Skip to content

Add support for extracting comments from Word documents (.docx, .docm, .doc, .dotx) #1850

Open

Description

@paullizer

Summary

Add support for extracting review comments from Word documents across both modern OOXML formats (.docx, .docm, .dotx) and legacy Word 97–2003 binary OLE documents (.doc).

User Value

Users and customers frequently store critical review feedback, clarifications, and Threaded Q&A in Word margin comments. Previously:

  • .docx files processed by Azure Document Intelligence (prebuilt-read) only extracted main document text, headers/footers, tables, and footnotes, ignoring word/comments.xml.
  • .docm files processed via docx2txt ignored word/comments.xml.
  • .doc (Word 97–2003 OLE) files flattened the piece table without extracting structured comment metadata (ccpAtn, PlcfandTxt, PlcfandRef, GrpXstAtnOwners), author names, or anchor positions.

Extracting and indexing Word comments alongside the document chunks where they are anchored makes review comments searchable and available to AI chat, workspace RAG, M365 extraction, and content screening.

Proposed Behavior

  • OOXML Word files (.docx, .docm, .dotx):
    • Safely parse word/comments.xml (author, initials, date, comment text), word/commentsExtended.xml (threaded replies via paraIdParent and resolved/done state), and word/document.xml (w:commentRangeStart, w:commentRangeEnd, w:commentReference for quoted anchor text and reading-order word_offset).
  • Legacy binary Word files (.doc):
    • Parse the MS-DOC File Information Block (FibRgLw97 ccpAtn and FibRgFcLcb97 PlcfandTxt, PlcfandRef, GrpXstAtnOwners) to extract structured comments with author/initials and anchor positions while keeping raw annotation subdocument text separate from body text.
  • Chunk & Text Integration:
    • Merge extracted comments into the chunk containing their anchored text during workspace ingestion (process_di_document, process_office_template), include them in extract_word_text() / direct chat uploads, and pass them to content screening.

Acceptance Criteria

  • .docx, .docm, and .dotx comments (including author, date, anchored text, threaded replies, and resolved status) are extracted and indexed.
  • Legacy .doc OLE comments are extracted with author/initials and anchor context without duplicating raw annotation text in the body.
  • Extracted comments in .docx documents are merged into their corresponding reading-order chunk (with overflow fallback).
  • Functional tests cover both OOXML .docx and legacy .doc comment extraction and chunk placement, and release notes are updated.

Notes

  • Uses local bounded XML (defusedxml) and OLE (olefile) parsing with zero additional Azure API calls.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

P2Priority 2: important, scheduled after P0/P1enhancementNew feature or request

Type

No type

Projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions