fix: improve error handling and status updates for document processing and add OCR requirement check for image files

[ricofurtado]

Supported File Types by Ingestion Path

There are currently two different supported file type lists depending on how content is ingested into OpenRAG.

Direct Upload (Frontend File Picker)

Defined in knowledge-dropdown.tsx (lines 60–76).

Category	Extensions
Documents	.txt, .md, .pdf, .docx, .html, .htm, .adoc, .asciidoc, .asc
Spreadsheets	.csv, .xlsx
Presentations	.pptx

---

Connector Ingestion (Backend Validator)

Defined in processors.py (lines 645–668).

Includes all file types supported by Direct Upload, plus additional Office formats, XHTML, and image formats.

Category	Extensions
Documents	.txt, .md, .pdf, .docx, .html, .htm, .adoc, .asciidoc, .asc
Additional Office Documents	.dotx, .dotm, .docm
Spreadsheets	.csv, .xlsx, .xls
Presentations	.pptx, .potx, .ppsx, .pptm, .potm, .ppsm
Web Documents	.xhtml
Images	.jpg, .jpeg, .png, .tiff, .bmp, .webp

---

Root Cause of the Reported Defect

The image formats (.jpg, .jpeg, .png, .tiff, .bmp, .webp) are the source of the reported issue.

What Happens

1. The connector ingestion validator accepts image files based on extension.
2. The files are sent through the ingestion pipeline.
3. Images only produce extractable text when Docling OCR is enabled.
4. When OCR is disabled, no text content is extracted.
5. No chunks are generated for indexing.
6. The updated ingestion logic now correctly marks these files as Failed rather than Active.

---

Existing Frontend Limitation

The frontend intentionally excludes several file types that are accepted by connector ingestion.

A TODO comment in knowledge-dropdown.tsx (line 57) states:

Re-add other MIME/extension groups (images, xlsx/xls/ppt, rtf/odt, etc.) once ingestion is verified end-to-end in Langflow.

This indicates that support for images and additional document types was intentionally withheld from the file picker until end-to-end ingestion behavior was fully validated.

---

Gap Between Frontend and Connector Validation

Area	Images Allowed?
Frontend File Picker	❌ No
Connector Ingestion Validator	✅ Yes

This inconsistency allows image files to be ingested through connectors while bypassing the frontend restrictions, creating the gap that exposed the bug.

Summary by CodeRabbit

Bug Fixes

• Improved validation of document processing results to ensure failures are correctly detected and reported when content is not properly indexed
• Enhanced error messages for image files to provide specific guidance when OCR processing is required for content extraction

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 2589f73b-0035-4332-8dc7-fa1048d1b7a1

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch cpd-files-with-unsupported-extensions-are-incorrectly-displayed-as-Active-Completed-during-connector-ingestion

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}