Assorted tools for enriching Thing-based metadata records with machine-retrievable information from 3rd-party services
- Python 100%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| .forgejo/workflows | ||
| docs | ||
| tools | ||
| README.md | ||
Knowledge Enrichment
This repository hosts tools and workflows to enrich dumpthings knowledge pools with additional machine-generated records.
available tools
tools/scrape-calendar.py(used by.forgejo/workflows/scrape.yml): scrapes three FZJ funding calendars and adds their events as XYZCompetition records (research information scheme)tools/enrich-via-doi.py(used by.forgejo/workflows/enrich_publications.yml): reads publication records and extends them with external metadata available via doi.org content negotiationtools/get-depiction-urls.py(used by the 'from-model'-websites, e.g., https://hub.psychoinformatics.de/www/www-from-model/src/branch/main/.forgejo/workflows/register-depictions.yaml): Given metadata input, this script extracts download URLs for eachDepictionof the record, provided the depiction has akindincluded in the depiction-type argument. For each depiction distribution, it will output the record curie, the file extension and the url, to stdout.tools/scrape-juser.py(used by.forgejo/workflows/scrape-juser.yml): Given an xml export of publications in MARC21 format as provided by Juser (juser.fz-juelich.de), as well as a local cache of XYZPerson and XYZPublication records, this script generates new or updates existing Publication and Person records. Person records are submitted into a 'protected' inbox, whereas Publication records are submitted into a 'public' inbox.