Skip to main content

Set up FAIR Data Point

Use FAIR Data Point (FDP) to publish GDI metadata in a standardised RDF format that can be harvested by the User Portal.

The operational FDP setup is maintained outside this repository:

Deploy FDP​

Deploy FDP by following the FAIR Data Point section of the starter-kit deployment guide. That guide covers the FDP Docker Compose setup, default accounts, production deployment references, Kubernetes resources, and the information that must be shared so the User Portal can harvest the FDP instance.

Update SHACL shapes​

The GDI SHACL shapes are maintained in the gdi-metadata repository. Do not update the FDP metadata schemas manually in the FDP UI unless you are explicitly testing a local draft.

Use the automated schema tool from gdi-metadata to publish the current shapes:

  1. Open schema-tool/.env.

  2. Set FDP_HOST to the FDP instance that should receive the update.

  3. Set FDP_USER and FDP_PWD to credentials for an FDP admin user.

  4. From schema-tool/, run:

    docker compose up

The tool uses schema-tool/Properties.yaml to combine the shape files, define parent-child relations, and publish the intended shapes to FDP.

Metadata model​

The current GDI metadata model is documented in gdi-metadata/Documentation:

  • gdi-metadata-overview.md describes the class hierarchy, namespaces, and validation approach.
  • Metadata_submission.csv is the field-by-field submission guide, including obligation level, cardinality, guidance notes, and example values.
  • choicelist.csv lists controlled vocabulary values used by enumerated fields.

Onboard metadata​

Automate large-scale metadata uploads using the SeMPyRO Jupyter notebook. Clone the SeMPyRO repository and run the notebook:

hatch run docs:jupyter lab

Use the Usage_example_GDI.ipynb notebook as an example for submitting GDI metadata to FDP.

Next steps

After setting up FAIR Data Point: