Meta Models
This page describes an ongoing discussion about the meta-model that describes all the aspects of the ReProTool project.
The whole !ReProTool meta-model consists of several sub-meta-models divided into separate packages. Each package should encapsulate a certain aspect of the intended application e.g. Input, Static Structure, Behaviour...
The following picture provides an overview of the main parts of the meta-model. In the first phase, we would like to limit the number of model elements. For example, we only consider plain text files and java source files as an input for the first prototype. Future releases of the application should, however, support more formats.
The core model describes the concept of traceability links which is used throughout the whole transformation pipeline. Outcomes of transformation tools will be linked to the original elements using traceability links.
Represents input files. In the initial phase, only plan text files and java source files will be supported.
Textual documents will be split into sections and further to paragraphs. Some paragraphs might be later identified as requirements or even as a special type of requirements such as use-cases. The identification may be automatic or manual.
In the initial phase, only Java source files will be parsed and the information about classes and methods will be extracted. These artefacts will be later connected with requirements.
Contains sub-models describing different aspects of the software engineering domain:
- Requirements Model
- Domain Model
- Behaviour Model
- Component Model
Requirements can be either a generic text paragraphs or can be further processed, such as Use-cases
TODO...