View on GitHub


A Modular Extension for Xtend

Download this project as a .zip file Download this project as a tar.gz file

Xtend2m turns Xtend into a model transformation language that promotes modular programming. Xtend is a versatile, extendable language with a functional flavor. With a concise syntax close to the Object Constraint Language, Xtend qualifies for model transformation development. To increase maintainability of larger and more complex transformations, this work adds a customized module concept to the Xtend language. It is the first model transformation language that features a module concept with information hiding support.


Development and maintenance of model transformations make up a substantial share of the lifecycle costs of software products that rely on model-driven techniques. In particular large and heterogeneous models lead to poorly understandable transformation code due to missing language concepts to master complexity.

Xtend2m embeds a module concept for model transformation languages into the Xtend language version 2 (as opposed to its obsolete version 1). Xtend2m allows programmers to control information hiding and strictly declare model and code dependencies at module interfaces. With Xtend2m, we can break down transformation logic into smaller parts, so that each part owns a clear interface for separating concerns. To meet the special demands of transformations, module interfaces give control over both model and code accessibility.

Designing model transformations in a modular fashion brings real benefits. In a case study, we examined the effort required to carry out two typical maintenance tasks on a real- world transformation written in Xtend. We are able to attest a significant reduction of effort, thereby demonstrating the practical effects of the interface concept on the maintainability of model transformations.


Xtend2m uses Xtend's concept of Active Annotations to non-invasively enhance the base language with several concepts:

If a module implementation violates its interface, e.g., if access to a model element id not allowed, errors are indicated in the editor (static type checking). The following code snipped gives an impression on how an example module interface and an implementation in Xtend2m is programmed.

interface IAction2Step {
    def Step mapAction2Step(Action self)
class Action2Step implements IAction2Step {
    def override Step mapAction2Step(Action self) { =
        self.succ.lateResolveOne [ = it ]
        result.isStart = self instanceof StartAction
        result.isStop = self instanceof StopAction

The module provides one mapping, mapAction2Step, that transforms instances of class Action to instances of class Step. Using annotations @ScopeIn and @ScopeIn, the implementation is not allowed to access any instances other than StartAction, StopAction, and Step.


The following steps describe a particular version of Eclipse, older/newer versions might work, as well:

  1. Install a Java SE runtime environment version 1.8.0_05-b13 (Newer and older versions, like Java 7, reportedly worked, as well);
  2. Download Eclipse 4.4.0 Luna Modeling Edition (Build 20140612-0600) and launch it;
  3. Install through menu Help > Install Modeling Components... Eclipse Xtext 2.6.1 (Xtend SDK 2.6.1.v201405210727) of the Model Development Tools (MDT) project;
  4. Choose Help > Check for Updates... to update to the latest Xtext. At the time of writing, it is required to add the Xtext Composite Update Site (Nightly) to the list of available update sites to obtain version 2.7.0 (Xtend SDK 2.7.0.v201407030400, earlier or later versions may require adaptations of our plugins);
  5. Download Xtend2m and import contained projects through File > Import > Existing Projects into Workspace… into your Eclipse workspace;
  6. Import project org.eclipse.xtend.core from origin/master (no tag exists for 2.7.0 yet as of Jul 14, 2014) of repository git:// and apply patch for model scope checking (see project org.eclipse.xtend.core-patch for instructions).

See Also



Work has partly been funded by the German Research Foundation (DFG) under grant No. RE 1674/5-1: Model-Driven Methods and Tools for Performance Prediction and Capacity Planning of Component-Based Software Systems.