Enrichment script: Wikidata (people)

[DutchJaFO]

Context

Split from #19. Implements IEnrichmentProvider for Wikidata, covering person quotes. Depends on the shared enrichment scaffold (#19); does not depend on TMDB (#35), Open Library (#36), or TVDB (#46).

API details

Property	Value
Base URL	https://www.wikidata.org
Auth	None required
Rate limit	No hard limit for modest use; set a User-Agent header per Wikidata bot policy
Endpoints	/w/api.php?action=wbsearchentities (search), /wiki/Special:EntityData/{id}.json (fetch)

What gets enriched

Field	How
date	Wikidata P569 (date of birth) → year; fallback only if no speech date is known
author	Confirm or normalise spelling against the Wikidata canonical English label

Implementation checklist

• Implement WikidataEnrichmentProvider : IEnrichmentProvider in Quotinator.Enrichment
• SupportedTypes: person
• Search wbsearchentities by author field value with type=item
• Fetch the top candidate entity and extract P569 (date of birth) → year string
• Use the entity's English label to confirm or normalise the author field spelling
• Only fill fields that are currently null or empty — never overwrite manually curated data
• Cache API responses in scripts/cache/wikidata/
• Set a User-Agent request header (e.g. Quotinator-Enrichment/1.0)
• Apply enrichment only on exactly one high-confidence match; flag ambiguous results
• Respects the shared --dry-run, --no-fetch, and --provider flags from the scaffold

Notes

• Depends on scaffold (Enrichment script: fill missing quote fields using TMDB, Open Library, and Wikidata #19) only
• For person quotes, date is the date the quote was said, not birth year — only use birth year as a fallback; log the distinction clearly