Skip to content

Dataset Website

Purpose

Every dataset repository on the ANC automatically gets a public-facing website, generated on every push to the main branch. The site is accessible as soon as any data is present on main — no further action is required.

Source code: ci-components/bids-dataset-website

The website is available at (once the datasteward sets a non-unique domain):

https://bids-datasets.data-pages.anc.plus.ac.at/<group>/<slug>

What it shows

The website is generated from the dataset repository and displays:

  • Dataset metadata from CITATION.cff (title, authors, DOI, license, description)
  • Participant statistics from participants.tsv (counts, sex distribution, age)
  • The full contents of README.md
  • ANC branding

How it works

The CI component runs website_generator.py on every push, which reads the BIDS directory and produces a static HTML site published via GitLab Pages. The BIDS Indexer is used to generate the majority of the metadata.

The component is added to all dataset repositories via .gitlab-ci.yml:

- component: $CI_SERVER_FQDN/ci-components/bids-dataset-website/dataset-website@main

Do not change the metadata.json download path

Researchers use the Data Download URL in metadata.json in their analysis scripts. The path must not be changed without coordinating with affected users.

Update workflow

Changes to the generator should be made via a merge request in the bids-dataset-website repository. Test on a dataset project before merging. A new container image is built and pushed automatically on merge to main.