Skip to content

fix(core): preserve OCR continuity for blank pages#135

Merged
jb-thery merged 1 commit into
mainfrom
release/2.19.1
Jul 18, 2026
Merged

fix(core): preserve OCR continuity for blank pages#135
jb-thery merged 1 commit into
mainfrom
release/2.19.1

Conversation

@jb-thery

Copy link
Copy Markdown
Member

Why

Ragmir previously treated empty Tesseract output as a missing page payload. A valid scanned PDF page with no recognizable text could therefore stop a complete reindex, a case that is especially visible when refreshing an existing project after an upgrade.

What changed

  • accept an empty OCR payload when exactly one requested page returns no text
  • keep strict page-count validation for malformed or incomplete OCR output
  • add regression coverage for the real blank scanned-page case

Upgrade continuity

Users upgrading an existing Ragmir project can run the normal upgrade diagnostic and reindex flow without excluding a document only because it contains a blank or image-only page. Source files remain untouched, the current atomic index-generation behavior is preserved, and this patch introduces no configuration or index-format migration.

Validation

  • full pnpm validate passed
  • 497 Core tests passed
  • Landing, Chat, and TTS test suites passed
  • Linux x64 and macOS ARM64 portability checks passed
  • CodeQL, security audit, smoke, packaging, and release artifact checks passed

@jb-thery
jb-thery merged commit bfe8ce0 into main Jul 18, 2026
15 checks passed
@jb-thery
jb-thery deleted the release/2.19.1 branch July 18, 2026 13:09
@github-actions

Copy link
Copy Markdown

🎉 This PR is included in version 2.19.1 🎉

The release is available on:

Your semantic-release bot 📦🚀

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant