Filter and restructure data from Wikimedia sources to prepare use for QwantMaps.

You first need to download a complete Wikidata dump from Wikimedia in JSON format.

Then you can generate CSV data for site links and labels:

src/main.py load-wikidata --dump latest-all.json.bz2

You can download and extract data from Wikimedia statistics with a single command:

# Omit `--download` if you already downloaded raw dumps
src/main.py load-stats --download

src/config.py holds configuration for the languages to include in the dumps and the list of the files to load for statistics.