Retriever

Reverse Engineering Technique with Refinement Integration of Extracted Views with Elements and Relationships

Reverse engineering of software component models from heterogeneous artifacts of the software development process. For this purpose, technology-specific rules are used within Retriever to automatically extract static components of a system. For this purpose, the heterogeneous artifacts of a software project are first transformed into a model representation. Based on the model, Retriever applies mappings specified in rule artifacts to each model instance to identify components for the final software architecture model.

This component extraction supports base components, composite structures, interfaces, ports, and connectors. The candidate components are then used to generate the elements of the target software architecture model, which can then be used for quality prediction purposes in the Palladio context. The extracted models are suitable for improving the understanding of existing software and enabling further quality analyses. Software performance, reliability, and maintenance analyses are already available as part of a complementary tool chain.

Status

The rules engine is currently under active development. If you are interested in further information or would like to contribute your personal thoughts or requirements, please do not hesitate to contact us.

Retriever Architecture

This UML diagram shows how the Retriever approach works with the Eclipse platform. The main part of the diagram is the Retriever component, which works with other processes such as Discoverer, Extraction, Finalization, and Refinement. These components work through an orchestrator that coordinates reverse engineering so that it can be used in two ways: programmatically and graphically. It also shows how the software interacts with external models and metamodels. The diagram shows how components such as SoMoX SEFF and Vulnerability connect to the main Retriever component to extract and display software model information.

Platform

Retriever is implemented with the Eclipse Modeling Framework (EMF). It is thus provided as a plug-in for the Eclipse platform.

Source code based on EMF metamodels is generated during the build process and is therefore not added to the repository. To generate the source code, either the Maven build or the EMF generation workflow must be run.

The rules for the model-to-model transformations are implemented in Xtend, which can be compiled into Java-compatible source code.

Discoverers

• CSV Discoverer
• ECMA Discoverer
• Java Discoverer
• JSON Discoverer
• Properties Discoverer
• SQL Discoverer
• XML Discoverer
• YAML Discoverer

Extraction

• Docker Container
• ECMAScript 6
• Gradle Build
• Jakarta Deployment
• Jakarta Services
• Maven Build
• Spring Boot
• Spring Gateway
• Zuul Gateway

Analyzer

• Behavior
• Vulnerability

Finalizer

• Quality Prediction
• Documentation

Getting Started

Preconditions

• Installation of PCM Nightly
  • Requires Java Development Kit 17 and Eclipse 2022-12 Modeling Tools
  • Install PCM Nightly from the update site
  • Do not install the deprecated categories

Installation

For Development

• Install the Fluent Api Model Generator from the update site
• Install the Xtend IDE from the update site
• Keep the current installation and, if necessary, change the items to be installed so that they are compatible
• Check out this repository and import the existing projects into Eclipse
• Make changes to the imported code in Eclipse
• Locally verify all projects in the repository with Maven: mvn clean verify
• Launch a new Eclipse runtime application

For Direct Use

• Adding these update sites, without any further installation:
  • The update site for the Fluent Api Model Generator
• Install Retriever from the update site
  • This will automatically install all the required dependencies

Usage

• Create a new run configuration to launch Retriever
  • Specify the root directory of the project to be analyzed
  • Select the technology used there
  • Specify the directory for saving the PCM instances
• If the PCM instances were not saved in the workspace, they can be imported later

Contributing

To contribute, please follow these steps:

1. Fork this repository.
2. Create a branch: git checkout -b <branch_name>.
3. Make your changes and commit them: git commit -ma '<commit_message>'
4. Push into the original branch: git push origin <project_name>/<location>
5. Create a new pull request.

Alternatively, you can read the GitHub documentation on how to create a pull request.

Links

• Palladio homepage: https://www.palladio-simulator.com/
• Comprehensive documentation: https://sdqweb.ipd.kit.edu/wiki/Palladio_Component_Model
• Issue tracker: https://palladio-simulator.atlassian.net/jira/
• Update site: https://updatesite.palladio-simulator.com/palladio-reverseengineering-retriever/nightly/
• Javadoc: https://updatesite.palladio-simulator.com/palladio-reverseengineering-retriever/nightly/javadoc/

Licensing

The code in this project is licensed under the EPL-2.0 License.

This site is open source. Improve this page.