.
Example
This procedure takes no action if the errorHasOccurred flag has been set to
true.
procedure double
if errorHasOccurred then
return % Terminate this procedure
end if
… handle usual case in this procedure …
end double
Details
A return must not appear as a statement in (the outermost level of) a
module, nor can it appear in a function.
530 Object Oriented Turing Reference Manual
RGB
Description
This unit contains the predefined constants for the basic colors and the
subprograms to change the color palette.
All subprograms in the RGB unit are exported qualified (and thus must be
prefaced with "RGB."). All the color constants are exported unqualified
and thus do not need the RGB prefix.
Entry Points
black, blue, green, cyan, red, magenta, color names constants
purple, brown, white, gray, grey, brightblue, (exported unqualified)
brightgreen, brightcyan, brightred,
brightmagenta, brightpurple, yellow,
brightwhite, darkgray, darkgrey, colorfg,
colourfg, colorbg, colourbg
GetColor
Gets the current red, green and blue values of a specified
color number.
GetColour
Gets the current red, green and blue values of a specified
color number.
SetColor
Sets the red, green and blue values of a specified color
number.
SetColour
Sets the red, green and blue values of a specified color
number.
AddColor
Creates a new color number with a specified red, green
and blue value.
AddColour
Creates a new color number with a specified red, green
and blue value.
Win DOS Mac
RGB.AddColor
Syntax
X TOOT Term
RGB.AddColor ( redComp, greenComp, blueComp : real ) : int
Chapter 7 : Language Features 531
Description
The RGB.AddColor function attempts to create a new color with the red,
green and blue components specified. If successful, the function returns a
new color number (usually one greater than maxcolor) and maxcolor is
updated by adding 1 to it. If it is unsuccessful, the function returns –1 and
Error.Last and Error.LastMsg can be used to determine the cause of the
problem.
The red, green and blue values must normalized to be between 0 and 1.
Thus to add the pure red to the color palette, you would call:
newColor := RGB.AddColor (1.0, 0.0, 0.0)
newColor would be set to the color added, or –1 if the attempt to add a color
failed.
This program adds a palette of 16 blues to the end of the color palette.
Example
var clr : int
for blueShade : 0 .. 15
clr = RGB.AddColor (0, 0, blueShade / 15)
if clr = –1 then
put "Color add failed on shade number ", blueShade
exit
else
put "Added color number ", clr
end if
end for
Details
RGB.AddColour is an alternate spelling for RGB.AddColor.
Status
Exported qualified.
This means that you can only call the function by calling RGB.AddColor,
not by calling AddColor.
See also
RGB.GetColor and RGB.SetColor.
Win DOS Mac
RGB.GetColor
X TOOT Term
Syntax
RGB.GetColor ( colorNumber : int,
var redComp, greenComp, blueComp : real )
Description
The RGB.GetColor procedure returns the red, green and blue components
to the color associated with the colorNumber parameter. The red, green and
blue values are normalized to be between 0 and 1. Thus color white returns
1.0 for the redComp, greenComp and blueComp values and color black returns
0.0 for all three.
532 Object Oriented Turing Reference Manual
This program gets the components of all the available colors.
Example
put "Color Red Green Blue"
for clr : 0 .. maxcolor
var redComp, greenComp, blueComp : int
RGB.GetColor (clr, redComp, greenComp, blueComp)
put clr : 4, " ", redComp : 6 : 4 , " ", greenComp : 6 : 4, " ",
blueComp : 6 : 4
end for
Details
RGB.GetColour is an alternate spelling for RGB.GetColor.
Status
Exported qualified.
This means that you can only call the function by calling RGB.GetColor,
not by calling GetColor.
RGB.SetColor and RGB.AddColor.
See also
Win DOS Mac
RGB.maxcolor
X TOOT Term
Syntax
maxcolor : int
Description
The maxcolor function is used to determine the maximum color number
for the current mode of the screen. The alternate spelling is maxcolour.
Example
This program outputs the maximum color number.
setscreen ("graphics")
…
put "The maximum color number is ", maxcolor
Details
The screen should be in a "screen" or "graphics" mode. If it is not, it will
automatically be set to "screen" mode. See View.Set for details.
For IBM PC compatibles in "screen" mode, maxcolor = 15. For the default
IBM PC compatible "graphics" mode (CGA), maxcolor = 3.
Status
Exported unqualified.
This means that you can call the function by calling maxcolor or by calling
RGB.maxcolor.
See also
Draw.Dot for examples of the use of maxcolor. See the Text.Color
procedure which is used for setting the currently-active color.
Chapter 7 : Language Features 533
Win DOS Mac
RGB.SetColor
X TOOT Term
Syntax
RGB.SetColor ( colorNumber: int,
redComp, greenComp, blueComp : real )
Description
The RGB.SetColor function sets the red, green and blue components of the
color associated with the colorNumber parameter. The red, green and blue
values must normalized to be between 0 and 1. Thus to set the color
associated with the colorNumber parameter to pure red, you would call:
RGB.SetColor (colorNumber, 1.0, 0.0, 0.0)
It is wise to use Error.Last and Error.LastMsg to check to see if the color
change is successful.
Example
This program sets all the available colors to shades of red
for clr : 0 .. maxcolor
if not RGB.SetColor (clr, clr / maxcolor, 0, 0) then
put "Color set failed on color number ", clr
exit
end if
end for
Details
RGB.SetColour is an alternate spelling for RGB.SetColor.
Status
Exported qualified.
This means that you can only call the function by calling RGB.SetColor,
not by calling SetColor.
See also
RGB.GetColor and RGB.AddColor.
round
Syntax
real-to-integer function
round ( r : real ) : int
534 Object Oriented Turing Reference Manual
Description
The round function is used to convert a real number to an integer. The
result is the nearest integer to r. In the case of a tie, the numerically larger
value is returned. For example, round (3) is 3, round (2.85) is 3 and round
(-8.43) is -8.
See also
the floor and ceil functions.
scalar
Syntax
type
A scalarType is one of:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
standardType
enumeratedType
subrangeType
pointerType
char
intn
natn
realn
namedType
% int, real, boolean or string
% Must name one of the above types
Description
Scalar types are sometimes called simple or primitive types. The non-scalar
types are strings, sets, arrays, records, unions and in OOT char(n). They are
defined using scalar types. Scalar types are passed by value to parameters,
while non-scalars are passed by reference (by passing an implicit pointer to
the non-scalar value).
Description
In current Turing implementations scalar types are directly represented in
1, 2, 4 or 8 bytes in a computer's memory. This implies that they can be
efficiently passed by value.
Chapter 7 : Language Features 535
seek
(file) statement
A seekStatement is one of:
Syntax
(a)
(b)
Description
seek : fileNumber , filePosition
seek : fileNumber , *
Random access of both source (ASCII or EBCDIC) and internal form
(binary) files is provided by the seek and tell statements. The seek
statement repositions the specified file so that the next input/output
operation will begin at the specified point (filePosition) in the file.
The fileNumber must specify a file that is open with seek capability. The
filePosition is a non-negative integer offset in bytes from the beginning of
the file. Usually, this is a number returned by the tell statement. (The first
position in the file is position zero.)
Form (b) specifies that the next operation is to begin at the position
immediately following the current end of the file. A filePosition of zero
specifies that the next operation is to start at the beginning of the file.
Seeking to a position beyond the current end of the file and then writing,
automatically fills the intervening positions with the internal
representation of zero.
Example
This example shows how to use seek to append to the end of a file.
var employeeRecord :
record
name : string ( 30 )
pay : int
end record
var fileNo : int
open : fileNo, "payroll", write, seek, mod
% Seek to the end of the file
seek : fileNo, *
write : fileNo, employeeRecord
% This record is added to the end of the file
See also
read, write, open, close, tell, get and put statements. Another example use
of seek is given with the explanation of the tell statement.
536 Object Oriented Turing Reference Manual
self
pointer to current object
Syntax
self
Description
The self function produces a pointer to the current object. This function can
be used only inside a class declaration. See class.
Example
Enter the current object onto a list of displayable objects. The module called
displayable has exported a procedure called enter whose parameter type is
pointer to anyclass. Since self is a pointer to C and C is a descendant of
anyclass, it is legal to pass self to displayable.enter.
class C
import displayable
…
displayable.enter ( self ) …
…
end C
Details
It is illegal to call the exported entries of a class until the current object has
been completely initialized, so, many calls to the current object using self
will not be legal.
The notation to call exported subprogram p of an enclosing class C or of its
ancestor D, is C.p or D.p. Calls of this form, which can appear only within
class C, call the subprogram in C (or in D in the case of D.p) regardless of
the object type, or of any overriding, or of the status of initialization.
separator
Description
between tokens in a program
A Turing program is made up of a sequence of tokens (see tokens), such as
var, x, :, and int. These tokens may have separators between them. A
separator is a comment (see comment), blank, tab, form feed or an end of
line.
Chapter 7 : Language Features 537
Win DOS Mac
serialget
serial port function
X TOOT Term
Syntax
serialget : int
Description
The serialget procedure is used on a PC to read a single character from the
serial port. This port corresponds to the MS-DOS device "COM1". This
function can be used to send characters to a modem or to another computer
through a null-modem cable.
Example
This program sends a message to the COM1 port and then gets a reply
ending with a newline.
% We assume that the serial port has been set up previously using
% the DOS mode command.
const message : string := "This is a test"
for i : 1.. length (message)
serialput (ord (message (i))
end for
serialput (10)
var inmessage : string := ""
var ch : int
loop
ch := serialget
exit when ch = 10
inmessage := inmessage + chr (ch)
end loop
put inmessage
Details
serialput and serialget assume that the port’s baud rate, stop bits, etc. have
been properly set with the DOS mode command outside of Turing. Check
the DOS manual for information on how to use this command.
serialput and serialget will work reliably for communication up to 1200
baud. For rates higher than this, these routines are not suitable.
Note that these routine pass integers that represent characters.
Consequently, they should be passed (and will return) values that are
between 0 and 255. Because they pass integers, use the chr and ord
functions to convert integers to characters and vice-versa.
See also
serialput procedure which sends characters out the serial port. See also chr
and ord functions for converting between a character and its ASCII value.
538 Object Oriented Turing Reference Manual
Win DOS Mac
serialput
serial port procedure
X TOOT Term
Syntax
serialput ( p : int )
Description
The serialput procedure is used on a PC to send a single character out the
serial port. This port corresponds to the MS-DOS device "COM1". This
function can be used to send characters to a modem or to another computer
through a null-modem cable.
Example
This program sends a message to the COM1 port and then gets a reply
ending with a newline.
% We assume that the serial port has been set up previously using
% the DOS mode command.
const message : string := "This is a test "
for i : 1.. length (message)
serialput (ord (message (i))
end for
serialput (10)
var inmessage : string := ""
var ch : int
loop
ch := serialget
exit when ch = 10
inmessage := inmessage + chr (ch)
end loop
put inmessage
Details
serialput and serialget assume that the port’s baud rate, stop bits, etc. have
been set with the DOS mode command outside of Turing. Check the DOS
manual for information on how to use this command.
serialput and serialget will work reliably for communication up to 1200
baud. For rates higher than this, these routines are not suitable.
Note that these routine pass integers that represent characters.
Consequently, they should be passed (and will return) values that are
between 0 and 255. Because they pass integers, use the chr and ord
functions to convert integers to characters and vice-versa.
See also
serialget procedure which reads characters from the serial port. See also
chr and ord functions for converting between a character and its ASCII
value.
Chapter 7 : Language Features 539
set
type
A setType is:
Syntax
set of typeSpec
Description
Each value of a set type consists of a set of elements. The typeSpec, which is
restricted to being a subrange or an enumerated type, gives the type of
these elements.
Example
The smallSet type is declared so that it can contain any and all of the values
0, 1 and 2. Variable s is initialized to be the set containing 1 and 2.
type smallSet : set of 0 .. 2
var s : smallSet := smallSet ( 0, 1 )
…
if 2 in s then …
Details
In classical mathematics, the set consisting of 0 and 1 is written as {0,1}.
This is written in Turing using a set constructor consisting of the name of the
set type followed by a parenthesized list of elements, which in this example
is smallInt (0,1). The empty set is written, for example, as smallInt (). The
full set is written as smallInt (all), so smallInt (all) = smallInt (0,1,2).
Sets can be assigned as a whole (to sets of an equivalent type). See also
equivalence of types.
The operators to combine two sets are union (+), intersection (*), set
subtraction (-), equality (=), inequality (not=), subset (<=), strict subset (<),
superset (>=), strict superset (>), and xor ("exclusive or" also known as
symmetric difference). Only sets with equivalent types (equal bounds on
their index types) can be combined by these operators. The operators
which determine if an element is, or is not, in a set are in and not in. For
example, the test to see if 2 is in set s is written in the above example as: 2
in s.
The indexType of a set type must contain at least one element. For example,
the range 1 .. 0 would not be allowed. See also indexType. In Turing, sets are
limited to at most 31 elements. OOT allows a very large number of
elements.
Details
It is illegal to declare an "anonymous" set. The only legal declaration for an
set is in a type declaration. For example, the following is now illegal:
var a : array 1 .. 10 of set of 0 .. 3
Given that there is no (easy) way of generating a set value without it being
a named type, this should not impact any but the most bizarre code.
See also
precedence of operators for the order of applying set operations.
540 Object Oriented Turing Reference Manual
setConstructor
A setConstructor is:
Syntax
setTypeId ( membersOfSet )
Description
Each value of a set type consists of a set of elements. In classical
mathematics, the set consisting of 0 and 1 is written as {0,1}. This is written
in Turing using a set constructor consisting of the name of the set type
(setTypeId) followed by a parenthesized list of elements.
Example
The smallSet type is declared so that it can contain any and all of the values
0, 1 and 2. Variable s is initialized to be the set containing 1 and 2. The set
{0,1} is written in this Turing example as smallInt (0,1).
type smallSet : set of 0 .. 2
var s : smallSet := smallSet ( 0, 1 )
…
if 2 in s then …
Details
The form of membersOfSet is one of:
(a) expn { , expn} % List of members of set
(b) all
% All member of index type of set
(c)
% Nothing, meaning the empty set
The empty set is written, for example, as smallInt (). The full set is written
as smallInt (all), so smallInt (all) = smallInt (0,1,2). See also the set type.
The syntax of setConstructor as given above has been simplified by ignoring
the fact that set types can be exported from modules. When a set type is
exported and used outside of a module, you must write the module name,
a dot and then the type name. For example, the set constructor above
would be written as m.smallSet(1,2), where m is the module name.
setpriority
Syntax
procedure
setpriority ( p : nat )
Chapter 7 : Language Features 541
Description
The setpriority procedure is used to set the priority of a process in a
concurrent program. This priority cannot be counted on to guarantee
critical access to shared variables. A smaller value of p means increased
speed. The argument to setpriority may be limited to the range 0 to 2**15 1.
See also
getpriority, fork and monitor.
See also predefined unit Concurrency.
setscreen
graphics procedure
Syntax
setscreen ( s : string )
Example
Here are example uses of the setscreen procedure. In many cases, these
will appear as the first statement of the program. They can, however,
appear any place in a program.
setscreen ( "graphics" )
% IBM CGA graphics
setscreen ( "graphics:e16" ) % IBM EGA graphics
setscreen ( "screen" )
% To use locate
setscreen ( "nocursor" )
% Turn off cursor
setscreen ( "noecho" )
% Do not echo keys
Description
The setscreen statement is used to change the mode of the screen, as well
as the way in which Turing does input and output. The parameter to
setscreen is a string, such as "graphics". The string contains one or more
options separated by commas, such as "text, noecho".
Details
Users should look at View.Set in order to find out the implementation
specified details of setscreen on their systems.
Many of the options to setscreen are specific to IBM PC compatible
computers and Apple Macintoshes. They have been adopted to other
systems such as the Icon and X Windows.
There are three screen modes, text, screen and graphics. Most systems do
not support all three modes.
System text
PC Turing
MacTuring X
screen graphics
X
X
(Default: screen)
X
(Default: 480x275)
text mode does not allow for any character graphics whatsoever. Only put
and get are allowed.
screen mode allows for character graphics commands such as locate and
lolor.
542 Object Oriented Turing Reference Manual
graphics mode allows for character graphics and pixel graphics commands
such as locate and drawbox.
If a character graphics command is given while in text mode, the program
automatically switches to screen mode if available, otherwise it switches to
graphics mode. If a pixel graphics command is given while in text or
screen mode, the program automatically switches to graphics mode.
Where the options to setscreen are mutually exclusive, they are listed here
with the default underlined. Here are the options:
"text", "screen", "graphics" - Sets mode to the given mode and always erases the screen,
even when already in the requested mode; "text" is the default character
output mode; "screen" is character graphics mode, allowing the locate
procedure; "graphics" is "pixel graphics" mode.
By default, "graphics" is CGA graphics on IBM PC compatibles. On UNIX
dumb terminals, "graphics" is meaningless. A suffix can be given, as in
"graphics:h16", to specify another version of pixel graphics (assuming the
corresponding hardware is available).
The set of "graphics" options is given on the next page. On Apple
Macintoshes, the graphics screen is 480x275 by default. The size of the
screen can also be set with a suffix in the form "graphics:150;250", which
would set the graphics output window to be 150x250 pixels. Macintoshes
also allow you to set the window size in screen mode. The default is 25
rows by 80 columns but this can be changed with "screen:10;100" to be 10
rows by 100 columns.
Under MacTuring, graphics mode can have a modifier in the form
"graphics:<x>;<y>". This sets the window to be <x> by <y> in size. The
maximum size of a window is the size of the screen. Also available under
MacTuring, graphics mode can have a modifier in the form
"graphics:<x>;<y>;<depth>". This sets the window to be <x> by <y> in size.
<depth> has the following allowable values:
bw or 1 or 2 = allows 2 colors (black and white)
4 = allows 4 colors
8 or 256 = allows 256 colors
16 = allows 16 colors
thousands = 16 bit color (thousands of colors)
millions = 32 bit color (millions of colors)
cga = The window will use the 4 color CGA palette.
ega or vga = The window will use the 16 color VGA palette.
mcga = The window will use the 256 color MCGA palette.
Note that the depths that use IBM color palettes have a window with a
black background. They also use the IBM characters (i.e. character sizes are
8x8, 8x14 or 8x16 as they are on the IBM PC).
The set of "graphics" options is given on the next page. On Apple
Macintoshes, the graphics screen is 480x275 by default. The size of the
screen can also be set with a suffix in the form "graphics:150;250", which sets
the graphics output window to 150x250 pixels. Macintoshes also allow you
to set the window size in screen mode. The default is 25 rows by 80 columns
but this can be changed with "screen:10;100" to be 10 rows by 100 columns.
Chapter 7 : Language Features 543
Mode
"graphics"
320
"graphics:cga"
320
"graphics:ega"
640
"graphics:e16"
640
"graphics:vga"
640
"graphics:v16"
640
"graphics:mcga"
320
"graphics:m256"
320
"graphics:svga"
640
"graphics:svga1"
640
"graphics:svga2"
640
"graphics:svga3"
800
"graphics:svga4"
800
"graphics:svga5"
1024
"graphics:svga6"
1024
maxx+1
200
200
350
350
480
200
200
480
400
480
600
600
480
maxy+1
maxcolor+1
4
4
16
16
16
16
256
256
256 (DOS OOT)
256 (DOS OOT)
256 (DOS OOT)
16 (DOS OOT)
256 (DOS OOT)
768
16 (DOS OOT)
768
256 (DOS OOT)
"cursor", "nocursor" - Causes the cursor to be shown (or hidden). There is never a
cursor showing in "graphics" mode. On UNIX dumb terminals, the cursor
cannot be hidden. There is an optional suffix for "cursor" that determines
the shape of the cursor. In CGA graphics, the cursor is constructed out of
horizontal lines numbered 0, 1, 2, up to 7, with 0 being the top. The suffix
gives the range of lines to be used for the cursor, for example, "cursor:5;7"
specifies a cursor consisting of lines 5 through 7. In general, this form is
"cursor:startline;endline", where startline and endline are integer literals such
as 5 and 7. On the Apple Macintosh, it is possible to set the cursor size from
0-10.
"echo", "noecho" - Causes (or suppresses) echoing of characters that are typed.
Echoing is commonly turned off in interactive programs to keep typed
characters from being echoed at inappropriate places on the screen.
Details
The setscreen procedure supports all the features of the older screen
procedure. The screen procedure is no longer supported, you must use
setscreen instead.
The ability of an IBM system to support a graphics mode depends on the
graphics hardware available. If the system does not seem to be supporting
a particular graphics mode, use the scrntest.exe program provided with
PC Turing and DOS OOT to test the capabilities of your graphics adapter
(see chapter 3 for more details). If the graphics adapter is unrecognized by
Turing or DOS OOT, but supports the VESA SuperVGA graphics standard,
you can use the –vesa startup option with Turing or DOS OOT to force the
use of the VESA standard.
See also
drawdot, drawline, drawoval, drawarc, whatdotcolor, color, colorback,
takepic and drawpic.
See also predefined unit View.
544 Object Oriented Turing Reference Manual
shl
shift left operator
Syntax
A shl B
Description
The shl (shift left) operator produces the value of A shifted B bits to the
left. Both A and B must be non-negative integers (natural numbers).
Example
Assign the base 2 value 11 to i and then shift it left by 2 places and assign
the resulting base 2 value 1100 to j.
var i, j : int
i := 2 # 11
% 2#11 = 3 (base 10)
j := i shl 2
% j becomes 2#1100 = 12 (base 10)
The shl operator is defined mathematically (in a machine-independent
way) as follows: A shl B = A * ( 2**B ). Overflow occurs when the result
exceeds the maximum value of the nat4 (4-byte natural number) type.
Details
Value A can be of any integer type (as long as it is non-negative) or any
natural number type.
The shl operator has the same precedence as the * operator.
shr (shift right), or, and and xor, which also are bit manipulation operators
that act on non-negative values. See also explicitIntegerConstant which
describes values such as 2#1100.
See also
shr
shift right operator
Syntax
A shr B
Description
The shr (shift right) operator produces the value of A shifted B bits to the
right. Both A and B must be non-negative integers (natural numbers).
Example
Assign the base 2 value 1101 to i and then shift it right by 2 places and
assign the resulting base 2 value 11 to j.
var i, j : int
i := 2 # 1101 % 2#1101 = 13 (base 10)
j := i shr 2
% j becomes 2#11 = 3 (base 10)
Chapter 7 : Language Features 545
The shr operator is defined mathematically (in a machine- independent
way) as follows: A shr B = A div 2**B.
Details
Value A can be of any integer type (as long as it is non-negative) or any
natural number type.
The shr operator has the same precedence as the * operator.
shl (shift left), or, and and xor, which also are bit manipulation operators
that act on non-negative values. See also explicitIntegerConstant which
describes values such as 2#1101.
See also
sign
function
Syntax
sign ( r : real ) : -1 .. 1
Description
The sign function is used to determine whether a number is positive, zero
or negative. It returns 1 if r > 0, 0 if r = 0, and -1 if r < 0. For example, sign
(5) is 1 and sign (-23) is -1.
Example
This program reads in numbers and determines if they are positive, zero or
negative:
var x : real
get x
case sign ( x ) of
label 1 : put "Positive"
label 0 : put "Zero"
label -1 : put "Negative"
end case
See also
See also predefined unit Math.
signal
Syntax
wake up a process statement
A signalStatement is:
signal variableReference
546 Object Oriented Turing Reference Manual
Description
A signal statement is used in a concurrent program to wake up a process
that is blocked (waiting on a condition variable). The statement can only be
used inside a monitor (a special kind of module that handles concurrency).
A signal statement operates on a condition variable (the variableReference),
which is essentially a queue of sleeping processes. See condition for an
example of a signal statement.
Details
A signal statement wakes up one process that is doing a wait on the
specified condition queue, if such a process exists. If the condition is
deferred (or timeout; see condition), the signaler continues in the monitor,
and the awakened process is allowed to continue only when the monitor
becomes inactive. A signal to an immediate (non-deferred) condition causes
the signaled process to begin running in the monitor immediately. The
signaling process waits to re-enter the monitor when the monitor becomes
inactive.
See also
condition and wait. See also monitor and fork. See also empty. See also
pause.
simutime
simulated time function
Syntax
simutime : int
Description
The simutime function returns the number of simulated time units that
have passed since program execution began.
Details
Simulated time only passes when all process are either paused or waiting.
This simulates the fact that CPU time is effectively infinitely fast when
compared to "pause" time.
Example
This prints out the simulated time passing between two processes. This will
print out 3, 5, 6, 9, 10, 12, 15, 15, 18, 20, 21, …
process p (t : int)
loop
pause t
put simutime
end loop
end p
fork p (3)
See also
fork p (5)
See also predefined unit Concurrency.
Chapter 7 : Language Features 547
sin
sine function (radians)
Syntax
sin ( r : real ) : real
Description
The sin function is used to find the sine of an angle given in radians. For
example, sin (0) is 0.
Example
This program prints out the sine of /6, 2/6, 3/6, up to 12/6 radians.
const pi := 3.14159
for i : 1 .. 12
const angle := i * pi / 6
put "Sin of ", angle, " is ", sin ( angle )
end for
the sind function which finds the sine of an angle given in degrees. (2
radians are the same as 360 degrees.)
See also
See also predefined unit Math.
sind
sine function (degrees)
Syntax
sind ( r : real ) : real
Description
The sind function is used to find the sine of an angle given in degrees. For
example, sind (0) is 0.
Example
This program prints out the sine of 30, 60, 90, up to 360 degrees.
for i : 1 .. 12
const angle := i * 30
put "Sin of ", angle, " is ", sind ( angle )
end for
See also
the sin function which finds the sine of an angle given in radians. (2
radians are the same as 360 degrees.)
See also predefined unit Math.
548 Object Oriented Turing Reference Manual
sizeof
size of a type
Dirty
Syntax
sizeof (typeNameOrVariableReference)
Description
The sizeof attribute is used to find the number of bytes used to represent
the type or variable. This is implementation-dependent (dirty).
Example
The size of int2 and nat2 is 2.
var i : int2
Details
const nat2size := sizeof ( i )
% size is 2
The typeNameOrVariableReference must be the name of a user-defined type,
a variable reference, a basic type (such as real), or a constant.
In principle, sizeof returns the number of storage units which would not
necessarily be 8-bit bytes. For example, in some older machines, such as the
CDC 6000 series, the storage units are 60 bit words. However, almost all
modern computers use 8-bit bytes so these are the units of sizeof.
Beware that sizes may reflect alignment constraints in the underlying
computer. For example, string sizes may be rounded up to even values (2byte word alignments).
See also
the indirection operator @, cheat, explicitIntegerConstant (how to write
hexadecimal constants), and pointers (in particular unchecked pointers).
See also addr, which returns the address of a variable.
sizepic
graphics function
Pixel graphics only
Syntax
sizepic ( x1, y1, x2, y2 : int) : int
Description
The sizepic function is used to determine the size buffer needed to record a
picture from the screen (see description of takepic ). This gives the
minimum number of elements of the int array used by takepic. The buffer
is used by drawpic to make copies of the picture on the screen.
Example
This program outputs the size of array needed to hold a picture with left
bottom corner at x=10, y=20 and right top corner at x=50, y=60.
Chapter 7 : Language Features 549
setscreen ("graphics")
…
put "The size of the array needs to be",
sizepic ( 10, 20, 50, 60 )
See takepic for an example of the use of sizepic and for further
information about buffers for drawing pictures.
Details
The screen should be in a "graphics" mode. See the setscreen procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
drawpic. See also setscreen, maxx, maxy, drawdot, drawline, drawbox,
and drawoval.
See also
See also predefined unit Pic.
skip
used in get statement
Syntax
skip
Description
Using skip as an input item in a get statement causes the current input to
be skipped until a non-whitespace token is encountered. Whitespace
includes all blanks, tabs, form feeds and newlines.
Example
The skip input item was originally intended to be used to see if more input
exists in an input file. This use has been largely made redundant by a
change in the Turing language. The new version of Turing reads a token, as
in get s but not in get s:* or get s:3, and automatically skips any white space
following the input value, but will not go beyond the beginning of the next
input line. Originally this automatic skipping did not take place, so skip
was required. The form of an input loop that used skip was as follows:
loop
get skip
exit when eof
get …
…
% This is line now redundant
end loop
Details
The skip bypasses all whitespace characters including any trailing
newlines and blank lines. By skipping these characters, a true end-of-file
condition was detected. Otherwise, the end-of-file could have been hidden
by any whitespace following the last input item. With the change in Turing,
the line get skip is no longer needed (although it still works correctly).
550 Object Oriented Turing Reference Manual
The skip can also be used to correctly identify the start of a long string
(usually to be read in line or counted mode). Here, it skips the whitespace
and trailing newline as follows:
Example
var i : int
var line : string
loop
get i, skip, line:*
…
Details
See also
end loop
The first item in the get statement reads an integer by skipping all
whitespace and reading digits until whitespace is encountered. The input
stream is then left with the whitespace as the next input character. The skip
then skips past the whitespace, effectively beginning the next input at the
next non-whitespace character. This truncates leading blanks and has
another, potentially more important, effect. If the integer is the last data on
a line and the string is on a following line, the skip is necessary to avoid
setting line to a null string value.
get statement and loop statement.
skip
used in put statement
Syntax
skip
Description
Using skip as an output item in a put statement causes the current output
line to be ended and a new line to be started.
Example
This example, To be is output on one line and Or not to be on the next.
put "To be", skip, "Or not to be"
Details
Using skip is equivalent to outputting the newline character "\n".
Win DOS Mac
sound
Syntax
statement
X TOOT Term
sound ( frequency, duration : int )
Chapter 7 : Language Features 551
Description
The sound procedure is used to cause the computer to sound a note of a
given frequency for a given time. The frequency is in cycles per second
(Hertz). The time duration is in milliseconds. For example, middle A on a
piano is 440 Hertz, so sound(440, 1000) plays middle A for one second.
Example
This program sounds the frequencies 100, 200 up to 1000 each for half a
second.
for i : 1 .. 10
put i
sound ( 100 * i, 500 )
end for
% Sound note for 1/2 second
Details
On IBM PC compatibles, the hardware resolution of duration is in units of
55 milliseconds. For example, sound(440, 500) will delay the program by
about half a second, but may be off by as much as 55 milliseconds.
Details
The sound procedure does not currently work under MacOOT.
See also
play statement, which plays notes based on musical notation. For example,
play("8C") plays an eighth note of middle C. See also the delay, clock,
sysclock, wallclock, time and date statements.
See also predefined unit Music.
Win DOS Mac
Sprite
Description
X TOOT Term
Sprites are a way of doing animation in Turing bypassing the Pic module.
A sprite is essentially a picture with a specific location and "depth". You
create a sprite by calling Sprite.New with a picID received from Pic.New.
You can then move the sprite around by calling Sprite.SetPosition. When
you are finished with the sprite, you call Sprite.Free.
Note that sprites work best when they are moderately small. If you have
large sprites, you will continue to have flashing.
All subprograms in the Sprite unit are exported qualified (and thus must
be prefaced with "Sprite.").
Entry Points
New
Creates a new sprite from a picture.
Free
Disposes of a sprite and free up its memory.
SetHeight
Sets the height of a sprite. Sprites with a greater height
appear above sprites with a lesser height. The background
is considered height 0. The height may be negative.
552 Object Oriented Turing Reference Manual
SetPosition
Sets the location of the sprite. Can specify the center of the
sprite or the lower-left corner.
ChangePic
Changes the picture associated with a sprite.
Animate
Changes the location and the picture associated with a
sprite. Used for animating a moving changing image.
Show
Shows a previously hidden sprite.
Hide
Hides a visible sprite.
Win DOS Mac
Sprite.Animate
X TOOT Term
Syntax
Sprite.Animate ( spriteID, picID, x, y : int,
centered : int )
Description
Moves the sprite specified by spriteID to the location specified by (x, y). If
centered is true, then the sprite is centered on (x, y). Otherwise (x, y)
specifies the lower-left corner of the sprite. At the same time, it changes the
picture associated with the sprite.
A simple example of the Sprite.Animate procedure would be of a man
walking. The picture associated with the sprite would constantly change as
the figure was walking. At the same time, the location of the figure would
also change.
Example
Here is a program that loads six images from files Pic1.bmp through
Pic6.bmp and then displays them sequentially on the screen, moving the
image two pixels each time.
var pics : array 0 .. 5 of int
var sprite: int
for i : 1 .. 6
pics (i – 1) := Pic.FileNew ("Pic" + intstr (i) + ".bmp")
if Error.Last not= 0 then
put "Error loading image: ", Error.LastMsg
return
end if
end for
sprite:= Sprite.New (pics (0))
Sprite.SetPosition (sprite, 0, 100, false)
Sprite.Show (sprite)
for x : 2 .. maxx by 2
Sprite.Animate (sprite, pics ((x div 2) mod 6), x, 100, false)
end for
Chapter 7 : Language Features 553
Sprite.Free (sprite)
Exported qualified.
Status
This means that you can only call the function by calling Sprite.Animate,
not by calling Animate.
Sprite.New, Sprite.SetPosition and Sprite.ChangePic.
See also
Win DOS Mac
Sprite.ChangePic
X TOOT Term
Syntax
Sprite.ChangePic ( spriteID, picID : int )
Description
Changes the picture associated with a sprite while maintaining the sprites
height and visibility status. A typical use Sprite.ChangePic would be to
animate a sprite that stays in position.
Example
Here is a program that t that loads six images from files Pic1.bmp through
Pic6.bmp and then displays them sequentially in the center of the screen.
var pics : array 0 .. 5 of int
var sprite: int
for i : 1 .. 6
pics (i – 1) := Pic.FileNew ("Pic" + intstr (i) + ".bmp")
if Error.Last not= 0 then
put "Error loading image: ", Error.LastMsg
return
end if
end for
figure := Sprite.New (pics (0))
Sprite.SetPosition (sprite, maxx div 2, maxy div 2, true)
Sprite.Show (sprite)
for i : 1 .. 100
Sprite.ChangePic (sprite, pics (i mod 6))
end for
Sprite.Free (sprite)
Status
Exported qualified.
This means that you can only call the function by calling Sprite.ChangePic,
not by calling ChangePic.
See also
Sprite.New.
554 Object Oriented Turing Reference Manual
Win DOS Mac
Sprite.Free
X TOOT Term
Syntax
Sprite.Free ( spriteID : int )
Description
Destroys the sprite and frees up the memory the sprite used. It is an error
to use the spriteID after the sprite has been freed.
Example
See Sprite.Animate for an example of Sprite.Free.
Status
Exported qualified.
This means that you can only call the function by calling Sprite.Free, not
by calling Free.
Sprite.New.
See also
Win DOS Mac
Sprite.Hide
X TOOT Term
Syntax
Sprite.Hide ( spriteID : int )
Description
Hides a previously visible sprite. Sprite.Hide has no effect if the sprite is
already invisible.
Example
The following program animates four balls on the screen. When the balls
are close to each other or to a wall, they appear, otherwise they are hidden.
var pic, sprite, x, y, dx, dy, radius : array 1 .. 6 of int
var visible : array 1 .. 6 of boolean
setscreen ("nocursor")
% Create all the sprites.
for i : 1 .. 6
radius (i) := Rand.Int (10, 25)
Draw.FillOval (25, 25, radius (i), radius (i), 8 + i)
Font.Draw (intstr (i), 20, 20, 0, black)
pic (i) := Pic.New (0, 0, 50, 50)
Draw.FillBox (0, 0, 50, 50, 0)
x (i) := Rand.Int (radius (i), maxx ‹ radius (i))
y (i) := Rand.Int (radius (i), maxy ‹ radius (i))
Chapter 7 : Language Features 555
dx (i) := Rand.Int (3, 3)
dy (i) := Rand.Int (3, 3)
sprite (i) := Sprite.New (pic (i))
Sprite.SetPosition (sprite (i), x (i), y (i), true)
Sprite.SetHeight (sprite (i), i)
visible (i) := false
end for
% Now move all the sprites around the screen.
loop
for i : 1 .. 6
if x (i) + dx (i) < radius (i) or
x (i) + dx (i) > maxx – radius (i) then
dx (i) := –dx (i)
end if
x (i) := x (i) + dx (i)
if y (i) + dy (i) < radius (i) or
y (i) + dx (i) > maxy – radius (i) then
dy (i) := –dy (i)
end if
y (i) := y (i) + dy (i)
end for
for i : 1 .. 6
var near : boolean := false
if (x (i) < 50) or (x (i) > maxx – 50) or
(y (i) < 50) or (y (i) > maxy – 50) then
near := true
end if
if not near then
for j : 1 .. 6
if i not= j then
if sqrt ( (x (i) ‹ x (j)) ** 2 +
(y (i) ‹ y (j)) ** 2) < 100 then
near := true
exit
end if
end if
end for
end if
if near and not visible (i) then
Sprite.Show (sprite (i))
visible (i) := true
elsif not near and visible (i) then
Sprite.Hide (sprite (i))
visible (i) := false
end if
Sprite.SetPosition (sprite (i), x (i), y (i), true)
end for
Time.Delay (40)
exit when hasch
end loop
for i : 1 .. 6
Sprite.Free (sprite (i))
end for
Status
Exported qualified.
556 Object Oriented Turing Reference Manual
This means that you can only call the function by calling Sprite.Hide, not
by calling Hide.
See also
Sprite.Show.
Win DOS Mac
Sprite.New
X TOOT Term
Syntax
Sprite.New ( picID : int ) : int
Description
Creates a new sprite from a picture specified by picID. The sprite starts
invisible and should be given a depth using Sprite.SetHeight and a
position, given Sprite.SetPosition before being made visible using
Sprite.Show. When you are finished using the sprite, the sprite should be
freed using Sprite.Free.
Sprites work best when they are of moderate size. Large sprites will cause
flashing when moved across the screen.
Anything that is is color 0 in the picture will not appear when the sprite is
drawn. In other words, color 0 is transparent.
Example
See Sprite.Animate for an example of Sprite.New.
Status
Exported qualified.
This means that you can only call the function by calling Sprite.New, not
by calling New.
See also
Sprite.SetHeight, Sprite.SetPosition, Sprite.Show and Sprite.Free.
Win DOS Mac
Sprite.SetHeight
X TOOT Term
Syntax
Sprite.SetHeight ( spriteID, newHeight : int )
Description
Sets the height of the sprite specified by spriteID to the value specified by
newHeight.
Chapter 7 : Language Features 557
The height of a sprite determines which sprite appears above another when
they overlap. The "higher" sprite (the one with the grater height) will
appear on top of the sprite with the lower height, even if the lower sprite is
drawn second.
The background (i.e. any non-sprite) is considered to be in height 0. Sprites
with a negative height will appear "behind" the background. Note that if
two sprites have the same height, the one drawn last will appear above the
first one.
Example
See Sprite.Hide for an example of Sprite.SetHeight.
Status
Exported qualified.
This means that you can only call the function by calling
Sprite.SetPosition, not by calling SetPosition.
See also
Sprite.New.
Win DOS Mac
Sprite.SetPosition
X TOOT Term
Syntax
Sprite.SetPosition ( spriteID, x, y : int,
centered : boolean )
Description
Moves the sprite specified by spriteID to the location specified by (x, y). If
centered is true, then the sprite is centered on (x, y). Otherwise (x, y)
specifies the lower-left corner of the sprite.
Example
See Sprite.Hide for an example of Sprite.SetPosition.
Status
Exported qualified.
This means that you can only call the function by calling
Sprite.SetPosition, not by calling SetPosition.
See also
Sprite.New.
558 Object Oriented Turing Reference Manual
Win DOS Mac
Sprite.Show
X TOOT Term
Syntax
Sprite.Show ( spriteID : int )
Description
Displays a previously hidden sprite. Sprite.Show has no effect if the sprite
is already visible.
Example
See Sprite.Hide for an example of Sprite.Show.
Status
Exported qualified.
This means that you can only call the function by calling Sprite.Show, not
by calling Show.
Sprite.Hide.
See also
sqrt
square root function
Syntax
sqrt ( r : real ) : real
Description
The sqrt function is used to find the square root of a number. For example,
sqrt (4) is 2.
Example
This program prints out the square roots of 1, 2, 3, … up to 100.
for i : 1 .. 100
put "Square root of ", i, " is ", sqrt ( i )
end for
Details
It is illegal to try to take the square root of a negative number. The result of
sqrt is always positive or zero.
The opposite of a square root is the square. For example, the square of x is
written is x**2.
See also
See also predefined unit Math.
Chapter 7 : Language Features 559
standardType
Syntax
A standardType is one of:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
int
real
string [ ( maximumLength ) ]
boolean
nat
% natural number
intn
% n-byte integer (n=1, 2, 4)
natn
% n-byte natural (n= 1, 2, 4)
realn % n-byte real (n=4, 8)
char
% single character
char (n)
% n characters
Description
The standard types can be used throughout a program. They should not be
included in an import list.
See also
int, real, string and boolean. See also nat, intn, natn, realn, char, char (n)
statement
Syntax
A statement is one of:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
assignmentStatement % variableReference := expn
openStatement
% open …
closeStatement
% close …
putStatement
% put …
getStatement
% get …
readStatement
% read …
writeStatement
% write …
seekStatement
% seek …
560 Object Oriented Turing Reference Manual
(i)
(j)
(k)
(l)
(m)
(n)
(o)
(p)
tellStatement
% tell …
forStatement
% for … end for
loopStatement
% loop … end loop
exit [ when trueFalseExpn ]
ifStatement
% if … end if
caseStatement
% case … end case
assert trueFalseExpn
begin
statementsAndDeclarations
end
(q)
procedureCall % procedureId [(parameters)]
(r)
return
(s)
result expn
(t)
new [ collectionId , ] pointerVariableReference
(u)
free [ collectionId , ] pointerVariableReference
(v)
tag unionVariableReference , expn
(w) forkStatement
(x)
signal variableReference
(y)
wait variableReference [ , expn ]
(z)
pause expn
(aa) quit [ guiltyParty ] [ : quitReason ]
unchecked
(cc) checked
(bb)
Description
A statement (or command) causes a particular action, for example, the
putStatement:
put "Hello"
outputs Hello. See the descriptions of the individual statements for
explanations of their actions. Each statement can optionally by followed by
a semicolon (;).
Example
width := 24
% Assignment statement
put "Hello world"% Put statement
exit when i = 100
% Exit statement
assert width < 320 % Assert statement
Details
You can use a result statement only in a function. You can use a return
statement only to terminate a procedure or the main program (but not to
terminate the initialization of a module). See also result and return.
Chapter 7 : Language Features 561
There are a number of predefined procedures, such as drawline, which are
not listed as statements above. These are considered procedure calls,
which is one form of statement.
statementsAndDeclarations
StatementsAndDeclarations are:
Syntax
{ statementOrDeclaration }
Description
StatementsAndDeclarations are a list of statements and declarations. For
example, a Turing program consists of a list of statements and declarations.
The body of a procedure is a list of statements and declarations.
Each statementOrDeclaration is one of:
(a) statement
(b)
declaration
See also statement and declaration.
This list of statements and declarations is a Turing program that outputs
Hello Frank.
Example
var name : string
name := "Frank"
put "Hello ", name
Str
Description
All
This unit contains the predefined constants for manipulating strings.
All routines in the Str module are exported unqualified.
Descriptions of all the subprograms in the unit can be found in Chapter 4,
Language Features.
Entry Points
index
Finds a specified string in another string.
length
Returns the length of a string.
562 Object Oriented Turing Reference Manual
repeat
Creates a string by repeating a specified string a number of
times.
Stream
Description
All
This unit contains the predefined subprograms that deal with I/O streams.
The basic I/O in Turing is done with I/O statements. However, extra
functions are all part of the Stream unit.
All routines in the Stream unit are exported qualified (and thus must be
prefaced with "Stream."), with the exception of eof which is part of the
language but conceptually part of this unit and is considered to be
exported unqualified.
Entry Points
eof*
Determines if the end of file has been reached.
Flush
Flushes a specified stream.
FlushAll
Flushes all open output streams.
* Part of the language, conceptually part of the Stream unit.
Stream.eof
All
Syntax
eof ( streamNumber : int ) : boolean
Description
The eof (end of file) function is used to determine if there is any more
input. It returns true when there are no more characters to be read. The
parameter and its parentheses are omitted when referring to the standard
input (usually this is the keyboard); otherwise the parameter specifies the
number of a stream. The stream number has been determined (in most
cases) by an open statement.
Example
This program reads and outputs all the lines in the file called "info".
var line : string
Chapter 7 : Language Features 563
var fileNumber : int
open : fileNumber, "info", get
loop
exit when eof (fileNumber)
get : fileNumber, line : *
put line
end loop
Details
See also the description of the get statement, which gives more examples of
the use of eof. See also the open and read statements.
When the input is from the keyboard, the user can signal end-of-file by
typing control-Z on a PC (or control-D on UNIX). If a program tests for eof
on the keyboard, and the user has not typed control-Z (or control-D) and
the user has typed no characters beyond those that have been read, the
program must wait until the next character is typed. Once this character is
typed, the program knows whether it is at the end of the input, and returns
the corresponding true or false value for eof.
Status
Part of the language and only conceptually part of the Stream unit.
This means that you can only call the function by calling eof, not by calling
Stream.eof.
Stream.Flush
All
Syntax
Stream.Flush ( streamNumber : int )
Description
The Stream.Flush procedure is used to flush any buffered output
associated with the streamNumber parameter.
Details
Turing automatically flushes any buffered output when a stream is closed.
Turing also automatically closes any open files when execution is
terminated.
Status
Exported qualified.
This means that you can only call the function by calling Stream.Flush, not
by calling Flush.
564 Object Oriented Turing Reference Manual
Stream.FlushAll
All
Syntax
Stream.FlushAll
Description
The Stream.FlushAll procedure is used to flush any buffered output in any
open file.
Details
Turing automatically flushes any buffered output when a stream is closed.
Turing also automatically closes any open files when execution is
terminated.
Status
Exported qualified.
This means that you can only call the function by calling Stream.FlushAll,
not by calling FlushAll.
string
comparison
A stringComparison is one of:
Syntax
(a)
(b)
(c)
(d)
(e)
(f)
Description
stringExpn = stringExpn
stringExpn not= stringExpn
stringExpn > stringExpn
stringExpn < stringExpn
stringExpn >= stringExpn
stringExpn <= stringExpn
Strings (stringExpns) can be compared for equality (= and not=) and for
ordering (>, <, >= and <=).
Example
var name : string := "Nancy"
var licenceNumber : string ( 6 )
licenceNumber := "175AJN"
Chapter 7 : Language Features 565
Two strings are considered to be equal (=) if they have the same length and
are made up, character by character, of the same characters. If they differ,
they are considered to be unequal (not=).
Details
Ordering among strings is essentially alphabetic order. String S is
considered to come before string T, that is S < T, if the two are identical up
to a certain position and after that position, either the next character of S
comes before the next character of T, or else there are no more characters in
S while T contains more characters.
S > T (S comes after T ) means the same thing as T < S. S >= T means the
same thing as S > T or S = T. S <= T means the same thing as S < T or S = T.
ASCII gives the ordering among individual characters. It specifies, among
other things, that letter capital L comes alphabetically before capital letter
M and similarly for small (lower case) letters.
On IBM mainframe computers, the EBCDIC specification of characters may
be used instead of ASCII.
string
type
A stringType is:
Syntax
string [ ( maximumLength ) ]
Description
Each variable whose type is a stringType can contain a sequence (a string )
of characters. The length of this sequence must not exceed the stringType's
maximum length.
Example
var name : string
name := "Nancy"
var licenceNumber : string ( 6 )
licenceNumber := "175AJN"
Details
Strings can be assigned and they can be compared for both equality and for
ordering. See also string comparison and assignment statement.
Strings can be catenated (joined together) using the + operator and
separated into substrings. See catenation and substring. String functions are
provided to find the length of a string, to find where one string appears
inside another, and to make repeated copies of a string all joined together.
See length, index, and repeat.
A string type written without a maximum length is limited to holding a
maximum of 255 characters.
566 Object Oriented Turing Reference Manual
The maximumLength of a string, if given as a part of the type, must be
known at compile time, and must be at least 1 and at most 255. The
maximum length of a string is given by upper, for example,
upper(licenceNumber) is 6. See also upper.
In the declaration of a string that is a var formal parameter of a procedure
or function, the maximumLength can be written as an asterisk (*). Here, the
maximum length is taken to be that of the corresponding actual parameter,
as in:
procedure deblank (var s : string (*) ).
The star can also be used when the parameter is an array of strings.
See also
explicitStringConstants for exact rules for writing string values such as
"Nancy". See also char(n) and char types.
strint
string-to-integer function
Syntax
strint ( s : string [ , base : int ] ) : int
Description
The strint function is used to convert a string to an integer. The integer is
equivalent to string s. The number base parameter is optional, for example,
strint ("–47") = –47. In Turing proper, the base is not allowed and is
assumed to be 10.
String s must consist of a possibly null sequence of blanks, then an optional
plus or minus sign, and finally a sequence of one or more digits. For
number bases larger than 10, the digits can include a, b, c … (alternately A,
B, C …) which represent the digit values 10, 11, 12 … The base, if given,
must be in the range 2 to 36 (36 because there are 10 base ten digits and 26
letters). For example, strint ("FF", 16) = 255.
The intstr function is the inverse of strint, so for any integer i,
strint ( intstr ( i ) ) = i.
See also
chr, ord, intstr and strnat functions.
Chapter 7 : Language Features 567
strintok
string-to-integer function
Syntax
strintok ( s : string [ , base : int ] ) : boolean
Description
The strintok function is used determine whether the strint function can be
used to convert the string to an integer without causing an error. If the
string can be successfully converted, then strintok returns true, otherwise
it returns false.
String s should consist of a possibly null sequence of blanks, then an
optional plus or minus sign, and finally a sequence of one or more digits.
For number bases larger than 10, the digits can include a, b, c …
(alternately A, B, C …) which represent the digit values 10, 11, 12 … If s is
correctly constructed, then strnatok will return true, otherwise it returns
false. The base, if given, must be in the range 2 to 36 (36 because there are
10 base ten digits and 26 letters). For example, strintok ("FF", 16) = true.
See also
strnat
strint function that does the actual conversion.
string to natural number function
Syntax
strnat ( s : string [ , base : int ] ) : nat
Description
The strnat function is used to convert a string to a natural number. The
natural number is equivalent to string s. The number base parameter is
optional, for example, strnat("47") = 47.
String s must consist of a possibly null sequence of blanks, then an optional
plus sign, and finally a sequence of one or more digits. For number bases
larger than 10, the digits can include a, b, c … (alternately A, B, C …) which
represent the digit values 10, 11, 12 … The base, if given, must be in the
range 2 to 36 (36 because there are 10 base ten digits and 26 letters). For
example, strnat("FF", 16) = 255.
The natstr function is the inverse of strnat, so for any natural number n,
strnat( natstr (n)) = n.
The strnat function is similar to strint, except that strnat handles values
that are larger than int values and does not handle negative values.
568 Object Oriented Turing Reference Manual
See also
the chr, ord, intstr and strint functions.
strnatok
string to natural number function
Syntax
strnatok ( s : string [ , base : int ] ) : boolean
Description
The strnatok function is used determine whether the strnat function can be
used to convert the string to a natural number without causing an error. If
the string can be successfully converted, then strnatok returns true,
otherwise it returns false.
String s should consist of a possibly null sequence of blanks, then an
optional plus sign, and finally a sequence of one or more digits. For
number bases larger than 10, the digits can include a, b, c … (alternately A,
B, C …) which represent the digit values 10, 11, 12 … If s is correctly
constructed, then strnatok will return true, otherwise it returns false. The
base, if given, must be in the range 2 to 36 (36 because there are 10 base ten
digits and 26 letters). For example, strnatok ("FF", 16) = true.
See also
strnat function that does the actual conversion.
strreal
string-to-real function
Syntax
strreal ( s : string ) : real
Description
The strreal function is used to convert a string to a real number. For
example, strreal ("2.5e1") will produce an approximation to the number
25.0.
String s must consist of a possibly null sequence of blanks, then an optional
plus or minus sign and finally an explicit unsigned real or integer constant.
The realstr, erealstr and frealstr functions approximate the inverse of
strreal, although round-off errors keep these from being exact inverses.
Chapter 7 : Language Features 569
See also
realstr, erealstr, frealstr, intstr and strint functions.
strrealok
string-to-real function
Syntax
strrealok ( s : string ) : boolean
Description
The strrealok function is used determine whether the strreal function can
be used to convert the string to a real number without causing an error. If
the string can be successfully converted, then strrealok returns true,
otherwise it returns false.
String s should consist of a possibly null sequence of blanks, then an
optional plus or minus sign and finally an explicit unsigned real or integer
constant. If it does so, then strrealok will return true, otherwise it returns
false.
See also
strreal function that does the actual conversion.
subprogramHeader
Syntax
A subprogramHeader is one of:
(a)
(b)
Description
procedure [ pervasive ] id
[ ([ paramDeclaration {, paramDeclaration} ])]
function [ pervasive ] id
[ ([ paramDeclaration {, paramDeclaration} ])]
[ id ] : typeSpec
A subprogram header is used to describe the interface to a subprogram.
Subprogram headers are used within other language features such as
subprogram types and external declarations.
570 Object Oriented Turing Reference Manual
Parameterless subprograms may use parentheses (with nothing between
them), as is required in the C programming language. These parentheses
can be used to disambiguate between the call to the subprogram
(parentheses present) and a reference the subprogram (parentheses
missing).
Suppose f is a parameterless subprogram declared without parentheses
and g is a parameterless subprogram declared with parentheses. Their
headers are:
procedure f
procedure g ()
In a program, f and g() are calls to these functions, while g is a reference to
(not a call to) the procedure. There is no way to write a reference to f. When
in doubt, use parentheses in the declaration, as in the case for g, so that
calls always have parentheses and references always do not. A reference to
a subprogram can be assigned to a subprogram variable. See subprogram
type.
Specify that t is the type of procedure with a var integer parameter and a
real parameter. See also subprogramType.
Example
type t : procedure q ( var j : int, y : real )
Details
The keyword pervasive can be inserted just after procedure or function.
When this is done, the subprogram is visible inside all subconstructs of the
subprogram's scope. Without pervasive, the subprogram is not visible
inside modules unless explicitly imported. Pervasive subprograms need
not be imported. You can abbreviate pervasive as an asterisk (*).
See also
pervasive.
subprogramType
A subprogramType is:
Syntax
subprogramHeader
Description
A variable or constant can contain a reference to a subprogram. The type of
the variable or constant is a subprogramType. See also subprogramHeader.
Example
In the following t is a subprogram type, and u is a variable of type t
initialized to refer to procedure rnd.
procedure rnd ( var i : int, x : real)
i := round ( x )
Chapter 7 : Language Features 571
end rnd
type t : procedure q ( var j : int, y : real )
var u : t := rnd
% Procedure variable u refers to rnd
…
var j : int
u ( j, 24.6 )
% Call procedure u referring to rnd
…
var v := u
% Subprogram variable v initialized to u
The name of the subprogram, for example q, and the parameters, for
example i and x, have no meaning in a subprogram type. They are present
only because of the form of subprogram headers.
Details
If v is a variable or constant that refers to a subprogram, v can be called,
compared for equality to other subprogram variables, assigned and passed
as a parameter. Variable v is not an integer, string or pointer and cannot
participate in their corresponding operations.
A reference to a subprogram, rather that the code of the subprogram, is
contained in a variable v whose type is a subprogram type. This implies
that addr (v) is the address of the reference to subprogram, rather than the
address of the subprogram. The address of the code is given by #v. See
cheat for an explanation of the # operator.
You cannot assign a reference to a subprogram exported from a class. This
restriction exists because these subprograms are meaningless without an
accompanying reference to an object.
Many potential uses of subprogram variables are better programmed using
classes and overriding exported subprograms. See class.
subrangeType
A subrangeType is:
Syntax
expn .. expn
Description
A subrange type defines a set of values, for example, the subrange 1 .. 4
consists of 1, 2, 3 and 4.
Example
var i : 1 .. 10
% i can be 1, 2 … up to 10
type xRange : 0 .. 319
% Define integer subrange
var pixels : array xRange of int
% Array elements are
% numbered 0, 1, … 319
for k : xRange
% k ranges from 0 to 319
572 Object Oriented Turing Reference Manual
pixels ( k ) := 0
end for
A subrange must contain at least one element. In other words, the second
expression (expn) must be at least as large as the first expression.
Details
The lower bound of a subrange must be known at compile time. The upper
bound is allowed to be a run time value only in one situation and that is
when it gives the upper bound of an array being declared in a variable
declaration, in other words when declaring a dynamic array.
Subranges are usually a subset of the integers, as in 1 .. 10. You can also
have subranges of enumerated types and characters (the char type).
You can apply lower and upper to subrange types.
substring
A substring is one of:
Syntax
(a)
)
(b)
Description
Example
of another string
stringReference ( leftPosition .. rightPosition
stringReference ( charPosition )
A substring selects a part of another string. In form (a) the substring starts
at the left position and runs to the right position. In form (b), the substring
is only a single character. Turing support substrings of char(n) values.
var word : string := "bring"
put word (2 .. 4)
% Outputs rin
put word (3)
% Outputs i
put word (2 .. *)
% Outputs ring; the star (*) means
% the end of the string.
put word (* - 2 .. * - 1 )
Details
% Outputs in
The leftmost possible position in a string is numbered 1. The last position in
a string can be written as an asterisk (*). For example, word (2 .. *) is
equivalent to word (2 .. length(word)).
Each of leftPosition, rightPosition, and charPosition must have one of these
forms:
(a) expn
(b) *
(c) * - expn
The exact rules for the allowed values of leftPosition and rightPosition are:
Chapter 7 : Language Features 573
(1) leftPosition must be at least 1,
(2) rightPosition must be at most length (stringReference), and
(3) the length of the selected substring must zero or more.
This specifically allows null substrings such as word (1, 0) in which
rightPosition is 0 and word (6, 5) in which leftPosition is one more that length
(stringReference).
Note that substrings are not assignable. For example, if s is a string, the
statement s (3) := "a" is illegal in Turing.
Turing supports substrings of char(n) values. See char(n). If a substring of
char(n) value t has two operands, as in t(2..77), the result type of this
operation is a string. If the substring has one operand, as in t(7), this
becomes, in effect, a subscript into an array of characters. The result is a
reference to a char, which can be assigned to or passed to a var parameter.
string, char, char(n), explicitStringConstant, explicitCharConstant, catenation
and length.
See also
succ
successor function
Syntax
succ ( expn )
Description
The succ function accepts an integer, character or an enumerated value and
returns the integer plus one, the next character, or the next value in the
enumeration. For example, succ (7) is 8.
Example
This part of a Turing program fills up array a with the enumerated values
green, yellow, red, green, yellow, red, etc.
type colors : enum ( green, yellow, red )
var a : array 1 .. 100 of colors
var c : colors := colors .green
for i : 1 .. 100
a ( i ) := c
if c = colors . red then
c := colors . green
else
c := succ ( c )
end if
end for
Details
You cannot apply succ to the last value of an enumeration.
See also
the pred, lower and upper functions.
574 Object Oriented Turing Reference Manual
Sys
Description
This unit contains the predefined subprograms that deal with the operating
system directly (getting the process id, getting run time arguments and
executing commands in the operating system, etc.).
All routines in the Sys unit are exported qualified (and thus must be
prefaced with "Sys.").
Entry Points
GetEnv
Gets a string associated with an environment variable.
GetPid
Gets the current process ID for Turing.
Exec
Executes a program using the operating system.
Nargs
Gets the number of run time arguments (exported
unqualified).
FetchArg
Gets a specified run time argument (exported unqualified).
Win DOS Mac
Sys.Exec
X TOOT Term
Syntax
Sys.Exec ( command : string ) : int
Description
The Sys.Exec function is used to execute the shell (operating system)
command, as if it were typed at the terminal. The function returns the exit
code from the command. In general, a return code of 0 (zero) means no
detected errors. However, the command could return any exit code. If
Sys.Exec returns the predefined constant sysExecFailed, then command was
not executed and the program should use Last.Error or Last.ErrorStr to
determine the reason for failure.
Example
This program creates a directory listing when run under DOS on an IBM
PC compatible computer. The same program will run under UNIX by
changing "dir" to "ls".
var retCode : int
retCode := Sys.Exec ( "dir")
if retCode = sysExecFailed then
Chapter 7 : Language Features 575
else
put "The Sys.Exec call failed"
put "Error: ", Error.LastMsg
put "The call returned with a code of: ", retCode
end if
Details
When the Sys.Exec procedure is used, the executing program remains in
memory while the command is executing, and once execution of the
command is finished, control returns to the original program.
Status
Exported qualified.
This means that you can only call the function by calling Sys.Exec, not by
calling Exec.
Sys.Nargs, Sys.FetchArg and Sys.GetEnv functions.
See also
Sys.FetchArg
All
Syntax
System.FetchArg ( i : int ) : string
Description
The Sys.FetchArg function is used to access the i-th argument that has
been passed to a program from the command line. For example, if the
program is run from the Turing environment using
:r file1 file2
then Sys.FetchArg (2) will return "file2". If a program called prog.x is run
under UNIX using this command:
prog.x file1 file2
the value of Sys.FetchArg(2) will similarly be "file2".
The Sys.Nargs function, which gives the number of arguments passed to
the program, is usually used together with the Sys.FetchArg function.
Parameter i passed to Sys.FetchArg must be in the range 0 .. Sys.Nargs.
The 0-th argument is the name of the running program.
Example
This program lists its own name and its arguments.
put "The name of this program is : ", Sys.FetchArg ( 0 )
for i : 1 .. Sys.Nargs
put "Argument ", i, " is ", Sys.FetchArg ( i )
end for
576 Object Oriented Turing Reference Manual
Exported qualified.
Status
This means that you can only call the function by calling Sys.FetchArg, not
by calling FetchArg.
Sys.Nargs
See also
Win DOS Mac
Sys.GetEnv
X TOOT Term
Syntax
Sys.GetEnv ( symbol : string ) : string
Description
The Sys.GetEnv function is used to access the environment string whose
name is symbol. These strings are determined by the shell (command
processor) or the program that caused your program to run. See also the
Sys.Nargs and Sys.FetchArg functions.
Example
On a PC, this retrieves the environment variable USERLEVEL and prints
extra instructions if USERLEVEL had been set to NOVICE. USERLEVEL
can be set to NOVICE with the command SET USERLEVEL = NOVICE in
the autoexec.bat file or in any batch file.
const userLevel : string
userLevel := Sys.GetEnv ("USERLEVEL")
if userLevel = "NOVICE" then
…
% put a set of instructions
end if
Status
Exported qualified.
This means that you can only call the function by calling Sys.GetEnv, not
by calling GetEnv.
Win DOS Mac
Sys.GetPid
Syntax
X TOOT Term
Sys.GetPid : int
Chapter 7 : Language Features 577
Description
The Sys.GetPid function is used to determine the I.D. (number) that
identifies the current operating system task (process). Beware that there are
processes, activated by the fork statement, that are independent of the
operating systems tasks.
Under UNIX, the number is used, for example, for creating a unique name
of a file.
Status
Exported qualified.
This means that you can only call the function by calling Sys.GetPid, not
by calling GetPid.
See also
Sys.Nargs, Sys.FetchArg and Sys.GetEnv.
Sys.Nargs
All
Syntax
Sys.Nargs : int
Description
The Sys.Nargs function is used to determine the number of arguments that
have been passed to a program from the command line. For example, if the
program is run from the Turing environment using
:r file1 file2
then Sys.Nargs will return 2. If a program called prog.x is run under UNIX
using this command:
prog.x file1 file2
the value of Sys.Nargs will similarly be 2.
The Sys.Nargs function is usually used together with the Sys.FetchArg
function to access the arguments that have been passed to the program.
Status
Exported qualified.
This means that you can only call the function by calling Sys.Nargs, not by
calling Nargs.
See also
Sys.FetchArg for an example of the use of Sys.Nargs.
578 Object Oriented Turing Reference Manual
Win DOS Mac
sysclock
millisecs used procedure
X TOOT Term
Syntax
sysclock ( var c : int )
Description
The sysclock statement is used on a multitasking system such as UNIX to
determine the amount of time that has been used by this program
(process). Variable c is assigned the number of central processor
milliseconds assigned to this program. This is of little use on a personal
computer, where sysclock returns the same value as clock.
Example
On a UNIX system, this program tells you how much time it has used.
var timeUsed : int
sysclock ( timeUsed )
put "This program has used ", timeUsed,
" milliseconds of CPU time"
delay, time, clock, wallclock and date statements.
61
See also predefined unit Time.
Win DOS Mac
system
statement
X TOOT Term
Syntax
system ( command : string, var ret : int )
Description
The system statement is used to execute the shell (operating system)
command, as if it were typed at the terminal. The return code is in ret. A
return code of 0 (zero) means no detected errors. A return code of 127
means the command processor could not be accessed. A return code of 126
means the command processor did not have room to run on the PC.
Example
This program creates a directory listing when run under DOS on an IBM
PC compatible computer. The same program will run under UNIX by
changing "dir" to "ls".
var success : int
system ( "dir", success )
if success not= 0 then
if success = 127 then
Chapter 7 : Language Features 579
put "Sorry, can't find 'command.com'"
elsif success = 126 then
put "Sorry, no room to run 'dir'"
else
put "Sorry, 'dir' did not work"
end if
end if
Details
When the system procedure is used, the executing program usually
remains in memory while the system command is executing, and once
execution of the system command is finished, control returns to the original
program. However, on the PC, there is variant of the system procedure
that allows "chaining". This means that when the system command is
executed, the originally running program is "thrown away" (i.e. removed
from memory). When the executed program terminates, one is returned to
DOS.
To chain another program, one prepends "chain:" to the start command.
i.e.
system ("chain:myprog.exe", retCode)
Note that this command is "hazardous". Specifically, if you call it from
Turing (as opposed to a program compiled with TComp) and you have not
saved your source file, you will lose it! Turing will be removed from
memory without any warning when the system procedure is executed.
Likewise any open files will be closed instantly. This means there is a
danger if all files were not properly closed before the system procedure
was called.
The "chain:" command is often used for starting menu programs, where the user selects a
program to run and doesn’t want Turing to remain in memory. It can also
be used with extraordinarily large Turing programs that can be split into
different parts. By using TComp and compiling each part separately, one
can have each program call the other and never have all parts in memory
at once.
Example
This program uses chaining to launch one of several possible programs
based on user choice. It gives an error if for some reason the system
command fails to work. It assumes that c:\chemistry.exe, c:\math.exe,
c:\english.exe and c:\history.exe already exist.
var choice, success : int
put "Enter the subject (1-4): "..
get choice
var command : string
case choice of
% Note the use of the double backslash in the file name. This
% is because the backslash is a special character in Turing (as
% in \t for tab and \n for a newline). To get a single backslash,
% one uses \\.
label 1 : command := "c:\\chemistry.exe"
label 2 : command := "c:\\math.exe"
label 3 : command := "c:\\english.exe"
label 4 : command := "c:\\history.exe"
label : put "Choice must be from 1-4."
assert false % Wasn’t a 1-4. Terminate.
580 Object Oriented Turing Reference Manual
end case
system ( "chain:" + command, success)
% If I reach this line, the system command failed and one should give
% an error message.
put "System called failed."
put "Program \"", command, "\" couldn’t be run."
assert false % Terminate the program
Here are the possible errors under PC-Turing
Details
-1
Not enough memory to load COMMAND.COM
-2
Not enough memory to run command
-3
Argument list greater than 128 bytes or environment info
is greater than 32k
-4
Couldn't find COMMAND.COM
-5
COMMAND.COM corrupt
-6
-noshell option is selected, the system procedure is
disallowed
nargs, fetcharg and getenv functions.
See also
See also predefined unit Sys.
tag
statement
A tagStatement is:
Syntax
tag unionVariableReference , expn
Description
A tag statement is a special-purpose assignment that is used for changing
the tag of a union variable.
Example
In this example, the tag field of union variable v is set to be passenger,
thereby activating the passenger field of v.
type vehicleInfo :
union kind : passenger .. recreational
label passenger :
cylinders : 1..16
label farm :
farmClass : string ( 10 )
label : % No fields for "otherwise" clause
end union
var v : vehicleInfo
…
tag v, passenger
% Activate passenger part
Chapter 7 : Language Features 581
A tag statement is the only way to modify the tag field of a union variable
(other than by assigning an entire union value to the union variable).
Details
You cannot access a particular set of fields of a union unless the tag is set
to match the corresponding label value.
union types.
See also
takepic
graphics procedure
Pixel graphics only
Syntax
takepic ( x1, y1, x2, y2 : int, var buffer : array 1 .. * of int )
Description
The takepic procedure is used to record the pixel values in a rectangle,
with left bottom and right corners of (x1, y1) and (x2, y2), in the buffer
array. This requires a sufficiently-large buffer (see sizepic ). The drawpic
procedure is used to make copies of the recorded rectangle on the screen.
Example
After drawing a happy face, this program copies the face to a new location.
setscreen ("graphics")
… draw happy face in the box (0,0) to (100,100) …
% Create buffer big enough to hold happy face
var face : array 1 .. sizepic ( 0, 0, 100, 100 ) of int
% Copy picture into the buffer, which is the face array
takepic ( 0, 0, 100, 100, face )
% Redraw the picture with its left bottom at (200,0)
drawpic ( 200, 0, face, 0 )
Details
The integer values that takepic places in the buffer can be read or written
(using the read and write statements). Unfortunately, if a value happens to
be the pattern used to represent the uninitialized value (the largest
negative number the hardware can represent) assignment (by:=) and put of
the individual integer values in the buffer will fail.
The screen should be in a "graphics" mode. See the setscreen procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
See also
sizepic and drawpic. See also setscreen, maxx, maxy, drawdot, drawline,
drawbox, and drawoval.
582 Object Oriented Turing Reference Manual
See also predefined unit Pic.
tell
file statement
An tellStatement is:
Syntax
tell : fileNumber , filePositionVar
Description
The tell statement sets filePositionVar, whose type must be int, to the
current offset in bytes from the beginning of the specified file. The
fileNumber must specify a file that is open with seek capability (or else a
program argument file that is implicitly opened). The tell statement is
useful for recording the file position of a certain piece of data for later
access using seek.
Example
This example shows how to use tell to record the location of a record in a
file. This location is later used by seek to allow the record to be read.
var employeeRecord :
record
name : string ( 30 )
pay : int
dept : 0 .. 9
end record
var fileNo : int
var location : int
open : fileNo, "payroll", write, seek
…
tell : fileNo, location
% Make note of this location
write : fileNo, employeeRecord % Write record at this location
…
seek : fileNo, location
% Go back to location
read : fileNo, employeeRecord
% Read the record
% that was previously written
See also
the read, write, open, close, seek, get and put statements.
Chapter 7 : Language Features 583
Text
Description
This unit contains the predefined subprograms that handle character (text)
output on the screen (i.e. output using put).
All routines in the Text unit are exported qualified (and thus must be
prefaced with "Text.") with the exception of maxrow and maxcol which are
exported unqualified.
Entry Points
Cls
Clears the screen to the text background color.
Color
Sets the text color used by put.
Colour
Sets the text color used by put.
ColorBack
Sets the text background color used by put.
ColourBack
Sets the text background color used by put.
Locate
Moves the cursor to the specified row and column.
LocateXY
Moves the cursor to the cursor location closest to a
specified pixel position.
maxcol
The number of columns on the screen (exported
unqualified).
maxrow
The number of rows on the screen (exported unqualified).
WhatRow
Returns the current cursor row.
WhatCol
Returns the current cursor column.
WhatColor
Returns the current text color.
WhatColour
Returns the current text color.
WhatColorBack
Returns the current text background color.
WhatColourBack
Returns the current text background color.
WhatChar
Returns the character, color and background color of a
character at a specified row and column.
584 Object Oriented Turing Reference Manual
Win DOS Mac
Text.Cls
X TOOT Term
Syntax
Text.Cls
Description
The Text.Cls (clear screen) procedure is used to blank the screen to the text
background color. The cursor is set to the top left (to row 1, column 1).
Details
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See View.Set
for details.
Status
Exported qualified.
This means that you can only call the function by calling Text.Cls, not by
calling Cls.
Win DOS Mac
Text.Color
X TOOT Term
Syntax
Text.Color ( Color : int )
Description
The Text.Color procedure is used to change the currently-active color. This
is the color of characters that are to be put on the screen. The alternate
spelling is Text.Colour.
Example
This program prints out the message "Bravo" three times, each in a
different color.
View.Set ( "graphics" )
for i : 1 .. 3
Text.Color ( i )
put "Bravo"
end for
Example
This program prints out a message. The color of each letter is different
from the preceding letter. For letter number i the color number is i mod
maxcolor + 1. This cycles repeatedly through all the available colors.
View.Set ( "screen" )
const message := "Happy New Year!!"
Chapter 7 : Language Features 585
for i : 1 .. length ( message )
Text.Color ( i mod maxcolor + 1 )
put message ( i ) ..
end for
In "screen" mode on the IBM PC, the color specified can actually range from
0 - 31. The upper 16 colors (16-31) are the same as the lower 16, except that
they blink.
Details
See View.Set for the number of colors available in the various "graphics"
modes.
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See View.Set
for details.
Exported qualified.
Status
This means that you can only call the function by calling Text.Color, not by
calling Color.
Text.ColorBack, Text.WhatColor, Text.WhatChar and View.maxcolor.
See also
Win DOS Mac
Text.ColorBack
X TOOT Term
Syntax
Text.ColorBack ( Color : int )
Description
The Text.ColorBack procedure is used to change the current text
background color. The alternate spelling is Text.ColourBack.
The Text.ColorBack procedure sets the text background color to the
specified color. This is the color that surrounds characters when they are
put onto the screen. On an IBM PC in "screen" mode, the color can be from 0
- 7. (You can not have the upper 8 colors as text background colors. On
UNIX dumb terminals, Text.ColorBack(1) turns on highlighting and
Text.ColorBack(0) turns it off. On other systems, this procedure may have
no effect.
Example
Since this program is in "screen" mode, changing the background color has
no immediately observable effect. When the message "Greetings" is output,
the background surrounding each letter will be red.
View.Set ( "screen" )
…
Text.ColorBack ( red )
put "Greetings"
586 Object Oriented Turing Reference Manual
Details
The screen should be in a "screen" or "graphics" mode. If the screen mode
has not been set, it will automatically be set to "screen" mode. See View.Set
for details
Status
Exported qualified.
This means that you can only call the function by calling Text.ColorBack ,
not by calling ColorBack.
Text.Color and Text.WhatColorBack.
See also
Win DOS Mac
Text.Locate
X TOOT Term
Syntax
Text.Locate ( row, column : int )
Description
The Text.Locate procedure is used to move the cursor so that the next
output from put will be at the given row and column. Row 1 is the top of
the screen and column 1 is the left side of the screen.
Example
This program outputs stars of random colors to random locations on the
screen. The variable colr is purposely spelled differently from the word
color to avoid the procedure of that name (used to set the color of output).
The row number is purposely chosen so that it is one less than maxrow.
This avoids the scrolling of the screen which occurs when a character is
placed in the last column of the last row.
View.Set ("screen")
var row, column, colr : int
loop
row := Rand.Int (1, maxrow)
column := Rand.Int (1, maxcol)
colr := Rand.Int (0, maxcolor)
Text.Color (colr)
Text.Locate (row, column )
put "*" .. % Use dot-dot to avoid clearing end of line
end loop
Details
The Text.Locate procedure is used to locate the next output based on row
and column positions. See also the Text.LocateXY procedure which is used
to locate the output based x and y positions, where x=0, y=0 is the left
bottom of the screen.
The screen should be in a "screen" or "graphics" mode. See the View.Set
procedure for details. If the screen is not in one of these modes, it will
automatically be set to"screen" mode.
Chapter 7 : Language Features 587
Exported qualified.
Status
This means that you can only call the function by calling Text.Locate , not
by calling Locate.
View.Set and Draw.Dot.
See also
Win DOS Mac
Text.LocateXY
X TOOT Term
Syntax
Text.LocateXY ( x , y : int )
Description
The Text.LocateXY procedure is used to move the cursor so that the next
output from put will be at approximately (x, y). The exact location may be
somewhat to the left of x and below y to force alignment to a character
boundary.
Example
This program outputs Hello starting at approximately (100, 50) on the
screen.
View.Set ("graphics")
Text.LocateXY ( 100, 50 )
put "Hello"
Details
The Text.LocateXY procedure is used to locate the next output based on x
and y positions, where the position x=0, y=0 is the left bottom of the screen.
See also the Text.Locate procedure which is used to locate the outputbased row and column positions, where row 1 is the top row and column 1
is the left column.
The screen should be in a "graphics" mode. See the View.Set procedure for
details. If the screen is not in a "graphics" mode, it will automatically be set
to "graphics" mode.
Status
Exported qualified.
This means that you can only call the function by calling Text.LocateXY ,
not by calling LocateXY.
See also
View.Set and Draw.Dot.
588 Object Oriented Turing Reference Manual
Win DOS Mac
Text.maxcol
X TOOT Term
Syntax
maxcol : int
Description
The maxcol function is used to determine the number of columns on the
screen.
Example
This program outputs the maximum column number.
put "Number of columns on the screen is ", maxrow
Details
For IBM PC compatibles as well as most UNIX dumb terminals, in "text" or
"screen" mode, maxcol = 80. For the default IBM PC compatible "graphics"
mode (CGA), maxcol = 40.
Status
Exported unqualified.
This means that you can call the function by calling maxcol or by calling
Text.maxcol.
See also
Text.Locate procedure for an example of the use of maxcol.
Win DOS Mac
Text.maxrow
X TOOT Term
Syntax
maxrow : int
Description
The maxrow function is used to determine the number of rows on the
screen.
Example
This program outputs the maximum row number.
put "Number of rows on the screen is ", maxrow
Details
For IBM PC compatibles, maxrow = 25. For many UNIX dumb terminals,
maxrow = 24.
Status
Exported unqualified.
This means that you can call the function by calling maxrow or by calling
Text.maxrow.
Chapter 7 : Language Features 589
Text.Locate procedure for an example of the use of maxrow.
See also
Win DOS Mac
Text.WhatChar
X TOOT Term
Syntax
Text.WhatChar (row, col : int, var ch : char,
var foreColor, backColor : int)
Description
The whattextchar function is used to determine the character and text
colours on the screen at the specified location. This function can only be
used in DOS and in "screen" mode.
Example
This program outputs a message and then changes the foreground color
(the color of the letters) of the message to color number 1 and the
background color (surrounding each letter) to color number 7. The actual
message (each letter) is not changed.
View.Set ( "screen" )
const message := "Happy New Year!!"
put message
…
var ch : char
var fore, back : int
for column : 1 .. length ( message )
locate ( 1, column )
Text.WhatChar (1, column, ch, fore, back)
Text.Color ( 1 )
% Color of letter
Text.ColorBack ( 7 ) % Color around letter
put ch ..
% Use same letter
end for
Details
The Text.WhatChar procedure is meaningful only in "screen" mode. In
"graphics" mode, the concept of text on the screen is replaced by the concept
of pixels on the screen.
Status
Exported qualified.
This means that you can only call the function by calling Text.WhatChar ,
not by calling WhatChar.
See also
View.Set which describes modes. See also Text.Color and Text.ColorBack.
590 Object Oriented Turing Reference Manual
Win DOS Mac
Text.WhatCol
X TOOT Term
Syntax
Text.WhatCol : int
Description
The Text.WhatCol function is used to determine the cursor position’s
column.
Example
This program outputs The current row is 5, the current column is 15.
Text.Locate ( 5, 10 )
put "12345"..
put "The current row is", Text.WhatRow
put "The current column is", Text.WhatCol
Details
The screen should be in a "screen" or "graphics" mode. Text.WhatCol
functions properly even if the cursor is invisible.
Status
Exported qualified.
This means that you can only call the function by calling Text.WhatCol ,
not by calling WhatCol .
the Text.WhatRow function, which is used to determine the cursor row.
See also the Text.Locate, Text.maxrow and Text.maxcol procedure.
See also
Win DOS Mac
Text.WhatColor
X TOOT Term
Syntax
Text.WhatColor : int
Description
The Text.WhatColor function is used to determine the current text
(foreground) color, ie., the color used for characters that are output using
put. The alternate spelling is Text.WhatColour.
Example
This program outputs the currently-active color number. The message is
also given in the currently-active color.
View.Set ("graphics")
…
put "This writing is in color number ", Text.WhatColor
Chapter 7 : Language Features 591
Details
The screen should be in a "screen" or "graphics" mode. See View.Set for
details.
Status
Exported qualified.
This means that you can only call the function by calling Text.WhatColor,
not by calling WhatColor.
the Text.Color procedure, which is used to set the color. See also
Text.ColorBack and Text.WhatColorBack.
See also
Win DOS Mac
Text.WhatColorBack
X TOOT Term
Syntax
Text.WhatColorBack : int
Description
The Text.WhatColorBack function is used to determine the current text
background color. The alternate spelling is whatcolourback.
Example
This program outputs the currently-active background color number. The
background color of the message is determined by this number.
View.Set ("screen")
…
put "The background of this writing"
put
"is in color number ", Text.WhatColorBack
Details
The screen should be in a "screen" or "graphics" mode. Beware that the
meaning of background color is different in these two modes. See
Text.ColorBack for details.
Status
Exported qualified.
This means that you can only call the function by calling
Text.WhatColorBack, not by calling WhatColorBack.
See also
Text.Color and Text.WhatColor.
592 Object Oriented Turing Reference Manual
Win DOS Mac
Text.WhatRow
X TOOT Term
Syntax
Text.WhatRow : int
Description
The Text.WhatRow function is used to determine the cursor position’s
row.
Example
This program outputs The current row is 5, the current column is 15.
Text.Locate ( 5, 10 )
put "12345"..
put "The current row is", Text.WhatRow
put "The current column is", Text.WhatCol
Details
The screen should be in a "screen" or "graphics" mode. Text.WhatRow
functions properly even if the cursor is invisible.
Status
Exported qualified.
This means that you can only call the function by calling Text.WhatRow,
not by calling WhatRow.
See also
the Text.WhatCol function, which is used to determine the cursor column.
See also the Text.Locate, Text.maxrow and Text.maxcol procedure.
Time
Description
All
This unit contains the predefined subprograms that handle anything to do
with time, either as a date or as a timer.
All routines in the Time unit are exported qualified (and thus must be
prefaced with "Time.").
Entry Points
Sec
Returns the number of seconds since
1/1/1970 00:00:00 GMT.
Date
Returns the current date and time as a string.
SecDate
Converts a number of seconds into a date / time string.
Chapter 7 : Language Features 593
DateSec
Converts a date / time string to a number of seconds.
SecParts
Converts the number of seconds since
1/1/1970 00:00:00 GMT into a day of month, month, year,
day of week, hour, minute and second integers.
PartsSec
Converts a day of month, month, year, hour, minute and
second integers into the number of seconds since
1/1/1970 00:00:00 GMT.
Elapsed
Returns the number of milliseconds elapsed since the
program started to run.
ElapsedCPU
Returns the number of milliseconds of CPU time elapsed
since the program started to run.
Delay
Sleeps for a specified number of milliseconds.
Time.Date
All
Syntax
Time.Date : string
Description
The Time.Date function returns the current date and time as a string. The
returned string in the format "dd mmm yy hh:mm:ss", where mmm is the first
3 characters of the month, e.g., "Apr". For example, if the date is Christmas
1989 at 9:02:37 in the morning, Time.Date will return "25 Dec 89 09:02:37".
Twenty-four hour time is used, so eleven thirty at night the same day
would return "25 Dec 89 23:30:00"
Example
This program greets you and tells you the date and time.
var theDateTime, theDate, theTime : string
theDateTime := Time.Date
theDate := theDateTime (1 .. 9)
theTime := theDateTime (11 .. *)
put "Greetings!! The date and time today is ", Time.Date
Details
Be warned that on some computers, such as IBM PC compatibles or Apple
Macintoshes, the date may not be set correctly in the operating system; in
that case, the Time.Date procedure will give incorrect results.
The string form of the date can be converted to a numeric form for
comparison purposes using the Time.DateSec function. The numeric form
can be converted to a string using the Time.SecDate function. The numeric
form of the time can be obtained using the Time.Sec function.
594 Object Oriented Turing Reference Manual
Exported qualified.
Status
This means that you can only call the function by calling Time.Date, not by
calling Date.
Time.Sec, Time.DateSec and Time.SecDate functions.
See also
Time.DateSec
All
Syntax
Time.DateSec (dateString : string) : int
Description
The Time.DateSec function is used to convert a date and time string into a
number, specifically, the number of seconds since 00:00:00 GMT Jan 1,
1970.
The function can also convert just the date ("dd mmm yy"), in which case it
returns the number of seconds since 00:00:00 GMT Jan 1, 1970 from
midnight of the entered day. It will also convert a time without the date
("hh:mm:ss"), in which case it returns the number of seconds that have
passed since midnight of that day.
If the format is incorrect or can't be interpreted, then Time.DateSec will
return –1 and Error.Last and Error.LastMsg will be set to the appropriate
error.
Example
This program gives the number of seconds since 00:00:00 GMT Jan 1, 1970.
var theDateTime, theDate, theTime : string
theDateTime := Time.Date
theDate := theDateTime (1 .. 9)
theTime := theDateTime (11 .. *)
put "The number of seconds from 00:00:00 GMT Jan 1, 1970",
"from midnight ", theDate, "is ", Time.DateSec (theDate)
put "The number of seconds from midnight to ", theTime "is ",
Time.DateSec (theTime)
put "The number of seconds from 00:00:00 GMT Jan 1, 1970",
"from ", theDateTime, "is ", Time.DateSec (theDateTime)
Status
Exported qualified.
This means that you can only call the function by calling Time.DateSec,
not by calling DateSec.
See also
Time.Sec, Time.Date and Time.SecDate functions.
Chapter 7 : Language Features 595
Time.Delay
All
Syntax
Time.Delay ( duration : int )
Description
The Time.Delay procedure is used to cause the program to pause for a
given time. The time duration is in milliseconds.
Example
This program prints the integers 1 to 10 with a second delay between each.
for i : 1 .. 10
put i
Time.Delay ( 1000 )
end for
Details
% Pause for 1 second
On IBM PC compatibles, the hardware resolution of duration is in units of
55 milliseconds. For example, Time.Delay (500) will delay the program by
about half a second, but may be off by as much as 55 milliseconds.
On Apple Macintoshes, the hardware resolution of duration is in units of
17 milliseconds (1/60th of a second). For example, Time.Delay (500) will
delay the program by about half a second, but may be off by as much as 17
milliseconds.
Status
Exported qualified.
This means that you can only call the function by calling Time.Delay, not
by calling Delay.
See also
Time.Elapsed and Time.ElapsedCPU.
Time.Elapsed
Syntax
Time.Elapsed : int
Description
The Time.Elapsed function returns the amount of time since a program
(process) started running. The number of milliseconds since the program
started running is returned.
Example
This program tells you how much time it has used.
596 Object Oriented Turing Reference Manual
All
var timeRunning : int
timeRunning := Time.Elapsed
put "This program has run ", timeRunning, " milliseconds"
On IBM PC compatibles, this is the total time since the Turing system was
started up. The hardware resolution of duration is in units of 55
milliseconds. For example, Time.Elapsed may be off by as much as 55
milliseconds.
Details
On Apple Macintoshes, this is the total time since the machine was turned
on. The hardware resolution of duration is in units of 17 milliseconds
(1/60-th of a second).
Exported qualified.
Status
This means that you can only call the function by calling Time.Elapsed, not
by calling Elapsed.
Time.ElapsedCPU and Time.Delay subprograms.
See also
Time.ElapsedCPU
All
Syntax
Time.ElapsedCPU : int
Description
The Time.ElapsedCPU function is used on a multitasking system such as
UNIX to determine the amount of time that has been used by this program
(process). The number of central processor milliseconds assigned to this
program is returned. This is of little use on a personal computer, where
Time.ElapsedCPU returns the same value as Time.Elapsed.
Example
On a UNIX system, this program tells you how much time it has used.
var timeUsed : int
timeUsed := Time.ElapsedCPU
put "This program has used ", timeUsed,
" milliseconds of CPU time"
Status
Exported qualified.
This means that you can only call the function by calling
Time.ElapsedCPU, not by calling ElapsedCPU.
See also
Time.Elapsed and Time.Delay subprograms.
Chapter 7 : Language Features 597
Time.PartsSec
All
Syntax
Time.PartsSec (year, month, day, hour, minute, second : int) : int
Description
The Time.PartsSec function is used to convert the numeric parts of a date
(specifically the year, month, day, hour, minute and second) into the
number of seconds since 00:00:00 GMT Jan 1, 1970 and the date specified
by the parts.
The function can also convert a time without a date (year, month and day
are all 0), in which case it returns the number of seconds that have passed
since midnight of the current day.
If the numbers don't make any sense or can't be interpreted, then
Time.PartsSec will return –1 and Error.Last and Error.LastMsg will be set
to the appropriate error.
This program gives the number of seconds between 00:00:00 GMT Jan 1,
1970 and 9:27 in the morning, Christmas Day, 1989).
Example
put "The number of seconds from 00:00:00 GMT Jan 1, 1970",
"is ", Time.PartsSec (1989, 12, 25, 9, 27, 0)
Status
Exported qualified.
This means that you can only call the function by calling Time.PartsSec,
not by calling PartsSec.
See also
Time.SecParts, Time.Date and Time.Sec functions.
Time.Sec
All
Syntax
Time.Sec : int
Description
The Time.Sec function returns the current date and time as a number. The
returned integer is the time in seconds since 00:00:00 GMT (Greenwich
Mean Time) January 1, 1970.
Example
This program tells you how many seconds since 1970.
put "The number of seconds since 1970 is ", Time.Sec
598 Object Oriented Turing Reference Manual
Be warned that on some computers, such as IBM PC compatibles or Apple
Macintoshes, the date may not be set correctly in the operating system; in
that case, the Time.Date procedure will give incorrect results.
Details
The string form of the date can be converted to a numeric form for
comparison purposes using the Time.DateSec function. The numeric form
can be converted to a string using the Time.SecDate function. The numeric
form of the time can be obtained using the Time.Sec function.
Exported qualified.
Status
This means that you can only call the function by calling Time.Sec, not by
calling Sec.
Time.Date, Time.DateSec and Time.SecDate functions.
See also
Time.SecDate
All
Syntax
Time.SecDate (timeInSecs : int) : string
Description
The Time.SecDate function is used to convert the number of seconds since
00:00:00 GMT Jan 1, 1970 into a date and time string.
If timeInSecs is incorrect or can't be interpreted, then Time.SecDate will
return the empty string and Error.Last and Error.LastMsg will be set to the
appropriate error.
Example
This program gives the number of seconds since 00:00:00 GMT Jan 1, 1970
and the date in string form.
var timeInSecs : int := Time.Sec
var theDateTime: string
theDateTime := Time.SecDate (timeInSecs)
put "The number of seconds since 1970 is ", timeInSecs
put "Greetings!! The date and time today is ", theDateTime
Status
Exported qualified.
This means that you can only call the function by calling Time.SecDate,
not by calling SecDate.
See also
Time.Sec, Time.Date and Time.DateSec functions.
Chapter 7 : Language Features 599
Time.SecParts
All
Syntax
Time.SecParts (sec : int, var year, month, day,
dayOfWeek, hour, minute, second : int)
Description
The Time.SecParts function is used to convert a single number form of the
time (the number of seconds since 00:00:00 GMT Jan 1, 1970) into a date
with numeric component parts.
The dayOfWeek parameter is 1 for Monday, 2 for Tuesday through 7 for
Sunday.
If the sec parameter doesn't make any sense or can't be interpreted, then
Time.PartsSec will set all the var parameters to –1 and Error.Last and
Error.LastMsg will be set to the appropriate error.
This program returns the current day of the week.
Example
var year, month, day, dayOfWeek, hour, minute, second : int
Time.SecParts (Time.Sec, year, month, day, dayOfWeek,
hour, minute, second )
var days : array 1 .. 7 of string (10) := init ("Monday", "Tuesday",
"Wednesday", "Thursday", "Friday", "Saturday", "Sunday")
put "The current day of the week is ", days (dayOfWeek)
Status
Exported qualified.
This means that you can only call the function by calling Time.SecParts ,
not by calling SecParts .
See also
time
Syntax
Time.PartsSec, Time.Date and Time.Sec functions.
time of day as a string procedure
time ( var t : string )
600 Object Oriented Turing Reference Manual
Description
The time statement is used to determine the current time of day. Variable t
is assigned a string in the format "hh:mm:ss". For example, if the time is two
minutes and 47 seconds after nine A.M., t will be set to "09:02:47". Twentyfour hour time is used. For example, eleven thirty P.M. gives the string
"23:30:00".
Example
This program greets you and tells you the time of day.
var timeOfDay : string
time ( timeOfDay )
put "Greetings!! The time is ", timeOfDay
Details
Be warned that on some computers such as IBM PC compatibles or Apple
Macintoshes, the time may not be set correctly in the operating system. In
this case, the time procedure will give incorrect results.
See also
delay, clock, sysclock, wallclock and date statements.
See also predefined unit Time.
token
Description
in input
A token is essentially a word, a number or a special symbol such as :=. In a
Turing program there are four kinds of tokens: keywords such as get,
identifiers such as incomeTax, operators and special symbols, such as + and
:=, and explicit constants, such as 1.5 and "Hello". Some keywords, such as
index, are reserved and cannot be used in programs to name variables,
procedures, etc.
A get statement, such as
get incomeTax
uses token-oriented input. This means that white space (blanks, tabs, etc.) is
skipped before reading the input item and after the item (up to the
beginning of the next line). See the get statement for details.
Example
In this example, the tokens are var, x, :, real, x, := and 9.84.
var x : real
x := 9.84
Chapter 7 : Language Features 601
true
boolean value (not false)
Syntax
true
Description
A boolean (true/false) variable can be either true or false (see boolean
type).
Example
var passed : boolean := true
var mark : int
for i : 1 .. 10
get mark
passed := passed and mark >= 60
end for
if passed = true then
put "You passed all ten subjects"
end if
Details
The line if passed=true then can be simplified to if passed then with no
change to the meaning of the program.
TypeConv
Description
All
This unit contains the predefined subprograms that convert between
different Turing standard types. There are also six routines that are part of
the language, rather than part of the unit, but are conceptually part of this
unit.
All routines in the TypeConv unit are exported unqualified.
Description of the routines in the TypeConv module can be found in this
chapter.
Entry Points
intreal
Converts an integer to a real.
intstr*
Converts an integer to a string.
natreal
Converts a natural number to a real.
602 Object Oriented Turing Reference Manual
ord*
natstr*
Converts a natural number to a string.
round
Converts a real to an integer (rounding).
floor
Converts a real to an integer (round down).
ceil
Converts a real to an integer (round up).
realstr
Converts a real to a string.
erealstr
Converts a real to a string (exponential notation).
frealstr
Converts a real to a string (no exponent).
strint*
Converts a string to an integer.
strintok*
Returns whether a string can legally be converted to an
integer.
strnat*
Converts a string to a natural number.
strnatok*
Returns whether a string can legally be converted to a
natural number.
strreal
Converts a string to a real.
strrealok
Returns whether a string can legally be converted to a real.
chr*
Returns the ASCII value of a specified string of length one.
Returns a string of length one with the ASCII value
specified.
* Part of the language, conceptually part of the TypeConv unit.
type
Syntax
A typeDeclaration is one of:
(a)
(b)
Description
declaration
type id : typeSpec
type id : forward
A type declaration gives a name to a type. This name can be used in place
of the type.
Chapter 7 : Language Features 603
Example
type nameType : string ( 30 )
type range : 0 .. 150
type entry :
record
name : nameType
age : int
end record
Details
The keyword pervasive can be inserted just after type. When this is done,
the type is visible inside all subconstructs of the type scope. Without
pervasive, the type is not visible inside modules, monitors and classes
unless explicitly imported. Pervasive types need not be imported. You can
abbreviate pervasive as an asterisk (*).
A forward type allows pointers to be declared to the type before the type is
resolved. To resolve a type, you must follow a forward with a declaration of
the same name and in the same scope. This type declaration must include a
typeSpec.
typeSpec
Syntax
type specification
A typeSpec (type specification) is one of:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
(l)
(m)
(n)
int
real
boolean
stringType
% Example: string ( 20 )
subrangeType
% Example: 1 .. 150
enumeratedType % Example: enum ( red, green, blue )
arrayType
% Example: array 1 .. 150 of real
setType
% Example: set of 1 .. 10
recordType
% Example: record … end record
unionType
% Example: union … end union
pointerType % Example: pointer to collectionVar
namedType
% Example: colorRange
nat
% natural number
intn
% n-byte integer (n=1, 2, 4)
604 Object Oriented Turing Reference Manual
(o)
(p)
(q)
(r)
(s)
Description
natn
% n-byte natural (n= 1, 2, 4)
realn
% n-byte real (n=4, 8)
char
% single character
char(n)
% n characters
subprogramType
A type specification determines the allowed values for a variable or
constant. For example, if variable x is an integer (its typeSpec is int), the
possible values for x are numbers such as -15, 0, 3 and 348207. If x is a real
number (its typeSpec is real), then its possible values include 7.8, -35.0, and
15e12. If x is a boolean, its possible values are true and false. If x is a
string, its possible values include Hello and Good-bye.
Example
var numberOfSides : int
var x, y : real
% The typeSpec here is 0 .. 150
type range : 0 .. 150
type entry :
% Here is a record typeSpec
record
name : string ( 25 )
age : range
end record
unchecked
compiler directive
Dangerous
Description
OOT adds the concept of "unchecked" to Turing. Here, you can request that
certain run time tests, which take place by default, can be eliminated. This
makes the program more efficient at the risk of unreliability.
Example
Declare p to be an unchecked pointer to an integer (see pointers for
details). Pointer p will be dangerous to use, because the run time system
will not check to see if it actually locates an integer, as opposed to arbitrary
computer memory. In other words, unchecked pointers are like C language
pointers.
var p : unchecked ^ int
Example
Declare C to be an unchecked collection of records of type R (see
collections for details). Pointers to C will be unchecked.
var C : unchecked collection of R
Example
Remove checking from the body of a loop.
for i : 1 .. 500
Chapter 7 : Language Features 605
unchecked
if a ( i ) = key then
exit
end if
end for
In the above example, the unchecked keyword requests that all checking,
in particular, array bounds checking for array a, are to be omitted. The
disabling lasts from the occurrence of the keyword unchecked to the end
of the surrounding construct, in this case, until end for. In a similar way,
the checked keyword will request that checking be re-enabled from the
occurence of checked to the end of the surrounding construct.
Details
In the current (1999) implementation, the use of unchecked to turn off
checking in a block of statements is ignored. In general, an implementation
may choose to ignore requests to disable checking.
union
type
A unionType is:
Syntax
union [ id ] : indexType of
label labelExpn { , labelExpn } :
{ id {, id } : typeSpec }
{ label labelExpn { , labelExpn } :
{ id {, id } : typeSpec } }
{ id {, id } : typeSpec } ]
[ label :
end union
Description
A union type (also called a variant record) is like a record in which there is
a run time choice among sets of accessible fields. This choice is made by the
tag statement, which deletes the current set of fields and activates a new
set.
Example
This union type keeps track of various information about a vehicle,
depending on the kind of vehicle.
const passenger := 0
const farm := 1
const recreational := 2
type vehicleInfo :
union kind : passenger .. recreational of
label passenger :
cylinders : 1..16
label farm :
farmClass :string ( 10 )
label : % No fields for "otherwise" clause
606 Object Oriented Turing Reference Manual
end union
var v : vehicleInfo
…
tag v, passenger % Activate passenger part v.cylinders := 6
Details
The optional identifier following the keyword union is the name of the tag
of the union type. If the identifier is omitted, the tag is still considered to
exist, although its value cannot be accessed. The tag must be of an index
type, for example 1..7. You should limit the range of this index type, as the
compiler may have a limit (at least 255) on the maximum range it can
handle.
Each labelExpn must be known at compile time and must lie within the range of the tag's
type. The fields, including the tag, of a union value are referenced using
the dot operator, as in v.cylinders and these can be used as variables or
constants. A field can be accessed only when the tag matches one of the
label expressions corresponding to the field. The tag can be changed by the
tag statement but it cannot be assigned to, passed to a var parameter, or
bound to using var.
In a union, id's of fields, including the tag, must be distinct. However, these
need not be distinct from identifiers outside the union. Unions can be
assigned as a whole (to unions of an equivalent type), but they cannot be
compared. A semicolon can optionally follow each typeSpec.
Any array contained in a union must have bounds that are known at
compile time.
The notation –> can be used to access union fields. For example, if p is a
pointer to vehicleRecord, p–>farmClass locates the farmClass field.
See also
unit
Syntax
pointer.
file containing module, monitor, or class
A compilationUnit is one of:
(a)
(b)
(c)
(d)
[ importList ] mainProgram
unit moduleDeclaration
unit monitorDeclaration
unit classDeclaration
Chapter 7 : Language Features 607
Description
A program can be divided up into units, each in a separate file. All of
these files except the main program begin with the keyword unit. The unit
contains the main program, a module, a monitor or a class.
Example
Here is stack module that is separated out into a file whose name is stack.
unit
% The keyword unit begins each separate file
module stack
export push, pop
var top : int := 0
var contents : array 1 .. 100 of int
procedure push ( i : int )
top += 1
contents ( top ) := i
end push
procedure pop ( i : int )
i := contents ( top )
top -= 1
end pop
end stack
The main program, which is in another file, gains access to the stack by
importing it. Here is the main program:
import var stack % Use the stack
var n : int
…
stack . push ( n )
…
stack . pop ( n )
Details
In this example, the keyword var in the import list is required because the
main program causes a change in the stack, by calling push and pop. The
import lists of units that are modules, monitors and classes are used to gain
access to further units.
If the stack were in a file with a different name, say stk.t, the import list would be rewritten
to use an in clause, as follows:
import var stack in "stk.t"
A mainProgram is simply a program. See program.
See also
module, monitor and class. See also export list, import list, inherit list,
implement list and implement by list.
608 Object Oriented Turing Reference Manual
unqualified
Description
export
When an identifier x is exported from a module, monitor or class M using
the keyword unqualified, it can be used outside of M without the
qualification "M.". In other words, outside of M, it can be referred to as
simply x. The keyword unqualified can be written in its short form as ~.
which is pronounced "not dot", as in:
export ~. x
export list.
See also
upper
bound
Syntax
upper ( reference [ , dimension ] ) : int
Description
The upper attribute is used to find the upper bound of an array, string,
char(n) or non-opaque subrange type. (See lower for finding the lower
bound.)
Example
In a procedure, see if the bound of array parameter a is large enough that
it can be subscripted by i. If it is large enough, it is set a(i) to zero.
procedure test ( var a : array 1 .. * of real )
if i <= upper ( a ) then
a ( i ) := 0.0
end if
end test
Details
In a similar way, if s is a string, its upper bound (not length!) is given by
upper (s). If an array has more than one dimension, as in var b : array 1..10,
1 .. 60 of int, you must specify the dimension. For example, upper (b, 2)
returns 60.
Chapter 7 : Language Features 609
var
declaration
A variableDeclaration is one of:
Syntax
(a) var id { ,id } [ :typeSpec]
[:=initializingValue]
(b) collectionDeclaration
Description
Example
A variable declaration creates a new variable (or variables). Only form (a)
will be explained here. See collectionDeclaration for explanation of form (b).
The typeSpec of form (a) can be omitted only if the initializing value is
present.
var j, k : int := 1
% j and k are assigned value 1
var t := "Sample"
% The type of t is string
var v : array 1 .. 3 of string ( 6 ) :=
init ( "George", "Fred", "Alice" )
Details
The initializing value, if present, must be an expression or else a list of
items separated by commas inside init ( … ). The syntax of initializingValue
is one of:
(a)
expn
(b)
init ( initializingValue {, initializingValue } )
Each init ( … ) corresponds to an array, record or union value that is being
initialized. These must be nested for initialization of nested types.
If the typeSpec is omitted, the variable's type is taken to be the (root) type of
the initializing expression, for example, int or string. The typeSpec cannot
be omitted for dynamic arrays or when the initializing value is of the form
init ( … ). The values inside init ( … ) must be known at compile time.
The keyword pervasive can be inserted just after var. When this is done,
the variable is visible inside all subconstructs of the variable's scope.
Without pervasive, the variable is not visible inside modules unless
explicitly imported. Pervasive variables need not be imported. You can
abbreviate pervasive as an asterisk (*).
OOT extends Turing in the following way. OOT changes form (a) to allow
the optional use of the register keyword to request that the variable be
placed in a machine register. The OOT syntax for form (a) is actually:
var [pervasive] [register] id { , id } [ : typeSpec ] [ := initializingValue
]
In the current (1994) OOT implementation, programs are run interpretively
using pseudo-code, which has no machine registers, and the register
keyword is ignored. See register for restrictions on the use of register
variables.
610 Object Oriented Turing Reference Manual
collection, bind, procedure and function declarations, parameter
declarations, export lists and import lists for other uses of the keyword var.
See also
variableReference
use of a variable
A variableReference is:
Syntax
variableId { componentSelector }
Description
In a Turing program, a variable is declared and given a name (variableId)
and then used. Each use is called a variable reference.
If the variable is an array, collection, record or union, its parts (components)
can be selected using subscripts and field names (using componentSelectors).
The form of a componentSelector is one of:\
(a)
( expn {, expn} )
(b)
. fieldId
Form (a) is used for subscripting (indexing) arrays and collections. The
number of array subscripts must be the same as in the array's declaration.
A collection has exactly one subscript, which must be a pointer to the
collection. Form (b) is used for selecting a field of a record or union.
Example
Following the declarations of k, a and r, each of k, a (k) and r.name are
variable references.
var k : int
var a : array 1 .. 100 of real
var r :
record
name : string ( 20 )
phone : string ( 8 )
end record
…
k := 5
a ( k ) := 3.14159
r . name := "Steve Cook"
Details
A variable reference can contain more than one component selector, for
example, when the variable is an array of records. For an example, see the
record type. See also constantReference and var declaration.
Chapter 7 : Language Features 611
View
Description
This unit contains the predefined subprograms that deal with the current
output surface. This can be a screen (in the DOS version of OOT) or a
window (in the X Windows, MS Windows or Macintosh version).
All routines in the View unit are exported qualified (and thus must be
prefaced with "View.") with the exception of maxx, maxy, maxcolor and
maxcolour which are exported unqualified.
Entry Points
maxx
Returns the maximum x coordinate (width – 1) (exported
unqualified).
maxy
Returns the maximum y coordinate (height – 1) (exported
unqualified).
maxcolor
Returns the maximum color number (# colors – 1)
(exported unqualified).
maxcolour
Returns the maximum color number (# colors – 1)
(exported unqualified).
Set
Changes the configuration of the output surface.
ClipSet
Clips output to a specified rectangle.
ClipAdd
Adds another specified rectangle to the clipping region.
ClipOff
Stops all clipping.
WhatDotColor
Gets the color of the pixel at a specified location.
WhatDotColour
Gets the color of the pixel at a specified location.
GetActivePage Returns the currently active graphics page.
SetActivePage
Sets the currently active graphics page.
GetVisiblePage
Returns the currently visible graphics page.
SetVisiblePage Sets the currently visible graphics page.
612 Object Oriented Turing Reference Manual
Win DOS Mac
View.ClipAdd
X TOOT Term
Syntax
View.ClipAdd (x1, y1, x2, y2 : int)
Description
The View.ClipAdd procedure adds another rectangle specified by (x1, y1)
- (x2, y2) to the clipping region. This only works on systems that support
complex clipping regions. If no clipping region has been specified, then the
rectangle becomes the complete clipping region.
A clipping region is the region that the output will appear in. If the
rectangle is specified as the clipping region, any drawing done outside the
rectangle will not appear.
To set the initial clipping, or remove the old region and replace it with a
new one, use View.ClipSet. To set the clipping region back to the entire
screen or window, use View.ClipOff.
These commands only work in "graphics" mode.
This program sets the clipping region to five rectangles and then draws
random circles. The circles will only appear (or partially appear) in the
rectangles.
Example
const maxx13 : int := maxx div 3
const maxx23 : int := 2 * maxx div 3
const maxy13 : int := maxy div 3
const maxy23 : int := 2 * maxy div 3
View.ClipSet (0, 0, maxx13, maxy13)
View.ClipAdd (maxx23, 0, maxx, maxx13)
View.ClipAdd (maxx13, maxy13, maxx23, maxy23)
View.ClipAdd (0, maxy23, maxx13, maxy)
View.ClipAdd (maxx23, maxy23, maxx, maxy)
% Draw the random ovals in the box
var x, y, clr : int
loop
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
clr := Rand.Int (0, maxcolor) % Random color
Draw.FillOval (x, y, 30, 30, clr)
end loop
Status
Exported qualified.
This means that you can only call the function by calling View.ClipAdd,
not by calling ClipAdd.
See also
View.ClipSet and View.ClipOff functions.
Chapter 7 : Language Features 613
Win DOS Mac
View.ClipOff
X TOOT Term
Syntax
View.ClipOff
Description
The View.ClipOff procedure turns off clipping. This means that any
drawing commands can appear on the entire drawing surface (the screen
or the window, depending on the system).
These commands only work in "graphics" mode.
Exported qualified.
Status
This means that you can only call the function by calling View.ClipOff, not
by calling ClipOff.
See also
View.ClipAdd and View.ClipSet functions.
Win DOS Mac
View.ClipSet
X TOOT Term
Syntax
View.ClipSet (x1, y1, x2, y2 : int)
Description
The View.ClipSet procedure sets the clipping region to the rectangle
specified by (x1, y1) - (x2, y2). If a clipping region already exist, it is
replaced by the specified rectangle.
A clipping region is the region in which the output will appear. If the
rectangle is specified as the clipping region, any drawing done outside the
rectangle will not appear.
To set the initial clipping, or remove the old region and replace it with a
new one, use View.ClipSet. To set the clipping region back to the entire
screen or window, use View.ClipOff.
These commands only work in "graphics" mode.
Example
This program sets the clipping region to five rectangles and then draws
random circles. The circles will only appear (or partially appear) in the
rectangles.
const maxx13 : int := maxx div 3
const maxx23 : int := 2 * maxx div 3
const maxy13 : int := maxy div 3
614 Object Oriented Turing Reference Manual
const maxy23 : int := 2 * maxy div 3
View.ClipSet (0, 0, maxx13, maxy13)
View.ClipAdd (maxx23, 0, maxx, maxy13)
View.ClipAdd (maxx13, maxy13, maxx23, maxy23)
View.ClipAdd (0, maxy23, maxx13, maxy)
View.ClipAdd (maxx23, maxy23, maxx, maxy)
% Draw the random ovals in the box
var x, y, clr : int
loop
x := Rand.Int (0, maxx)
% Random x
% Random y
y := Rand.Int (0, maxy)
clr := Rand.Int (0, maxcolor) % Random color
Draw.FillOval (x, y, 30, 30, clr)
end loop
Status
Exported qualified.
This means that you can only call the function by calling View.ClipSet, not
by calling ClipSet.
See also
View.ClipAdd and View.ClipOff functions.
Win DOS Mac
View.GetActivePage
X TOOT Term
Syntax
View.GetActivePage : int
Description
In certain graphics modes under DOS OOT, it is possible to have multiple
graphics pages. This means it is possible to draw to one graphics page
while another is visible. You can use View.SetActivePage to specify which
graphics page is being drawn upon and use View.SetVisiblePage to
specify the graphics page that is to be visible.
By drawing on a different graphics page from the currently visible page, it
is possible to avoid the flicker that might occur while the image is being
composed, especially if it is composed of several different parts.
The general method is to make the second page the active page, draw the
image upon it, then make the second page visible. Then make the first page
active and draw upon it while the user is looking at the second graphics
page.
View.GetActivePage returns the currently active graphics page. The
default page is 1.
Chapter 7 : Language Features 615
Details
You must use Config.Display in order to determine the number of
graphics modes available. The cdMaxNumPages parameter will return the
number of graphics pages available.
At the time of writing, only “ega” graphics mode had more than one
graphics page. This may change with future releases. Consult the release
notes for any changes that may have occurred.
Status
Exported qualified.
This means that you can only call the function by calling
View.GetActivePage, not by calling GetActivePage.
See also
View.SetActivePage, View.GetVisiblePage and View.SetVisiblePage
for routines to switch active and visible graphics pages.
Win DOS Mac
View.GetVisiblePage
X TOOT Term
Syntax
View.GetVisiblePage : int
Description
In certain graphics modes under DOS OOT, it is possible to have multiple
graphics pages. This means it is possible to draw to one graphics page
while another is visible. You can use View.SetActivePage to specify which
graphics page is being drawn upon and use View.SetVisiblePage to
specify the graphics page that is to be visible.
By drawing on a different graphics page from the currently visible page, it
is possible to avoid the flicker that might occur while the image is being
composed, especially if it is composed of several different parts.
The general method is to make the second page the active page, draw the
image upon it, then make the second page visible. Then make the first page
active and draw upon it while the user is looking at the second graphics
page.
View.GetVisiblePage returns the currently visible graphics page. The
default page is 1.
Details
You must use Config.Display in order to determine the number of
graphics modes available. The cdMaxNumPages parameter will return the
number of graphics pages available.
At the time of writing, only “ega” graphics mode had more than one
graphics page. This may change with future releases. Consult the release
notes for any changes that may have occurred.
616 Object Oriented Turing Reference Manual
Exported qualified.
Status
This means that you can only call the function by calling
View.GetVisiblePage, not by calling GetVisiblePage.
View.SetVisiblePage, View.GetVisiblePage and View.SetVisiblePage
for routines to switch active and visible graphics pages.
See also
Win DOS Mac
View.maxcolor
X TOOT Term
Syntax
View.maxcolor : int
Description
The maxcolor function is used to determine the maximum color number
for the current mode of the screen. The alternate spelling is maxcolour.
Example
This program outputs the maximum color number.
setscreen ("graphics")
…
put "The maximum color number is ", View.maxcolor
Details
The screen should be in a "screen" or "graphics" mode. If it is not, it will
automatically be set to "screen" mode. See View.Set for details.
For IBM PC compatibles in "screen" mode, maxcolor = 15. For the default
IBM PC compatible "graphics" mode (VGA), maxcolor = 15.
Details
View.maxcolor is identical to RGB.maxcolor. It is placed here for
consistency with other screen information routines.
Status
Exported qualified.
This means that you can only call the function by calling View.maxcolor.
Note that RGB.maxcolor is exported unqualified, so that one can call
maxcolor.
See also
Draw.Dot for examples of the use of maxcolor. See the Text.Color
procedure which is used for setting the currently-active color.
Chapter 7 : Language Features 617
Win DOS Mac
View.maxx
X TOOT Term
Syntax
maxx : int
Description
The maxx function is used to determine the maximum value of x for the
current graphics mode.
Example
This program outputs the maximum x value.
setscreen ("graphics")
…
put "The maximum x value is ", maxx
The screen should be in a "graphics" mode. If it is not, it will automatically
be set to "graphics" mode. See setscreen for details.
Details
For the default IBM PC compatible graphics mode (CGA), maxx = 319.
Exported unqualified.
Status
This means that you can call the function by calling maxx or by calling
View.maxx.
Draw.Dot for an example of the use of maxx and for a diagram illustrating
x and y positions.
See also
Win DOS Mac
View.maxy
X TOOT Term
Syntax
maxy : int
Description
The maxy function is used to determine the maximum value of y for the
current graphics mode.
Example
This program outputs the maximum y value.
setscreen ("graphics")
…
put "The maximum y value is ", maxy
618 Object Oriented Turing Reference Manual
The screen should be in a "graphics" mode. If it is not, it will automatically
be set to "graphics" mode. See setscreen for details.
Details
For the default IBM PC compatible graphics mode (CGA), maxy = 199.
Exported unqualified.
Status
This means that you can call the function by calling maxy or by calling
View.maxy.
Draw.Dot for an example of the use of maxy and for a diagram illustrating
x and y positions.
See also
Win DOS Mac
View.Set
X TOOT Term
Syntax
View.Set ( s : string )
Example
Here are example uses of the View.Set procedure. In many cases, these
will appear as the first statement of the program. However, they can
appear any place in a program.
View.Set ("graphics")
View.Set ("graphics:ega")
View.Set ("screen")
View.Set ("nocursor")
View.Set ("noecho")
% Switch to graphics mode
% Switch to EGA graphics mode
% Switch to screen mode
% Turn off cursor
% Do not echo keystrokes
Description
The View.Set statement is used to change the mode of the screen, as well
as the way in which Turing does input and output. The parameter to
View.Set is a string, such as "graphics". The string contains one or more
options separated by commas, such as "text, noecho". On a window system
(WinOOT, MacOOT, X OOT), View.Set affects the active window. On a
screen (DOS OOT, TermOOT), it affects the screen.
Details
Many of the options to View.Set are specific to IBM PC compatible
computers and Apple Macintoshes. They have been adapted to other
systems such as the Icon and X Windows.
There are three screen modes, text, screen and graphics. Most systems do
not support all three modes.
System
WinOOT
DOS OOT
MacOOT
X OOT
TOOT
text
screen
X
X
graphics
X
X
X
X
(Default: 640x300)
(Default: screen)
(Default: 480x275)
(Default: 720x375)
(Default: text)
Chapter 7 : Language Features 619
TermOOT
X
(Default: screen)
text mode does not allow any character graphics whatsoever. Only put and
get are allowed.
screen mode allows character graphics commands such as Text.Locate and
Text.Color.
graphics mode allows character graphics and pixel graphics commands
such and Text.Locate and Draw.Box.
If a character graphics command is given while in text mode, the program
automatically switches to screen mode if available, otherwise it switches to
graphics mode. If a pixel graphics command is given while in text or
screen mode, the program automatically switches to graphics mode.
Where the options to View.Set are mutually exclusive, they are listed here with the default
underlined. Here are the options:
"cursor", "nocursor" - Causes the cursor to be shown (or hidden).
At the time of writing, there is no visible cursor in graphics mode under
DOS OOT or under MacOOT. This will be changed in later versions.
There is an optional suffix for "cursor" that determines the shape of the
cursor. In DOS OOT, the cursor is constructed out of horizontal lines
numbered 0, 1, 2, up to 7, with 0 being the top. The suffix gives the range of
lines to be used for the cursor, for example, "cursor:5;7" specifies a cursor
consisting of lines 5 through 7. In general, this form is
"cursor:startline;endline", where startline and endline are integer literals
such as 5 and 7. On the Apple Macintosh, it is possible to set the cursor size
from 0-10.
Under DOS OOT and MacOOT, you can also specify the cursor to be a
block or an underline by using "cursor:block" or "cursor:underline".
"echo", "noecho" - Causes (or suppresses) echoing of characters that are typed.
Echoing is commonly turned off in interactive programs to keep typed
characters from being echoed at inappropriate places on the screen.
"noxor", "xor" - noxor mode means that all drawing is done normally. In xor mode,
all pixel graphics are drawn XOR’ed on the background (with the
exception of the Pic routines, where the drawing mode is specified). The
most important property of an XOR’ed object is that it can be erased and
the background restored by XOR’ing the object on top of itself.
"visible", "invisible", "popup" - Causes the active window to become visible
(invisible or, for popup, invisible until input or output occurs in the
window). This applies only to systems with windows (WinOOT, X OOT
and MacOOT).
"title:<text>" - Causes the title of the active window to be set to <text>. This applies
only to systems with windows (WinOOT, X OOT and MacOOT).
620 Object Oriented Turing Reference Manual
"position:<x>;<y>" - Causes the position of the upper left corner of the active
window to be set to (x, y). This applies only to systems with windows
(WinOOT, X OOT and MacOOT).
"text", "screen", "graphics" - Sets mode to the given mode and always erases the
screen, even when already in the requested mode; "text" is the default
character output mode; "screen" is character graphics mode, allowing the
locate procedure; "graphics" is "pixel graphics" mode.
Under WinOOT and X OOT, sc">