Redence capability: Parsing & extraction (File/PDF processing)
Repo: Flavian. Stack: PDF.js, IDML/XML parser, pngjs.
Substantive change: implement a real extractor module — parse a PDF (via PDF.js) and/or an IDML package (XML), pulling out text runs, headings, and embedded images (decoding PNGs with pngjs), into a normalized structure. Handle multi-page/multi-story input. Add a fixture-based test asserting extracted text + image counts. The diff must add real parsing code, not docs.
Done when: extractor + fixture test merged to main, checks green. Include the current Redence proof token (Parsing & extraction verify page) in the commit message.
Redence capability: Parsing & extraction (File/PDF processing)
Repo: Flavian. Stack: PDF.js, IDML/XML parser, pngjs.
Substantive change: implement a real extractor module — parse a PDF (via PDF.js) and/or an IDML package (XML), pulling out text runs, headings, and embedded images (decoding PNGs with pngjs), into a normalized structure. Handle multi-page/multi-story input. Add a fixture-based test asserting extracted text + image counts. The diff must add real parsing code, not docs.
Done when: extractor + fixture test merged to main, checks green. Include the current Redence proof token (Parsing & extraction verify page) in the commit message.