Skip to content

Latest commit

 

History

History
141 lines (108 loc) · 6.45 KB

CONTRIBUTING.md

File metadata and controls

141 lines (108 loc) · 6.45 KB
Raw

Contributing

Thank you for taking an interest in contributing to CWL Viewer.

The following is a set of guidelines for contribution and helpful tips and resources for working with the code base.

Code of Conduct

This project and everyone participating in it is governed by the CWL Project Code of Conduct. By participating, you are expected to uphold this code. Please report unacceptable behavior to leadership@commonwl.org.

Issue Contribution

Issues for both bug reports and feature requests are welcome. In the case of bug reports, please try to provide a verifiable example on the production instance if this is possible.

Before you submit an issue, please search the issue tracker, maybe an issue for your problem already exists and the discussion might inform you of workarounds readily available.

Code Contribution

Workflow

The preferred workflow for contributing to CWL Viewer is to fork the main repository on GitHub, clone, and develop on a branch. Steps:

  1. Fork the project repository by clicking on the 'Fork' button near the top right of the page. This creates a copy of the code under your GitHub user account. For more details on how to fork a repository see this guide.

  2. Clone your fork of the cwlviewer repo from your GitHub account to your local disk:

    $ git clone git@github.com:YourLogin/cwlviewer.git
$ cd cwlviewer

  3. Create a feature branch to hold your development changes:

    $ git checkout -b my-feature

    Always use a feature branch. It's good practice to never work on the master branch!

  4. Develop the feature on your feature branch. Add changed files using git add and then git commit files:

    $ git add modified_files
$ git commit

    to record your changes in Git, then push the changes to your GitHub account with:

    $ git push -u origin my-feature

  5. Follow these instructions to create a pull request from your fork.

Running the Application and Tests

See README.md for details on running the application with dependencies. The tests can be run using the standard mvn test command.

Code Structure and Dependencies

This project uses the Maven standard directory layout and is a Model-view-controller application built with Spring Boot.

Packaging is done by feature and all Spring configuration is Java annotation based.

Templates for the view use Thymeleaf, which allows them to be displayed in browsers as static prototypes.

MongoDB is used to store information about Workflow and QueuedWorkflow objects using Spring Data JPA.

The application also uses a triple store to keep the RDF representing workflows (gathered from cwltool's --print-rdf functionality).

In general, the controller classes call services which have the main logic for the application. These controllers are:

  • PageController - handles basic static pages such as the index and documentation
  • workflow.WorkflowController - the main controller class for the application
  • workflow.WorkflowJSONController - handles API functionality
  • workflow.PermalinkController - handles permalinks with content negotiation for retrieving different formats

Notable services include

  • workflow.WorkflowService - Handles workflow related functionality
  • researchobject.ROBundleService - Creates research object bundles
  • graphviz.GraphVizService - A wrapper for com.github.jabbalaci.graphviz.GraphViz to generate images from DOT source code
  • git.GitService - Builds on JGit to provide Git functionality
  • cwl.CWLService - Implements parsing of cwl files

Note: For the async operations, Spring does not support the calling of a method within the same class (as a proxy needs to kick in to spawn a new thread). For this reason some extra classes such as researchobject.ROBundleFactory and cwl.CWLToolRunner are used when they would otherwise not be required.

Basic Application Flow

  1. User fills in the form on the front page. This is represented by workflow.WorkflowForm and consists of just a URL to Github/Gitlab, or a URL to a git repository as well as the branch and path of a workflow within the repository.

  2. This is submitted and picked up by a method in workflow.WorkflowController. The form is validated and parsed by workflow.WorkflowFormValidator to produce a git.GitDetails object with a repository URL, branch and path. The MongoDB database is checked for already pending workflow.QueuedWorkflow or created workflow.Workflow objects based on this (but this flow assumes they do not already exist).

  3. A new workflow.QueuedWorkflow object is created by cloning the repository locally (if it does not already exist), checking out the new branch and parsing the file using built-in YAML parsing code. Intermediate visualisations and models are produced which may not yet be complete.

  4. cwltool is run on the workflow using the --print-rdf option to produce the RDF representation. The RDF will be stored in the SPARQL store and queries will extract the information required. Afterwards this is used to construct a Research Object Bundle for the workflow. This is an asynchronous operation so meanwhile...

  5. The user is redirected to the main workflow page, which will use the loading template for now until cwltool has finished running. The background is the intermediate visualisation. An AJAX call repeatedly checks the status of cwltool saved in the workflow.QueuedWorkflow object in MongoDB.

  6. The page either displays an error on the loading page or reloads to view the parsed workflow on the workflow template. An AJAX call checks if the Research Object Bundle has been created and adds it to the page when it has.