OEMC FAIR Checker

A tool for assessing FAIR (Findable, Accessible, Interoperable, Reusable) compliance of datasets published via GeoKnowledge Hub and archived on Zenodo.

---

How it works

1. Fetches a GeoKnowledge Hub package and lists all linked knowledge resources (datasets, publications, software).
2. For each dataset with a Zenodo DOI, retrieves metadata and files from the Zenodo API.
3. Runs FAIR checks and produces a scored report with actionable recommendations.
4. Optionally writes the full report to a Markdown file.

FAIR Criteria

Principle	Checks
Findable	Has DOI, title, description, keywords, authors, publication date, version
Accessible	Files downloadable, DOI resolves, open access, CC-BY / CC-BY-4.0 license
Interoperable	Uses preferred geographic formats, avoids proprietary formats, links to publications, belongs to communities, follows OEMC filename convention
Reusable	License present and open, detailed description (>200 chars), funding info, links to code, version

Preferred geographic formats

.geojson, .tif / .tiff, .nc, .zarr, .shp, .parquet, .h5 / .hdf5, .gpkg

OEMC filename convention

Geographic files are validated against the OEMC naming convention:

{variable}_{method}_{var_type}_{spatial_support}_{depth_ref}_{date_start}_{date_end}_{bbox}_{epsg}_{version}.{ext}

Example: lccs_rf_m_1km_s_20000101_20001231_eu_epsg.3035_v20240101.tif

Usage

Open OEMC_FAIRChecker.ipynb in Jupyter and run the cells in order:

1. Fetch package — set PACKAGE_ID and fetch GKHub metadata
2. Parse resources — extract datasets, publications, software
3. Assess FAIR — run checks on all Zenodo datasets
4. Export report — write a Markdown report file

from fair_checker import GKHubClient, parse_knowledge_resources, \
    assess_all_knowledge_resources_fair, write_report

client = GKHubClient()
pkg = client.get_package("your-package-id")
resources = parse_knowledge_resources(html_text, "your-package-id")
results = assess_all_knowledge_resources_fair(resources)
write_report(pkg, resources, results)          # writes fair_report_<timestamp>.md

Requirements

requests
beautifulsoup4
colorama

Install with:

pip install requests beautifulsoup4 colorama