Check / Xtend / Xpand Reference

Introduction
Type System
Expressions
Check
Xtend
Xpand2

Introduction

The oAW4 generator framework provides textual languages, that are useful in different contexts in the MDSD process (e.g. checks, extensions, code generation, model transformation). Each oAW language (Check, Xtend, and Xpand) is built up on a common expression language and type system. Therefore, they can operate on the same models, metamodels and meta-metamodels and you do not need to learn the syntax again and again, because it is always the same.

The expressions framework provides a uniform abstraction layer over different meta-meta-models (e.g. EMF Ecore, Eclipse UML, JavaBeans, XML Schema etc.). Additionally, it offers a powerful, statically typed expressions language, which is used in the various textual languages.

Type System

The abstraction layer on API basis is called a type system. It provides access to built-in types and different registered metamodel implementations. These registered metamodel implementations offer access to the types they provide. The first part of this documentation describes the type system. The expression sub-language is described afterwards in the second part of this documentation. This differentiation is necessary because the type system and the expression language are two different things. The type system is a kind of reflection layer, that can be extended with metamodel implementations. The expression language defines a concrete syntax for executable expressions, using the type system.

The Java API described here is located in the org.openarchitectureware.type package and is a part of the subproject core.expressions.

Types

Every object (e.g. model elements, values, etc.) has a type. A type contains properties and operations. In addition it might inherit from other types (multiple inheritance).

Type Names

Types have a simple name (e.g. String) and an optional namespace used to distingish between two types with the same name (e.g. my::metamodel). The delimiter for name space fragments is a double colon "::". A fully qualified name looks like this:

my::fully::qualified::MetaType

The namespace and name used by a specific type is defined by the corresponding MetaModel implementation. The EmfMetaModel, for instance, maps EPackages to namespace and EClassifiers to names. Therefore, the name of the Ecore element EClassifier is called:

ecore::EClassifier

If you do not want to use namespaces (for whatever reason), you can always implement your own metamodel and map the names accordingly.

Collection Type Names

The built-in type system also contains the following collection types: Collection, List and Set. Because the expressions language is statically type checked and we do not like casts and ClassCastExceptions, we introduced the concept of parameterized types. The type system does not support full featured generics, because we do not need them.

The syntax is:

Collection[my::Type]
List[my::Type]
Set[my::Type]

Features

Each type offers features. The type (resp. the metamodel) is responsible for mapping the features. There are three different kinds of features:

  • Properties
  • Operations
  • Static properties

Properties are straight forward: They have a name and a type. They can be invoked on instances of the corresponding type. The same is true for Operations. But in contrast to properties, they can have parameters. Static properties are the equivalent to enums or constants. They must be invoked statically and they do not have parameters.

Built-In Types

As mentioned before, the expressions framework has several built-in types that define operations and properties. In the following, we will give a rough overview of the types and their features. We will not document all of the operations here, because the built-in types will evolve over time and we want to derive the documentation from the implementation (model-driven, of course). For a complete reference, consult the generated API documentation (http://www.openarchitectureware.org/api/built-ins/).

Object

Object defines a couple of basic operations, like equals(). Every type has to extend Object.

Void

The Void type can be specified as the return type for operations, although it is not recommended, because whenever possible expressions should be free of side effects whenever possible.

Simple types (Data types)

The type system doesn't have a concept data type. Data types are just types. As in OCL, we support the following types: String, Boolean, Integer, Real.

  • String : A rich and convenient String library is especially important for code generation. The type system supports the '+' operator for concatenation, the usual java.lang.String operations ( length() , etc.) and some special operations (like toFirstUpper() , toFirstLower() , regular expressions, etc. often needed in code generation templates).
  • Boolean : Boolean offers the usual operators (Java syntax): &&, ||, !, etc.
  • Integer and Real : Integer and Real offer the usual compare operators (<,>,<=,>=) and simple arithmetics (+,-,*,/). Note that Integer extends Real !

Collection types

The type system has three different Collection types. Collection is the base type, it provides several operations known from java.util.Collection. The other two types (List, Set) correspond to their java.util equivalents, too.

Type system types

The type system describes itself, hence, there are types for the different concepts. These types are needed for reflective programming. To avoid confusion with metatypes with the same name (it is not unusual to have a metatype called Operation, for instance) we have prefixed all of the types with the namespace oaw. We have:

  • oaw::Type
  • oaw::Feature
  • oaw::Property
  • oaw::StaticProperty
  • oaw::Operation

Metamodel Implementations (also known as Meta-Metamodels)

By default, the type system only knows the built-in types. In order to register your own metatypes (e.g. Entity or State), you need to register a respective metamodel implementation with the type system. Within a metamodel implementation the oAW type system elements (Type, Property, Operation) are mapped to an arbitrary other type system (Java reflections, Ecore or XML Schema).

Example JavaMetaModel

For instance, if you want to have the following JavaBean act as a metatype (i.e. your model contains instances of the type):

public class Attribute {
   private String name;
   private String type;
   public String getName() {
      return name;
   }
   public void setName(String name) {
      this.name = name;
   }
   public String getType() {
      return type;
   }
   public void setType(String type) {
      this.type = type;
   }
}

You need to use the JavaMetaModel implementation which uses the ordinary Java reflection layer in order to map access to the model.

So, if you have the following expression in e.g. Xpand:

myattr.name.toFirstUpper()

and myattr is the name of a local variable pointing to an instance of Attribute. The oAW type system asks the metamodel implementations, if they 'know' a type for the instance of Attribute. If you have the JavaMetaModel registered it will return an oaw::Type which maps to the underlying Java class. When the type is asked if it knows a property 'name', it will inspect the Java class using the Java reflection API.

The JavaMetaModel implementation shipped with oAW can be configured with a strategy [GOF95-Pattern] in order to control or change the mapping. For instance, the JavaBeansStrategy maps getter and setter methods to simple properties, so we would use this strategy for the example above.

Eclipse IDE MetaModelContributors

You should know that for each Metamodel implementation you use at runtime, you need to have a so called MetamodelContributor extension for the plugins to work with. If you just use one of the standard metamodel implementations (EMF, UML2 or Java) you don't have to worry about it, since oAW is shipped with respective MetamodelContributors (see the corresponding docs for details). If you need to implement your own MetamodelContributor you should have a look at the Eclipse plug-in reference doc.

Configuring Metamodel implementations with the workflow

You need to configure your oAW language components with the respective metamodel implementations.

A possible configuration of the Xpand2 generator component looks like this:

<component class="oaw.xpand2.Generator">
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelPackage value="my.generated.MetaModel1Package"/>
   </metaModel>
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="my/java/package/metamodel2.ecore"/>
   </metaModel>
   ...
</component>

In this example the EmfMetaModel implementation is configured two times. This means that we want to use two metamodels at the same time, both based on EMF. The metaModelPackage property is a property that is specific to the EmfMetaModel (located in the core.emftools project). It points to the generated EPackages interface. The second meta model is configured using the Ecore file. You do no need to have a generated Ecore model for oAW in order to work. The EmfMetaModel works with dynamic EMF models just as it works with generated EMF models.

Using different Metamodel implementations (also known as Meta-Metamodels)

With oAW you can work on different kinds of Model representations at the same time in a transparent manner. One can work with EMF models, XML DOM models, and simple JavaBeans in the same Xpand-Template. You just need to configure the respective MetaModel implementations.

If you want to do so you need to know how the type lookup works. Let us assume that we have an EMF metamodel and a model based on some Java classes. Then the following would be a possible configuration:

<component class="oaw.xpand2.Generator">
   <metaModel class="oaw.type.impl.java.JavaMetaModel"/>
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="my/java/package/metamodel.ecore"/>
   </metaModel>

   ...
</component>

When the oAW runtime needs to access a property of a given object, it asks the metamodels in the configured order. Let us assume that our model element is an instance of the Java type org.eclipse.emf.ecore.EObject and it is a dynamic instance of an EMF EClass MyType.

We have three Metamodels:

  1. Built-Ins (always the first one)
  2. JavaMetaModel
  3. EMFMetaModel – metamodel.ecore

The first one will return the type Object (not java.lang.Object but Object of oAW). At this point the type Object best fits the request, so it will act as the desired type.

The second metamodel returns an oAW type called oaw::eclipse::emf::ecore::EObject The type system will check if the returned type is a specialization of the current 'best-fit' type (Object). It is, because it extends Object (Every metatype has to extend Object). At this time the type system assumes oaw::eclipse::emf::ecore::EObject to be the desired type.

The third metamodel will return metamodel::MyType which is the desired type. But unfortunately it doesn't extend org::eclipse::emf::ecore::EObject as it has nothing to do with those Java types. Instead it extends emf::EObject which extends Object.

We need to swap the configuration of the two metamodels to get the desired type.

<component class="oaw.xpand2.Generator">
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="my/java/package/metamodel.ecore"/>
   </metaModel>
   <metaModel class="oaw.type.impl.java.JavaMetaModel"/>

   ...
</component>

Expressions

The oAW expression sub-language is a syntactical mixture of Java and OCL. This documentation provides a detailed description of each available expression. Let us start with some simple examples.

Accessing a property:

myModelElement.name

Accessing an operation:

myModelElement.doStuff()

simple arithmetic:

1 + 1 * 2

boolean expressions (just an example:-)):

!('text'.startsWith('t') && ! false)

Literals and special operators for built-in types

There are several literals for built-in types:

Object

There are naturally no literals for object, but we have two operators:

equals:

obj1 == obj2

not equals:

obj1 != obj2

Void

The only possible instance of Void is the null reference. Therefore, we have one literal:

null

Type literals

The literal for types is just the name of the type (no '.class' suffix, etc.). Example:

String // the type string
my::special::Type // evaluates to the type 'my::special::Type'

StaticProperty literals

The literal for static properties (aka enum literals) is correlative to type literals:

my::Color::RED

String

There are two different literal syntaxes (with the same semantics):

'a String literal'
"a String literal" // both are okay

For Strings the expression sub-language supports the plus operator that is overloaded with concatenation:

'my element '+ ele.name +' is really cool!'

Note, that multi-line Strings are supported.

Boolean

The boolean literals are:

true
false

Operators are:

true && false // AND
true || false // OR
! true        // NOT

Integer and Real

The syntax for integer literals is as expected:

// integer literals
3
57278
// real literals
3.0
0.75

Additionally, we have the common arithmetic operators:

3 + 4  // addition
4 - 5  // subtraction
2 * 6  // multiplication
3 / 64 // divide
// Unary minus operator
- 42
- 47.11

Furthermore, the well known compare operators are defined:

4 > 5 // greater than
4 < 5 // smaller than
4 >= 23 // greater equals than
4 <= 12 // smaller equals than

Collections

There is a literal for lists:

{1,2,3,4} // a list with four integers

There is no other special concrete syntax for collections. If you need a set, you have to call the toSet() operation on the list literal:

{1,2,4,4}.toSet() // a set with 3(!) integers

Special Collection operations

Like OCL, the oAW expression sub-language defines several special operations on collections. However, those operations are not members of the type system, therefore you cannot use them in a reflective manner.

select

Sometimes, an expression yields a large collection, but one is only interested in a special subset of the collection. The expression sub-language has special constructs to specify a selection out of a specific collection. These are the select and reject operations. The select specifies a subset of a collection. A select is an operation on a collection and is specified as follows:

collection.select(v | boolean-expression-with-v)

select returns a sublist of the specified collection. The list contains all elements for which the evaluation of boolean-expression-with-v results is true. Example:

{1,2,3,4}.select(i | i >= 3) // returns {3,4}

typeSelect

A special version of a select expression is typeSelect. Rather than providing a boolean expression a class name is here provided.

collection.typeSelect(classname) 

typeSelect returns that sublist of the specified collection, that contains only objects which are an instance of the specified class (also inherited).

reject

The reject operation is similar to the select operation, but with reject we get the subset of all the elements of the collection for which the expression evaluates to false. The reject syntax is identical to the select syntax:

collection.reject(v | boolean-expression-with-v)

Example:

{1,2,3,4}.reject(i | i >= 3) // returns {1,2}

collect

As shown in the previous section, the select and reject operations always result in a sub-collection of the original collection. Sometimes one wants to specify a collection which is derived from another collection, but which contains objects that are not in the original collection (it is not a sub-collection). In such cases, we can use a collect operation. The collect operation uses the same syntax as the select and reject and is written like this:

collection.collect(v | expression-with-v)

collect again iterates over the target collection and evaluates the given expression on each element. In contrast to select, the evaluation result is collected in a list. When an iteration is finished the list with all results is returned. Example:

namedElements.collect(ne | ne.name) // returns a list of strings 

Shorthand for collect (and more than that)

As navigation through many objects is very common, there is a shorthand notation for collect that makes the expressions more readable. Instead of

self.employee.collect(e | e.birthdate) 

one can also write:

self.employee.birthdate

In general, when a property is applied to a collection of Objects, it will automatically be interpreted as a collect over the members of the collection with the specified property.

The syntax is a shorthand for collect, if the feature does not return a collection itself. But sometimes we have the following:

self.buildings.rooms.windows // returns a list of windows

This syntax works, but one cannot express it using the collect operation in an easy way.

forAll

Often a boolean expression has to be evaluated for all elements in a collection. The forAll operation allows specifying a Boolean expression, which must be truefor all objects in a collection in order for the forAll operation to return true:

collection.forAll(v | boolean-expression-with-v)

The result of forAll is true if boolean-expression-with-v is true for all the elements contained in a collection. If boolean-expression-with-v is false for one or more of the elements in the collection, then the forAll expression evaluates to false.

Example:

{3,4,500}.forAll(i | i < 10) // evaluates to false (500 < 10 is false)

exists

Often you will need to know whether there is at least one element in a collection for which a boolean is true. The exists operation allows you to specify a Boolean expression which must be true for at least one object in a collection:

collection.exists(v | boolean-expression-with-v)

The result of the exists operation is true if boolean-expression-with-v is true for at least one element of collection. If the boolean-expression-with-v is false for all elements in collection, then the complete expression evaluates to false.

Example:

{3,4,500}.exists(i | i < 10) // evaluates to true (e.g. 3 < 10 is true)

sortBy[8]

If you want to sort a list of elements, you can use the higher order function sortBy . The list you invoke the sortBy operation on, is sorted by the results of the given expression.

Example:

myListOfEntity.sortBy(entity | entity.name)

In the example the list of entities is sorted by the name of the entities. Note that there is no such Comparable type in oaw. If the values returned from the expression are instances of java.util.Comparable the compareTo method is used, otherwise toString() is invoked and the the result is used.

More Examples – all the following expressions return true:

{'C','B','A'}.sortBy(e | e) == {'A','B','C'}
{'AAA','BB','C'}.sortBy(e | e.length) == {'C','BB','AAA'}
{5,3,1,2}.sortBy(e | e) == {1,2,3,5}
{5,3,1,2}.sortBy(e | e - 2 * e) == {5,3,2,1}
...

if expression

There are two different forms of conditional expressions. The first one is the so-called if expression. Syntax:

condition ? thenExpression : elseExpression

Example:

name != null ? name : 'unknown'

switch expression

The other one is called switch expression. Syntax:

switch (expression) {
   (case expression : thenExpression)*
   default : catchAllExpression
}

The default part is mandatory, because switch is an expression, therefore it needs to evaluate to something in any case. Example:

switch (person.name) {
   case 'Hansen' : 'Du kanns platt schnacken'
   default : 'Du kanns mi nech verstohn!'
}

There is an abbreviation for Boolean expressions:

switch {
   case booleanExpression : thenExpression
   default : catchAllExpression
} 

Chain expression

Expressions and functional languages should be free of side effects as far as possible. But sometimes there you need invocations that do have side effects. In some cases expressions even don not have a return type (i.e. the return type is Void). If you need to call such operations, you can use the chain expression. Syntax:

anExpr ->
anotherExpr ->
lastExpr 

Each expression is evaluated in sequence, but only the result of the last expression is returned. Example:

pers.setName('test') ->
pers

This chain expression will set the name of the person first, before it returns the person object itself.

create expression

The create expression is used to instantiate new objects of a given type:

new TypeName

let expression

The let expression lets you define local variables. Syntax is as follows:

let v = expression : expression-with-v 

This is especially useful together with a chain- and a create expressions. Example:

let p = new Person :
   p.name('John Doe') ->
   p.age(42) ->
   p.city('New York') ->
   p

'GLOBALVAR' expression

Sometimes you don't want to pass everything down the call stack by parameter. Therefore, we have the GLOBALVAR expression. There are two things you need to do, to use global variables within one of the openArchitectureWare languages (Check, Xtend or Xpand).

Using GLOBALVARS to configure workflows

Each workflow component using the expression framework (Xpand, Check and Xtend) can be configured with global variables. Here is an example:

<workflow>
   .... stuff
   <component class="oaw.xpand2.Generator">
      ... usual stuff (see ref doc)
      <globalVarDef name="MyPSM" value="slotNameOfPSM"/>
      <globalVarDef name="ImplClassSuffix" value="'Impl'"/>
   </component>
</workflow>

If you have injected global variables into the respective component, you can call them using the following syntax:

GLOBALVAR ImplClassSuffix

Note, we don't have any static type information. Therefore Object is assumed. So, you have to down cast the global variable to the intended type:

((String) GLOBALVAR ImplClassSuffix)

It is good practice to type it once, using an Extension and then always refer to that extension:

String implClassSuffix() : GLOBALVAR ImplClassSuffix;
// usage of the typed global var extension
ImplName(Class c) :
   name+implClassSuffix();

Multi methods (multiple dispatch)

The expressions language supports multiple dispatching . This means that when there is a bunch of overloaded operations, the decision which operation has to be resolved is based on the dynamic type of all parameters (the implicit 'this' included).

In Java only the dynamic type of the 'this' element is considered, for parameters the static type is used. (this is called single dispatch)

Here is a Java example:

class MyClass {
   boolean equals(Object o) {
      if (o instanceof MyClass) {
         return equals((MyClass)o);
      }
      return super.equals(o);
   }
   boolean equals(MyType mt) {
      //implementation...
   }
} 

The method equals(Object o) would not have to be overwritten, if Java would support multiple dispatch.

Casting

The expression language is statically type checked. Although there are many concepts that help the programmer to have really good static type information, sometimes. one knows more about the real type than the system. To explicitly give the system such an information casts are available. Casts are 100% static, so you do not need them, if you never statically typecheck your expressions!

The syntax for casts is very Java-like:

((String)unTypedList.get(0)).toUpperCase()

Check

openArchitectureWare also provides a language to specify constraints that the model has to fulfill in order to be correct. This language is very easy to understand and use. Basically, it is built around the expression syntax that has been discussed in detail in the previous section. Constraints specified in the Check language have to be stored in files with the file extension .chk . Furthermore, these files have to be on the Java classpath, of course, in order to be found. Let us look at an example, in order to understand, what these constraints look like and what they do:

import data;
context Attribute ERROR
  "Names have to be more than one character long." :
  name.length > 1;

Now, let us look at the example line by line:

  1. First, the metamodel has to be imported.

  2. Then, the context is specified for which the constraint applies. In other words, after the context keyword, we put the name of the metaclass that is going to be checked by the constraint. Then, there follows either ERROR or WARNING, These keywords specify what kind of action will be taken in case the constraint fails:

    Table 1. Types of action for Check constraints

    WARNINGIf the constraint fails, the specified message is printed, but the workflow execution is not stopped.
    ERRORIf the constraint fails, the specified message is printed and all further processing is stopped.


  3. Now, the message that is put in case that the constraint fails is specified as a string. It is possible to include the value of attributes or the return value of functions into the message in order to make the message more clear. For example, it would be possible to improve the above example by rewriting it like this:

    import data;
    context Attribute ERROR
      "Name of '" + name + "too short. Names have to be more than one character long." :
      name.length > 1;
  4. Finally, there is the condition itself, which is specified by an expression, which has been discussed in detail in the previous section. If this expression is true, the constraint is fulfilled.

Important

Please always keep in mind that the message that is associated with the constraint is printed, if the condition of the constraint is false! Thus, if the specified constraint condition is true, nothing will be printed out and the constraint will be fulfilled.

Guard Conditions

In version 4.3 of openArchitectureWare, so called guard conditions have been added to the Check language. These conditions allow to apply a check constraint only to model elements that meet certain criteria. Specifying such a guard condition is done by adding an if clause to the check constraint. The if clause has to be added after the context clause as demonstrated by the following example:

import data;
context Attribute if name.length > 1 ERROR
  "Attribute names have to start with an 'a'" :
  name.startsWith("a");

Xtend

Like the expressions sublanguage that summarizes the syntax of expressions for all the other textual languages delivered with the openArchitectureWare framework, there is another commonly used language called Xtend.

This language provides the possibility to define rich libraries of independent operations and non-invasive metamodel extensions based on either Java methods or oAW expressions. Those libraries can be referenced from all other textual languages, that are based on the expressions framework.

Extend files

An extend file must reside in the Java class path of the used execution context. Additionally it is file extension must be *.ext. Let us have a look at an extend file.

import my::metamodel;extension other::ExtensionFile;

/**
  * Documentation
  */
anExpressionExtension(String stringParam) :
  doingStuff(with(stringParam))
;

/**
  * java extensions are just mappings
  */
String aJavaExtension(String param) : JAVA
  my.JavaClass.staticMethod(java.lang.String)
;

The example shows the following statements:

  1. import statements

  2. extension import statements

  3. expression or java extensions

Comments

We have single- and multi-line comments. The syntax for single line comments is:

// my comment

Multi line comments are written like this:

/* My multi line comment */

Import Statements

Using the import statement one can import name spaces of different types.(see expressions framework reference documentation).

Syntax is:

import my::imported::namespace;

Extend does not support static imports or any similar concept. Therefore, the following is incorrect syntax:

import my::imported::namespace::*; // WRONG! import my::Type; // WRONG!

Extension Import Statement

You can import another extend file using the extension statement. The syntax is:

extension fully::qualified::ExtensionFileName;

Note, that no file extension (*.ext) is specified.

Reexporting Extensions

If you want to export extensions from another extension file together with your local extensions, you can add the keyword 'reexport' to the end of the respective extension import statement.

extension fully::qualified::ExtensionFileName reexport;

Extensions

The syntax of a simple expression extension is as follows:

ReturnType extensionName(ParamType1 paramName1, ParamType2...): expression-using-params;

Example:

String getterName(NamedElement ele) : 'get'+ele.name.firstUpper();

Extension Invocation

There are two different ways of how to invoke an extension. It can be invoked like a function:

getterName(myNamedElement)

The other way to invoke an extension is through the "member syntax":

myNamedElement.getterName()

For any invocation in member syntax, the target expression (the member) is mapped to the first parameter. Therefore, both syntactical forms do the same thing.

It is important to understand that extensions are not members of the type system, hence, they are not accessible through reflection and you cannot specialize or overwrite operations using them.

The expression evaluation engine first looks for an appropriate operation before looking for an extension, in other words operations have higher precedence.

Type Inference

For most extensions, you do not need to specify the return type, because it can be derived from the specified expression. The special thing is, that the static return type of such an extension depends on the context of use.

For instance, if you have the following extension

asList(Object o): {o};

the invocation of

asList('text')

has the static type List[String]. This means you can call

asList('text').get(0).toUpperCase()

The expression is statically type safe, because its return type is derived automatically.

There is always a return value, whether you specify it or not, even if you specify explicitly 'Void'.

See the following example.

modelTarget.ownedElements.addAllNotNull(modelSource.contents.duplicate())

In this example duplicate() dispatches polymorphically. Two of the extensions might look like:

Void duplicate(Realization realization):
   realization.Specifier().duplicate()->
   realization.Realizer().duplicate()
;

create target::Class duplicate(source::Class):
   ...
;

If a 'Realization' is contained in the 'contents' list of 'modelSource', the 'Realizer' of the 'Realization' will be added to the 'ownedElements' list of the 'modelTarget'. If you do not want to add in the case that the contained element is a 'Realization' you might change the extension to:

Void duplicate(Realization realization):
   realization.Specifier().duplicate()->
   realization.Realizer().duplicate() ->
   {}
;

Recursion

There is only one exception: For recursive extensions the return type cannot be inferred, therefore you need to specify it explicitly:

String fullyQualifiedName(NamedElement n) : n.parent == null ? n.name :
   fullyQualifiedName(n.parent)+'::'+n.name
;

Recursive extensions are non-deterministic in a static context, therefore, it is necessary to specify a return type.

Cached Extensions

If you call an extension without side effects very often, you would like to cache the result for each set of parameters, in order improve the performance. You can just add the keyword 'cached' to the extension in order to achieve this:

cached String getterName(NamedElement ele) :
   'get'+ele.name.firstUpper()
;

The getterName will be computed only once for each NamedElement.

Private Extensions

By default all extensions are public, i.e. they are visible from outside the extension file. If you want to hide extensions you can add the keyword 'private' in front of them:

private internalHelper(NamedElement ele) :
   // implementation....
;

Java Extensions

In some rare cases one does want to call a Java method from inside an expression. This can be done by providing a Java extension:

Void myJavaExtension(String param) :
   JAVA my.Type.staticMethod(java.lang.String)
;

The signature is the same as for any other extension. The implementation is redirected to a public static method in a Java class.

Its syntax is:

JAVA fully.qualified.Type.staticMethod(my.ParamType1,
                                       my.ParamType2,
                                       ...)
;

Note that you cannot use any imported namespaces. You have to specify the type, its method and the parameter types in a fully qualified way.

Example:

If you have defined the following Java extension:

Void dump(String s) :
   JAVA my.Helper.dump(java.lang.String)
;

and you have the following Java class:

package my;

public class Helper {
   public final static void dump(String aString) {
      System.out.println(aString);
   }
}

the expressions

dump('Hello world!')
'Hello World'.dump()

both result are invoking the Java method void dump(String aString)

Create Extensions (Model Transformation)

Since Version 4.1 the Xtend language supports additional support for model transformation. The new concept is called create extension and it is explained a bit more comprehensive as usual.

Elements contained in a model are usually referenced multiple times. Consider the following model structure:

    P
   / \
  C1 C2
   \ /
    R

A package P contains two classes C1 and C2. C1 contains a reference R of type C2 (P also references C2).

We could write the following extensions in order to transform an Ecore (EMF) model to our metamodel (Package, Class, Reference).

toPackage(EPackage x) :
   let p = new Package :
      p.ownedMember.addAll(x.eClassifiers.toClass()) ->
      p;

toClass(EClass x) :
   let c = new Class :
      c.attributes.addAll(x.eReferences.toReference()) ->
      c;

toReference(EReference x) :
   let r = new Reference :
      r.setType(x.eType.toClass()) ->
      r;

For an Ecore model with the above structure, the result would be:

    P
   / \
  C1 C2
  |
  R - C2

What happened? The C2 class has been created 2 times (one time for the package containment and another time for the reference R that also refers to C2). We can solve the problem by adding the 'cached' keyword to the second extension:

cached toClass(EClass x) :
   let c = new Class :
      c.attributes.addAll(c.eAttributes.toAttribute()) ->
      c;

The process goes like this:

  1. start create P

    1. start create C1 (contained in P)

      1. start create R (contained in C1)

        1. start create C2 (referenced from R)

        2. end (result C2 is cached)

      2. end R

    2. end C1

    3. start get cached C2 (contained in P)

  2. end P

So this works very well. We will get the intended structure. But what about circular dependencies? For instance, C2 could contain a Reference R2 of type C1 (bidirectional references):

The transformation would occur like this:

  1. start create P

    1. start create C1 (contained in P)

      1. start create R (contained in C1)

        1. start create C2 (referenced from R)

          1. start create R2 (contained in C2)

            1. start create C1 (referenced from R1)... OOPS!

C1 is already in creation and will not complete until the stack is reduced. Deadlock! The problem is that the cache caches the return value, but C1 was not returned so far, because it is still in construction. The solution: create extensions

The syntax is as follows:

create Package toPackage(EPackage x) :
   this.classifiers.addAll(x.eClassifiers.toClass());

create Class toClass(EClass x) :
   this.attributes.addAll(x.eReferences.toReference());

create Reference toReference(EReference x) :
   this.setType(x.eType.toClass());

This is not only a shorter syntax, but it also has the needed semantics: The created model element will be added to the cache before evaluating the body. The return value is always the reference to the created and maybe not completely initialized element.

Calling Extensions From Java

The previous section showed how to implement Extensions in Java. This section shows how to call Extensions from Java.

// setup
XtendFacade f = XtendFacade.create("my::path::MyExtensionFile");

// use
f.call("sayHello",new Object[]{"World"});

The called extension file looks like this:

sayHello(String s) :
   "Hello " + s;

This example uses only features of the BuiltinMetaModel, in this case the "+" feature from the StringTypeImpl.

Here is another example, that uses the JavaBeansMetaModel strategy. This strategy provides as additional feature: the access to properties using the getter and setter methods.

For more information about type systems, see the Expressions reference documentation.

We have one JavaBean-like metamodel class:

package mypackage;
public class MyBeanMetaClass {
   private String myProp;
   public String getMyProp() { return myProp; }
   public void setMyProp(String s) { myProp = s;}
}

in addition to the built-in metamodel type system, we register the JavaMetaModel with the JavaBeansStrategy for our facade. Now, we can use also this strategy in our extension:

// setup facade

XtendFacade f = XtendFacade.create("myext::JavaBeanExtension");

// setup additional type system
JavaMetaModel jmm =
   new JavaMetaModel("JavaMM", new JavaBeansStrategy());

f.registerMetaModel(jmm);

// use the facade
MyBeanMetaClass jb = MyBeanMetaClass();
jb.setMyProp("test");
f.call("readMyProp", new Object[]{jb}));

The called extension file looks like this:

import mypackage;

readMyProp(MyBeanMetaClass jb) :
   jb.myProp
;

WorkflowComponent

With the additional support for model transformation, it makes sense to invoke Xtend within a workflow. A typical workflow configuration of the Xtend component looks like this:

<component class="oaw.xtend.XtendComponent">
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="metamodel1.ecore"/>
   </metamodel>
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="metamodel2.ecore"/>
   </metaModel>
   <invoke value="my::example::Trafo::transform(inputSlot)"/>
   <outputSlot value="transformedModel"/>
</component>

Note that you can mix and use any kinds of metamodels (not only EMF metamodels).

Aspect-Oriented Programming in Xtend (since 4.2)

Using the workflow engine, it is now possible to package (e.g. zip) a written generator and deliver it as a kind of black box. If you want to use such a generator but need to change some things without modifying any code, you can make use of around advices that are supported by Xtend.

The following advice is weaved around every invocation of an extension whose name starts with 'my::generator::':

around my::generator::*(*) :
  log('Invoking ' + ctx.name) -> ctx.proceed()
;

Around advices let you change behaviour in an non-invasive way (you do not need to touch the packaged extensions).

Join Point and Point Cut Syntax

Aspect orientaton is basically about weaving code into different points inside the call graph of a software module. Such points are called join points. In Xtend the join points are the extension invocations (Note that Xpand offers a similar feature, see the Xpand documentation).

One specifies on which join points the contributed code should be executed by specifying something like a 'query' on all available join points. Such a query is called a point cut.

around [pointcut] :
   expression;

A point cut consists of a fully qualified name and a list of parameter declarations.

Extensions Name

The extension name part of a point cut must match the fully qualified name of the definition of the join point. Such expressions are case sensitive. The asterisk character is used to specify wildcards. Some examples:

my::Extension::definition // extensions with the specified name
org::oaw::* //extensions prefixed with 'org::oaw::'
*Operation* // extensions containing the word 'Operation' in it.
* // all extensions

Warning

Be careful when using wildcards, because you will get an endless recursion, in case you weave an extension, which is called inside the advice.

Parameter Types

The parameters of the extensions that we want to add our advice to, can also be specified in the point cut. The rule is, that the type of the specified parameter must be the same or a supertype of the corresponding parameter type (the dynamic type at runtime) of the definition to be called.

Additionally, one can set the wildcard at the end of the parameter list, to specify that there might be none or more parameters of any kind.

Some examples:

my::Templ::extension() // extension without parameters
my::Templ::extension(String s) // extension with exactly one parameter of type String
my::Templ::extension(String s,*) // templ def with one or more parameters,
                                 // where the first parameter is of type String
my::Templ::extension(*) // templ def with any number of parameters
Proceeding

Inside an advice, you might want to call the underlying definition. This can be done using the implicit variable ctx, which is of the type xtend::AdviceContext and provides an operation proceed() which invokes the underlying definition with the original parameters (Note that you might have changed any mutable object in the advice before).

If you want to control what parameters are to be passed to the definition, you can use the operation proceed(List[Object] params). You should be aware, that in advices, no type checking is done.

Additionally, there are some inspection properties (like name, paramTypes, etc.) available.

Workflow configuration

To weave the defined advices into the different join points, you need to configure the XtendComponent with the qualified names of the Extension files containing the advices.

Example:

<component class="oaw.xtend.XtendComponent">
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="metamodel1.ecore"/>
   </metamodel>
   <metaModel class="oaw.type.emf.EmfMetaModel">
      <metaModelFile value="metamodel2.ecore"/>
   </metaModel>

   <invoke value="my::example::Trafo::transform(inputSlot)"/>
      <outputSlot value="transformedModel"/>
   <advices value="my::Advices,my::Advices2"/>
</component>

Model-to-Model transformation with Xtend

This example uses Eclipse EMF as the basis for model-to-model transformations. It builds on the emfExample documented elsewhere. Please read and install the emfExample first.

You can download the example from http://www.eclipse.org/gmt/oaw/download.

The idea in this example is to transform the data model introduced in the EMF example into itself. This might seem boring, but the example is in fact quite illustrative.

Workflow

By now, you should know the role and structure of workflow files. Therefore, the interesting aspect of the workflow file below is the XtendComponent.

<workflow>
   <property file="workflow.properties"/>
      ...
   <component class="oaw.xtend.XtendComponent">
      <metaModel class="oaw.type.emf.EmfMetaModel">
         <metaModelPackage value="data.DataPackage"/>
      </metaModel>
      <invoke value="test::Trafo::duplicate(rootElement)"/>
      <outputSlot value="newModel"/>
   </component>
   ...
</workflow>

As usual, we have to define the metamodel that should be used, and since we want to transform a data model into a data model, we need to specify only the data.DataPackage as the metamodel.

We then specify which function to invoke for the transformation. The statement test::Trafo::duplicate(rootElement) means to invoke:

  • the duplicate function taking the contents of the rootElement slot as a parameter
  • the function can be found in the Trafo.ext file
  • and that in turn is in the classpath, in the test package

The transformation

The transformation, as mentioned above, can be found in the Trafo.ext file in the test package in the src folder. Let us walk through the file.

So, first we import the metamodel.

import data;

The next function is a so-called create extension. Create extensions, as a side effect when called, create an instance of the type given after the create keyword. In our case, the duplicate function creates an instance of DataModel. This newly created object can be referred to in the transformation by this (which is why this is specified behind the type). Since this can be omitted, we do not have to mention it explicitly in the transformation.

The function also takes an instance of DataModel as its only parameter. That object is referred to in the transformation as s. So, this function sets the name of the newly created DataModel to be the name of the original one, and then adds duplicates of all entities of the original one to the new one. To create the duplicates of the entities, the duplicate() operation is called for each Entity. This is the next function in the transformation.

create DataModel this duplicate(DataModel s):
   entity.addAll(s.entity.duplicate()) ->
   setName(s.name);

The duplication function for entities is also a create extension. This time, it creates a new Entity for each old Entity passed in. Again, it copies the name and adds duplicates of the attributes and references to the new one.

create Entity this duplicate(Entity old):
   attribute.addAll(old.attribute.duplicate()) ->
   reference.addAll(old.reference.duplicate()) ->
   setName(old.name);

The function that copies the attribute is rather straight forward, but ...

create Attribute this duplicate(Attribute old):
   setName(old.name) ->
   setType(old.type);

... the one for the references is more interesting. Note that a reference, while being owned by some Entity, also references another Entity as its target. So, how do you make sure you do not duplicate the target twice? Xtend provides explicit support for this kind of situation. Create extensions are only executed once per tuple of parameters! So if, for example, the Entity behind the target reference had already been duplicated by calling the duplicate function with the respective parameter, the next time it will be called the exact same object will be returned. This is very useful for graph transformations.

create EntityReference this duplicate(EntityReference old):
   setName( old.name ) ->
   setTarget( old.target.duplicate() );

For more information about the Xtend language please see the Xtend reference documentation.

Xpand2

The openArchitectureWare framework contains a special language called Xpand Xpand that is used in templates to control the output generation. This documentation describes the general syntax and semantics of the Xpand language.

Typing the guillemets (« and ») used in the templates is supported by the Eclipse editor: which provides keyboard shortcuts with Ctrl+< and Ctrl+>.

Template files and encoding

Templates are stored in files with the extension .xpt. Template files must reside on the Java classpath of the generator process.

Almost all characters used in the standard syntax are part of ASCII and should therefore be available in any encoding. The only limitation are the tag brackets (guillemets), for which the characters "«" (Unicode 00AB) and "»" (Unicode 00BB) are used. So for reading templates, an encoding should be used that supports these characters (e.g. ISO-8859-1 or UTF-8).

Names of properties, templates, namespaces etc. must only contain letters, numbers and underscores.

General structure of template files

Here is a first example of a template:

«IMPORT meta::model»
«EXTENSION my::ExtensionFile»

«DEFINE javaClass FOR Entity»
   «FILE fileName()»
      package «javaPackage()»;

      public class «name» {
         // implementation
      }
   «ENDFILE»
«ENDDEFINE»

A template file consists of any number of IMPORT statements, followed by any number of EXTENSION statements, followed by one or more DEFINE blocks (called definitions).

Statements of the Xpand language

IMPORT

If you are tired of always typing the fully qualified names of your types and definitions, you can import a namespace using the IMPORT statement.

«IMPORT meta::model»

This one imports the namespace meta::model. If your template contains such a statement, you can use the unqualified names of all types and template files contained in that namespace. This is similar to a Java import statement import meta.model.*.

EXTENSION

Metamodels are typically described in a structural way (graphical, or hierarchical, etc.) . A shortcoming of this is that it is difficult to specify additional behaviour (query operations, derived properties, etc.). Also, it is a good idea not to pollute the metamodel with target platform specific information (e.g. Java type names, packages, getter and setter names, etc.).

Extensions provide a flexible and convenient way of defining additional features of metaclasses. You do this by using the Xtend language. (See the corresponding reference documentation for details)

An EXTENSION import points to the Xtend file containing the required extensions:

«EXTENSION my::ExtensionFile»

Note that extension files have to reside on the Java classpath, too. Therefore, they use the same namespace mechanism (and syntax) as types and template files.

DEFINE

The central concept of Xpand is the DEFINE block, also called a template. This is the smallest identifiable unit in a template file. The tag consists of a name, an optional comma-separated parameter list, as well as the name of the metamodel class for which the template is defined.

«DEFINE templateName(formalParameterList) FOR MetaClass»
   a sequence of statements
«ENDDEFINE»

To some extent, templates can be seen as special methods of the metaclass – there is always an implicit this parameter which can be used to address the "underlying" model element; in our example above, this model element is "MetaClass".

As in Java, a formal parameter list entry consists of the type followed by the name of that parameter.

The body of a template can contain a sequence of other statements including any text.

A full parametric polymorphism is available for templates. If there are two templates with the same name that are defined for two metaclasses which inherit from the same superclass, Xpand will use the corresponding subclass template, in case the template is called for the superclass. Vice versa, the template of the superclass would be used in case a subclass template is not available. Note that this not only works for the target type, but for all parameters. Technically, the target type is handled as the first parameter.

So, let us assume you have the following metamodel:

Figure 25. Sample metamodel

Sample metamodel


Assume further, you would have a model which contains a collection of A, B and C instances in the property listOfAs. Then, you can write the following template:

«DEFINE someOtherDefine FOR SomeMetaClass»
   «EXPAND implClass FOREACH listOfAs»
«ENDDEFINE»

«DEFINE implClass FOR A»
   // this is the code generated for the superclass A
«ENDDEFINE»

«DEFINE implClass FOR B»
   // this is the code generated for the subclass B
«ENDDEFINE»

«DEFINE implClass FOR C»
   // this is the code generated for the subclass C
«ENDDEFINE»

So for each B in the list, the template defined for B is executed, for each C in the collection the template defined for C is invoked, and for all others (which are then instances of A) the default template is executed.

FILE

The FILE statement redirects the output generated from its body statements to the specified target.

«FILE expression [outletName]»
   a sequence of statements
«ENDFILE»

The target is a file in the file system whose name is specified by the expression (relative to the specified target directory for that generator run). The expression for the target specification can be a concatenation (using the + operator). Additionally, you can specify an identifier (a legal Java identifier) for the name of the outlet. (See the configuration section for a description of outlets).

The body of a FILE statement can contain any other statements. Example:

«FILE InterfaceName + ".java"»
   package «InterfacePackageName»;

   /* generated class! Do not modify! */
   public interface «InterfaceName» {
      «EXPAND Operation::InterfaceImplementation FOREACH  Operation»
   }
«ENDFILE»


«FILE ImplName + ".java" MY_OUTLET»
   package «ImplPackageName»;

   public class «ImplName» extends «ImplBaseName»
                           implements «InterfaceName» {
   //TODO: implement it
   }
«ENDFILE»

EXPAND

The EXPAND statement "expands" another DEFINE block (in a separate variable context), inserts its output at the current location and continues with the next statement. This is similar in concept to a subroutine call.

«EXPAND definitionName [(parameterList)]
   [FOR expression | FOREACH expression [SEPARATOR expression] ]»

The various alternative syntaxes are explained below.

Names

If the definitionName is a simple unqualified name, the corresponding DEFINE block must be in the same template file.

If the called definition is not contained in the same template file, the name of the template file must be specified. As usual, the double colon is used to delimit namespaces.

«EXPAND TemplateFile::definitionName FOR myModelElement»

Note that you would need to import the namespace of the template file (if there is one). For instance, if the template file resides in the java package my.templates, there are two alternatives. You could either write

«IMPORT my::templates»
...
«EXPAND TemplateFile::definitionName FOR myModelElement»

or

«EXPAND my::templates::TemplateFile::definitionName
        FOR myModelElement»

FOR vs. FOREACH

If FOR or FOREACH is omitted the other template is called FOR this.

«EXPAND TemplateFile::definitionName»

equals

«EXPAND TemplateFile::definitionName FOR this»

If FOR is specified, the definition is executed for the result of the target expression.

«EXPAND myDef FOR entity»

If FOREACH is specified, the target expression must evaluate to a collection type. In this case, the specified definition is executed for each element of that collection.

«EXPAND myDef FOREACH entity.allAttributes»  
Specifying a Separator

If a definition is to be expanded FOREACH element of the target expression it is possible to specify a SEPARATOR expression:

«EXPAND paramTypeAndName FOREACH params SEPARATOR ','»

The result of the separator expression will be written to the output between each evaluation of the target definition (not after each one, but rather only in between two elements. This comes in handy for things such as comma-separated parameter lists).

An EvaluationException will be thrown if the specified target expression cannot be evaluated to an existing element of the instantiated model or no suitable DEFINE block can be found.

FOREACH

This statement expands the body of the FOREACH block for each element of the target collection that results from the expression. The current element is bound to a variable with the specified name in the current context.

«FOREACH expression AS variableName [ITERATOR iterName] [SEPARATOR expression]»
   a sequence of statements using variableName to access the
   current element of the iteration
«ENDFOREACH»

The body of a FOREACH block can contain any other statements; specifically FOREACH statements may be nested. If ITERATOR name is specified, an object of the type xpand2::Iterator (see API doc for details) is accessible using the specified name. The SEPARATOR expression works in the same way as the one for EXPAND.

Example:

«FOREACH {'A','B','C'} AS c ITERATOR iter SEPARATOR ','»
   «iter.counter1» : «c»
«ENDFOREACH»

The evaluation of the above statement results in the following text:

1 : A,
2 : B,
3 : C

IF

The IF statement supports conditional expansion. Any number of ELSEIF statements are allowed. The ELSE block is optional. Every IF statement must be closed with an ENDIF. The body of an IF block can contain any other statement, specifically, IF statements may be nested.

«IF expression»
   a sequence of statements
[ «ELSEIF expression» ]
   a sequence of statements ]
[ «ELSE»
   a sequence of statements ]
«ENDIF»

PROTECT

Protected Regions are used to mark sections in the generated code that shall not be overridden again by the subsequent generator run. These sections typically contain manually written code.

«PROTECT CSTART expression CEND expression ID expression (DISABLE)?»
   a sequence of statements
«ENDPROTECT»

The values of CSTART and CEND expressions are used to enclose the protected regions marker in the output. They should build valid comment beginning and end strings corresponding to the generated target language (e.g. "/*" and "*/" for Java). The following is an example for Java:

«PROTECT CSTART "/*" CEND "*/" ID ElementsUniqueID»
   here goes some content
«ENDPROTECT»

The ID is set by the ID expression and must be globally unique (at least for one complete pass of the generator).

Generated target code looks like this:

public class Person {
/*PROTECTED REGION ID(Person) ENABLED START*/
   This protected region is enabled, therefore the contents will
   always be preserved. If you want to get the default contents
   from the template you must remove the ENABLED keyword (or even
   remove the whole file :-))
/*PROTECTED REGION END*/
}

Protected regions are generated in enabled state by default. Unless you manually disable them, by removing the ENABLED keyword, they will always be preserved.

If you want the generator to generate disabled protected regions, you need to add the DISABLE keyword inside the declaration:

«PROTECT CSTART '/*' CEND '*/' ID this.name DISABLE»

LET

LET lets you specify local variables:

«LET expression AS variableName»
   a sequence of statements
«ENDLET»

During the expansion of the body of the LET block, the value of the expression is bound to the specified variable. Note that the expression will only be evaluated once, independent from the number of usages of the variable within the LET block. Example:

«LET packageName + "." + className AS fqn»
   the fully qualified name is: «fqn»;
«ENDLET»

ERROR

The ERROR statement aborts the evaluation of the templates by throwing an XpandException with the specified message.

«ERROR expression»

Note that you should use this facility very sparingly, since it is better practice to check for invalid models using constraints on the metamodel, and not in the templates.

Comments

Comments are only allowed outside of tags.

«REM»
   text comment
«ENDREM»

Comments may not contain a REM tag, this implies that comments are not nestable. A comment may not have a white space between the REM keyword and its brackets. Example:

«REM»«LET expression AS variableName»«ENDREM»
   a sequence of statements
«REM»  «variableName.stuff»
«ENDLET»«ENDREM»

Expression Statement

Expressions support processing of the information provided by the instantiated metamodel. Xpand provides powerful expressions for selection, aggregation, and navigation. Xpand uses the expressions sublanguage in almost any statement that we have seen so far. The expression statement just evaluates the contained expression and writes the result to the output (using the toString() method of java.lang.Object). Example:

public class «this.name» {

All expressions defined by the oArchitectureWare expressions sublanguage are also available in Xpand. You can invoke imported extensions. (See the Expressions and Xtend language reference for more details).

Controlling generation of whitespace

If you want to omit the output of superfluous whitespace you can add a minus sign just before any closing bracket. Example:

«FILE InterfaceName + ".java"-»
«IF hasPackage-»
package «InterfacePackageName»;
«ENDIF-»
...
«ENDFILE»

The generated file would start with two new lines (one after the FILE and one after the IF statement) if the minus characters had not been set.

In general, this mechanism works as follows: If a statement (or comment) ends with such a minus all preceding whitespace up to the newline character (excluded!) is removed. Additionally all following whitespace including the first newline character (\r\n is handled as one character) is also removed.

Aspect-Oriented Programming in Xpand

Using the workflow engine it is now possible to package (e.g. zip) a written generator and deliver it as a kind of black box. If you want to use such a generator but need to change some small generation stuff, you can make use of the AROUND aspects.

«AROUND qualifiedDefinitionName(parameterList)? FOR type»
   a sequence of statements
«ENDAROUND» 

AROUND lets you add templates in an non-invasive way (you do not need to touch the generator templates). Because aspects are invasive, a template file containing AROUND aspects must be wrapped by configuration (see next section).

Join Point and Point Cut Syntax

AOP is basically about weaving code into different points inside the call graph of a software module. Such points are called Join Points. In Xpand, there is only one join point so far: a call to a definition.

You specify on which join points the contributed code should be executed by specifying something like a 'query' on all available join points. Such a query is called a point cut.

«AROUND [pointcut]»
   do stuff
«ENDAROUND»

A pointcut consists of a fully qualified name, parameter types and the target type.

Definition Name

The definition name part of a point cut must match the fully qualified name of the join point definition. Such expressions are case sensitive. The asterisk character is used to specify wildcards.

Some examples:

my::Template::definition // definitions with the specified name
org::oaw::* // definitions prefixed with 'org::oaw::'
*Operation* // definitions containing the word 'Operation' in it.
*           // all definitions
Parameter Types

The parameters of the definitions we want to add our advice to, can also be specified in the point cut. The rule is that the type of the specified parameter must be the same or a supertype of the corresponding parameter type (the dynamic type at runtime!) of the definition to be called.

Additionally, one can set a wildcard at the end of the parameter list, to specify that there might be an arbitrary number of parameters of any kind.

Some examples:

my::Templ::def() // templ def without parameters
my::Templ::def(String s) // templ def with exactly one parameter
                         // of type String
my::Templ::def(String s,*) // templ def with one or more parameters,
                           // where the first parameter is of type String
my::Templ::def(*) // templ def with any number of parameters
Target Type

Finally, we have to specify the target type. This is straightforward:

my::Templ::def() FOR Object// templ def for any target type
my::Templ::def() FOR Entity// templ def objects of type Entity

Proceeding

Inside an advice, you might want to call the underlying definition. This can be done using the implicit variable targetDef, which is of the type xpand2::Definition and which provides an operation proceed()that invokes the underlying definition with the original parameters (Note that you might have changed any mutable object in the advice before).

If you want to control, what parameters are to be passed to the definition, you can use the operation proceed(Object target, List params). Please keep in mind that no type checking is done in this context.

Additionally, there are some inspection properties (like name, paramTypes, etc.) available.

Generator Workflow Component

This section describes the workflow component that is provided to perform the code generation, i.e. run the templates. You should have a basic idea of how the workflow engine works. (see Workflow reference). A simple generator component configuration could look as follows:

<component class="oaw.xpand2.Generator">
   <fileEncoding value="ISO-8859-1"/>
   <metaModel class="oaw.type.emf.EmfMetaModel">
       <metaModelPackage value="org.eclipse.emf.ecore.EcorePackage"/>
   </metaModel>
   <expand value="example::Java::all FOR myModel"/>

   <!-- aop configuration -->
   <advices value='example::Advices1, example::Advices2'/>

   <!--  output configuration -->
   <outlet path='main/src-gen'/>
   <outlet name='TO_SRC' path='main/src' overwrite='false'/>
   <beautifier class="oaw.xpand2.output.JavaBeautifier"/>
   <beautifier class="oaw.xpand2.output.XmlBeautifier"/>

   <!-- protected regions configuration -->
   <prSrcPaths value="main/src"/>
   <prDefaultExcludes value="false"/>
   <prExcludes value="*.xml"/>
</component>

Now, let us go through the different properties one by one.

Main configuration

The first thing to note is that the qualified Java name of the component is org.openarchitectureware.xpand2.Generator. One can use the shortcut oaw instead of a preceding org.openarchitectureware. The workflow engine will resolve it.

Encoding

For Xpand, it is important to have the file encoding in mind because of the guillemet characters used to delimit keywords and property access. The fileEncoding property specifies the file encoding to use for reading the templates, reading the protected regions and writing the generated files. This property defaults to the default file encoding of your JVM.

Metamodel

The property metaModel is used to tell the generator engine on which metamodels the Xpand templates should be evaluated. One can specify more than one metamodel here. Metamodel implementations are required by the expression framework (see Expressions) used by Xpand2. In the example above we configured the Ecore metamodel using the EMFMetaModel implementation shipped with the core part of the openArchitectureWare 4 release.

A mandatory configuration is the expand property. It expects a syntax similar to that of the EXPAND statement (described above). The only difference is that we omit the EXPAND keyword. Instead, we specify the name of the property. Examples:

<expand value="Template::define FOR mySlot"/>

or:

<expand value="Template::define('foo') FOREACH {mySlot1,mySlot2}"/>

The expressions are evaluated using the workflow context. Each slot is mapped to a variable. For the examples above the workflow context needs to contain elements in the slots 'mySlot', 'mySlot1' and 'mySlot2'. It is also possible to specify some complex expressions here. If, for instance, the slot myModel contains a collection of model elements one could write:

<expand value="Template::define FOREACH myModel.typeSelect(Entity)"/>

This selects all elements of type Entity contained in the collection stored in the myModel slot.

Output configuration

The second mandatory configuration is the specification of so called outlets (a concept borrowed from AndroMDA). Outlets are responsible for writing the generated files to disk. Example:

<component class="oaw.xpand2.Generator2">
   ...
   <outlet path='main/src-gen'/>
   <outlet name='TO_SRC' path='main/src' overwrite='false'/>
   ...
</component>

In the example there are two outlets configured. The first one has no name and is therefore handled as the default outlet. Default outlets are triggered by omitting an outlet name:

«FILE 'test/note.txt'»
# this goes to the default outlet
«ENDFILE»

The configured base path is 'main/src-gen', so the file from above would go to 'main/src-gen/test/note.txt'.

The second outlet has a name ('TO_SRC') specified. Additionally the flag overwrite is set to false (defaults to true). The following Xpand fragment

«FILE 'test/note.txt' TO_SRC»
# this goes to the TO_SRC outlet
«ENDFILE»

would cause the generator to write the contents to 'main/src/test/note.txt' if the file does not already exist (the overwrite flag).

Another option called append (defaults to false) causes the generator to append the generated text to an existing file. If overwrite is set to false this flag has no effect.

Beautifier

Beautifying the generated code is a good idea. It is very important that generated code looks good, because developers should be able to understand it. On the other hand template files should look good, too. It is thus best practice to write nice looking template files and not to care how the generated code looks – and then you run a beautifier over the generated code to fix that problem. Of course, if a beautifier is not available, or if white space has syntactical meaning (as in Python), you would have to write your templates with that in mind (using the minus character before closing brackets as described in a preceding section).

The Xpand workflow component can be configured with multiple beautifiers:

<beautifier
   class="org.openarchitectureware.xpand2.output.JavaBeautifier"/>
<beautifier
   class="org.openarchitectureware.xpand2.output.XMLBeautifier"/>

These are the two beautifiers delivered with Xpand. If you want to use your own beautifier, you would just need to implement the PostProcessor Java interface:

package org.openarchitectureware.xpand2.output;

public interface PostProcessor {
   public void beforeWriteAndClose(FileHandle handle);
   public void afterClose(FileHandle handle);
}

The beforeWriteAndClose method is called for each ENDFILE statement.

JavaBeautifier

The JavaBeautifier is based on the Eclipse Java formatter provides base beautifying for Java files.

XmlBeautifier

The XmlBeautifier is based on dom4j and provides a single option fileExtensions (defaults to ".xml, .xsl, .wsdd, .wsdl") used to specify which files should be pretty-printed.

Protected Region Configuration

Finally, you need to configure the protected region resolver, if you want to use protected regions.

<prSrcPaths value="main/src"/>
<prDefaultExcludes value="false"/>
<prExcludes value="*.xml"/>

The prSrcPaths  property points to a comma-separated list of directories. The protected region resolver will scan these directories for files containing activated protected regions.

There are several file names which are excluded by default:

RCS, SCCS, CVS, CVS.adm, RCSLOG, cvslog.*, tags, TAGS, .make.state, .nse_depinfo, *~, #*,
.#*, ',*', _$*,*$, *.old, *.bak, *.BAK, *.orig, *.rej, .del-*, *.a, *.olb, *.o, *.obj,
 *.so, *.exe, *.Z,* .elc, *.ln, core, .svn

If you do not want to exclude any of these, you must set prDefaultExcludes to false.

<prDefaultExcludes value="false"/>

If you want to add additional excludes, you should use the prExcludes property.

<prExcludes value="*.xml,*.hbm"/>

Note

It is bad practice to mix generated and non-generated code in one artifact. Instead of using protected regions, you should try to leverage the extension features of the used target language (inheritance, inclusion, references, etc.) wherever possible. It is very rare that the use of protected regions is an appropriate solution.

VetoStrategy

The Xpand engine will generate code for each processed FILE statement. This implies that files are written that might not have changed to the previous generator run. Normally it does not matter that files are rewritten. There are at least two good reasons when it is better to avoid rewriting of files:

  1. The generated source code will be checked in. In general it is not the recommended way to go to check in generated code, but sometimes you will have to. Especially with CVS there is the problem that rewritten files are recognized as modified, even if they haven't changed. So the problem arises that identical files get checked in again and again (or you revert it manually). When working in teams the problem even becomes worse, since team members will have conflicts when checking in.

  2. When it can be predicted that the generator won't produce different content before a file is even about to be created by a FILE statement then this can boost performance. Of course it is not trivial to predict that a specific file won't result in different content before it is even created. This requires information from a prior generator run and evaluation against the current model to process. Usually a diff model would be used as input for the decision.

Case 1) will prevent file writing after a FILE statement has been evaluated, case 2) will prevent creating a file at all.

To achieve this it is possible to add Veto Strategies to the generator, which are implementations of interface org.openarchitectureware.xpand2.output.VetoStrategy or org.openarchitectureware.xpand2.output.VetoStrategy2. Use VetoStrategy2 if you implement your own.

VetoStrategy2 declares two methods:

  • boolean hasVetoBeforeOpen (FileHandle)

    This method will be called before a file is being opened and generated. Return true to suppress the file creation.

  • boolean hasVeto (FileHandle)

    This method will be called after a file has been produced and after all configured PostProcessors have been invoked. Return true to suppress writing the file.

Veto Strategies are configured per Outlet. It is possible to add multiple stratgy instances to each Outlet.

  <component id="generator" class="oaw.xpand2.Generator" skipOnErrors="true">
    <metaModel class="oaw.uml2.UML2MetaModel"/>
    <expand value="templates::Root::Root FOR model"/>
    <fileEncoding value="ISO-8859-1"/>
      <outlet path="src-gen">
         <postprocessor class="oaw.xpand2.output.JavaBeautifier"/>
         <vetoStrategy class="oaw.xpand2.output.NoChangesVetoStrategy"/>
      </outlet>
  </component>

One VetoStrategy is already provided. The org.openarchitectureware.xpand2.output.NoChangesVetoStrategy is a simple implementation that will compare the produced output, after it has been postprocessed, with the target file. If the content is identical the strategy vetoes the file writing. This strategy is effective, but has two severe drawbacks:

  1. The file has been created at least in memory before. This consumes time and memory. If applying code formatting this usually implies that the file is temporarily written.

  2. The existing file must be read into memory. This also costs time and memory.

Much better would be to even prevent the creation of files by having a valid implementation for the hasVetoBeforeOpen() method. Providing an implementation that predicts that files do not have to be created requires domain knowledge, thus a standard implementation is not available.

The number of skipped files will be reported by the Generator component like this:

2192 INFO  - Generator(generator): generating <...>
3792 INFO  - Skipped writing of 2 files to outlet [default](src-gen)

Example for using Aspect-Oriented Programming in Xpand

This example shows how to use aspect-oriented programming techniques in Xpand templates. It is applicable to EMF based and Classic systems. However, we explain the idea based on the emfExample – hence you should read that before.

The Problem

There are many circumstances when template-AOP is useful. Here are two examples:

Scenario 1: Assume you have a nice generator that generates certain artifacts. The generator (or cartridge) might be a third party product, delivered in a single JAR file. Still you might want to adapt certain aspects of the generation process – without modifying the original generator.

Scenario 2: You are building a family of generators that can generate variations of the generate code, e.g. Implementations for different embedded platforms. In such a scenario, you need to be able to express those differences (variabilities) sensibly without creating a non-understandable chaos of if statements in the templates.

Example

To illustrate the idea of extending a generator without "touching" it, let us create a new project called oaw4.demo.emf.datamodel.generator-aop. The idea is that it will "extend" the original oaw4.demo.emf.datamodel.generator project introduced in the emfExample. So this new projects needs to have a project dependency to the former one.

Templates

An AOP system always needs to define a join point model; this is, you have to define, at which locations of a (template) program you can add additional (template) code. In Xpand, the join points are simply templates (i.e. DEFINE .. ENDDEFINE) blocks. An "aspect template" can be declared AROUND previously existing templates. If you take a look at the oaw4.demo.emf.datamodel.generator source folder of the project, you can find the Root.xpt template file. Inside, you can find a template called Impl that generates the implementation of the JavaBean.

«DEFINE Entity FOR data::Entity»
   «FILE baseClassFileName() »
      // generated at «timestamp()»
      public abstract class «baseClassName()» {
         «EXPAND Impl»
      }
   «ENDFILE»
«ENDDEFINE»

«DEFINE Impl FOR data::Entity»
   «EXPAND GettersAndSetters»
«ENDDEFINE»

«DEFINE Impl FOR data::PersistentEntity»
   «EXPAND GettersAndSetters»
    public void save() {

   }
«ENDDEFINE»

What we now want to do is as follows: Whenever the Impl template is executed, we want to run an additional template that generates additional code (for example, some kind of meta information for frameworks – the specific code is not important for the example here).

So, in our new project, we define the following template file:

«AROUND Impl FOR data::Entity»
   «FOREACH attribute AS a»
      public static final AttrInfo «a.name»Info = new AttrInfo(
         "«a.name»", «a.type».class );
   «ENDFOREACH»
   «targetDef.proceed()»
«ENDAROUND»

So, this new template wraps around the existing template called Impl It first generates additional code and then forwards the execution to the original template using targetDef.proceed(). So, in effect, this is a BEFORE advice. Moving the proceed statement to the beginning makes it an AFTER advice, omitting it, makes it an override.

Workflow File

Let us take a look at the workflow file to run this generator:

<workflow>
   <cartridge file="workflow.oaw"/>
   <component adviceTarget="generator"
              id="reflectionAdvice"
              class="oaw.xpand2.GeneratorAdvice">
      <advices value="templates::Advices"/>
   </component>
</workflow>

Mainly, what we do here, is to call the original workflow file. It has to be available from the classpath. After this cartridge call, we define an additional workflow component, a so called advice component. It specifies generator as its adviceTarget. That means, that all the properties we define inside this advice component will be added to the component referenced by name in the adviceTarget instead. In our case, this is the generator. So, in effect, we add the <advices value="templates::Advices" /> to the original generator component (without invasively modifying its own definition). This contributes the advice templates to the generator.

Running the new generator

Running the generator produces the following code:

public abstract class PersonImplBase {
   public static final AttrInfo
      nameInfo = new AttrInfo("name", String.class);
   public static final AttrInfo
      name2Info = new AttrInfo("name2", String.class);
   private String name;
   private String name2;

   public void setName(String value) {
      this.name = value;
   }

   public String getName() {
      return this.name;
   }

   public void setName2(String value) {
      this.name2 = value;
   }

   public String getName2() {
      return this.name2;
   }
}

More Aspect Orientation

In general, the syntax for the AROUND construct is as follows:

<<AROUND fullyQualifiedDefinitionNameWithWildcards
      (Paramlist (*)?) FOR TypeName>>
   do Stuff
<<ENDAROUND>>

Here are some examples:

<<AROUND *(*) FOR Object>>

matches all templates

<<AROUND *define(*) FOR Object>>

matches all templates with define at the end of its name and any number of parameters

<<AROUND org::oaw::* FOR Entity>>

matches all templates with namespace org::oaw:: that do not have any parameters and whose type is Entity or a subclass

<<AROUND *(String s) FOR Object>>

matches all templates that have exactly one String parameter

<<AROUND *(String s,*) FOR Object>>

matches all templates that have at least one String parameter

<<AROUND my::Template::definition(String s) FOR Entity>>

matches exactly this single definition

Inside an AROUND, there is the variable targetDef, which has the type xpand2::Definition. On this variable, you can call proceed, and also query a number of other things:

<<AROUND my::Template::definition(String s) FOR String>>
   log('invoking '+<<targetDef.name>>+' with '+this)
   <<targetDef.proceed()>>
<<ENDAROUND>>


[8] since 4.1.2