Apply provenance-related features from upstream

[jsheunis]

The latest updates to provenance metadata extraction and rendering in a catalog are contained in datalad/datalad-catalog#433.

We should ensure that previously existing features and the latest changes are available to this catalog instance, so that provenance info can be rendered.

[jsheunis]

Latest update:

• datalad-catalog main already had the ability to understand runprov records (based on it being incorporated into the catalog schema) and render them, for a long time
• The runprov extractor from metalad was updated and incorporated into datalad-catalog as a standalone script in Runprov extractor datalad/datalad-catalog#433. This PR also added a script to datalad-catalog/tools demonstrate how the extractor can be used.
• The run-record rendering was tweaked in Small tweaks to rendering of runprov records datalad/datalad-catalog#460

A demo for how this works for the clustered-network-pub dataset:

1. Clone the dataset: git clone https://github.com/psychoinformatics-de/ClusteredNetwork_pub.git
2. Create a new empty catalog: datalad catalog-create --catalog clustercat
3. Run the tool script: python datalad-catalog/tools/extract_core_runprov.py ClusteredNetwork_pub clustercat
   This script:
   • takes paths to the DataLad dataset and the empty catalog as arguments
   • extracts core metadata as well as runprov metadata from the dataset, using the relevant extractor modules in datalad-catalog/extractors
   • translates these records to catalog-ready records
   • adds the records to the catalog
4. Set the catalog home page

   datalad catalog-set --catalog clustercat --dataset-id bfe12f51-3f30-4ceb-b4de-9b6f85f2ef6c --dataset-version 62e2fe8326350e27a381fa4b28afc4f83fa6b893 home
   

5. serve the catalog: datalad catalog-serve -c clustercat

video:

clusterdemo.mp4

[mih]

This has been resolved. Discussed in lab meeting 2024-06-06