v0.1.0
The first release: read, search, structure, history, Wikidata, feeds, metrics, geo, and the XML dumps, from one binary.
The first public release of wiki.
One binary, no credentials, every public Wikipedia and Wikimedia surface:
- Read an article as clean text, Markdown, HTML, or wikitext; pull a one-paragraph summary; open it in the browser.
- Search the full text, autocomplete titles, fetch a random article, and list related pages.
- Structure: links, backlinks, categories, category members, media, references, citations, interlanguage links, and page info.
- History: revisions and unified diffs.
- Wikidata: entities by title or id, and raw SPARQL against the Query Service.
- Feeds and metrics: the daily featured content, on-this-day events, the most-viewed articles, and pageview time series.
- Geo: articles near a coordinate or near another article.
- Bulk: list, download, page through, grep, and export the public XML dumps into a clean Markdown or text corpus.
Every command shares one output contract (table, json, jsonl, csv,
tsv, url, and Go --template), targets any wiki with -l/--lang,
--project, or --site, and doubles as a resource-URI driver for a host that
addresses pages as wikipedia:// URIs.