Home 

Introduction

Operation Schemas

User Manual

Directory Structure

Parser

Model Extractor

Type Checker

Modifications

Extending the compiler

Outstanding Issues

Current Status

References and Links

Javadoc
Model Extractor

Any model information source will have to implement the lglepfl.ocl.check.types.ModelFacade interface. This interface contains just one method, Type getClassifier(String), which takes a String as a parameter and returns a Type object corresponding to the type for the parameter string in the model.

There are at present two implementations for ModelFacade:

  1. A hard-coded implementation of the RoyLoy model contained in the lglepfl.ocl.check.types.testfacade package. 
  2. The implementation in lglepfl.ocl.check.types.xmifacade, which obtains model information by parsing an XMI file. 
The class which implements ModelFacade for the XMI parser implementation is lglepfl.ocl.check.types.xmifacade.Model. The XMI file itself is parsed by the class lglepfl.ocl.check.types.xmifacade.XmiParser, which adds the operations and attributes to a lglepfl.ocl.check.types.xmifacade.ModelClass object, and the ModelClass objects are themselves added to a lglepfl.ocl.check.types.xmifacade.Model object. ModelClass implements the Type interface, so if the getClassifier method of Model is called, it looks for the Type object in the set of ModelClasses it has, and returns the appropriate Type object. ModelClass has sets containing all the attributes and operations, so it can easily implement the navigateParametrized() and navigateQualified() methods of the Type interface.

The XmiParser class uses the standard Apache Xerces XMI parser v1.4 (contained in the jar archive xerces1.4.jar) and obtains the parsed document, from which it can extract the XML elements corresponding to various element names. The DOMParser class parses XMI files.

See Modifications to see the changes incorporated in XMIParser.java so that tool accepts Rose 2001 generated files.