All pull requests

Pull requests

• fix(extractor): read the text of a stream over the operator cap without its path operators

  #660·freyzzzz opened Nov 12, 2026·

  ·
  2

• feat(api): load_document_mem and the _from_doc functions read one loaded document

  #659·freyzzzz opened Nov 12, 2026·

  ·
  5

• feat(extractor): PositionOptions::include_invisible keeps invisible text in the position and region functions

  #658·freyzzzz opened Nov 12, 2026·

  ·
  2

• fix(markdown): write page markers for pages without text lines

  #657·rth opened Nov 11, 2026Contributor·

  ·
  1

  2

• perf(extract): per-page Markdown gathers each page's items, rects and lines in one pass

  #655·BukeLy opened Nov 10, 2026·3 tasks done·

  ·

• fix(detector): bound content stream decompression across page scan and font loading

  #654·ravi-arnan opened Nov 9, 2026·

  ·
  1

  12

• perf(detector): skip OCR reasons in lightweight classification

  #649·fradesa opened Oct 31, 2026·

  ·

• fix(tables): keep the first row of a continuation table without a repeated header

  #648·Aschenbath opened Oct 31, 2026·

  ·
  2

• Expose per-page raster coverage

  #645·LydiaTools opened Oct 27, 2026·

  ·
  1

  5

• Feat: add LoadedPdf handle and PdfDocument handle (wasm) for repeated page processing

  #642·ruihe774 opened Oct 23, 2026·

  ·
  5

• Pin pip tooling version in CI

  #640·dtaivpp opened Oct 22, 2026·

  ·
  3

• fix(underline): detect sub-point filled strikeouts and struck paragraphs

  #639·jvitormelo opened Oct 20, 2026·

  ·
  6

• Cap form resource walk depth, budget content-stream decode, index line grouping

  #638·abimaelmartell opened Oct 20, 2026Member·

  ·
  18

• fix(page_box): resolve the visible page box as PDFium does; expose it per page

  #637·PhilLehmann opened Oct 18, 2026·

  ·
  4

• fix(tables): keep super- and subscripts in the cell of the value they belong to

  #629·wenzhengjiang opened Oct 12, 2026·

  ·
  25

• feat: read a scan's invisible OCR text layer when asked

  #628·logancyang opened Oct 10, 2026·

  ·
  1

  99

• fix(extractor): skip comment stripping inside inline image data

  #626·deepanshupal opened Oct 10, 2026·

  ·
  7

• fix(extractor): a simple font's sparse ToUnicode CMap reads the codes its program does not

  #621·tlq5l opened Oct 6, 2026·

  ·
  1

  10

• feat: add best_effort option and text_unreliable flag

  #619·abimaelmartell opened Oct 6, 2026Member·

  ·
  10

• Decode AcroForm field names and values as PDF text strings

  #616·litoosh13 opened Oct 5, 2026·

  ·
  1

• fix(markdown): a run of no letters or digits never opens bold or italic

  #615·Toasty360 opened Oct 4, 2026·

  ·
  5

• fix(markdown): end a list item at a paragraph gap

  #614·Toasty360 opened Oct 4, 2026·

  ·
  29

• perf(tables): cut chart detection and layout complexity off quadratic paths

  #613·shaheeranser opened Oct 2, 2026·

  ·
  4

• fix(detector): cap Form-resource walk depth at 64

  #611·abimaelmartell opened Sep 30, 2026Member·

  ·
  26

• fix(detector): bound page-content and bound-form decode by byte budget

  #610·abimaelmartell opened Sep 30, 2026Member·

  ·
  15