// the find
BuffettCode/edinet_xbrl
A Python xbrl file downloader and parser for EDINET
A Python library for downloading and parsing XBRL filings from EDINET, Japan's financial disclosure platform run by the FSA. Targets quant researchers and finance developers who need structured access to Japanese corporate financial data. The audience is narrow but the need is real — EDINET's native interface is painful.
Solves a genuinely annoying problem: EDINET XBRL files use Japan-specific taxonomy namespaces (jppfs_cor, etc.) that generic XBRL parsers handle poorly. The API is minimal and clear — download by ticker, parse file, get value by key+context. Ships with real test data from Yahoo Japan and Kakaku.com, which is more than most financial data parsers bother with.
Last commit was October 2020 — five years dead. EDINET's API has changed since (v2 launched in 2019, further updates followed), so the downloader is likely broken against current endpoints. No type annotations, no async support, and the data model is a raw dict wrapper rather than typed financial statement objects, so you're doing a lot of string key spelunking. The UFO Catcher dependency (a now-defunct Japanese scraper service) is a red flag for long-term viability.