Repository Issue Activity (beta)

#3759 DOCX headings incorrectly numbered when numFmt is none

#3739 DOCX table rows with `w:gridBefore` are silently truncated (cells dropped)

#3645 HybridChunker: Option to use markdown table format instead of flat text serialization

#3740 ImportError: cannot import name 'LatexCharsNode' from 'pylatexenc.latexwalker'

#3758 HF_TOKEN not being able to be pass trough docling CLI

#1920 ConversionError doesn't preserve underlying exception details for error classification (PDF Encrypted

#3755 `Bus error` of `torchao` installed, on Termux. Must get rid od `torchao` to resolve

#3512 ThreadedDoclingParseDocumentBackend drops tables that DoclingParseDocumentBackend detects (same pipeline options, reproduced at parser_threads=1)

#3473 OCR'd `TextItem`s inside a TABLE region are never referenced from `body.children` and therefore vanish from `md_content`

#3736 MarkdownTableSerializer missing |-| separator on chunked table continuations

#1014 Automatically detect PDFs requiring force OCR

#3749 TableFormer fails on borderless tables with whitespace-delimited columns

#3508 bug: HTML to markdown lose table content and the "li" serial number is error

#3321 VlmConvertModel drops generation settings when building VlmEngineInput

#1692 "Strict" DOCX conversion does not work

#3652 feat: Add PaddlePaddle v6 as an OCR preset

#2495 Integrate PaddleOCR-VL for SOTA OCR

#2381 `document_timeout` doesn't work on `docling-parse` processing a PDF page with 40k XObjects

#3743 msword/docx: malformed hyperlink address aborts the whole conversion

#3205 Differentiate between conversion whre `document_timeout` is reached versus those for which individual pages failed

#3564 Citation question: Is there a peer-reviewed version of the Docling paper?

#1945 Detect code in docx

#3402 [BEE] Table model missing the last part of the table in the Japanese document

#3733 AttributeError: module 'docling' has no attribute '__version__'

#2733 Add code language detection