Skip to content

Load models to your workspace automatically

Migration from v1.X.X to v2.X.X and later

This feature replaces the read-only image of version v1.X.X. Before starting the new image, you have to clone the Git repositories that you've passed to the read-only image manually. The path with all repositories can then be mounted to the new image as described below.

Technical prerequisites

The feature relies on an Eclipse plugin that is not part of Capella. The plugin is called models-from-directory-importer and is available on GitHub: https://github.com/DSD-DBS/capella-addons.

The plugin is part of all Capella based pre-built images on GitHub. If you build it manually, make sure that you follow the "Install dropins" instructions.

To load models to your workspace automatically, you can mount a volume to the container.

docker run -d \
    -v path/to/models/on/host:/models \
    -e ECLIPSE_PROJECTS_TO_LOAD='[]' \
    capella/base

The ECLIPSE_PROJECTS_TO_LOAD environment variable is a JSON array that contains:

[
  {
    "revision": "master", // (1)
    "nature": "project", // (2)
    "path": "/models/directory", // (3)
    "entrypoint": "test.aird" // (4)
  }
]
  1. The revision of the Eclipse project. In case of duplicated project names, the revision is added as suffix to the project name.
  2. Optional: Can be either 'project' or 'library'. Defaults to 'project'. Ignored if the the directory provided in the path attribute contains a .project file.
  3. Path to the directory where the project should be loaded from.
  4. Path to the aird file, starting from the directory provided in the path attribute. Required if the .aird is not placed directly in the directory provided as path. If None, the aird is searched in the path directory without recursion.

All additional attributes are ignored.

You can use all images that are based on the capella/base image for this feature.