[codex] Handle empty document ingestion gracefully#399
Merged
Conversation
ArnavAgrawal03
marked this pull request as ready for review
May 11, 2026 00:48
Adityav369
added a commit
that referenced
this pull request
Jul 23, 2026
Blank pages were dropped during ColPali rendering (since #399), so chunk numbers stopped matching source page numbers for any document containing blank pages, breaking anything that maps a chunk back to its page. Render every page: blank pages become normal chunks, and a page that fails to render now yields a white placeholder instead of being dropped. Documents where every page fails still error out as before. Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR makes document ingestion tolerant of blank or unextractable pages by skipping blank/failed rendered pages while preserving valid pages for ColPali. It adds a deep parser fallback that retries empty text extraction with Office-to-PDF conversion and OCR-oriented Docling settings before giving up. If no text or image chunks can still be extracted, ingestion now completes with explicit non-searchable metadata instead of failing the whole document. Validation:
python3.11 -m py_compile core/parser/morphik_parser.py core/workers/ingestion_worker.py core/services/ingestion_service.py core/tests/unit/test_ingestion_colpali_rendering.pyandgit diff --check; targeted pytest remains blocked by the local morphik-rust/Python environment issue.