User Manual
OPENRULES
®
Open Source Business
Decision Management System
Release 6.2.3
User Manual
OpenRules, Inc.
www.openrules.com
January-2013
OpenRules® User Manual
OpenRules, Inc.
Table of Contents
Introduction ................................................................................................................6
Brief History .......................................................................................................................6
OpenRules® Components ....................................................................................................7
Document Conventions.......................................................................................................7
Core Concepts .............................................................................................................8
Spreadsheet Organization and Management ..............................................................9
Workbooks, Worksheets, and Tables...................................................................................9
How OpenRules® Tables Are Recognized ............................................................................ 10
OpenRules® Rule Table Example ........................................................................................ 12
Business and Technical Views ........................................................................................... 13
Decision Modeling and Execution .............................................................................. 14
Starting with Decision ....................................................................................................... 15
Defining Decision Tables ................................................................................................... 18
Decision Table Execution Logic .................................................................................. 20
AND/OR Conditions .................................................................................................... 20
Decision Table Operators ........................................................................................... 21
Using Regular Expressions in Decision Table Conditions ........................................... 24
Conditions and Conclusions without Operators ........................................................ 24
Using Decision Variable Names inside Decision Table Cells ...................................... 25
Using Formulas inside Decision Tables ...................................................................... 26
Direct References to Decision Variables .................................................................... 27
Defining Business Glossary ................................................................................................ 28
Defining Test Data ............................................................................................................ 29
Connecting the Decisions with Business Objects ................................................................ 31
Decision Execution............................................................................................................ 32
Decision Analysis .............................................................................................................. 33
Decision Testing ......................................................................................................... 33
Decision Syntax Validation ......................................................................................... 34
Decision Execution Reports ........................................................................................ 35
Decision Tracing ......................................................................................................... 38
Rules Repository Search ............................................................................................. 39
Consistency Checking ................................................................................................. 40
Advanced Decision Tables ......................................................................................... 40
Specialized Conditions and Conclusions ............................................................................. 40
Specialized Decision Tables ............................................................................................... 41
DecisionTable1 ........................................................................................................... 42
2
OpenRules, Inc.
OpenRules® User Manual
DecisionTable2 ........................................................................................................... 43
Rule Tables ............................................................................................................... 44
Simple Rule Table ............................................................................................................. 44
How Rule Tables Are Organized ........................................................................................ 47
Separating Business and Technical Information ................................................................. 50
How Rule Tables Are Executed .......................................................................................... 53
Relationships between Rules inside Rule Tables ....................................................... 53
Multi-Hit Rule Tables .................................................................................................. 54
Rules Overrides in Multi-Hit Rule Tables ................................................................... 55
Single-Hit Rule Tables ................................................................................................. 57
Rule Sequences .......................................................................................................... 58
Relationships among Rule Tables............................................................................... 59
Simple AND / OR Conditions in Rule Tables ....................................................................... 60
Horizontal and Vertical Rule Tables ................................................................................... 61
Merging Cells.................................................................................................................... 61
Sub-Columns and Sub-Rows for Dynamic Arrays ................................................................ 62
Using Expressions inside Rule Tables ................................................................................. 63
Integer and Real Intervals .......................................................................................... 63
Comparing Integer and Real Numbers ....................................................................... 65
Using Comparison Operators inside Rule Tables ....................................................... 66
Comparing Dates ........................................................................................................ 67
Comparing Boolean Values ........................................................................................ 68
Representing String Domains ..................................................................................... 69
Representing Domains of Numbers ........................................................................... 70
Using Java Expressions ............................................................................................... 70
Expanding and Customizing Predefined Types .......................................................... 72
Performance Considerations...................................................................................... 72
Rule Templates ......................................................................................................... 72
Simple Rules Templates .................................................................................................... 73
Defining Default Rules within Templates ........................................................................... 74
Templates with Default Rules for Multi-Hit Tables .................................................... 74
Templates with Default Rules for Single-Hit Tables ................................................... 75
Partial Template Implementation ..................................................................................... 76
Templates with Optional Conditions and Actions............................................................... 78
Templates for the Default Decision Tables ......................................................................... 79
Decision Templates .................................................................................................... 82
Decision Table Templates .......................................................................................... 82
Customization............................................................................................................. 83
3
OpenRules, Inc.
OpenRules® User Manual
OpenRules® API ......................................................................................................... 84
OpenRulesEngine API........................................................................................................ 85
Engine Constructors ................................................................................................... 85
Engine Runs ................................................................................................................ 87
Undefined Methods ................................................................................................... 87
Accessing Password Protected Excel Files ................................................................. 89
Engine Attachments ................................................................................................... 89
Engine Version ............................................................................................................ 89
Dynamic Rules Updates.............................................................................................. 89
Decision API ..................................................................................................................... 90
Decision Example ....................................................................................................... 90
Decision Constructors ................................................................................................ 91
Decision Parameters .................................................................................................. 91
Decision Runs ............................................................................................................. 92
Executing Decision Methods From Excel ................................................................... 93
Decision Glossary ....................................................................................................... 94
Business Concepts and Decision Objects ................................................................... 95
Changing Decision Variables Types between Decision Runs ..................................... 96
Decision Execution Modes ......................................................................................... 97
JSR-94 Implementation ..................................................................................................... 97
Multi-Threading................................................................................................................ 98
Integration with Java and XML.................................................................................. 98
Java Classes ................................................................................................................ 98
XML Files................................................................................................................... 100
Data Modeling ........................................................................................................ 102
Datatype and Data Tables ............................................................................................... 103
How Datatype Tables Are Organized ............................................................................... 106
How Data Tables Are Organized ...................................................................................... 108
Predefined Datatypes ..................................................................................................... 110
Accessing Excel Data from Java - Dynamic Objects ........................................................... 112
How to Define Data for Aggregated Datatypes ................................................................ 113
Finding Data Elements Using Primary Keys ...................................................................... 114
Cross-References Between Data Tables ........................................................................... 114
OpenRules® Repository ............................................................................................ 116
Logical and Physical Repositories .................................................................................... 116
Hierarchies of Rule Workbooks ....................................................................................... 118
Included Workbooks ................................................................................................ 118
Include Path and Common Libraries of Rule Workbooks ........................................ 119
4
OpenRules, Inc.
OpenRules® User Manual
Using Regular Expressions in the Names of Included Files ...................................... 119
Imports from Java .................................................................................................... 120
Imports from XML .................................................................................................... 120
Parameterized Rule Repositories..................................................................................... 121
Rules Version Control ..................................................................................................... 122
Rules Authoring and Maintenance Tools ......................................................................... 123
Database Integration .............................................................................................. 124
External Rules ......................................................................................................... 124
OpenRules® Projects ................................................................................................ 125
Pre-Requisites ................................................................................................................ 125
Sample Projects .............................................................................................................. 125
Main Configuration Project ............................................................................................. 126
Supporting Libraries ................................................................................................. 126
Predefined Types and Templates ............................................................................. 127
Technical Support ................................................................................................... 127
5
OpenRules, Inc.
OpenRules® User Manual
INTRODUCTION
OpenRules® was developed in 2003 by OpenRules, Inc. as an open source
Business Rules Management System (BRMS) and since then has become one of
the most popular BRMS on the market. Over these years OpenRules® has been
naturally transformed in a Business Decision Management System (BDMS) with
proven records of delivering and maintaining reliable decision support software.
OpenRules® is a winner of several software awards for innovation and is used
worldwide by multi-billion dollar corporations, major banks, insurers, health
care providers, government agencies, online stores, universities, and many other
institutions.
Brief History
From the very beginning, OpenRules® was oriented to subject matter experts
(business analysts) allowing them to work in concert with software developers to
create, maintain, and efficiently execute business rules housed in enterpriseclass rules repositories. OpenRules® avoided the introduction of yet another “rule
language” as well as another proprietary rules management GUI. Instead,
OpenRules® relied on commonly used tools such as MS Excel, Google Docs and
Eclipse integrated with the standard Java. This approach enabled OpenRules®
users to create and maintain inter-related decision tables directly in Excel.
Initially each rules table included several additional rows, in which a software
developer could place Java snippets to specify the exact semantics of rule
conditions and actions.
In March of 2008, OpenRules® Release 5 introduced Rule Templates. Templates
allowed a business analyst to create hundreds and thousands of business rules
based on a small number of templates supported by software developers. Rule
templates minimized the use of Java snippets and hid them from business users.
Rule templates were a significant step in minimizing rule repositories and
clearly separating the roles of business analysts and software specialists in
maintaining the rules.
In March of 2011 OpenRules® introduced Release 6, which finally moved control
over business logic to business users. OpenRules® 6 effectively removed any
Java coding from rules representation allowing business analysts themselves to
specify their decisions and supporting decision tables directly and completely in
Excel. Business users can also create business glossaries and test cases in Excel
tables. They may then test the accuracy of execute their decisions without the
need for any coding at all.
Once a decision has been tested it can be easily incorporated into any Java or
.NET environment. This process may involve IT specialists but only to integrate
the business glossary with a specific business object model. The business logic
remains the complete prerogative of subject matter experts.
6
OpenRules, Inc.
OpenRules® User Manual
OpenRules® Components
OpenRules® offers the following decision management components:

Rule Repository for management of enterprise-level decision rules

Rule Engine for execution of decisions and different business rules

Rule Dialog for building rules-based Web questionnaires

Rule Learner for rules discovery and predictive analytics

Rule Solver for solving constraint satisfaction and optimization problems

Finite State Machines for event processing and “connecting the dots”.
Integration of these components with executable decisions has effectively
converted OpenRules® from a BRMS to a BDMS, Business Decision
Management System, oriented to “decision-centric” application development.
OpenRules, Inc. is a professional open source company that provides software,
product documentation and technical support and other services that are highly
praised by our customers. You may start learning about product with the
document “Getting Started” which describes how to install OpenRules® and
includes simple examples. Then you may look at a more complex example in the
tutorial “Calculating Tax Return”. This user manual covers the core OpenRules®
concepts in greater depth. Additional OpenRules® components are described in
separate user manuals: see Rule Learner, Rule Solver, and Rule Dialog.
Document Conventions
-
The regular Century Schoolbook font is used for descriptive information.
-
The italic Century Schoolbook font is used for notes and fragments
clarifying the text.
-
The Courier New font is used for code examples.
7
OpenRules® User Manual
OpenRules, Inc.
CORE CONCEPTS
OpenRules® is a BDMS, Business Decision Management System, oriented to
“decision-centric” application development.
OpenRules® utilizes the well-
established spreadsheet concepts of workbooks, worksheets, and tables to build
enterprise-level rule repositories. Each OpenRules® workbook is comprised of
one or more worksheets that can be used to separate information by types or
categories.
To create and edit rules and other tables presented in Excel-files you can use any
standard spreadsheet editor such as:

MS Excel™

OpenOffice™

Google Docs™
Google Docs™ is especially useful for collaborative rules management.
OpenRules® supports different types of spreadsheets that are defined by their
keywords. Here is the list of OpenRules® tables along with brief description of
each:
Table Type
(Keyword)
Decision
DecisionTable or DT
Glossary
DecisionObject
Comment
Defines a decision that may consist of multiple
sub-decisions associated with different decision
tables
This is a single-hit decision table that uses
multiple conditions on different defined on
variables to reach conclusions about the
decision variables
For each decision variable used in the decision
tables, the glossary defines related business
concepts, as well as related implementation
attributes and their possible domain
Associates business concepts specified in the
glossary with concrete objects defined outside
the decision (i.e. as Java objects or Excel Data
8
OpenRules® User Manual
OpenRules, Inc.
Rules
Datatype
Data
Variable
Environment
Method
DecisionTable1 or DT1
DecisionTable2 or DT2
Layout
tables)
Defines a decision table that includes Java
snippets that specify custom logic for
conditions and actions. Read more. Some Rules
tables may refer to templates that hide those
Java snippets.
Defines a new data type directly in Excel that
can be used for testing
Creates an array of test objects
Creates one test object
This table defines the structure of a rules
repository by listing all included workbooks,
XML files, and Java packages
Defines expressions using snippets of Java
code and known decision variables and objects
A multi-hit decision table that allows rule
overrides
A multi-hit decision table that like
DecisionTable2 executes all rules in top-down
order but results of the execution of previous
rules may affect the conditions of rules that
follow
A special table type used by OpenRules®
Forms and OpenRules® Dialog
The following section will provide a detailed description of these concepts.
SPREADSHEET ORGANIZATION AND MANAGEMENT
OpenRules® uses Excel spreadsheets to represent and maintain business rules,
web forms, and other information that can be organized using a tabular format.
Excel is the best tool to handle different tables and is a popular and widely used
tool among business analysts.
Workbooks, Worksheets, and Tables
OpenRules® utilizes commonly used concepts of workbooks and worksheets.
These can be represented and maintained in multiple Excel files. Each
OpenRules® workbook is comprised of one or more worksheets that can be used
to separate information by categories. Each worksheet, in turn, is comprised of
one or more tables. Decision tables are the most typical OpenRules® tables and
9
OpenRules, Inc.
OpenRules® User Manual
are used to represent business rules. Workbooks can include tables of different
types, each of which can support a different underlying logic.
How OpenRules® Tables Are Recognized
OpenRules® recognizes the tables inside Excel files using the following parsing
algorithm.
1. The OpenRules® parser splits spreadsheets into “parsed tables”. Each logical
table should be separated by at least one empty row or column at the start of
the table. Table parsing is performed from left to right and from top to
bottom. The first non-empty cell (i.e. cell with some text in it) that does not
belong to a previously parsed table becomes the top-left corner of a new
parsed table.
2. The parser determines the width/height of the table using non-empty cells as
its clues. Merged cells are important and are considered as one cell. If the
top-left cell of a table starts with a predefined keyword (see the table below),
then such a table is parsed into an OpenRules® table.
3. All other "tables," i.e. those that do not begin with a keyword are ignored and
may contain any information.
The list of all keywords was described above. OpenRules® can be extended with
more table types, each with their own keyword.
While not reflected in the table recognition algorithm, it is good practice to use a
black background with a white foreground for the very first row. All cells in this
row should be merged, so that the first row explicitly specifies the table width.
We call this row the "table signature". The text inside this row (consisting of
one or more merged cells) is the table signature that starts with a keyword. The
information after the keyword usually contains a unique table name and
additional information that depends on the table type.
If you want to put a table title before the signature row, use an empty row
between the title and the first row of the actual table. Do not forget to put an
10
OpenRules® User Manual
OpenRules, Inc.
empty row after the last table row. Here are examples of some typical tables
recognized by OpenRules®.
OpenRules® table with 3 columns and 2 rows:
Keyword <some text>
Something
Something
Something
Something
Something
Something
OpenRules® table with 3 columns and still 2 rows:
Keyword
Something
Something
Something
Something
Something
Something
Something
Something
OpenRules® table with 3 columns and 3 rows (empty initial cells are acceptable):
Keyword <some text>
Something
Something
Something
Something
Something
OpenRules® table with 3 columns and 2 rows (the empty 3rd row ends the table):
Keyword <some text>
Something
Something
Something
Something
Something
Something
Something
Something
Something
OpenRules® table with 2 columns and 2 rows (the empty cell in the 3rd column of
the title row results in the 4th columns being ignored. This also points out the
importance of merging cells in the title row):
Keyword
Something
Something
Something
Something
Something
Something
Something
Something
Something
Something
OpenRules® will not recognize this table (there is no empty row before the
signature row):
11
OpenRules® User Manual
OpenRules, Inc.
Table Title
Keyword <some text>
Something
Something
Something
Something
Something
Fonts and coloring schema are a matter of the table designer's taste. The
designer has at his/her disposal the entire presentation power of Excel (including
comments) to make the OpenRules® tables more self-explanatory.
OpenRules® Rule Table Example
Here is an example of a worksheet with two rules tables:
This workbook is comprised of three worksheets:
1. Worksheet "Decision Tables" - includes rule tables
2. Worksheet "Launcher" - includes a method that defines an order and
conditions under which rules will be executed
12
OpenRules® User Manual
OpenRules, Inc.
3. Worksheet "Environment" - defines the structure of a rules repository by
listing all included workbooks, XML files, and Java packages (if any).
The worksheet "Decision Tables" is comprised of two rule tables "defineGreeting"
and "defineSalutation". Rule tables are a traditional way to represent business
decision tables.
Rule tables are decision tables that usually describe
combinations of conditions and actions that should be taken when all of the
conditions have been satisfied. In the table "defineGreeting", the action "Set
Greeting" will be executed when an "hour," passed to this table as a parameter,
is between "Hour From" and "Hour To". In the table "defineSalutation", an action
"Set Salutation" will be executed when a customer's Gender and Marital Status
correspond to the proper row.
These tables start with signature rows that are determined by a keyword in the
first cell of the table. A table signature in general has the following format:
Keyword return-type table-name(type1 par1, type2 par2,..)
where table-name is a one-word function name and return-type, type1, and type
2 are types defined in the current OpenRules® configuration. For example, type
may be any basic Java type such as int, double, Date, or String.
The rule tables above are recognized by the keyword "Rules". All of the columns
have been merged into a single cell in the signature rows. Merging cells B3, C3,
and D3 specifies that table "defineGreeting" has 3 columns. A table includes all
those rows under its signature that contain non empty cells: in the example
above, an empty row 12 indicates the end of the table "defineGreeting".
Limitation. Avoid merging rule rows in the very first column (or in the very first
row for horizontal tables) - it may lead to invalid logic.
Business and Technical Views
OpenRules® tables such as “Rules” and “Data” may have two views:
[1] Business View
13
OpenRules, Inc.
OpenRules® User Manual
[2] Technical View
These two views are implemented using Excel's outline buttons [1] and [2] at the
top left corner of every worksheet - see the figure below. This figure represents a
business view - no technical details about the implementation are provided. For
example, from this view it is hard to tell for sure what greeting will be generated
at 11 o'clock: "Good Morning" or "Good Afternoon"? If you push the Technical
View button [2] (or the button "+" on the left), you will see the hidden rows with
the technical details of this rules table:
The technical view opens hidden rows 4-6 that contain the implementation
details. In particular, you can see that both "Hour From" and "Hour To" are
included in the definition of the time intervals. Different types of tables have
different technical views.
Note. Using Rules Templates you may completely split business and technical
information between different Excel tables. Decisions do not use technical views
at all because they do not require any coding and rely on predefined templates.
DECISION MODELING AND EXECUTION
OpenRules® methodological approach allows business analysts to develop their
executable decisions with underlying decision tables without (or only with a
limited) help from software developers. You may become familiar with the major
14
OpenRules® User Manual
OpenRules, Inc.
decision modeling concepts from simple examples provided in the document
“Getting Started” and several associated tutorials. First we will consider the
simple implementation options for decision modeling, and later on we will
describe more advanced OpenRules® concepts.
Starting with Decision
From the OpenRules® perspective a decision contains:
-
a set of decision variables that can take specific values from domains of
values
-
a set of decision rules (frequently expressed as decision tables) that
specify relationships between decision variables.
Some decision variables are known (decision input) and some of them are
unknown (decision output).
A decision may consist of other decisions (sub-
decisions). To execute a decision means to assign values to unknown decision
variables in such a way that satisfies the decision rules.
This approach
corresponds to the oncoming OMG standard known as “DMN”.
OpenRules® applies a top-down approach to decision modeling. This means that
we usually start with the definition of a Decision and not with rules or data.
Only then we will define decision tables, a glossary, and then data. Here is an
example of a Decision:
Here the decision “DeterminePatientTherapy” consists of four sub-decisions:
15
OpenRules® User Manual
OpenRules, Inc.

“Define Medication” that is implemented using a decision table
“DefineMedication”

“Define Creatinine Clearance” that is implemented using a decision table
“DefineCreatinineClearance”

“Define
Dosing”
that
is
implemented
using
a
decision
table
“DefineDosing”

“Check Drug Interaction” that is implemented using a decision table
“WarnAboutDrugInteraction”.
The table “Decision” has two columns “Decisions” and “Execute Decision Tables”
(those are not keywords and you can use any other titles for these columns). The
first column contains the names of all our sub-decisions - here we can use any
combinations of words as decision names. The second column contains exact
names of decision tables that implement these sub-decisions. The decision table
names cannot contain spaces or special characters (except for “underscore”) and
they should always be preceded by “:=”, which indicates the decision tables will
actually be executed by the OpenRules® engine.
OpenRules® allows you to use multiple (embedded) tables of the type “Decision”
to define more complex decisions. For example, a top-level decision, that defines
the main decision variable, may be defined through several sub-decisions about
related variables:
Decision DecisionMain
Decisions
Execute Rules / Sub-Decisions
Define Variable 1
:= DecisionTableVariable1()
Define Variable 2
:= DecisionTableVariable21()
Define Variable 2
:= DecisionTableVariable22()
Define Variable 3
:= DecisionVariable3(decision)
Define Variable 4
:= DecisionTableVariable4()
In order to Define Variable 2 it is necessary to execute two decision tables. Some
decisions, like "Define Variable 3", may require their own separate sub-decisions
such as described in the following table:
16
OpenRules® User Manual
OpenRules, Inc.
Decision DecisionVariable3
Decisions
Execute Rules
Define Variable 3.1
:= DecisionTableVariable31()
Define Variable 3.2
:= DecisionTableVariable32()
Define Variable 3.3
:= DecisionTableVariable33()
These tables can be kept in different files and can be considered as building
blocks for your decisions. This top-down approach with Decision Tables and
dependencies between them allows you to represent even quite complex decision
logic in an intuitive, easy to understand way.
Some decisions may have a more complex structure than the just described
sequence of sub-decisions. You can even use conditions inside decision tables. For
example, consider a situation when the first sub-decision validates your data and
a second sub-decision executes complex calculations but only if the preceding
validation was successful. Here is an example of such a decision from the tax
calculation tutorial:
Since this table “Decision Apply1040EZ” uses an optional column “Condition”, we
have to add a second row.
“ActionExecute”
are
defined
The keywords “Condition”, “ActionPrint”, and
in
the
standard
OpenRules®
template
“DecisionTemplate” – see the configuration file “DecisionTemplates.xls” in the
folder “openrules.config”. This table uses a decision variable “1040EZ Eligible”
that is defined by the first (unconditional) sub-decision “Validate”. We assume
that the decision “ValidateTaxReturn” should set this decision variable to TRUE
or FALSE. Then the second sub-decision “Calculate” will be executed only when
17
OpenRules, Inc.
OpenRules® User Manual
“1040EZ Eligible” is TRUE. When it is FALSE, this decision, “Apply1040EZ”,
will simply print “Do Not Calculate”. In our example the reason will be printed
by the decision table “ValidateTaxReturn”.
Note. You may use many conditions of the type “Condition” defined on different
decision variables. Similarly, you may use an optional condition “ConditionAny”
which instead of decision variables can use any formulas defined on any known
objects.
It is also possible to add custom actions using an optional action
“ActionAny” – see “DecisionTemplates.xls” in the folder “openrules.config”.
When you have completed defining all decision and sub-decisions, you may define
decision tables.
Defining Decision Tables
OpenRules® decision modeling approach utilizes the classical decision tables that
were in the heart of OpenRules® BDMS from its introduction in 2003.
OpenRules® uses the keyword “Rules” to represent different types of classical
decision tables. Rules tables rely on Java snippets to specify execution logic of
multiple conditions and actions. In 2011 OpenRules® version 6 introduced a
special type of decision tables with the keyword “DecisionTable” (or “DT”) that
do not need Java snippets and rely on the predefined business logic for its
conditions and conclusions defined on already known decision variables. For
example, let’s consider a very simple decision “DetermineCustomerGreeting”:
Decision DetermineCustomerGreeting
Decisions
Execute Rules
Define Greeting Word
:= DefineGreeting()
Define Salutation Word
:= DefineSalutation()
It refers to two decision tables. Here is an example of the first decision table:
18
OpenRules® User Manual
OpenRules, Inc.
DecisionTable DefineGreeting
Condition
Condition
Current Hour
Current Hour
>=
0
<=
11
>=
>=
>=
11
17
22
<=
<=
<=
17
22
24
Is
Conclusion
Greeting
Good Morning
Is
Is
Is
Good Afternoon
Good Evening
Good Night
Its first row contains a keyword “DecisionTable” and a unique name (no spaces
allowed). The second row uses keywords “Condition” and “Conclusion” to specify
the types of the decision table columns. The third row contains decision variables
expressed in plain English (spaces are allowed but the variable names should be
unique).
The columns of a decision table define conditions and conclusions using different
operators and operands appropriate to the decision variable specified in the
column headings. The rows of a decision table specify multiple rules. For
instance, in the above decision table “DefineGreeting” the second rule can be
read as:
“IF Current Hour is more than or equal to 11 AND Current Hour is less
than or equal to 17 THEN Greeting is Good Afternoon”.
Similarly, we may define the second decision table “DefineSalutation” that
determines a salutation word (it uses the keyword “DT” that is a synonym for
“DecisionTable”):
DT DefineSalutation
Condition
Condition
Gender
Is
Male
Is
Is
Female
Female
Conclusion
Marital Status
Is
Is
Married
Single
Salutation
Is
Mr.
Is
Is
Mrs.
Ms.
19
OpenRules® User Manual
OpenRules, Inc.
If some cells in the rule conditions are empty, it is assumed that this condition is
satisfied. A decision table may have no conditions but it always should contain
at least one conclusion.
Decision Table Execution Logic
OpenRules® executes all rules within DecisionTable in a top-down order. When
all conditions inside one rule (row) are satisfied the proper conclusion(s) from the
same row will be executed, and all other rules will be ignored.
Note. OpenRules® decision tables can be used to implement a methodological
approach described in the book “The Decision Model”. It relies on a special type
of decision tables called “Rule Families” that require that the order of rules
inside a decision table should not matter. It means that to comply with the
Decision Model principles, you should not rely on the default top-down rules
execution order of OpenRules® decision tables. Instead, you should design your
decision table (you even may use the keyword “RuleFamily” instead of “DT”) in
such a way that all rules are mutually exclusive and cover all possible
combinations of conditions. The advantage of this approach is that when you
decide to add new rules to your rule family you may place them in any rows
without jeopardizing the execution logic. However, in some cases, this approach
may lead to much more complex organization of rule families to compare with
the standard decision tables.
AND/OR Conditions
The conditions in a decision table are always connected by a logical operator
“AND”. When you need to use “OR”, you may add another rule (row) that is an
alternative to the previous rule(s).
However, some conditions may have a
decision variable defined as an array, and within such array-conditions “ORs”
are allowed. Consider for example the following, more complex decision table:
20
OpenRules® User Manual
OpenRules, Inc.
Here the decision variables “Customer Profile”, “Customer Product”, and
“Offered Products” are arrays of strings. In this case, the second rule can be read
as:
IF Customer Profile Is One Of New or Bronze or Silver
AND Customer Products Include Checking Account and
Overdraft Protection
AND Customer Products Do Not Include CD with 25 basis point
increase, Money Market Mutual Fund, and Credit Card
THEN Offered Products ARE CD with 25 basis point increase,
Money Market Mutual Fund, and Credit Card
Decision Table Operators
OpenRules® supports multiple ways to define operators within decision table
conditions and conclusions. When you use a text form of operators you can freely
use upper and lower cases and spaces. The following operators can be used inside
decision table conditions:
Operator
Synonyms
Is
=, ==
Is Not
>
!=, isnot, Is Not Equal To, Not, Not
Equal., Not Equal To
Is More, More, Is More Than, Is
Greater, Greater, Is Greater Than
Comment
When you use “=” or “==”
inside Excel write”’=” or”’==”
to avoid confusion with
Excel’s own formulas
Defines an inequality operator
For integers and real numbers,
and Dates
21
OpenRules® User Manual
OpenRules, Inc.
>=
<=
<
Is More Or Equal. Is More Or Equal
To, Is More Than Or Equal To, Is
Greater Or Equal To, Is Greater Than
Or Equal To
Is Less Or Equal, Is Less Than Or
Equal To, Is Less Than Or Equal To, Is
Smaller Or Equal To, Is Smaller Than
Or Equal To, Is Smaller Than Or Equal
To,
Is Less, Less, Is Less Than, Is Smaller,
Smaller, Is Smaller Than
Is True
Is False
Is Empty
Contains
Contain
Starts
With
Start with, Start
Match
Matches, Is Like, Like
No
Match
NotMatch, Does Not Match, Not Like,
Is Not Like, Different, Different From
Within
Inside, Inside Interval, Interval
Is One Of
Is One, Is One of Many, Is Among,
Among
Is Not
One Of
Is not among, Not among
Include
Include All
For integers and real numbers,
and Dates
For integers and real numbers,
and Dates
For integers and real numbers,
and Dates
For booleans
For booleans
A string is considered “empty”
if it is either “null” or contains
only zero or more whitespaces
For strings only, e.g. “House”
contains “use”. The
comparison is not casesensitive
For strings only, e.g. “House”
starts with “ho”. The
comparison is not casesensitive
Compares if the string matches
a regular expression
Compares if a string does not
match a regular expression
For integers and real numbers.
The interval can be defined as:
[0;9], (1;20], 5–10, between 5
and 10, more than 5 and less or
equals 10 – see more
For integer and real numbers,
and for strings. Checks if a
value is among elements of the
domain of values listed
through comma
For integer and real numbers,
and for strings. Checks if a
value is NOT among elements
of the domain of values listed
through comma
To compare two arrays.
Returns true when the first
array (decision variable)
include all elements of the
second array (value within
decision table cell)
22
OpenRules® User Manual
OpenRules, Inc.
Exclude
Do Not Include, Exclude One Of
Does Not
Include
Include Not All
Intersect
Intersect With, Intersects
To compare an array or value
with an array
To compare two arrays.
Returns true when the first
array (decision variable) does
not include all elements of the
second array (value within
decision table cell)
To compare an array with an
array
If the decision variables do not have an expected type for a particular operator,
the proper syntax error will be diagnosed.
The following operators can be used inside decision table conclusions:
Operator
Synonyms
Is
=, ==
Are
Add
Assign
Plus
+=
Assign
Minus
-=
Assign
Multiply
*=
Assign
Divide
/=
Comment
Assigns one value to the conclusion
decision variable. When you use “=” or
“==” inside Excel write”’=” or”’==” to
avoid confusion with Excel’s own
formulas.
Assigns one or more values listed
through commas to the conclusion
variable that is expected to be an array
Adds one or more values listed through
commas to the conclusion variable that is
expected to be an array
Takes the conclusion decision variable,
adds to it a value from the rule cell, and
saves the result in the same decision
variable.
Takes the conclusion decision variable,
subtracts from it a value from the rule
cell, and saves the result in the same
decision variable.
Takes the conclusion decision variable,
multiplies it by a value from the rule cell,
and saves the result in the same decision
variable.
Takes the conclusion decision variable,
divides it by a value from the rule cell,
and saves the result in the same decision
variable.
23
OpenRules® User Manual
OpenRules, Inc.
Using Regular Expressions in Decision Table Conditions
OpenRules® allows you to use standard regular expressions. Operators "Match"
and "No Match" (and their synonyms from the above table) allow you to match
the content of your text decision variables with custom patterns such as phone
number or Social Security Number (SSN). Here is an example of a decision table
that validates SSN:
DecisionTable testSSN
Condition
SSN
No
\d{3}-\d{2}-\d{4}
Match
Match
\d{3}-\d{2}-\d{4}
Message
Message
Invalid SSN
Valid SSN
The use of this decision table is described in the sample project
“DecisionHelloJava”.
Conditions and Conclusions without Operators
Sometimes the creation of special columns for operators seems unnecessary,
especially for the operators “Is” and “Within”. OpenRules® allows you to use a
simpler format as in this decision table:
DT DefineGreeting
If
Then
Current
Greeting
Hour
0-11
Good Morning
11-17
17-22
22-24
Good Afternoon
Good Evening
Good Night
As you can see, instead of keywords “Condition” and “Conclusion” we use the
keywords “If” and “Then” respectively.
While this decision table looks much
simpler in comparison with the functionally identical decision table defined
24
OpenRules, Inc.
OpenRules® User Manual
above, we need to make an implicit assumption that the lower and upper bounds
for the intervals “0-11”, “11-17”, etc. are included.
Using Decision Variable Names inside Decision Table Cells
When your decision table contains too many columns it may become too wide and
unmanageable. In practice large decision tables have many empty cells because
not all decision variables participate in all rule conditions even if the proper
columns are reserved or all rules. To make your decision table more compact,
OpenRules® allows you to move a variable name from the column title to the rule
cells. To do that, instead of the standard column’s structure with two subcolumns
you may use another column representation with 3 sub-columns:
This way you may replace a wide table with many condition columns like the one
below:
to a much more compact table that may look as follows:
25
OpenRules® User Manual
OpenRules, Inc.
You simply replace a column of the type “Condition” to the one that has the
standard type “ConditionVarOperValue”. Similarly, instead of a column of the
type “Conclusion” you may use a column of the type “ConclusionVarOperValue”
with 3 sub-columns that represent:
- Decision variable name
- Operator
- Value.
Using Formulas inside Decision Tables
OpenRules® allows you to use formulas in the rule cells instead of constants. The
formulas usually have the following format:
::= ( expression )
where an “expression” can be written using standard Java expressions. Here is
an example:
This decision table has only one conclusion that simply calculates a value for the
decision variable “Adjusted Gross Income” as a sum of values for the decision
variables “Wages”, “Taxable Interest”, and “Unemployment Compensation”. This
example also demonstrates how to gain access to different decision variables –
you may write getReal(“VARIABLE_NAME”) for real decision variables.
Similarly, you may use methods getInt(…), getBool(…), getDate(…), and
getString(…).
26
OpenRules, Inc.
OpenRules® User Manual
You may also put your formula in a specially defined Method and then refer to
this method from the decision table – observe how it is done in the following
example:
Here we defined a new method “taxableIncome()” that returns a real value using
the standard Java type “double”. Then we used this method inside both
conditions and one conclusion of this decision table.
Note. Actually the formula format ::= ( expression ) is a shortcut for a more
standard OpenRules® formula format := “” +( expression ) that also can be used
inside decision tables.
Direct References to Decision Variables
You may want to refer to values of some decision variables inside cells for
different tables. To do that, you may simply put a dollar sign (“$”) in front
of the variable name. For example, in the following table
the conclusion-column contains references $Vendor and $Provider to
the values of decision variables Vendor and Provider. The reference
$Vendor is similar to the formula ::= getString(“Vendor”). You may
also use similar references inside arrays. For example, to express a
27
OpenRules® User Manual
OpenRules, Inc.
condition that a Vendor should not be among providers, you may use the
operator “Is Not One Of” with an array “ABC, $Vendor, XYZ”.
Defining Business Glossary
While defining decision tables, we freely introduced different decision variables
assuming that they are somehow defined. The business glossary is a special
OpenRules® table that actually defines all decision variables. The Glossary table
has the following structure:
Glossary glossary
Variable
Business Concept
Attribute
Domain
The first column will simply list all of the decision variables using exactly the
same names that were used inside the decision tables.
The second column
associates different decision variables with the business concepts to which they
belong. Usually you want to keep decision variables that belong to the same
business concept together and merge all rows in the column “Business Concept”
that share the same concept. Here is an example of a glossary from the standard
OpenRules® example “DecisionLoan”:
28
OpenRules, Inc.
OpenRules® User Manual
All rows for the concepts such as “Customer” and “Request” are merged.
The third column “Attribute” contains “technical” names of the decision variables
– these names will be used to connect our decision variables with attributes of
objects used by the actual applications, for which a decision has been defined.
The application objects could be defined in Java, in Excel tables, in XML, etc.
The decision does not have to know about it: the only requirement is that the
attribute names should follow the usual naming convention for identifiers in
languages like Java: it basically means no spaces allowed. The last column,
“Domain”, is optional, but it can be useful to specify which values are allowed to
be used for different decision variables. Decision variable domains can be
specified using the naming convention for the intervals and domains described
below. The above glossary provides a few intuitive examples of such domains.
These domains can be used during the validation of a decision.
Defining Test Data
29
OpenRules, Inc.
OpenRules® User Manual
OpenRules® provides a convenient way to define test data for decisions directly
in Excel without the necessity of writing any Java code. A non-technical user
can define all business concepts in the Glossary table using Datatype tables. For
example, here is a Datatype table for the business concept “Customer” defined
above:
The first column defines the type of the attribute using standard Java types such
as “int”, “double”, “Boolean”, “String”, or “Date”. The second column contains the
same attribute names that were defined in the Glossary. To create an array of
objects of the type “Customer” we may use a special “Data” table like the one
below:
This table is too wide (and difficult to read), so we could actually transpose it to a
more convenient but equivalent format:
30
OpenRules, Inc.
OpenRules® User Manual
Now, whenever we need to reference the first customer we can refer to him as
customers[0]. Similarly, if you want to define a doubled monthly income for the
second custromer, “Mary K. Brown”, you may simply write
::= (customers[1].monthlyIncome * 2)
You can find many additional details about data modeling in this section.
Connecting the Decisions with Business Objects
To tell OpenRules® that we want to associate the object customers[0] with our
business concept “Customer” defined in the Glossary, we need to use a special
table “DecisionObject” that may look as follows:
Here we also associate other business concepts namely Request and Internal
with the proper business objects – see how they are defined in the standard
example “DecisionLoan”.
31
OpenRules, Inc.
OpenRules® User Manual
The above table connects a decision with test data defined by business users
directly in Excel. This allows the decision to be tested. However, after the
decision is tested, it will be integrated into a real application that may use
objects defined in Java, in XML, or in a database, etc. For example, if there are
instances of Java classes Customer and LoanRequest, they may be put in the
object “decision” that is used to execute the decision. In this case, the proper
table “decisionObjects” may look like:
It is important that Decision does not “know” about a particular object
implementation: the only requirement is that the attribute inside these objects
should have the same names as in the glossary.
Note. You cannot use the predefined function “decision()” within the table
“decisionObjects” because its content is be not defined yet. You need to use the
internal variable “decision” directly.
Decision Execution
OpenRules® provides a template for Java launchers that may be used to execute
different decisions. There are OpenRules® API classes OpenRulesEngine and
Decision. Here is an example of a decision launcher for the sample project
“DecisionLoan”:
32
OpenRules, Inc.
OpenRules® User Manual
Actually, it just creates an instance of the class Decision. It has only two
parameters:
1)
a path to the main Excel file “Decision.xls”
2)
a name of the main Decision inside this Excel file.
When you execute this Java launcher using the provided batch file “run.bat” or
execute it from your Eclipse IDE, it will produce output that may look like the
following:
*** Decision DetermineLoanPreQualificationResults ***
Decision has been initialized
Decision DetermineLoanPreQualificationResults: Calculate Internal
Variables
Conclusion: Total Debt Is 165600.0
Conclusion: Total Income Is 360000.0
Decision DetermineLoanPreQualificationResults: Validate Income
Conclusion: Income Validation Result Is SUFFICIENT
Decision DetermineLoanPreQualificationResults: Debt Research
Conclusion: Debt Research Result Is Low
Decision DetermineLoanPreQualificationResults: Summarize
Conclusion: Loan Qualification Result Is NOT QUALIFIED
ADDITIONAL DEBT RESEARCH IS NEEDED from DetermineLoanQualificationResult
*** OpenRules made a decision ***
This output shows all sub-decisions and conclusion results for the corresponding
decision tables.
Decision Analysis
Decision Testing
OpenRules® provides an ability to create a test harness comprised of an
executable set of different test cases. It is important that the same people who
design rules (usually business analysts) are able to design tests for them.
33
OpenRules® User Manual
OpenRules, Inc.
Usually they create test cases directly in Excel by specifying their own
data/object types and creating instances of test objects of these types. Read more
at the section Defining Test Data.
Decision Syntax Validation
OpenRules® allows you to validate your decision by checking that:
-
there are no syntax error in the organization of all decision tables
-
values inside decision variable cells correspond to the associated domains
defined in the glossary.
The validation template is described in the standard file
“DecisionTableValidateTemplates.xls”.
If you use the Eclipse Plugin, it will display…
OpenRules®
also
provides
a
special
plugin
for Eclipse
IDE, a
de-facto
standard project management tools for software developers within a Java-based
development environment. Eclipse is used for code editing, debugging, and
testing of rule projects within a single commonly known integrated development
environment. OpenRules® has been designed to catch as many errors as possible
in design-rime vs. run-time when it is too late. OpenRules® Plugin automatically
diagnoses errors in the Excel-files and displays the proper error messages inside
Eclipse views like at the picture below:
34
OpenRules® User Manual
OpenRules, Inc.
Eclipse Plugin diagnoses any errors in Excel-files before you even deploy or run
your OpenRules-based application. To make sure that Eclipse controls your
OpenRules® project, you have first to right-click to your project folder and "Add
OpenRules Nature". You always can similarly "Remove OpenRules Nature". To
be validated, your main xls-files should be placed into an Eclipse source folder
while all included files should be kept in regular (non-source) folders.
OpenRules® Plugin displays a diagnostic log with possible errors inside the
Eclipse Console view. The error messages include hyperlinks that will open the
proper Excel file with a cursor located in a cell where the error occurred.
Decision Execution Reports
OpenRules® provides an ability to generate decision execution reports in the
HTML-format. To generate an execution report, you should add the following
setting to the decision’s Java launcher:
decision.put("report", "On");
35
OpenRules, Inc.
OpenRules® User Manual
before calling decision.execute(). By default, execution reports are not
generated as they are needed mainly for decision analysis. Reports are
regenerated for every decision run.
During decision execution, OpenRules® automatically creates a sub-directory
“report” in your main project directory and generates a report inside this subdirectory. For every decision table, including single-hit, multi-hit, and rule
sequencing tables, OpenRules® generates a separate html-file with the name
Report<n>.<DecisionTableName>.html, where n is an execution order
number for this particular decision table. For example, for the sample project
“DecisionLoan” OpenRules® will generate the following files:
The first file contains a list of links to all executed decision tables:
36
OpenRules, Inc.
OpenRules® User Manual
Below are other generated files (one per decision table) with lists of rules (rows)
that were actually executed:
37
OpenRules® User Manual
OpenRules, Inc.
These reports help a rule designer to analyze which rules were actually executed
and in which order. The “Executed Rule #” corresponds to the sequential number
of a rule inside its decision table.
Note. Execution reports are intended to explain the behavior of certain decision
tables and are used mainly for analysis and not for production. If you turn on
report generation mode in a multi-threaded environment that shares the same
instance of OpenRulesEngine, the reports will be produced only for the first
thread.
Decision Tracing
OpenRules® relies on the standard Java logging facilities for the decision output.
They can be controlled by the standard file “log4j.properties” that by default
looks like below:
log4j.rootLogger=INFO, stdout
log4j.appender.stdout=org.apache.log4j.ConsoleAppender
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout.layout.ConversionPattern=%m%n
#log4j.logger.org.openl=DEBUG
You may replace INFO to DEBUG and uncomment the last line to see OpenRules
debugging information.
You may control how “talkative” your decision is by setting decision’s parameter
“Trace”. For example, if you add the following setting to the above Java launcher
decision.put("trace", "Off");
just before calling decision.execute(), then your output will be much more
compact:
38
OpenRules, Inc.
OpenRules® User Manual
*** Decision DetermineLoanPreQualificationResults ***
Decision DetermineLoanPreQualificationResults: Calculate Internal
Variables
Decision DetermineLoanPreQualificationResults: Validate Income
Decision DetermineLoanPreQualificationResults: Summarize
ADDITIONAL DEBT RESEARCH IS NEEDED from DetermineLoanQualificationResult
*** OpenRules made a decision ***
You may also change output by modifying the tracing details inside the proper
decision templates in the configuration files “DecisionTemplates.xls” and
“DecisionTableExecuteTemplates.xls”.
Rules Repository Search
To analyze rules within one Excel files you may effectively use familiar Search
and Replace features provided by Excel or Google Docs.
When you want to search across multiple Excel files and folders, you may use a
free while powerful tool called “IceTeaReplacer” that can be very useful for doing
search & replace in OpenRules repositories. The following options are available:

Perform search before replacing

Match whole word only

Ignore word case

Do backup before replace

Deselect files on which you don’t want to perform replace.
Here is an example of its graphical interface:
39
OpenRules® User Manual
OpenRules, Inc.
Consistency Checking
OpenRules® provides a special component Rule Solver™ that along with powerful
optimization features allow a user to check consistency of the decision models
and find possible conflicts within decision tables and across multiple decision
tables.
The
detail
description
of
the
product
can
be
found
at
http://openrules.com/pdf/RulesSolver.UserManual.pdf.
ADVANCED DECISION TABLES
In real-world project you may need more complex representations of rule sets
and the relationships between them than those allowed by the default decision
tables. OpenRules® allows you to use advanced decision tables and to define
your own rule sets with your own logic.
Specialized Conditions and Conclusions
The standard columns of the types “Condition” and “Conclusion” always have
two sub-columns: one for operators and another for values. OpenRules® allows
you to specify columns of the types “If” and “Then” that do not require subcolumns. Instead, they allow you to use operators or even natural language
expressions together with values to represent different intervals and domains of
40
OpenRules, Inc.
OpenRules® User Manual
values. Read about different ways to represent intervals and domains in this
section below.
Sometimes your conditions or actions are not related to a particular decision
variable and can be calculated using formulas. For example, a condition can be
defined based on combination of several decision variables, and you would not
want to artificially add an intermediate decision variable to your glossary in
order to accommodate each needed combination of existing decision variables. In
such a case, you may use a special type “ConditionAny” like in the example
below:
Here the word “Condition” does not represent any decision variable and instead
you may insert any text, i.e. “Compare Adjusted Gross Income with Dependent
Amount”. When your conclusion, does not set a value for a single decision
variable but rather does something that is expressed in the formulas within the
cells of this column, you should use a column of type “ActionAny”. It does not
have sub-columns because there is no need for an operator.
Note. There is also a column of type “Action” that is equivalent to type “Then”.
Specialized Decision Tables
Sometimes the default behavior of a DecisionTable (as single-hit rules tables) is
not sufficient. OpenRules® provide two additional types of decision tables
DecisionTable1 (or DT1) and DecisionTable2 (or DT2). While we recommend
avoiding these types of decision tables, in certain situations they provide a
convenient way around the limitations imposed by the standard DecisionTable.
41
OpenRules, Inc.
OpenRules® User Manual
DecisionTable1
Contrary to the standard DecisionTable that is implemented as a single-hit rules
table, decision tables of type “DecisionTable 1” are implemented as multi-hit
decision tables. “DecisionTable 1” supports the following rules execution logic:
1. All rules are evaluated and if their conditions are satisfied, they will be
marked as “to be executed”
2. All actions columns (of the types “Conclusion”, “Then”, “Action”,
“ActionAny”, or “Message”) for the “to be executed” rules will be executed
in top-down order.
Thus, we can make two important observations about the behavior of the
“DecisionTable1”:

Rule actions cannot affect the conditions of any other rules in the decision
table – there will be no re-evaluation of any conditions

Rule overrides are permitted. The action of any executed rule may
override the action of any previously executed rule.
Let’s consider an example of a rule that states: “A person of age 17 or older is
eligible to drive. However, in Florida 16 year olds can also drive”. If we try to
present this rule using the standard DecisionTable, it may look as follows:
Using a non-standard DecisionTable1 we may present the same rule as:
42
OpenRules, Inc.
OpenRules® User Manual
In the DecisionTable1 the first unconditional rule will set “Driving Eligibility” to
“Eligible”. The second rule will reset it to “Not Eligible” for all people younger
than 17. But for 16 year olds living in Florida, the third rule will override the
variable again to “Eligible”.
DecisionTable2
There is one more type of decision table, “DecisionTable2,” that is similar to
“DecisionTable1” but allows the actions of already executed rules to affect the
conditions of rules specified below them. “DecisionTable2” supports the following
rules execution logic:
1. Rules are evaluated in top-down order and if a rule condition is satisfied,
then the rule actions are immediately executed.
2. Rule overrides are permitted. The action of any executed rule may
override the action of any previously executed rule.
Thus, we can make two important observations about the behavior of the
“DecisionTable2”:

Rule actions can affect the conditions of other rules

There could be rule overrides when rules defined below already executed
rules could override already executed actions.
Let’s consider the following example:
43
OpenRules, Inc.
OpenRules® User Manual
Here the first (unconditional) rule will calculate and set the value of the decision
variable “Taxable Income”. The second rule will check if the calculated value is
less than 0. If it is true, this rule will reset this decision variable to 0.
RULE TABLES
OpenRules® supports several ways to represent business rules inside Excel
tables. Default decision table is the most popular way to present sets of related
business rules because they do not require any coding. However, there classical
decision tables can represent more complex execution logic that is frequently
custom for different conditions and actions.
Actually, standard DecisionTable is a special case of an OpenRules® single-hit
decision table that is based on a predefined template (see below). Since 2003,
OpenRules® allows its users to configure different types of custom decision tables
directly in Excel. In spite of the necessity to use Java snippets to specify custom
logic, these tables are successfully used by major corporations in real-world
decision support applications. This chapter describes different decision tables
that go beyond the default decision tables. It will also describe how to use simple
IF-THEN-ELSE statements within Excel-based tables of type "Method".
Simple Rule Table
Let's consider a simple set of HelloWorld rules that can be used to generate a
string like "Good Morning, World!" based on the actual time of the day. How one
understands such concepts as "morning", "afternoon", "evening", and "night" is
defined in this simple rules table:
44
OpenRules, Inc.
OpenRules® User Manual
Hopefully, this rule table is not much more difficult to compare with the default
DecisonTable. It states that if the current hour is between 0 and 11, the greeting
should be "Good Morning", etc. You may change Hour From or Hour To if you
want to customize the definition of "morning" or "evening". This table is also
oriented to a business user. However, its first row already includes some
technical information (a table signature):
Rules void helloWorld(int hour)
Here "Rules" is an OpenRules® keyword for this type of tables. "helloWorld" is
the name of this particular rules table. It tells to an external program or to other
rules how to launch this rules table. Actually, this is a typical description of a
programming method (its signature) that has one integer parameter and returns
nothing (the type "void"). The integer parameter "hour" is expected to contain the
current time of the day. While you can always hide this information from a
business user, it is an important specification of this rule table.
You may ask: where is the implementation logic for this rule table? All rule
tables include additional hidden rows (frequently password protected) that you
can see if you click on the buttons "+" to open the Technical View below:
45
OpenRules, Inc.
OpenRules® User Manual
This part of the rule table is oriented to a technical user, who is not expected to
be a programming guru but rather a person with a basic knowledge of the "C"
family of languages which includes Java. Let's walk through these rows step by
step:
-
Row "Condition and Action Headers" (see row 4 in the table above). The
initial columns with conditions should start with the letter "C", for example
"C1", "Condition 1". The columns with actions should start with the letter
"A", for example "A1", "Action 1".
-
Row "Code" (see row 5 in the table above). The cells in this row specify the
semantics of the condition or action associated with the corresponding
columns. For example, the cell B5 contains the code min <= hour. This
means that condition C1 will be true whenever the value for min in any cell
in the column below in this row is less than or equals to the parameter hour.
If hour is 15, then the C1-conditions from rows 8 and 9 will be satisfied.
The code in the Action-columns defines what should be done when all
conditions are satisfied. For example, cell D5 contains the code:
System.out.println(greeting + ", World!")
This code will print a string composed of the variable greeting and ", World!",
where greeting will be chosen from a row where all of the conditions are
46
OpenRules, Inc.
OpenRules® User Manual
satisfied. Again, if hour is 15, then both conditions C1 and C2 will be
satisfied only for row 9 (because 9 <= 15 <= 17). As a result, the words "Good
Afternoon, World!" will be printed. If the rule table does not contain a row
where all conditions have been satisfied, then no actions will be executed.
Such a situation can be diagnosed automatically.
-
Row "Parameters" (see row 6 in the table above). The cells in this row specify
the types and names of the parameters used in the previous row.
-
Row "Display Values" (see row 7 in the table above). The cells in this row
contain a natural language description of the column content.
The same table can be defined a little bit differently using one condition code for
both columns "min" and "max":
How Rule Tables Are Organized
As you have seen in the previous section, rule tables have the following
structure:
Row 1: Signature
Rules void tableName(Type1 par1, Type2 par2, ..) - Multi-Hit Rule Table
Rules <JavaClass> tableName(Type1 par1, Type2 par2, ..) - Single-Hit Rule
Table
Row 2: Condition/Action Indicators
47
OpenRules, Inc.
OpenRules® User Manual
The condition column indicator is a word starting with “C”.
The action column indicator is a word starting with “A”.
All other starting characters are ignored and the whole column is considered
as a comment
Row 3: Code
The cells in each column (or merged cells for several columns) contain Java
Snippets.
Condition codes should contain expressions that return Boolean values.
If an action code contains any correct Java snippet, the return type is
irrelevant.
Row 4: Parameters
Each condition/action may have from 0 to N parameters. Usually there is
only one parameter description and it consists of two words:
parameterType parameterName
Example: int min
parameterName is a standard one word name that corresponds to Java
identification rules.
parameterType can be represented using the following Java types:
-
Basic Java types: boolean, char, int, long, double,
String, Date
-
Standard Java classes: java.lang.Boolean,
java.lang.Integer, java.lang.Long, java.lang.Double,
java.lang.Character, java.lang.String, java.util.Date
-
Any custom Java class with a public constructor that has a String
parameter
-
One-dimensional arrays of the above types.
Multiple parameters can be used in the situations when one code is used for
several columns. See the standard example “Loan1.xls”.
48
OpenRules, Inc.
OpenRules® User Manual
Row 5: Columns Display Values
Text is used to give the column a definition that would be meaningful to
another reader (there are no restrictions on what text may be used).
Row 6 and below: Rules with concrete values in cells
Text cells in these rows usually contain literals that correspond to the
parameter types.
For Boolean parameters you may enter the values "TRUE" or "FALSE" (or
equally "Yes" or "No") without quotations.
Cells with Dates can be specified using java.util.Date. OpenRules® uses
java.text.DateFormat.SHORT to convert a text defined inside a cell into
java.util.Date. Before OpenRules® 4.1 we recommended our
customers not to use Excel's Date format and define Date fields in Excel as
Text fields. The reason was the notorious Excel problem inherited from a
wrong assumption that 1900 was a leap year. As a result, a date entered in
Excel as 02/15/2004 could be interpreted by OpenRules® as 02/16/2004.
Starting with release 4.1 OpenRules® correctly interprets both Date and Text
Excel Date formats.
Valid Java expression (Java snippets) may be put inside table cells by one
of two ways:
-
by surrounding the expression in curly brackets, for example: {
driver.age+1; }
-
by putting ":=" in front of your Java expression, for example:
:=driver.age+1
Make sure that the expression's type corresponds to the parameter
type.
Empty cells inside rules means "whatever" and the proper condition is
automatically considered satisfied. An action with an empty value will be
49
OpenRules, Inc.
OpenRules® User Manual
ignored. If the parameter has type String and you want to enter a space
character, you must explicitly enter one of the following expressions:
:= " "
'= " "
{ " "; }
Note. Excel is always trying to "guess" the type of text is inside its cells and
automatically converts the internal representation to something that may not be
exactly what you see. For example, Excel may use a scientific format for certain
numbers. To avoid a "strange" behavior try to explicitly define the format "text"
for the proper Excel cells.
Separating Business and Technical Information
During rules harvesting, business specialists initially create rule tables using
regular Excel tables. They put a table name in the first row and column names in
the second row. They start with Conditions columns and end with Action
columns. For example, they can create a table with 5 columns [C1,C2,C3,A1,A2]
assuming the following logic:
IF conditions C1 and C2 and C3 are satisfied
THEN execute actions A1 and A2
Then, a business specialist provides content for concrete rules in the rows below
the title rows.
As an example, let's consider the rule table "defineSalutation" with the rules that
define how to greet a customer (Mr., Ms, or Mrs.) based on his/her gender and
marital status. Here is the initial business view (it is not yet syntactically
correct):
50
OpenRules® User Manual
OpenRules, Inc.
A business analyst has initially created only five rows:
-
A signature "Rules defineSalutation" (it is not a real signature yet)
-
A row with column titles: two conditions "Gender", "Marital Status" and one
action "Set Salutation"
-
Rows with three rules that can be read as:
1)
IF Gender is “Male” THEN Set Salutation “Mr."
2)
IF Gender is “Female” and Marital Status is “Married” THEN Set Salutation “Mrs.”
3)
IF Gender is “Female” and Marital Status is “Single” THEN Set Salutation “Ms.”
While business specialists continue to define such rule tables, at some point a
technical specialist should take over and add to these tables the actual
implementation.
The
technical
specialist
(familiar
with
the
software
environment into which these rules are going to be embedded) talks to the
business specialist (author of the rule table) about how the rules should be used.
In the case of the "defineSalutation" rule table, they agree that the table will be
used to generate a salutation to a customer. So, the technical specialist decides
that the table will have two parameters:
1) a customer of the type Customer
2) a response of the type Response
The technical specialist will modify the signature row of the table to look like
this:
Rules void defineSalutation(Customer customer, Response response)
Then she/he inserts three more rows just after the first (signature) row:
51
OpenRules, Inc.
OpenRules® User Manual
-
Row 2 with Condition/Action indicators
-
Row 3 with Condition/Action implementation
-
Row 4 with the type and name of the parameters entered in the proper column.
Here is a complete implementation of this rule table:
The rules implementer will decide that to support this rule table, type Customer
should have at least two attributes, "gender" and "maritalStatus", and the type
Response should be able somehow to save different pairs (names,value)
like("salutation","Mr."). Knowing the development environment, s/he will decide
on the types of attributes. Let's assume that both types Customer and Response
correspond to Java classes, and the attributes have the basic Java type of String.
In this case, the column "Gender" will be marked with a parameter "String
gender" and the condition will be implemented as a simple boolean expression:
customer.gender.equals(gender)
The second column "C2" is implemented similarly with a String attribute and a
parameter maritalStatus. Finally (to make it a little bit more complicated), we
will assume that the class Response contains an attribute map of the predefined
Java type HashMap, in which we can put/get pairs of Strings. So, the
implementation of the action "Set Salutation" will look like:
response.map.put("salutation",salutation)
52
OpenRules, Inc.
OpenRules® User Manual
How Rule Tables Are Executed
The rules inside rule tables are executed one-by-one in the order they are placed
in the table. The execution logic of one rule (row in the vertical table) is the
following:
IF ALL conditions are satisfied THEN execute ALL actions.
If at least one condition is violated (evaluation of the code produces false), all
other conditions in the same rule (row) are ignored and are not evaluated. The
absence of a parameter in a condition cell means the condition is always
true. Actions are evaluated only if all conditions in the same row are evaluated
to be true and the action has non-empty parameters. Action columns with no
parameters are ignored.
For the default vertical rule tables, all rules are executed in top-down order.
There could be situations when all conditions in two or more rules (rows) are
satisfied. In that case, the actions of all rules (rows) will be executed, and the
actions in the rows below can override the actions of the rows above.
For horizontal rule tables, all rules (columns) are executed in left-to-right order.
Relationships between Rules inside Rule Tables
OpenRules® does not assume any implicit ("magic") execution logic, and executes
rules in the order specified by the rule designer. All rules are executed one-byone in the order they are placed in the rule table. There is a simple rule that
governs rules execution inside a rules table:
The preceding rules are evaluated and executed first!
OpenRules® supports the following types of rule tables that offer different
execution logic to satisfy different practical needs:
-
Multi-hit rule tables
Single-hit rule tables
Rule Sequences.
53
OpenRules, Inc.
OpenRules® User Manual
Multi-Hit Rule Tables
A multi-hit rule table evaluates conditions in ALL rows before any action is
executed. Thus, actions are executed only AFTER all conditions for all rules
have already been evaluated. From this point of view, the execution logic is
different from traditional programming if-then logic. Let us consider a simple
example. We want to write a program "swap" that will do the following:
If x is equal to 1 then make x to be equal to 2.
If x is equal to 2 then make x to be equal to 1.
Suppose you decided to write a Java method assuming that there is a class App
with an integer variable x. The code may (but should not) look like this:
void swapX(App app) {
if (app.x == 1) app.x = 2;
if (app.x == 2) app.x = 1;
}
Obviously, this method will produce an incorrect result because of the missing
"else". This is “obvious” to a software developer, but may not be at all obvious to
a business analyst. However, in a properly formatted rule table the following
representation would be a completely legitimate:
It will also match our plain English description above. Here is the same table
with an extended technical view:
54
OpenRules® User Manual
OpenRules, Inc.
Rules Overrides in Multi-Hit Rule Tables
There could be situations when all conditions in two or more rules (rows) are
satisfied at the same time (multiple hits). In that case, the actions of all rules
(rows) will be executed, but the actions in the rows below can override the
actions of the rows above. This approach also allows a designer to specify a very
natural requirement:
More specific rules should override more generic rules!
The only thing a designer needs to guarantee is that "more specific" rules are
placed in the same rule table after "more generic" rules. For example, you may
want to execute Action-1 every time that Condition-1 and Condition-2 are
satisfied. However, if additionally, Condition-3 is also satisfied, you want to
execute Action-2. To do this, you could arrange your rule table in the following
way:
Condition-1
Condition-2
X
X
X
X
Condition-3
Action-1
Action-2
X
X
X
In this table the second rule may override the first one (as you might naturally
expect).
55
OpenRules® User Manual
OpenRules, Inc.
Let's consider the execution logic of the following multi-hit rule table that defines
a salutation "Mr.", "Mrs.", or "Ms." based on a customer's gender and marital
status:
Rules void defineSalutation(Customer customer, Response response)
Gender
Marital Status
Set Salutation
Male
Mr.
Female
Married
Mrs.
Female
Single
Ms.
If a customer is a married female, the conditions of the second rules are satisfied
and the salutation "Mrs." will be selected. This is only a business view of the
rules table. The complete view including the hidden implementation details
("Java snippets") is presented below:
Rules void defineSalutation(Customer customer, Response response)
C1
C2
A1
customer.gender. customer.maritalStatus. response.map.put("salutation",s
equals(gender)
equals(status)
alutation);
String gender
Gender
String status
Marital Status
Male
String salutation
Set Salutation
Mr.
Female
Married
Mrs.
Female
Single
Ms.
The OpenRulesEngine will execute rules (all 3 "white" rows) one after another.
For each row if conditions C1 and C2 are satisfied then the action A1 will be
executed with the selected "salutation". We may add one more rule at the very
end of this table:
Rules void defineSalutation(Customer customer, Response
response)
Gender
Marital Status
Set Salutation
Male
Mr.
Female
Married
Mrs.
Female
Single
Ms.
???
56
OpenRules® User Manual
OpenRules, Inc.
In this case, after executing the second rule OpenRules® will also execute the
new, 4th rule and will override a salutation "Mrs." with "???". Obviously this is
not a desirable result. However, sometimes it may have a positive effect by
avoiding undefined values in cases when the previous rules did not cover all
possible situations. What if our customer is a Divorced Female?! How can this
multi-hit effect be avoided? What if we want to produce "???" only when no other
rules have been satisfied?
Single-Hit Rule Tables
To achieve this you may use a so-called "single-hit" rule table, which is specified
by putting any return type except "void" after the keyword "Rules". The
following is an example of a single-hit rule table that will do exactly what we
need:
Rules String defineSalutation(Customer customer, Response
response)
Gender
Marital Status
Set Salutation
Male
Mr.
Female
Married
Mrs.
Female
Single
Ms.
???
Another positive effect of such "single-hitness" may be observed in connection
with large tables with say 1000 rows. If OpenRules® obtains a hit on rule #10 it
would not bother to check the validity of the remaining 990 rules.
Having rule tables with a return value may also simplify your interface. For
example, we do not really need the special object Response which we used to
write our defined salutation. Our simplified rule table produces a salutation
without an additional special object:
57
OpenRules, Inc.
Rules String defineSalutation(Customer customer)
C1
C2
customer.gender.
customer.maritalStatus
equals(gender)
.equals(status)
String gender
String status
Gender
Marital Status
Male
Female
Married
Female
Single
OpenRules® User Manual
A1
return salutation;
String salutation
Set Salutation
Mr.
Mrs.
Ms.
???
Please note that the last action in this table should return a value that has the
same type as the entire single-hit table. The single-hit table may return any
standard or custom Java class such as String or Customer. Instead of basic Java
types such as "int" you should use the proper Java classes such as Integer in the
table signature.
Here is an example of Java code that creates an OpenRulesEngine and executes
the latest rules table "defineSalutation":
public static void main(String[] args) {
String fileName = "file:rules/main/HelloCustomer.xls";
OpenRulesEngine engine =
new OpenRulesEngine(fileName);
Customer customer = new Customer();
customer.setName("Robinson");
customer.setGender("Female");
customer.setMaritalStatus("Married");
String salutation =
(String)engine.run("defineSalutation", customer);
System.out.println(salutation);
}
Rule Sequences
There is one more type of rule tables called “Rule Sequence” that is used mainly
internally within templates. Rule Sequence can be considered as a multi-hit rule
table with only one difference in the execution logic, conditions are not evaluated
before execution of the actions. So, all rules will be executed in top-down order
with possible rules overrides. Rule actions are permitted to affect the conditions
of any rules that follow the action. The keyword “Rules” should be replaced with
58
OpenRules, Inc.
OpenRules® User Manual
another keyword “RuleSequence”. Let’s get back to our “swapX” example. The
following multi-hit table will correctly solve this problem:
However, a similar rule sequence
will fail because when x is equal to 1, the first rule will make it 2, and
then the second rules will make it 1 again.
Relationships among Rule Tables
In most practical cases, business rules are not located in one file or in a single
rule set, but rather are represented as a hierarchy of inter-related rule tables
located in different files and directories - see Business Rules Repository.
Frequently, the main Excel-file contains a main method that specifies the
execution logic of multiple decision tables. You may use the table “Decision” for
the same purposes. In many cases, the rule engine can execute decision tables
directly from a Java program – see API.
59
OpenRules® User Manual
OpenRules, Inc.
Because OpenRules® interprets rule tables as regular methods, designers of rules
frequently create special "processing flow" decision tables to specify the
conditions under which different rules should be executed. See examples of
processing flow rules in such sample projects as Loan2 and LoanDynamics.
Simple AND / OR Conditions in Rule Tables
All conditions inside the same row (rule) are considered from left to right using
the AND logic. For example, to express
if (A>5 && B >10) {do something}
you may use the rule table:
Rules void testAND(int a, int b)
C1
C2
a>5
b>10
String x
String x
A>5
B > 10
X
X
A1
System.out.println(text)
String text
Do
Something
To express the OR logic
if (A>5 || B >10) {do something}
you may use the rules table:
Rules void testOR(int a, int b)
C1
C2
a>5
b>10
String x
String x
A>5
B > 10
X
X
A1
System.out.println(text)
String text
Do
Something
Sometimes instead of creating a decision table it is more convenient to represent
rules using simple Java expressions inside Method tables. For example, the
above rules table may be easily represented as the following Method table:
60
OpenRules, Inc.
OpenRules® User Manual
Method void testOR(int a, int b)
if (a > 5 || b>10) System.out.println("Something");
Horizontal and Vertical Rule Tables
Rule tables can be created in one of two possible formats:
-
Vertical Format (default)
-
Horizontal Format.
Based on the nature of the rule table, a rules creator can decide to use a vertical
format (as in the examples above where concrete rules go vertically one after
another) or a horizontal format where Condition and Action are located in the
rows and the rules themselves go into columns. Here is an example of the proper
horizontal format for the same rule table "helloWorld":
OpenRules® automatically recognizes that a table has a vertical or a horizontal
format. You can use Excel's Copy and Paste Special feature to transpose a rule
table from one format to another.
Note. When a rule table has too many rules (more than you can see on one page)
it is better to use the vertical format to avoid Excel's limitations: a worksheet has
a maximum of 65,536 rows but it is limited to 256 columns.
Merging Cells
OpenRules® recognizes the powerful Cell Merging mechanism supported by
Excel and other standard table editing tools. Here is an example of a rule table
with merged cells:
61
OpenRules® User Manual
OpenRules, Inc.
Rules void testMerge(String value1, String value2)
Rule
C1
C2
A1
A2
value1.equals(val)
value2.equals(val)
out("A1: " + text);
out("A2: " +
text);
String val
String val
String text
String text
#
1
Name
Value
One
Text 1
Text 2
12
2
B
Two
3
4
11+21
Three
D
22
31
32
41
42
The semantics of this table is intuitive and described in the following table:
Value Value
1
2
Applied
Rules
B
One
1
B
Two
2
B
Three
3
D
Three
4
A
D
Two
Two
none
none
Printed
Results
A1: 11+21
A2: 12
A1: 11+21
A2: 22
A1: 31
A2: 32
A1: 41
A2: 42
Restriction. We added the first column with rules numbers to avoid the known
implementation restriction that the very first column (the first row for horizontal
rule tables) cannot contain merged rows. More examples can be found in the
standard rule project "Merge" - click here to analyze more rules. When you use
the standard decision tables, you may put the standard condition “C#” in the
very first column and use numbers to mark each table’s row.
Sub-Columns and Sub-Rows for Dynamic Arrays
One table column can consist of several sub-columns (see sub-columns "Min" and
"Max" in the example above).
You may efficiently use the Excel merge
62
OpenRules, Inc.
OpenRules® User Manual
mechanism to combine code cells and to present them in the most intuitive way.
Here is an example with an unlimited number of sub-columns:
As you can see, condition C6 contains 4 sub-columns for different combinations of
rates. The cells in the Condition, code, parameters and display values, rows are
merged. You can insert more sub-columns (use Excel's menu "Insert") to handle
more rate combinations if necessary without any changes in the code. The
parameter row is defined as a String array, String[] rates. The actual values
of the parameters should go from left to right and the first empty value in a subcolumn should indicate the end of the array "rates". You can see the complete
example in the rule table "Rule Family 212" in the file Loan1.xls.
If your rule table has a horizontal format, you may use multiple sub-rows in a
similar way (see the example in file UpSell.xls).
Using Expressions inside Rule Tables
OpenRules® allows a rules designer to use “almost” natural language expressions
inside rule tables to represent intervals of numbers, strings, dates, etc. You also
may use Java expressions whenever necessary.
Integer and Real Intervals
You may use plain English expressions to define different intervals for integer
and real decision variables inside rule tables. Instead of creating multiple
columns for defining different ranges for integer and real values, a business user
63
OpenRules® User Manual
OpenRules, Inc.
may define from-to intervals in practically unlimited English using such phrases
as: "500-1000", "between 500 and 1000", "Less than 16", "More or equals to 17",
"17 and older", "< 50", ">= 10,000", "70+", "from 9 to 17", "[12;14)", etc.
You also may use many other ways to represent an interval of integers by
specifying their two bounds or sometimes only one bound. Here are some
examples of valid integer intervals:
Cell Expression
Comment
5
equals to 5
[5,10]
contains 5, 6, 7, 8, 9, and 10
5;10
contains 5, 6, 7, 8, 9, and 10
[5,10)
contains 5 but not 10
5 - 10
contains 5 and 10
5-10
contains 5 and 10
5- 10
contains 5 and 10
-5 - 20
contains -5 and 20
-5 - -20
error: left bound is greater than the right
one
-5 - -2
contains -5 , -4, -3, -2
from 5 to 20
contains 5 and 20
less 5
does not contain 5
less than 5
does not contain 5
less or equals 5
contains 5
less or equal 5
contains 5
less or equals to 5
contains 5
smaller than 5
does not contain 5
more 10
does not contain 10
more than 10
does not contain 10
10+
more than 10
>10
does not contain 10
>=10
contains 10
between 5 and 10
contains 5 and 10
no less than 10
contains 10
no more than 5
contains 5
64
OpenRules® User Manual
OpenRules, Inc.
equals to 5
equals to 5
greater or equal than 5
and less than 10
contains 5 but not 10
more than 5 less or
equal than 10
does not contain 5 and contains 10
more than 5,111,111
and less or equal than
10,222,222
does not contain 5,111,111 and contains
10,222,222
[5'000;10'000'000)
contains 5,000 but not 10,000,000
[5,000;10,000,000)
contains 5,000 but not 10,000,000
(5;100,000,000]
contains 5,000 and 10,000,000
You may use many other ways to represent integer intervals as you usually do in
plain English. The only limitation is the following: min should always go
before max!
Similarly
to
integer
intervals,
one
may
use
the
predefined
type FromToDouble to represent intervals of real numbers. The bounds of
double intervals could be integer or real numbers such as [2.7; 3.14).
Comparing Integer and Real Numbers
You may use the predefined type CompareToInt to compare a decision variable
with an integer number that is preceded by a comparison operator. Examples of
acceptable operators:
Cell Expression
Comment
<= 5
less or equals to 5
<5
strictly less than 5
>5
strictly more than 5
>= 5
more or equals to 5
!=
not equal to 5
5
equals to 5.
Note that absence of a comparison operator means
equality. You cannot use an explicit operator "=" (not to
be confused with Excel's formulas).
65
OpenRules, Inc.
OpenRules® User Manual
Similarly to CompareToInt one may use the predefined type CompareToDouble
to represent comparisons with real numbers. The comparison values may be
presented as integer or real numbers, e.g. "<= 25.4" and "> 0.5".
Using Comparison Operators inside Rule Tables
A user can employ a comparison operators such as "<" for "less" or ">" for "more"
directly inside the rules. There are several ways to accomplish this. Here is an
example from the rule table "Rule Family 212" (Loan1.xls):
You may use the Excel Data Validation menu to limit the choice of the operators:
66
OpenRules® User Manual
OpenRules, Inc.
Here the sign "==" has an apostrophe in front to differentiate it from an Excel
formula.
provided
The actual implementation of possible comparison operators is
as
an
example
in
the
project
"com.openrules.tools"
(see com.openrules.tools.Operator.java). You may change them or add
other operators. In addition to values of the type "int" you may also use Operator
to compare long, double, and String types.
Comparing Dates
You may use the standard java.util.Date or any other Java Comparable type.
Here is an example of comparing Dates:
C1
op.compare(visit.date,date)
Operator op
Date date
Operator
Date
==
2/15/2007
!=
1/1/2007
<=
2/15/2007
>
2/15/2007
<
2/15/2007
67
OpenRules® User Manual
OpenRules, Inc.
Please note that the current implementation compares dates without time.
Another way to use operators directly inside a table is to use expressions. In the
example above, instead of having two sub-columns "Operator" and "Value" we
could use one column and put an expression inside the cell:
{ c.creditCardBalance <= 0; }
The use of expressions is very convenient when you do not know ahead of time
which operator will be required for which columns.
Comparing Boolean Values
If a parameter type is defined as "boolean", you are allowed to use the following
values inside rule cells:
-
True, TRUE, Yes, YES
False, FALSE, No, NO
You also may use formulas that produces a Boolean, .e.g.
{ loan.additionalIncomeValidationNeeded; }
Sometimes, you want to indicate that a condition is satisfied or an action should
be executed. You may use any character like X or * without checking its actual
value – the fact that the cell is not empty indicates that the condition is true. For
example, in the following table (from the standard project VacationDays)
68
OpenRules, Inc.
OpenRules® User Manual
only actions marked with "X" will be executed. You can use any other character
instead of "X".
Representing String Domains
Let's express a condition that validates if a customer's internal credit score is one
of several acceptable rates such as "A B C" and "D F". To avoid the necessity to
create multiple sub-columns for similar conditions, we may put all possible string
values inside the same cell and separate them by spaces or commas. Here is an
example of such a condition:
Condition
domain.contains(customer.internalCreditRating)
DomainString domain
Internal Credit Rating
ABC
DF
DF
ABC
Here we use the predefined type DomainString that defines a domain of strings
(words) separated by whitespaces. The method "contains(String string)" of
the class DomainString checks if the parameter "string" is found among all
strings listed in the current "domain". You also may use the method
69
OpenRules® User Manual
OpenRules, Inc.
"containsIgnoreCase(String
string)" that allows you to ignore case
during the comparison.
If possible values may contain several words, one may use the predefined
type DomainStringC where "C" indicates that commas will be used as a string
separator. For example, we may use DomainStringC to specify a domain such
as "Very Hot, Hot, Warm, Cold, Very Cold".
Representing Domains of Numbers
If you need to represent domains of integer or double values, there are several
predefined types similar to DomainString:

DomainInt

DomainIntC

DomainDouble

DomainDoubleC
For example, here is a condition column with eligible loan terms:
Condition
domain.contains(c.loanTerm)
DomainIntC domain
Eligible Loan Terms
24,36,72
36,72
72
Using Java Expressions
The use of Java expressions provides the powerful ability to perform calculations
and test for complex logical conditions. While the writing of expressions requires
some technical knowledge, it does not require the skills of a programmer. Realworld experience shows that business analysts frequently have a need to write
these expressions themselves. It is up to the rule table designer to decide
70
OpenRules® User Manual
OpenRules, Inc.
whether to show the expressions to business people or to hide them from view.
Let's consider a decision table for "Income Validation" from the provided
standard example “Loan1”:
Rules void ValidateIncomeRules(LoanRequest loan, Customer customer)
C1
A1
customer.monthlyIncome * 0.8 loan.incomeValidationResult
customer.monthlyDebt > loan.amount/loan.term
= result;
boolean condition
String result
IF
Income is Sufficient for the Loan
THEN
Set Income Vaidation
Result
No
UNSUFFICIENT
Yes
SUFFICIENT
Here the actual income validation expression is hidden from business people inside "gray"
technical rows, and a business person would only be able to choose between "Yes" or "No".
However, the same table could be presented in this way:
Rules void ValidateIncomeRules(LoanRequest loan, Customer customer)
C1
A1
loan.incomeValidationResult
condition == true
= result;
boolean condition
String result
IF
Condition is True
THEN
Set Income Validation
Result
UNSUFFICIENT
:= customer.monthlyIncome * 0.8 customer.monthlyDebt > loan.amount/loan.term
SUFFICIENT
Now, a user can both see and change the actual income validation condition.
71
OpenRules® User Manual
OpenRules, Inc.
Notes: Do not use Excel's formulas if you want the content to be recognized by
the OpenRules® engine: use OpenRules® expressions instead.
If you want to start your cell with "=" you have to put an apostrophe in front of it
i.e. '= to direct Excel not to attempt to interpret it as a formula.
Expanding and Customizing Predefined Types
All the predefined types mentioned above are implemented in the Java
package com.openrules.types. You may get the source code of this package
and
expand
and/or
customize
the
proper
classes.
In
particular,
for
internationalization purposes you may translate the English key words into your
preferred
language.
You
may
change
the
default
assumptions
about
inclusion/exclusion of bounds inside integer and real intervals. You may add
new types of intervals and domains.
Performance Considerations
The use of expressions inside OpenRules® tables comes with some price - mainly
in performance, for large rule tables. This is understandable because for every
cell with an expression OpenRules® will create a separate instance of the proper
Java class during rules execution. However, having multiple representation
options allows a rule designer to find a reasonable compromise between
performance and expressiveness.
RULE TEMPLATES
OpenRules® provides a powerful yet intuitive mechanism for compactly
organizing enterprise-level business rules repositories. Rule templates allow
rule designers to write the rules logic once and use it many times. With rule
templates you may completely hide rules implementation details from business
users. OpenRules® supports several rule templatization mechanisms from simple
rule tables that inherit the exact structure of templates to partial template
implementations.
72
OpenRules® User Manual
OpenRules, Inc.
Simple Rules Templates
Rule templates are regular rule tables that serve as structural prototypes for
many other rule tables with the same structure but different content (rules). A
simple rule template usually does not have rules at all but only specifies the
table structure and implementation details for conditions and actions. Thus, a
simple rule template contains the first 5 rows of a regular decision table as in the
following example:
Rules void defineGreeting(App app, int hour)
C1
A1
min <= hour && hour <=
max
app.greeting =
greeting;
int min
int max
String greeting
Hour From
Hour To
Set Greeting
Signature with
parameters
Conditions and Actions
identifiers
Java snippets describe
condition/action
semantics
Parameter types and
names
Business names for
conditions and actions
We may use this rule table as a template to define different greeting rules for
summer
and
winter
time.
The
actual
decision
tables
will implement (or extend) the template table with particular rules:
Rules summerGreeting template defineGreeting
Hour
Hour To
Set Greeting
From
0
10
Good Morning
11
18
Good Afternoon
19
22
Good Evening
23
24
Good Night
and
Rules winterGreeting template defineGreeting
Hour
Hour To
Set Greeting
From
0
11
Good Morning
12
17
Good Afternoon
73
OpenRules® User Manual
OpenRules, Inc.
18
23
22
24
Good Evening
Good Night
Note that rule tables "summerGreeting" and "winterGreeting" do not have
technical information at all - Java snippets and a signature are defined only once
and reside in the template-table "defineGreeting".
Along with the keyword "template" you may use other keywords:

implements

implement

extends

extend
We will refer to these rule tables created based on a template as "template
implementations".
Simple templates require that the extended tables should have exactly the same
condition and action columns.
Defining Default Rules within Templates
When many rule tables are created based on the same rule template, it could be
inconvenient to keep the same default rules in all extended tables. As an
alternative you may add the default rules directly to the template. The location
of the default rules depends on the types of your rules tables.
Templates with Default Rules for Multi-Hit Tables
Multi-hit rule tables execute all their rules that are satisfied, allowing rules
overrides. However, when conditions in all specified rules are not satisfied then a
multi-hit table usually uses the first (!) rules to specify the default action. The
74
OpenRules® User Manual
OpenRules, Inc.
rules from the template will be executed before the actual rules defined inside
the extended tables.
Let's consider an example. You may notice that the rules tables above would not
produce any greeting if the parameter "hour" is outside of the interval [0;24].
Let's assume that in this case we want to always produce the default greeting
"How are you". To do this, simply add one default rule directly to the template:
Rules void defineGreeting(App app, int hour)
C1
A1
min <= hour &&
hour <= max
app.greeting = greeting;
int min
String greeting
int max
How are you
This rule will be added at
the beginning of all
template implementations.
This greeting will be
produced if all other rules
in the rule tables fail
A template for multi-hit tables could include more than one default rule each
with different conditions - they all will be added to the beginning of the template
implementation tables and will execute different default actions.
Templates with Default Rules for Single-Hit Tables
Single-hit rule tables usually end their execution when at least one rules is
satisfied. However, when conditions in all specified rules are not satisfied then a
single-hit table usually uses the last rule(s) to specify the default action(s). The
rules from the template will be executed after the actual rules defined inside
the template implementation.
Let's consider an example. We have shown that without modification, the rule
tables above would not produce any greeting if the parameter "hour" is outside of
the interval [0;24]. Instead of adding the same error message in both "summer"
and "winter" rules, we could do the following:
75
OpenRules® User Manual
OpenRules, Inc.
-
make our "defineGreeting" template a single-hit table by changing a return
type from "void" to "String"
-
- add the default reaction to the error in "hour" directly to the template:
Signature now returns
String
Conditions and Actions
identifiers
Rules String defineGreeting(App app, int hour)
C1
A1
min <= hour &&
hour <= max
int min
int max
Hour
Hour To
From
app.greeting =
greeting; return greeting;
String greeting
"return greeting;" has been
added
Parameter types and names
Business names for
conditions and actions
This rule will be added at
the end of all template
implementations tables. The
error message will be return
instead of a greeting when
all other rules fail.
Set Greeting
ERROR: Invalid Hour
A template for single-hit tables could include more than one rule with different
conditions
-
they
all
will
be
added
at
the
end
of
the template
implementation tables to execute different default actions.
Partial Template Implementation
Usually template implementation tables have exactly the same structure as the
rule templates they extend. However, sometimes it is more convenient to build
your own rule table that contains only some conditions and actions from already
predefined rule templates. This is especially important when a library of rule
templates for a certain type of business is used to create a concrete rules-based
application. How can this be achieved?
The template implementation table uses its second row to specify the names of
the used conditions and actions from the template. Let's consider an example.
The DebtResearchRules from the standard OpenRules® example "Loan
Origination" may be used as the following template:
76
OpenRules® User Manual
OpenRules, Inc.
Rules void DebtResearchRules(LoanRequest loan, Customer c)
C1
C2
C3
c.outsideCredit
c.mortgag
c.loanHol
Score>min &&
eHolder.eq
der.equal
c.outsideCredit
uals(YN)
s(YN)
Score<=max
String YN int min
IF
Mortgage
Holder
AND
Loan
Holder
Max
C5
C6
C7
A1
contains(r c.internalA loan.debt
op.compare(c. op.compare(c.e
ates,c.inte nalystOpin Research
creditCardBala ducationLoanBa
rnalCredit ion.equals Result =
nce,value)
lance,value)
Rating) (level)
level;
int
Opera
String YN
max
tor op
AND
Outside
Credit Score
Min
C4
int Operat
int value
value or op
AND
Education
Loan
Balance
AND
Credit Card
Balance
Oper
Value
Oper
Value
String[]
rates
AND
Internal
Credit
Rating
String
level
String
level
THEN
AND
Debt
Internal
Research
Analyst
Recomme
Opinion
ndations
We may create a rule table that implements this template using only conditions
C1, C2, C5, C6 and the action A1:
Rules MyDebtResearchRules template DebtResearchRules
C1
IF
Mortgage
Holder
Yes
No
No
No
No
No
No
No
No
No
C2
AND
Outside Credit
Score
Min
100
550
550
550
550
550
550
550
550
Max
550
900
900
900
900
900
900
900
900
C5
AND
Education Loan
Balance
Oper
Value
>
<=
<=
0
0
0
<=
>
>
0
0
0
C6
A1
THEN
Debt
AND
Research
Internal Credit Rating
Recommen
dations
High
High
Mid
High
High
A
B
C
Mid
D
F
Low
Low
High
D
F
Low
A
B
C
The additional second row specifies which conditions and actions from the
original template are selected by this rule table. The order of conditions and
actions may be different from the one defined in the template. Only names like
"C2", "C6", and "A1" should be the same in the template and in its
implementation. It is preferable to use unique names for conditions and actions
inside templates. If there are duplicate names inside templates the first one
77
OpenRules® User Manual
OpenRules, Inc.
(from left to right) will be selected. You may create several columns using the
same condition and/or action names.
Templates with Optional Conditions and Actions
There is another way to use optional conditions and actions from the templates.
If the majority of the template implementations do not use a certain condition
from the template, then this condition may be explicitly marked as optional by
putting the condition name in brackets, e.g. "[C3]" or "[Conditon-5]". In this
case it is not necessary to use the second row to specify the selected conditions in
the majority of
the extended
tables.
For
example,
let's
modify the
DebtResearchRules template making the conditions C3, C4, and C7 optional:
Rules void DebtResearchRules(LoanRequest loan, Customer c)
C1
C2
[C3]
[C4]
C5
C6
[C7]
A1
Now we can implement this template as the following rule table without the
necessity to name all of the conditions and actions in the second row:
Rules MyDebtResearchRules template DebtResearchRules
AND
THEN
AND
Outside
IF
Debt
Education Loan
AND
Mortgag Credit
Research
Balance
Internal Credit Rating
e Holder Score
Recommend
ations
Min Max Oper
Value
High
Yes
High
No
100 550
Mid
No
550 900
High
No
550 900
>
0
High
No
550 900
<=
0
A
B
C
Mid
No
550 900
<=
0
D
F
Low
No
550 900
Low
No
550 900
<=
0
High
No
550 900
>
0
D
F
55
90
Low
No
>
0
A
B
C
0
0
78
OpenRules® User Manual
OpenRules, Inc.
However, a template implementation that does want to use optional conditions
will have to specify them explicitly using the second row:
Rules MyDebtResearchRules template DebtResearchRules
C1
C2
C3
C4
C5
C6
A1
THEN
Debt
Research
Recomm
endations
High
High
Mid
High
High
Mid
Low
AND
AND
AND
IF
AND
AND
Outside
Credit Card
Education
Mortgag Credit Score Loan
Balance
Loan Balance Internal Credit
e Holder
Holder
Rating
Min Max
Oper Value Oper Value
Yes
No
No
No
No
No
No
100
550
550
550
550
550
550
900
900
900
900
900
Yes
Yes
Yes
Yes
No
<=
>
>
>
>
0
0
0
0
0
>
<=
<=
0
0
0
A B C
D F
Similarly, optional actions may be marked as [A1]" or "[Action3]".
Implementation Notes:
o
Rule templates are supported for both vertical and horizontal rule tables.
o
The keywords "extends" or "implements" may be used instead of the
keyword "template"
o
Template implementations cannot be used as templates themselves.
Templates for the Default Decision Tables
The rule tables of the type “DecisionTable” are implemented using several
templates located in the following files inside the configuration project
“openrules.config”:
-
DecisionTemlates.xls: contains the following rule templates and
methods for the decision tables:
o
DecisionTemplate(Decision
decision): a template for the
tables of type “Decision”
o
initializeDecision(): the method that initializes the current decision
o
decision(): the method that returns the current decision
79
OpenRules® User Manual
OpenRules, Inc.
o
getGlossary(): the method that returns the glossary
o
getDecisionObject(String
method
that
returns
a
nameofBusinessConcept): the
business
object
associated
with
the
BusinessConcept
o
isTraceOn(): returns true if the tracing of the decision is on
o
DecisionObjectTemplate(Decision decision): a template for
the table of the type “DecisionObject”
o
GlossaryTemplate(Decision decision): a template for the table
of type “Glossary”
o
o
o
Methods that return values of decision variables based on their names:

int getInt(String name)

double getReal(String name)

String getString(String name)

Date getDate(String name)

boolean getBool(String name)
Methods that set values of decision variables based on their names:

void getInt(String name, int value)

void getReal(String name, double value)

void getString(String name, String value)

void getDate(String name, Date value)

void getBool(String name, Boolean value)
Comparison methods that compare a decision variable with a given “name”,
against a given “value”, or another decision variable using a given operator,
“op”:

boolean compareInt(String name, String op, int
value)

boolean
compareInt(String
name1,
String
op,
name,
String
op,
String name2)

boolean
compareReal(String
double value)

boolean compareReal(String name1, String op,
String name2)
80
OpenRules® User Manual
OpenRules, Inc.

boolean
compareBool(String
name,
String
op,
boolean value)

boolean compareBool(String name1, String op,
String name2)

boolean
compareDate(String
name,
String
op,
Date date)

boolean compareDate(String name1, String op,
String name2)

boolean compareString(String name, String op,
String value)

boolean compareDomain(String name, String op,
String domain)
o
the Environment table that includes the following references:

DecisionTable${OPENRULES_MODE}Templates.xls:
where ${OPENRULES_MODE} is an environment variable that has
one of the following values:

Execute – the default value for Decision Table execution
templates

Validate –for Decision Table validation templates

Solve – for execution of decision models using Rule
Solver.

DecisionTableExecuteTemplates.xls:
templates
for
templates
for
execution

DecisionTableValidateTemplates.xls:
validation
-
DecisionTableExecuteTemplates.xls: contains the following rule templates:
o
DecisionTableTemplate(): a template for execution of the single-hit
tables of the type “DecisionTable”
o
DecisionTable1Template(): a template for execution of the multi-hit
tables of the type “DecisionTable1”
o
DecisionTable2Template(): a template for execution of the rule
sequence tables of the type “DecisionTable2”
81
OpenRules® User Manual
OpenRules, Inc.
o
customInitializeDecision(): the method that can be used for
initialization of custom objects
-
DecisionTableValidateTemplates.xls: contains the following rule templates:
o
DecisionTableTemplate(): a template for validation of the tables of
type “DecisionTable” against the domains defined in the glossary
o
customInitializeDecision(): the method that can be used for the
initialization of custom objects.
Decision Templates
The template “DecisionTemplate” contains two mandatory action columns with
names “ActionPrint” and “ActionExecute” and three optional columns with the
names “Condition”, “ConditionAny”, and “ActionAny”. Here is an example of this
template:
RuleSequence void DecisionTemplate(Decision decision)
[Condition]
[ConditionAny]
ActionPrint
getGlossary().comp
are(
$COLUMN_TITLE,
op,value);
Oper
op
String
value
Decision variable
op.compare(val
ue);
boolea
n
value
Dynamic
Condition
Oper
op
ActionExecute
[ActionAny]
Log.info("Decision "
+ $TABLE_TITLE +
": " + name);
String name
Decisions
Object object
Execute
Rules
Object
value
Title for
Action Any
Because you can use the same column “Condition” or “ConditionAny” many times
in your own decision and sub-decision tables, you may create tables of type
“Decision” that are based on this template with virtually unlimited complexity.
Decision Table Templates
The template “DecisionTableTemplate” serves as a template for all standard
decision tables. All columns in this template are conditional meaning their
names are always required. Here are the first two rows of this template:
Rules String DecisionTableTemplate()
[Condition]
[ConditionAny]
[If]
[Conclusion]
[Action]
[ActionAny]
[Then]
82
[Message]
OpenRules® User Manual
OpenRules, Inc.
The template “DecisionTable1Template” serves as a template for all decision
tables of type “DecisionTable1”. Here are the first two rows of this template:
Rules void DecisionTable1Template()
[Condition]
[ConditionAny]
[If]
[Conclusion]
[Action]
[ActionAny]
[Then]
[Message]
The template “DecisionTable2Template” serves as a template for all decision
tables of type “DecisionTable2”. Here are the first two rows of this template:
RuleSequence void DecisionTable2Template()
[Condition]
[ConditionAny]
[If]
[Conclusion]
[Action]
[ActionAny]
[Then]
[Message]
You can use all these columns as many times as you wish when you may create
concrete decision tables based on these templates. Please check the file
“DecisionTableExecuteTemplates.xls” in your standard configuration project
“openrules.config” to see the latest version of the decision templates.
Customization
Customizing Default Decision Tables
A user may move the above files from “openrules.config” to different locations
and modify the decision table templates (and possible other templates). For
example, to have different types of messaging inside a custom decision, a user
may add two more columns to the template “DecisionTableTemplate”:
-
Warning: similar to Message but can use a different log for warning only
-
Errors: similar to Message but can use a different log for errors only.
Adding Custom Decision Tables
Users may add their own decision tables with conditions and actions specific to
their applications by defining their own keywords by simply extending the
keyword "DecisionTable" with they own identifier. For example, a user may add
a new decision table type called "DecisionTableMy" by defining the proper
custom
conditions
and
actions
inside
a
template
with
the
name
"DecisionTableMyTemplate". The standard installation includes a project
83
OpenRules® User Manual
OpenRules, Inc.
"DecisionCustom"
that
"DecisionTableCustom"
demonstrates
created
based
a
custom
on
a
decision
table
project-specific
called
template
"DecisionTableCustomTemplate". This template is placed in the project file
"DecisionTableCustomTemplates.xls".
Adding Custom Methods to Decision and Decision Runs
The file "DecisionTemplates.xls" contains the default methods:
-
customInitializeDecision
-
customInitializeDecisionRun
that may be replaced by your own methods. For example, rewriting the method
“customInitializeDecision“ allows a user to
initialize custom objects.
These and other methods are described below. For a good example of
customization look at the file "DecisionTableSolveTemplates.xls" that is used by
Rule Solver instead of the file "DecisionTableExecuteTemplates.xls". Contact
[email protected] if you need help with more complex customization of the
decision templates.
OPENRULES® API
OpenRules® provides an Application Programming Interface (API) that defines a
set of commonly-used functions:
-
Creating a rule engine associated with a set of Excel-based rules
-
Creating a decision associated with a set of Excel-based rules
-
Executing different rule sets using application specific business objects
-
Creating a web session and controlling client-server interaction.
84
OpenRules® User Manual
OpenRules, Inc.
OpenRulesEngine API
OpenRulesEngine is a Java class provide by OpenRule® to execute different rule
sets and methods specified in Excel files using application-specific business
objects. OpenRulesEngine can be invoked from any Java application using a
simple Java API or a standard JSR-94 interface.
Engine Constructors
OpenRulesEngine provides an interface to execute rules and methods defined in
Excel tables. You can see examples of how OpenRulesEngine is used in basic rule
projects such as HelloJava, DecisionHellJava, HelloJsr94 and web applications
such as HelloJsp, HelloForms, and HelloWS. To use OpenRulesEngine inside
your
Java
code
you
need
to
add
an
import
statement
and
com.openrules.ruleengine.OpenRulesEngine
for
make
sure
that openrules.all.jar is in the classpath of your application. This jar and
all 3rd party jar-files needed for OpenRules® execution can be found in the
subdirectory openrules.config/lib of the standard OpenRules® installation.
You may create an instance of OpenRulesEngine inside of your Java program
using the following constructor:
public OpenRulesEngine(String xlsMainFileName)
where xlsMainFileName parameter defines the location for the main xls-file. To
specify
a
file
location,
OpenRules®
uses
notation with prefixes such as "file:",
an URL
pseudo-protocol
"classpath:",
"http://",
"ftp://", "db:", etc. Typically, your main xls-file Main.xls is located in the
subdirectory "rules/main" of your Java project. In this case, its location may be
defined as "file:rules/main/Main.xls". If your main xls-file is located
directly
in
the
project
classpath,
you
may
define
its
location
"classpath:Main.xls". Use a URL like
http://www.example.com/rules/Main.xls
85
as
OpenRules, Inc.
OpenRules® User Manual
when Main.xls is located at a website. All other xls-files that can be invoked
from this main file are described in the table "Environment" using includestatements.
You may also use other forms of the OpenRulesEngine constructor. For example,
the constructor
OpenRulesEngine(String xlsMainFileName, String methodName)
allows you to also define the main method from the file xlsMainFileName that
will be executed during the consecutive runs of this engine.
Here is a complete example of a Java module that creates and executes a rule
engine (see HelloJava project):
package hello;
import com.openrules.ruleengine.OpenRulesEngine;
public class RunHelloCustomer {
public static void main(String[] args) {
String fileName = "file:rules/main/HelloCustomer.xls";
String methodName = "helloCustomer";
OpenRulesEngine engine = new OpenRulesEngine(fileName);
Customer customer = new Customer();
customer.setName("Robinson");
customer.setGender("Female");
customer.setMaritalStatus("Married");
Response response = new Response();
Object[] objects = new Object[] { customer, response };
engine.run(methodName,objects);
System.out.println("Response: " +
response.getMap().get("greeting") + ", " +
response.getMap().get("salutation") +
customer.getName() + "!" );
}
}
86
OpenRules, Inc.
OpenRules® User Manual
As you can see, when an instance "engine" of OpenRulesEngine is created, you
can create an array of Java objects and pass it as a parameter of the method
"run".
Engine Runs
The same engine can run different rules and methods defined in its Excel-files.
You may also specify the running method using
setMethod(String methodName);
or use it directly in the engine run:
engine.run(methodName,businessObjects);
If you want to pass to OpenRulesEngine only one object such as "customer", you
may write something like this:
engine.run("helloCustomer",customer);
If you do not want to pass any object to OpenRulesEngine but expect to receive
some results from the engine's run, you may use this version of the method
"run":
String[] reasons = (String[]) engine.run("getReasons");
Undefined Methods
OpenRulesEngine checks to validate if all Excel-based tables and methods are
actually defined. It produces a syntax error if a method is missing. Sometimes,
you want to execute a rule method/table from an Excel file but only if this
method is actually present. To do this, you may use this version of the method
"run":
boolean mayNotDefined = true;
engine.run(methodName, businessObjects, mayNotDefined);
87
OpenRules® User Manual
OpenRules, Inc.
In this case, if the method "methodName" is not defined, the engine would not
throw a usual runtime exception "The method <name> is not defined" but rather
will
produce
a
warning
and
will
continue
to
work.
The
parameter
"mayNotDefined" may be used similarly with the method "run" with one
parameter or with no parameters, e.g.
engine.run("validateCustomer", customer, true);
How to invoke rules from other rules if you do not know if these rules are
defined? It may be especially important when you use some predefined rule
names in templates. Instead of creating an empty rules table with the needed
name, you want to use the above parameter "mayNotDefined" directly in Excel.
Let's say you need to execute rules tables with names such as "NJ_Rules" or
"NY_Rules" from another Excel rules table but only if the proper state rules are
actually defined. You may do it by calling the following method from your rules:
Method void runStateRules(OpenRulesEngine engine, Customer customer, Response
response)
String methodName = customer.state + "_Rules";
Object[] params = new Object[2];
params[0] = customer;
params[1] = response;
engine.run(methodName, params, true);
We assume here that all state-specific rules ("NJ_Rules", "NY_Rules", etc.) have
two parameters, "customer" and "response". To use this method you need to pass
the current instance of OpenRulesEngine from your Java code to your main
Excel file as a parameter "engine". If you write an OpenRules Forms application,
this
instance
of
the
OpenRulesEngine
is
always
available
as dialog().getEngine(), otherwise you have to provide access to it, e.g. by
attaching it to one of your own business objects such as Customer.
By default OpenRules will produce a warning when the required Excel rules
table or method is not available. You may suppress such warnings by calling:
88
OpenRules® User Manual
OpenRules, Inc.
engine.turnOffNotDefinedWarning();
Accessing Password Protected Excel Files
Some Excel workbooks might be encrypted (protected by a password) to prevent
other people from opening or modifying these workbooks. Usually it's done using
Excel Button
and then Prepare plus Encrypt Document.
OpenRules
Engine may access password-protected workbooks by calling the following
method just before creating an engine instance:
OpenRulesEngine.setCurrentUserPassword("password");
Instead of "password" you should use the actual password that protects your
main and/or other Excel files. Only one password may be used by all protected
Excel files that will be processed by one instance of the OpenRulesEngine
created after this call. This call does not affect access to unprotected files. The
standard project "HelloJavaProtected" provides an example of the protected
Excel file - use the word "password" to access the file "HelloCustomer.xls".
Note.
The
static
method
"setCurrentUserPassword"
of
the
class
OpenRulesEngine actually sets the BIFF8 encryption/decryption password for
the current thread. The use of a "null" string will clear the password.
Engine Attachments
You may attach any Java object to the OpenRulesEngine using
methods setAttachment(Object attachment) and getAttachment().
Engine Version
You may receive a string with the current version number of the
OpenRulesEngine using the method getVersion().
Dynamic Rules Updates
89
OpenRules® User Manual
OpenRules, Inc.
If a business rule is changed, OpenRulesEngine automatically reloads the rule
when necessary. Before any engine's run, OpenRulesEngine checks to determine
if the main Excel file associated with this instance of the engine has been
changed. Actually, OpenRulesEngine looks at the latest modification dates of
the file xlsMainFileName.
If it has been modified, OpenRulesEngine re-
initializes itself and reloads all related Excel files. You can shut down this
feature by executing the following method:
engine.setCheckRuleUpdates(false);
Decision API
Decision Example
OpenRules® provides a special API for decision execution using the Java class
“Decision”. The following example from the standard project “Decision1040EZ”
demonstrates the use of this API.
public class Main {
public static void main(String[] args) {
String fileName = "file:rules/main/Decision.xls";
OpenRulesEngine engine =
new OpenRulesEngine(fileName);
Decision decision =
new Decision("Apply1040EZ",engine);
DynamicObject taxReturn =
(DynamicObject) engine.run("getTaxReturn");
engine.log("=== INPUT:\n" + taxReturn);
decision.put("taxReturn",taxReturn);
decision.execute();
engine.log("=== OUTPUT:\n" + taxReturn);
}
}
Here we first created an instance engine of the class OpenRulesEngine and used
it to create an instance decision of the class Decision. We used the engine to get
an example of the object taxReturn that was described in Excel data tables:
DynamicObject taxReturn =
(DynamicObject) engine.run("getTaxReturn");
90
OpenRules, Inc.
OpenRules® User Manual
Then we added this object to the decision:
decision.put("taxReturn",taxReturn);
and simply executed decision:
decision.execute();
The Decision described in “Decision.xls” is supposed to modify certain attributes
inside the object decision and objects which were put inside the decision after its
execution.
Decision Constructors
The class Decision provides the following constructor:
public Decision(String decisionName, String xlsMainFileName)
where “decisionName” is the name of the main table of the type “Decision” and
“xlsMainFileName” is the same parameter as in the OpenRulesEngine’s
constructor that defines a location for the main xls-file.
There is also another constructor:
public Decision(String decisionName, OpenRulesEngine engine)
where the parameter OpenRulesEngine engine refers to an already created
instance of the OpenRulesEngine as in the above example.
Each decision has an associated object of type Glossary. When a decision is created, it
first executes the table “glossary” that must be defined in our rules repository. It fills out the
glossary, a step that applies to all consecutive decision executions. You may always access
the glossary by using the method
Glossary glossary = decision.getGlossary();
Decision Parameters
91
OpenRules, Inc.
OpenRules® User Manual
The class Decision is implemented as a subclass of the standard Java class
HashMap. Thus, you can put any object into the decision similarly as we did
above:
decision.put("taxReturn",taxReturn);
You may access any object previously put into the decision by calling the method
get(name) as in the following example:
TaxReturn taxReturn = (TaxReturn)decision.get("taxReturn");
You may set a special parameter
decision.put("trace",”Off”);
to tell your decision to turn off the tracing . You may use “On” to turn it on again.
Decision Runs
After defining decision parameters, you may execute the decision as follows:
decision.execute();
This method will execute your decision starting from the table of type “Decision”
whose name was specified as the first parameter of the decision’s constructor.
You may reset the parameters of your decision and execute it again without the
necessity of constructing a new decision. This is very convenient for multitransactional systems where you create a decision once by instantiating its
glossary, and then you execute the same decision multiple times but with
different parameters. To make sure that it is possible, the Decision’s method
execute() calls Excel’s method “decisionObjects” each time before actually
executing the decision.
If you know that the types of decision parameters are changed between different
decision runs you may use the following variation of the method “execute”:
92
OpenRules, Inc.
OpenRules® User Manual
decision.execute(true);
The actual execution of “this” decision involves engine runs for the following
Excel methods (in this order):
-
engine.run("decisionObjects",this);
-
engine.run("initializeDecision",this);
-
engine.run("initializeDecisionRun",this);
-
engine.run(this); // run the main decision
-
engine.run("finalizeDecision",this);
All these methods are described in the standard file “DecisionTemplates.xls”.
The method "initializeDecision" is executed only during the first decision run. It
calls the method "customInitializeDecision" that may include an application
specific decision initialization.
The method "initializeDecisionRun" is executed during every decision run. It
calls the method "customInitializeDecisionRun" that may include a code that is
specific for every decision run, e.g. it may analyze the parameters of this run and
redefine some decision variables.
The method "finalizeDecision" is executed after the main Excel table of the type
“Decision” that was specified in the decision’s constructor.
Executing Decision Methods From Excel
There is one more form of this method:
decision.execute(String methodName);
It is used within Excel when you want to execute another Excel method. It is
implemented as follows:
public Object execute(String methodName) {
return getEngine().run(methodName);
}
93
OpenRules, Inc.
OpenRules® User Manual
Decision Glossary
Every decision has an associated business glossary – see above. Glossaries are
usually presented in Excel tables that may look like this table "glossary":
In large, real-world projects the actual content of business concepts such as the
above "Customer" can be defined in external applications using Java-based
Business Object Models or they may come from XML files, a database table, etc.
The list of attributes inside business objects can be very large and/or to be
defined dynamically. In such cases, you do not want to repeat all attributes in
your Excel-based glossary and then worry about keeping the glossary
synchronized with an IT implementation.
It is possible to programmatically define/extend the definition of the Glossary.
For example, we may leave in the Excel's glossary only statically defined
business concepts and their variables, e.g. in the above table we may keep only
the variables of the concept "Response" and remove all rows related to the
concept "Customer". Then in the Java module that creates an object "decision" of
the predefined type Decision we may add the following code:
Decision decision = new Decision(fileName);
String[] attributes = getCustomerAttributes();
String businessConcept = "Customer";
for (int i = 0; i < attributes.length; i++) {
String varName = attributes[i].getName();
94
OpenRules, Inc.
OpenRules® User Manual
decision.getGlossary().put(varName,businessConcept,varName);
}
...
decision.put("customer", customer);
decision.execute();
Here we assume that the method getCustomerAttributes() returns the
names of attributes defined in the class Customer. The variable name and the
attribute name are the same for simplicity - of course you may define them
differently.
You may add multiple concepts to the Glossary in a similar way. In all cases
keep in mind that the table "Glossary glossary" always has to be present in your
Excel repository even when it contain no rows. You also may find that the same
method put(variableName, businessConcept, attributeName) of the
class Glossary is used in the Glossary Template definition in the standard file
"DecisionTemplates.xls".
Business Concepts and Decision Objects
OpenRules® Glossary specifies names of business concepts that contain decision
variables. The connection (mapping) between business concepts and actual
objects that implement these concepts (decision objects) is usually specified in
the Excel table “decisionObjects” that may look like:
The standard mapping is implemented in the DecisionObjectTemplate using the
following Glossary’s method:
95
OpenRules® User Manual
OpenRules, Inc.
void useBusinessObject(String businessConcept, Object object)
What if you want to change actual business objects on the fly during the decision
execution? You can do it by using the same method inside your Excel rules. For
example, you may want to apply the following decision table “EvaluateAssets” for
all elements of an array “assets” of a given customer:
DecisionTable EvaluateAsset
Condition
Asset Name
Is One
Of
Asset12, Asset21, Asset23
Is
Condition
Conclusion
Asset Status
Customer's Assets
Status
Active
Is
Sufficient
In this case you still may specify the business concept “Asset” in your glossary
only once, but you may associate different elements of an array “assets” with the
concept Asset multiple times in the loop similar to the one below:
Method void evaluateCustomerAssets(Customer customer)
Asset[] assets = customer.getAssets();
customer.customerAssetsStatus = "Insufficient";
for(int i=0; i<assets.length; i++) {
getGlossary().useBusinessObject("Asset",customer.assets[i]);
EvaluateAsset();
if ("Sufficient".equals(customer.customerAssetsStatus))
return;
}
Changing Decision Variables Types between Decision Runs
OpenRules® Glossary does not require a user to specify actual types of the
variables - they are automatically defined from the actual types of decision
96
OpenRules, Inc.
OpenRules® User Manual
parameters. It allows you to change types of decision parameters between
decision runs without necessity to download all rules again. If you know that
some attributes corresponding to your decision variables may change their types
between different runs of the same decision, you may use the following Decision's
method:
execute(boolean objectTypesVary)
If the parameter "objectTypesVary" is true then before executing the
decision, the OpenRulesEngine will re-evaluate the decision's glossary and will
reset types of all object attributes based on the actual type of objects passed to
the decision as parameters. By default, the parameter "objectTypesVary" is
false.
Decision Execution Modes
Before executing a decision you may validate it by setting a special “validation”
mode. Here is a code example:
String fileName = "file:rules/main/Decision.xls";
System.setProperty("OPENRULES_MODE", "Validate");
Decision decision = new
Decision("DetermineDecisionVariable",fileName);
During the validation along with regular syntax check OpenRules® will validate
if the values for conditions and actions inside all decision tables correspond to
their glossary domains (if they are defined).
As you can see, the system property "OPENRULES_MODE" defines which mode
to use. By default this property is set to "Execute". If you create an
OpenRulesEngine before creation a Decision, you need to set this property first.
JSR-94 Implementation
OpenRules® provides a reference implementation of the JSR94 standard known
as Java Rule Engine API (see http://www.jcp.org/en/jsr/detail?id=94).
The
complete OpenRules® installation includes the following projects:
97
OpenRules® User Manual
OpenRules, Inc.
JSR-94 Project
Description
This project contains the standard jsr941.0 library
This is an OpenRules®'s reference
implementation for the JSR94 standard
com.openrules.jsr94 and includes the source code. It uses
OpenRulesEngine to implement
RuleExecutionSet
This is an example of using JSR94 for
HelloJsr94
simple rules that generate customized
greetings
HelloJspJsr94 is similar to HelloJsp but
uses the OpenRules® JSR-94 Interface to
HelloJspJsr94
create and run OpenRulesEngine for a web
application.
lib.jsr94
Multi-Threading
OpenRulesEngine is thread-safe and works very efficiently in multi-threaded
environments supporting real parallelism. OpenRulesEngine is stateless, which
allows a user to create only one instance of the class OpenRulesEngine, and then
share this instance between different threads. There are no needs to create a
pool of rule engines. A user may also initialize the engine with application data
common for all threads, and attach this data directly to the engine using the
methods setAttachment(Object attachment). Different threads will receive
this instance of the rule engine as a parameter, and will safely run various rules
in parallel using the same engine.
The complete OpenRules® installation includes an example "HelloFromThreads"
that demonstrates how to organize a parallel execution of the same
OpenRulesEngine's instance in different threads and how to measure their
performance.
INTEGRATION WITH JAVA AND XML
Java Classes
98
OpenRules® User Manual
OpenRules, Inc.
OpenRules® allows you to externalize business logic into xls-files. However,
these files can still use objects and methods defined in your Java environment.
For example, in the standard example “RulesRepository” all rule tables deal
with the Java object Appl defined in the Java package myjava.package1.
Therefore, the proper Environment table inside file Main.xls (see above) contains
a property "import.java" with the value "myjava.package1.*":
The property "import.java" allows you to define all classes
from the package following the standard Java notation, for
example
the
specific
class your rules may need, as in the example above.
You can
define
"hello.*".
a
separate
You
may
property
also
import
only
"import.java"
for
every
Java
package used or merge the property "import.java" into one cell
with many rows for different Java packages.
Here is a more
complex example:
Environment
import.static
import.java
include
com.openrules.tools.Methods
my.bom.*
my.impl.*
my.inventory.*
com.openrules.ml.*
my.package.MyClass
com.3rdparty.*
../include/Rules1.xls
../include/Rules2.xls
Naturally the proper jar-files or Java classes should be in the classpath of the
Java application that uses these rules.
If you want to use static Java methods defined in some standard Java libraries
and you do not want to specify their full path, you can use the property
99
OpenRules, Inc.
OpenRules® User Manual
"import.static". The static import declaration imports static members from
Java classes, allowing them to be used in Excel tables without class
qualification.
For example, many OpenRules® sample projects use static
methods from the standard Java library com.openrules.tools that includes class
Methods.
So, many Environment tables have property "import.static"
defined as "com.openrules.tools.Methods". This allows you to write
out("Rules 1")
instead of
Methods.out("Rules 1")
XML Files
Along with Java classes, OpenRules® tables can use objects defined in XML files.
For example, the standard sample project HelloXMLCustomer uses an object of
type Customer defined in the file Customer.xml located in the project classpath:
<Customer
name="Robinson"
gender="Female"
maritalStatus="Married"
age="55"
/>
The xls-file, HelloXmlCustomer.xls, that deals with this object includes the
following Environment table:
The property, "import.schema", specifies the location of the proper xml-file, in
this case "classpath:/Customer.xml". Of course, you can use any other
100
OpenRules® User Manual
OpenRules, Inc.
location in your local file system that starts with the prefix "file:". This
example also tells you that this Excel file uses:
1. static
Java
methods
defined
in
the
standard
OpenRules®
package
"com.openrules.tools.Methods"
2. xml-file "classpath:/Customer.xml"
3. Java class "Response" from a package "hello"
4. include-file "HelloRules.xls" which is located in the subdirectory "include" of the
directory where the main xls file is located.
The object of the type "Customer" can be created using the following API:
Customer customer = Customer.load("classpath:/Customer.xml");
You may use more complex structures defined in xml-files. For example, the
project HelloXMLPeople uses the following xml-file:
<?xml version="1.0" encoding="UTF-8"?>
<People type="Array of Person(s)">
<Person name="Robinson" gender="Female" maritalStatus="Married"
age="55" />
<Person name="Robinson" gender="Female"
maritalStatus="Single" age="23" />
<Person name="Robinson" gender="Male"
maritalStatus="Single" age="17" />
<Person name="Robinson" gender="Male"
maritalStatus="Single" age="3" />
</People>
The method that launches greeting rules for every Person from an array
People is defined as:
101
OpenRules® User Manual
OpenRules, Inc.
DATA MODELING
OpenRules® includes an ability to define new data/object types and creates the
objects of these types directly in Excel. It allows business analysts to do Rule
Harvesting by defining business terms and facts without worrying about their
implementation in Java, C#, or XML. It also provides the ability to test the
business rules in a pre-integrated mode. To do standalone rule testing, a
designer of rules and forms specifies his/her own data/object types as Excel
tables and creates instances of objects of these types passing them to the rule
tables. We describe how to do it in the sections below.
There is one more important reason why a business or even a technical specialist
may
need
data
modeling
abilities
without
knowing
complex
software
development techniques. In accordance with the SOA principle of loosely coupled
services, rule services have to specify what they actually need from the objects
defined in an external environment. For example, if an object "Insured" includes
attributes related to a person's military services, it does not require that all
business rules that deal with the insured be interested in those attributes. Such
encapsulation of only the essential information in the Excel-based data types,
together with live process modeling, allows OpenRules® to complete the rule
modeling cycle without leaving Excel.
102
OpenRules, Inc.
OpenRules® User Manual
OpenRules® provides the means to make business rules and forms independent
of a concrete implementation of such concepts. The business logic expressed in
the decision tables should not depend on the implementation of the objects these
rules are dealing with. For example, if a rule says: “If driver's age is less than 17
then reject the application” the only thing this business rule should "know" about
the object “driver” is the fact that it has a property “age” and this property has a
type that support a comparison operator “<” with an integer. It is a question of
configuration whether the Driver is a Java class or an XML file or a DB table
from a legacy system. Similarly, if a form has an input field "Driver's Age", the
form should be able to accept a user's input into this field and automatically
convert it into the proper object associated with this field independently of how
this object was implemented.
Thus, OpenRules® supports data source independent business rules (decision
tables) and web forms. Your business rules can work with an object of type
Customer independently of the fact that this type is defined as a Java class, as
an XML file or as an Excel table. You can see how it can be done using examples
HelloJava, HelloXML, and HelloRules from the OpenRules®'s standard
installation. It is a good practice to start with Excel-based data types. Even if you
later on switch to Java classes of other data types, you would always be able to
reuse Excel-based types for standalone testing of your rules-based applications.
Datatype and Data Tables
OpenRules® allows a non-technical user to represent different data types directly in Excel and
to define objects of these types to be used as test data. Actually, it provides the ability to
create Excel-based Data Models, which, in turn, define problem specific business terms and
facts. At the same time, a data model can include data types specified outside Excel, for
example in Java classes or in XML files. Here is an example of a simple data type
"PersonalInfo":
103
OpenRules, Inc.
OpenRules® User Manual
Now we can create several objects of this type "PersonalInfo" using the following data
table:
We can reference to these objects inside rules or forms as in the following snippets:
out(personalInformation["He"].lastName);
if (personalInformation["She"].state.equals("NJ")) ...
You may use one datatype (such as PersonalInfo) to define a more complex aggregate
datatype, like TaxReturn in this example:
104
OpenRules, Inc.
OpenRules® User Manual
You may even create an object of the new composite type "TaxReturn" using references to
the objects "He" and "She" as in this example:
Now we can reference these objects from inside rules or forms as in the following snippet:
out(taxReturn[0].taxPayer.lastName);
The above tables may remind you of traditional database tables simply presented in Excel.
While these examples give you an intuitive understanding of OpenRules® Datatype and Data
tables, the next sections will provide their formal descriptions.
105
OpenRules® User Manual
OpenRules, Inc.
You may use a type of table "Variable". These tables are similar to the Data tables but
instead of arrays of variables they allow you to create separate instances of objects directly in
Excel files. Here is a simple example:
Variable Customer mary
name
age
Name
Age
Mary Brown
5
gender
Gender
Female
maritalStatus
Marital Status
Single
The variable "mary" has type Customer and can be used inside rules or passed back from an
OpenRulesEngine to a Java program as a regular Java object. As usual, the object type
Customer can be defined as a Java class, an Excel Datatype, or an xml structure.
How Datatype Tables Are Organized
Every Datatype table has the following structure:
Datatype tableName
AttributeType1
AttrubuteName1
AttributeType2
AttrubuteName2
..
..
..
..
The first "signature" row consists of two merged cells and starts with the
keyword "Datatype". The "tableName" could be any valid one word identifier of
the table (a combination of letters and numbers). The rows below consist of two
cells with an attribute type and an attribute name. Attribute types can be the
basic Java types:
-
boolean
-
char
-
int
-
double
-
long
-
String (java.lang.String)
-
Date
(java.util.Date)
106
OpenRules, Inc.
OpenRules® User Manual
You may also use data types defined:
-
in other Excel Datatype tables
-
in any Java class with a public constructor with a single parameter of the type String
-
as one-dimensional arrays of the above types.
The datatype "PersonalInfo" gives an example of a very simple datatype. We can
define another datatype for a social security number (SSN):
and add a new attribute of this type to the datatype "PersonalInfo":
It is interesting that these changes do not affect the already existing data
objects defined above (like personalInformation["He"]) - their SSNs just
will not be defined.
Implementation Restriction. Make sure that the very first attribute in a Datatype
table has type String or your own type but not a basic Java type like int.
The following example demonstrates how to create a Data table for a Datatype
that includes one-dimensional arrays:
107
OpenRules® User Manual
OpenRules, Inc.
Datatype Order
String
number
String[]
selectedItems
String[]
double
String
offeredItems
totalAmount
status
Here is an example of the proper Data table:
Data Order orders
number
selectedItems
totalAmount
status
Number
Selected Items
Total
Amount
Status
3700
In Progress
INTRS-PGS394
6P-U01
INTRS-PGS456
Paste-ARMC-2150
You may also present the same data in the following way:
Data Order orders
number
Number
6P-U01
Item 1
INTRSPGS394
selectedItems
totalAmount
Selected Items
Total Amount
Item 2
Item 3
INTRSPaste-ARMC3700
PGS456
2150
How Data Tables Are Organized
Every Datatype table has a vertical or horizontal format. A typical vertical Data table has the
following structure:
Data datatypeName tableName
AttributeName1 AttributeName2
from
from
"datatypeName" "datatypeName"
Display value of Display value of
the
the
AttributeName3
from
...
"datatypeName"
Display value of
...
the
108
OpenRules, Inc.
AttributeName1 AttributeName2 AttributeName3
data
data
data
data
data
data
...
...
...
OpenRules® User Manual
...
...
...
The first "signature" row consists of two merged cells and starts with the
keyword "Data". The next word should correspond to a known datatype: it can be
an already defined Excel Datatype table or a known Java class or an XML file.
The "tableName" is any one word valid identifier of the table (a combination of
letters and numbers).
The second row can consists of cells that correspond to attribute names in the
data type "datatypeName". It is not necessary to define all attributes, but at
least one should be defined. The order of the columns is not important.
The third row contains the display name of each attribute (you may use
unlimited natural language).
All following rows contain data values with types that correspond to the types of
the column attributes.
Here is an example of the Data table for the datatype "PersonalInfo" defined
in the previous section (with added SSN):
The table name is "personalInformation" and it defines an array of objects of
the type PersonalInfo. The array shown consists only of two elements
personalInformation[0] for John and personalInformation[1] for Mary.
You may add as many data rows as necessary.
109
OpenRules, Inc.
OpenRules® User Manual
The attributes after the SSN attribute have not been defined. Please, note that
the references to the aggregated data types are defined in a natural way
(ssn.ssn1, ssn.ssn2, ssn.ssn3) using the dot-convention.
As you can see from this example, the vertical format may not be very convenient
when there are many attributes and not so many data rows. In this case, it could
be preferable to use a horizontal format for the data tables:
Data datatypeName tableName
AttributeName1
Display value of the
data data data ...
from "datatypeName" AttributeName1
AttributeName2 from Display value of the
data data data ...
"datatypeName"
AttributeName2
AttributeName3 from Display value of the
data data data ...
"datatypeName"
AttributeName3
...
..
... ... ... ...
Here is how our data table will look when presented in the horizontal format:
Predefined Datatypes
OpenRules® provides predefined Java classes to create data tables for arrays of
integers, doubles, and strings. The list of predefined arrays includes:
110
OpenRules® User Manual
OpenRules, Inc.
1. ArrayInt - for arrays of integer numbers, e.g.:
Method int[] getTerms()
return ArrayInt.getValues(terms);
Data ArrayInt terms
value
Term
36
72
108
144
2. ArrayDouble - for arrays of real numbers, e.g.:
Method double[] getCosts()
return ArrayDouble.getValues(costs);
Data ArrayDouble costs
value
Costs
$295.50
$550.00
$1,000.00
$2,000.00
$3,295.00
$5,595.00
$8,895.00
3. ArrayString - for arrays of strings, e.g.:
Method String[] getRegions()
return ArrayString.getValues(regions);
Data ArrayString regions
value
Region
NORTHEAST
MID-ATLANTIC
SOUTHERN
MIDWEST
MOUNTAIN
PACIFIC-COAST
111
OpenRules® User Manual
OpenRules, Inc.
These arrays are available from inside an OpenRules® table by just calling their
names: getTerms(), getCosts(), getRegions(). You may also access these
arrays from a Java program, using this code:
OpenRulesEngine engine =
new OpenRulesEngine("file:rules/Data.xls");
int[] terms = (int[])engine.run("getTerms");
The standard installation includes a sample project "DataArrays", that shows
how to deal with predefined arrays.
Accessing Excel Data from Java - Dynamic Objects
You can access objects created in Excel data tables from your Java program.
These objects have a predefined type DynamicObject. Let's assume that you
defined your own Datatype, Customer, and created an array of customers in
Excel:
Data Customer customers
name
maritalStatus
Customer
Marital Status
Name
Robinson
Married
Smith
Single
gender
age
Gender
Age
Female
Male
24
19
Method Customer[] getCustomers()
return customers;
In you Java program you may access these objects as follows:
OpenRulesEngine engine =
new OpenRulesEngine("file:rules/Data.xls");
DynamicObject[] customers =
(DynamicObject[])engine.run("getCustomers");
System.out.println("\nCustomers:");
for(int i=0; i<customers.length; i++)
System.out.println("\t"+customers[i]);
112
OpenRules® User Manual
OpenRules, Inc.
This code will print:
Customer(id=0){
name=Robinson
age=24
gender=Female
maritalStatus=Married
}
Customer(id=1){
name=Smith
age=19
gender=Male
maritalStatus=Single
}
You may use the following methods of the class DynamicObject:
public Object getFieldValue(String name);
public void setFieldValue(String name, Object value);
For example,
String gender = (String) customers[0].getFieldValue("gender");
will return "Female", and the code
customer.setFieldValue("gender", "Male");
customer.setFieldValue("age", 40);
will change the gender of the object customers[0] to "Male" and his age to 40.
How to Define Data for Aggregated Datatypes
When one Datatype includes attributes of another Datatype, such datatypes are
usually known as aggregated datatypes. You have already seen an example of an
aggregated type, PersonalInfo, with the subtype SSN. Similarly, you may
have two datatypes, Person and Address, where type Person has an attribute
"address" of the type Address. You may create a data table with type Person
using aggregated field names such as "address.street", "address.city",
"address.state", etc. The subtype chain may have any length, for example
"address.zip.first5"
or
"address.zip.last4".
This
feature
very
113
OpenRules, Inc.
OpenRules® User Manual
conveniently allows a compact definition of test data for complex interrelated
structures.
Finding Data Elements Using Primary Keys
You may think about a data table as a database table. There are a few things
that make them different from traditional relational tables, but they are
friendlier and easier to use in an object-oriented environment. The very first
attribute in a data table is considered to be its primary key. For example, the
attribute "id" is a primary key in the data table "personalInformation" above.
You may use values like "He" or "She" to refer to the proper elements of this
table/array. For example, to print the full name of the person found in the array
"personalInformation", you may write the following snippet:
PersonalInfo pi = personalInformation["He"];
out(pi.fisrtName + " " + pi.middeInitial + ". "
+ pi.lastName);
Cross-References Between Data Tables
The primary key of one data table could serve as a foreign key in another table
thus providing a cross-reference mechanism between the data tables. There is a
special format for data tables to support cross-references:
Data datatypeName tableName
AttributeName2
AttributeName1 from
AttributeName3 from
from
"datatypeName"
"datatypeName"
"datatypeName"
>referencedDataTable1
>referencedDataTable2
Display value
Display value of the
Display value of the
of the
AttributeName1
AttributeName3
AttributeName2
data
data
data
data
data
data
...
...
...
...
...
...
...
...
114
OpenRules® User Manual
OpenRules, Inc.
This format adds one more row, in which you may add references to the other
data tables, where the data entered into these columns should reside. The sign
">" is a special character that defines the reference, and "referencedDataTable"
is the name of another known data table. Here is an example:
Both
columns
"TaxPayer"
and
"Spouse"
use
the
reference
">personalInformation". It means that these columns may include only primary
keys from the table, "personalInformation". In our example there are only two
valid keys, He or She. If you enter something else, for example "John" instead of
"He" and save your Excel file, you will receive a compile time (!) error "Index Key
John not found" (it will be displayed in your Eclipse Problems windows). It is
extremely important that the cross-references are automatically validated
at compile time in order to prevent much more serious problems at run-time.
Multiple examples of complex inter-table relationships are provided in the
sample rule project AutoInsurance. Here is an intuitive example of three related
data tables:
115
OpenRules, Inc.
OpenRules® User Manual
See more complex examples in the standard project “AutoInsurance”.
OPENRULES® REPOSITORY
To represent business rules OpenRules® utilizes a popular spreadsheet
mechanism and places rules in regular Excel files. OpenRules® allows users to
build enterprise-level rules repositories as hierarchies of inter-related xls-files.
The OpenRules® Engine may access these rules files directly whether they are
located in the local file system, on a remote server, in a standard version control
system or in a relational database.
Logical and Physical Repositories
The following picture shows the logical organization of an OpenRules® repository
and its possible physical implementations:
116
OpenRules, Inc.
OpenRules® User Manual
Logically, OpenRules® Repository may be considered as a hierarchy of rule
workbooks. Each rule workbook is comprised of one or more worksheets that can
be used to separate information by types or categories. Decision tables are the
most typical OpenRules® tables and are used to represent business rules. Along
with rule tables, OpenRules® supports tables of other types such as: Form
Layouts, Data and Datatypes, Methods, and Environment tables. A detailed
description of OpenRules® tables can be found here.
Physically, all workbooks are saved in well-established formats, namely as
standard xls- or xml-files. The proper Excel files may reside in the local file
system, on remote application servers, in a version control system such as
Subversion, or inside a standard database management system.
OpenRules® uses an URL pseudo-protocol notation with prefixes such
as "file:", "classpath:", "http://", "ftp://", "db:", etc.
117
OpenRules® User Manual
OpenRules, Inc.
Hierarchies of Rule Workbooks
An OpenRules® repository usually consists of multiple Excel workbooks
distributed between different subdirectories. Each rule workbook may include
references to other workbooks thus comprising complex hierarchies of interrelated workbooks and rule tables.
Included Workbooks
Rules workbooks refer to other workbooks using so called "includes" inside the
OpenRules® "Environment" tables. To let OpenRules® know about such includerelationships, you have to place references to all included xls-files into the table
"Environment". Here is an example of an OpenRules® repository that comes
with the standard sample project "RuleRepository":
The main xls-file "Main.xls" is located in the local directory
"rules/main". To invoke any rules associated with this file,
the proper Java program creates an OpenRulesEngine using
a string "file:rules/main/Main.xls" as a parameter.
There are many other xls-files related to the Main.xls and
located in different subdirectories of "rules". Here is a
fragment of the Main.xls "Environment" table:
As you can guess, in this instance all included files are defined relative to the
directory "rules/main" in which “Main.xls” resides. You may notice that files
“RulesA11.xls” and “RulesA12.xls” are not included. The reason for this is that
only “RulesA1.xls” really "cares" about these files. Naturally its own table
"Environment" contains the proper "include":
118
OpenRules® User Manual
OpenRules, Inc.
Here, both "includes" are defined relative to the directory "CategoryA" of their
"parent" file “RulesA1.xls”. As an alternative, you may define your included files
relative to a so called "include.path" - see sample in the next section.
Include Path and Common Libraries of Rule Workbooks
Includes provide a convenient mechanism to create libraries of frequently used
xls-files and refer to them from different rule repositories. You can keep these
libraries in a file system with a fixed "include.path". You may even decide to
move such libraries with common xls-files from your local file system to a remote
server. For instance, in our example above you could move a subdirectory "libA"
with
all
xls-files
to
a
new
location
with
an
http
address http://localhost:8080/my.common.lib. In this case, you should first define
a so-called "include.path" and then refer to the xls-files relative to this
include.path using angle brackets as shown below:
Here we want to summarize the following important points:
-
The structure of your rule repository can be presented naturally inside xlsfiles themselves using "includes"
-
The rule repository can include files from different physical locations
-
Complex branches on the rules tree can encapsulate knowledge about their
own organization.
Using Regular Expressions in the Names of Included Files
Large rule repositories may contain many files (workbooks) and it is not
convenient to list all of them by name. In this case you may use regular
expression inside included file names within the Environment table. For
example, consider in the following Environment table:
119
OpenRules® User Manual
OpenRules, Inc.
Environment
include
include
include
../category1/*.xls
../category2/XYZ*.xls
../category3/A?.xls
The first line will include all files with an extension “xls” from the folder
“category1”. The second line will include all files with an extension “xls” and
which names start with “XYZ” from the folder “category2”. The third line will
include all files with an extension “xls” that start with a letter “A” following
exactly one character from the folder “category1”.
Actually along with wildcard characters “*” or “?” you may use any standard
regular expressions to define the entire path to different workbooks.
Imports from Java
OpenRules® allows you to externalize business logic into xls-files. However,
these files still can use objects and methods defined in your Java environment.
For example, in the standard example “RulesRepository” all rule tables deal with
Java objects defined in the Java package myjava.package1. Therefore, the
proper Environment table inside file Main.xls (see above) contains a property
"import.java" with value "myjava.package1.*".
Usually, you only place common Java imports inside the main xls-file. If some
included xls-files use special Java classes you can reference them directly from
inside their own Environment tables.
Imports from XML
Along with Java, OpenRules® allows you to use objects defined in XML files. For
example, the standard sample project “HelloXMLCustomer” uses an object of the
type, Customer, defined in the file Customer.xml located in the project classpath:
<Customer
name="Robinson"
gender="Female"
maritalStatus="Married"
120
OpenRules, Inc.
OpenRules® User Manual
age="55"
/>
The xls-file “HelloCustomer.xls” that deals with this object includes the following
Environment table:
The property "import.schema" specifies the location of the proper xml-file, in
this case "classpath:/Customer.xml". Of course, it could be any other
location in the file system that starts with the prefix "file:". This example also
tells you that this Excel file uses:
1. static Java methods defined in the standard OpenRules® package
"com.openrules.tools.Methods"
2. xml-file "classpath:/Customer.xml"
3. Java class "Response" from a package "hello"
4. include-file "HelloRules.xls" that is located in the subdirectory "include" of
the directory where the main xls file is located.
Parameterized Rule Repositories
An OpenRules® repository may be parameterized in such a way that different
rule workbooks may be invoked from the same repository under different
circumstances. For example, let's assume that we want to define rules that offer
different travel packages for different years and seasons. We may specify a
concrete year and a season by using environment variables YEAR and SEASON.
Our rules repository may have the following structure:
rules/main/Main.xls
rules/common/CommonRules.xls
121
OpenRules® User Manual
OpenRules, Inc.
rules/2007/SummerRules.xls
rules/2007/WinterRules.xls
rules/2008/SummerRules.xls
rules/2008/WinterRules.xls
To make the OpenRulesEngine automatically select the correct rules from such a
repository, we may use the following parameterized include-statements inside
the Environment table of the main xls-file rules/main/Main.xls:
Environment
import.java
season.offers.*
include
../common/SalutationRules.xls
include
../${YEAR}/${SEASON}Rules.xls
Thus, the same rules repository will handle both WinterRules and SummerRules
for different years. A detailed example is provided in the standard project
SeasonRules.
Rules Version Control
For rules version control you can choose any standard version control system
that works within your traditional software development environment. We
would recommend using an open source product "Subversion" that is a
compelling replacement for CVS in the open source community. For business
users, a friendly web interface is provided by a popular open source product
TortoiseSVN. For technical users, it may be preferable to use a Subversion
incorporated into Eclipse IDE. One obvious advantage of the suggested approach
is the fact that both business rules and related Java/XML files will be handled by
the same version control system.
You may even keep your Excel files with rules, data and other OpenRules® tables
directly in Subversion. If your include-statements use http-addresses that point
to a concrete Subversion repository then the OpenRulesEngine will dynamically
122
OpenRules, Inc.
OpenRules® User Manual
access SVN repositories without the necessity to move Excel files back into a file
system.
Another way to use version control is to place your rule workbooks in a database
and use DBV-protocol to access different versions of the rules in run-time read more.
Rules Authoring and Maintenance Tools
OpenRules® relies on standard commonly used tools (mainly from Open Source)
to organize and manage a Business Rules Repository:
123
OpenRules® User Manual
OpenRules, Inc.
To create and edit rules and other tables presented in Excel-files you may use
any standard spreadsheet editors such as:
-
MS Excel™
-
OpenOffice™
-
Google Spreadsheets™
Google Spreadsheets are especially useful for sharing spreadsheet editing - see
section Collaborative Rules Management with Google Spreadsheets.
For technical people responsible for rules project management OpenRules
provides an Eclipse Plug-in that allows them to treat business rules as a natural
part of complex Java projects.
DATABASE INTEGRATION
OpenRules® provides a user with ability to access data and rules defined in
relational databases. There are two aspects of OpenRules® and database
integration:
1. Accessing data located in a database
2. Saving and maintaining rules in a database as Blob objects.
The
detailed
description
of
database
integration
in
provided
at
http://openrules.com/pdf/OpenRulesUserManual.DB.pdf.
EXTERNAL RULES
OpenRules® allows a user to create and maintain their rules outside of Excelbased rule tables. It provides a generic Java API for adding business rules from
different external sources such as:
1. Database tables created and modified by the standard DB management
tools
2. Live rule tables in memory dynamically modified by an external GUI
124
OpenRules® User Manual
OpenRules, Inc.
3. Java objects of the predefined type “RuleTable”
4. Problem-specific rule sources that implement a newly offered rules
provider interface.
With external rules you may keep the business parts of your rules in any
external source while the technical part (Java snippets) will remain in an Excelbased template, based on which actual rules will be created by the
OpenRulesEngine. For example, you may keep your rules in a regular database
table as long as its structure corresponds to the columns (conditions and actions)
of the proper Excel template. Thus, the standard DB management tools, or your
own GUI that maintains these DB-based rule tables, de-facto become your own
rules management environment.
The external rules may also support a preferred distribution of responsibilities
between technical and business people. The business rules can be kept and
maintained in a database or other external source by business analysts while
developers can continue to use Excel and Eclipse to maintain rule templates and
related software interfaces.
The
detailed
description
of
external
rules
in
provided
http://openrules.com/pdf/OpenRulesUserManual.ExternalRules.pdf.
OPENRULES® PROJECTS
Pre-Requisites
OpenRules® requires the following software:
-
Java SE JDK 1.5 or higher
-
Apache Ant 1.6 or higher
-
MS Excel or OpenOffice or Google Docs (for rules and forms editing only)
-
Eclipse SDK (optional, for complex project management only)
Sample Projects
125
at
OpenRules, Inc.
OpenRules® User Manual
The complete OpenRules® installation includes the following workspaces:
openrules.decisions - decision projects
openrules.rules - various rules projects
openrules.dialog – rules-based web questionnaires
openrules.web - rules-based web applications & web services
openrules.solver - constraint-based applications.
Each project has its own subdirectory, e.g. "DecisionHello". OpenRules libraries
®
and related templates are located in the main configuration project,
“openrules.config”, included in each workspace. A detailed description of the
sample projects is provided in the Installation Guide.
Main Configuration Project
OpenRules® provides a set of libraries (jar-files) and Excel-based templates in the folder
“openrules.config” to support different projects.
Supporting Libraries
All OpenRules® jar-files are included in the folder, “openrules.config/lib”.
For the decision management projects you need at least the following jars:







openrules.all.jar
poi-3.6-20091214.jar
commons-logging-1.1.jar (or higher)
commons-logging-api-1.1.jar (or higher)
commons-lang-2.3.jar (or higher)
log4j-1.2.15.jar (or higher)
commons-beanutils.jar (or higher)
There is a supporting library

com.openrules.tools.jar
contains the following optional facilities:
-
operators described in the Java class Operator that can be used inside your
own Rules tables and templates
126
OpenRules® User Manual
OpenRules, Inc.
-
convenience methods like “out(String text)” described in the Java class
Methods
-
a simple JDBC interface DbUtil
-
text
validation
methods
like
“isCreditCardValid(String
text)”
described in the Java class Validator.
If you use the JSR-94 interface you will also need

com.openrules.jsr94.jar
If you use external rules from a database you will also need




openrules.db.jar
openrules.dbv.jar
derby.jar
commons-cli-1.1.jar.
Different workspaces like “openrules.decisions”, “openrules.rules”, etc.
include the proper versions of the folder “openrules.config”.
Predefined Types and Templates
The Excel-based templates that support Decisions and Decision Tables included
in the folder, “openrules.config”:



DecisionTempaltes.xls
DecisionTableExecuteTemplates.xls
DecisionTableValidateTemplates.xls
Sample decision projects include Excel tables of the type “Environment” that
usually refer to “../../../openrules.config/DecisionTemplates.xls”.
You may move all templates to another location and simply modify this reference
making it relative to your main xls-file.
TECHNICAL SUPPORT
Direct all your technical questions to [email protected] or to this
Discussion Group. Read more at http://openrules.com/services.htm.
127
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement