Preserve cross-page PDF table rows during indexing - #830
Draft
hedhoud wants to merge 3 commits into
Draft
Conversation
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Context
Large PDF tables can split one logical row over several pages. Continuation pages often omit the identifying cells, so retrieval loses the relationship between the row identity and its supporting text. Marker can also represent part of such a table as plain text or an image caption, which meant the first implementation only worked reliably with PyMuPDF parser output.
Expected behavior
Indexing presets can opt into conservative automatic reconstruction. Marker remains the primary parser, while a CPU-only PyMuPDF sidecar supplies layout evidence for matched table regions. Confirmed continuations are indexed as self-contained
column: valuerow chunks; the original Marker blocks remain available for debugging. Ambiguous captions, headers, boundaries, or overlapping regions preserve the complete parser output unchanged. The feature remains disabled by default.Chunk provenance records both the original parser anchor and the layout provider. Oversized rows split at semantic and source-page boundaries, repeat their identity columns, and cite the page containing each part's content.
Validation
CST salarié/L. 421-1across all three pages and keepsL. 421-3as a separate row.No database or vector-store migration is required.