Grule是Golang实现的规则引擎库， 受到著名的JBOSS Drools的启发

"Gopher Holds The Rules"

Grule

import "github.com/hyperjumptech/grule-rule-engine"

Rule Engine for Go

Grule is a Rule Engine library for Golang programming language. Inspired by the acclaimed JBOSS Drools, done in a much simple manner.

Like Drools, Grule have its own DSL comparable as follows.

Drools's DRL be like :

rule "SpeedUp"
    salience 10
    when
        $TestCar : TestCarClass( speedUp == true && speed < maxSpeed )
        $DistanceRecord : DistanceRecordClass()
    then
        $TestCar.setSpeed($TestCar.Speed + $TestCar.SpeedIncrement);
        update($TestCar);
        $DistanceRecord.setTotalDistance($DistanceRecord.getTotalDistance() + $TestCar.Speed)
        update($DistanceRecord)
end

And Grule's GRL be like :

rule SpeedUp "When testcar is speeding up we keep increase the speed." salience 10  {
    when
        TestCar.SpeedUp == true && TestCar.Speed < TestCar.MaxSpeed
    then
        TestCar.Speed = TestCar.Speed + TestCar.SpeedIncrement;
        DistanceRecord.TotalDistance = DistanceRecord.TotalDistance + TestCar.Speed;
}

What is RuleEngine?

There are no better explanation compared to the article authored by Martin Fowler. You can read the article here (RulesEngine by Martin Fowler).

Taken from TutorialsPoint website (with very slight modification),

Grule is Rule Engine or a Production Rule System that uses the rule-based approach to implement and Expert System. Expert Systems are knowledge-based systems that use knowledge representation to process acquired knowledge into a knowledge base that can be used for reasoning.

A Production Rule System is Turing complete with a focus on knowledge representation to express propositional and first-order logic in a concise, non-ambiguous and declarative manner.

The brain of a Production Rules System is an Inference Engine that can scale to a large number of rules and facts. The Inference Engine matches facts and data against Production Rules – also called Productions or just Rules – to infer conclusions which result in actions.

A Production Rule is a two-part structure that uses first-order logic for reasoning over knowledge representation. A business rule engine is a software system that executes one or more business rules in a runtime production environment.

A Rule Engine allows you to define “What to Do” and not “How to do it.”

What is a Rule?

(also taken from TutorialsPoint)

Rules are pieces of knowledge often expressed as, "When some conditions occur, then do some tasks."

When
   <Condition is true>
Then
   <Take desired Action>

The most important part of a Rule is its when part. If the when part is satisfied, the then part is triggered.

rule  <rule_name> <rule_description>
   <attribute> <value> {
   when
      <conditions>

   then
      <actions>
}

Advantages of a Rule Engine

Declarative Programming

Rules make it easy to express solutions to difficult problems and get the solutions verified as well. Unlike codes, Rules are written in less complex language; Business Analysts can easily read and verify a set of rules.

Logic and Data Separation

The data resides in the Domain Objects and the business logic resides in the Rules. Depending upon the kind of project, this kind of separation can be very advantageous.

Centralization of Knowledge

By using Rules, you create a repository of knowledge (a knowledge base) which is executable. It is a single point of truth for business policy. Ideally, Rules are so readable that they can also serve as documentation.

Agility To Change

Since business rules are actually treated as data. Adjusting the rule according to business dynamic nature become trivial. No need to re-build codes, deploy as normal software development do, you only need to roll out sets of rule and apply them to knowledge repository.

Hello Grule

Grule Rule Language (GRL)

The GRL is a DSL (Domain Specific Language) designed for Grule. Its a simplified language to be used for defining rule evaluation criteria and action to be executed if the criteria were met.

Generally, the language have the following structure :

RuleName identify a specific rule. The name should be unique in the entire knowledge base, consist of one word thus it should not contains white-space.

RuleDescription describes the rule. The description should be enclosed with a double-quote.

Salience defines the importance of the rule. Its an optional rule configuration, and by default, when you don't specify them, all rule have the salience of 0 (zero). The lower the value, the less important the rule. Whenever multiple rule are a candidate for execution, highest salience rule will be executed first. You may define negative value for the salience, to make the salience even lower. Like any implementation of Rule-Engine, there are no definitive algorithm to specify which rule to be execute in case of conflicting candidate, the engine may run which ever they like. Salience is one way of hinting the rule engine of which rule have more importance compared to the other.

Boolean Expression is an expression that will be used by rule engine to identify if that specific rule are a candidate for execution for the current facts.

Assignment or Operation Expression contains list of expressions (each expression should be ended with ";" symbol.) The expression are designed to modify the current fact values, making calculation, make some logging, etc.

Boolean Expression

Boolean expression comes natural for java or golang developer in GRL.

Constants and Literals

13442344-553234.4553-234.3trueTRUEFalse

+-/*&&||<<=>>===!=

Comments

You can always put a comment inside your GRL script. Such as :

Examples

Loading GRL on to Knowledge

One knowledge base may consist of many rules. Tens to hundreds of rules. They may be loaded from multiple sources. Those rules will go to the same knowledge as long as you use the same knowledge when loading each of the resource.

Before you load any rule, you need to have your own knowledge

From File

From String or ByteArray

From URL

Preparing Facts

struct

struct

And then you instantiate those struct.

Add those struct instances (fact) into data context.

Now your fact is ready to be executed in the rule engine that already prepared with some knowledge.

Executing A Knowledge On Facts and get result

You already know how to load rules into knowledge base, and you also know how to prepare your fact. Now we can execute the knowledge base agains facts.

The rule engine will use loaded knowledge base to work upon sets of fact data in data context.

Calling Function in Grule

All invocable functions which are invocable from the DataContext is Invocable from within the rule, both in the "When" scope and "Then" scope.

Assuming you have a struct with some functions.

And add the struct into knowledge base

You can call the fuction within the rule

Default Functions

model/GruleFunctions.go

model/GruleFunctions.go

Important Thing you must know about Custom Function in Grule

When you make your own function to be called from rule engine, you need to know the following rules.

int64float64int64float64

RETE Algorithm

From Wikipedia : The Rete algorithm (/ˈriːtiː/ REE-tee, /ˈreɪtiː/ RAY-tee, rarely /ˈriːt/ REET, /rɛˈteɪ/ reh-TAY) is a pattern matching algorithm for implementing rule-based systems. The algorithm was developed to efficiently apply many rules or patterns to many objects, or facts, in a knowledge base. It is used to determine which of the system's rules should fire based on its data store, its facts.

grule-rule-engine1.1.0ConflictSet

ExpressionAtom

Class

The expression above involve attribute/function result comparison and math operation from 3 different class. This makes RETE's class separation of expression token difficult.

You can read about RETE algorithm here:

Tasks and Help Wanted.

• Need to do more and more and more test.
• Better code coverage test.
• Better commenting for go doc best practice.
• Improve function argument handling to be more fluid and intuitive.