Issues

• A simple font's short ToUnicode CMap is ignored for codes the embedded program does not name: BIÊN BẢN reads as BIÊ! BẢ!

  firecrawl/pdf-inspector#620 · Filter by author tlq5l opened on Oct 6, 2026

• One page whose content stream does not decode fails extraction of the whole document

  firecrawl/pdf-inspector#617 · Filter by author ghempton opened on Oct 5, 2026

• Headings and paragraph above a line-ruled table are absorbed into the table's first row (LibreOffice/Word export, single full-page clip)

  firecrawl/pdf-inspector#612 · Filter by author closytalk-daniele-sarnari opened on Oct 1, 2026

• Table cells split across multiple pages result in data loss

  firecrawl/pdf-inspector#588 · Filter by author beerendlauwers opened on Sep 25, 2026

• Expose a one-parse structured page result

  firecrawl/pdf-inspector#587 · Filter by author jnssksk opened on Sep 25, 2026

• High extraction time and memory usage when using @firecrawl/pdf-inspector from an npm package

  firecrawl/pdf-inspector#585 · Filter by author shaheeranser opened on Sep 24, 2026

• Vertical writing mode text is assembled row-wise across columns

  bug
  firecrawl/pdf-inspector#575 · Filter by author abimaelmartell opened on Sep 22, 2026

• Follow-ups from #567 (control-character ToUnicode destinations)

  enhancement
  firecrawl/pdf-inspector#574 · Filter by author abimaelmartell opened on Sep 22, 2026

• Bundled Adobe-*-UCS2.bcmap files fail to parse (unexpected EOF in bcmap), so the CID collection → Unicode fallback is never built

  bug
  firecrawl/pdf-inspector#573 · Filter by author abimaelmartell opened on Sep 22, 2026
  2

• Text rendering mode is reset at every BT and ignored when set outside a text object

  bug
  firecrawl/pdf-inspector#572 · Filter by author abimaelmartell opened on Sep 22, 2026

• pdf-inspector doesn't HTML-escape literal <, >, & in extracted body text

  firecrawl/pdf-inspector#565 · Filter by author vpckso opened on Sep 21, 2026

• Zero text extracted from IA Type0 /GlyphLessFont layers (ToUnicode present) and WinAnsi TrueType without ToUnicode — while classification reports text_based with pages flagged (1.17.0 and 1.21.0)

  firecrawl/pdf-inspector#554 · Filter by author cmdaltctr opened on Sep 19, 2026

• Link annotations extracted but not rendered in markdown output

  firecrawl/pdf-inspector#545 · Filter by author hoffrocket opened on Sep 18, 2026

• Feature request: Markdown source spans linked to PDF bounding boxes

  firecrawl/pdf-inspector#542 · Filter by author SkyeNygaard opened on Sep 17, 2026

• Unable to handle merged cells or complex table scenarios.

  firecrawl/pdf-inspector#537 · Filter by author HedgeLinn opened on Sep 16, 2026

• extract_pages_markdown flags a native text page as suspected_garbled_text (empty markdown) while classify_pdf / detect_pdf call it text_based; other extractors read the page fine

  firecrawl/pdf-inspector#528 · Filter by author jmfv1986 opened on Sep 12, 2026
  1

• Code listings lose all newlines and indentation: a 9-line Python block extracts as one line

  firecrawl/pdf-inspector#526 · Filter by author juanpons opened on Sep 10, 2026

• processPdf misclassifies a correctly-password-decrypted RC4 V2/R3/128 PDF as Scanned (0 chars extracted)

  firecrawl/pdf-inspector#524 · Filter by author chintans opened on Sep 10, 2026
  1

• [Bindings] Expose positioned OCR spans in OcrPageResult

  firecrawl/pdf-inspector#513 · Filter by author yitom486 opened on Sep 8, 2026
  1

• acroform field names and values are decoded as utf-8, mangling utf-16be text strings

  firecrawl/pdf-inspector#504 · Filter by author 8bittts opened on Sep 6, 2026

• Feature: report bench scores per document type, not one composite

  firecrawl/pdf-inspector#503 · Filter by author al-ashalash opened on Sep 6, 2026

• pdf2md returns empty markdown for OCR'd PDFs whose only text is an invisible (Tr 3) layer

  firecrawl/pdf-inspector#500 · Filter by author enomado opened on Sep 4, 2026
  1

• is_complex_layout/pages_with_columns reported but multi-column pages still flattened into one interleaved stream (bogus table)

  firecrawl/pdf-inspector#498 · Filter by author yonikremer opened on Sep 4, 2026

• Native extraction splits Hebrew words into per-character fragments (spurious spaces inside words)

  firecrawl/pdf-inspector#497 · Filter by author yonikremer opened on Sep 4, 2026
  1

• Why does this PDF prompt 'anydoc. Unsupported Error: unsupported input: PDF has no extractable text (ImageBased, 1 pages)'

  firecrawl/pdf-inspector#491 · Filter by author luzhihaoTestingLab opened on Sep 3, 2026