Paperless-ngx v3.0.4

v3.0.4

Paperless-ngx v3.0.4

Changed 5
  • More efficient diacritic normalization in dropdown filtering
  • Pass LLM output language to chat if specified
  • Prefetch notes and custom fields for LLM index text building
  • Scope llm index updates to actually modified documents
  • Resolve npm provenance issues with chokidar and semver
Fixed 9
  • Prevent pdfjs highlight scrolling from affecting the entire page
  • Don't skip OCR/archive for tagged PDFs with no actual text
  • Dedupe permission-visible documents when combined with multi-tag filters
  • Ensure preview reload on live changes
  • Clamp all out-of-range possible fields
  • Add docstring to DocumentClassifierSchema for cleaner LLM tool description
paperless-ngx 3.0.4
Bug Fixes
  • Fix: prevent pdfjs highlight scrolling from affecting the entire page @shamoon (#13355)
  • Fix: don't skip OCR/archive for tagged PDFs with no actual text @stumpylog (#13351)
  • Performance: more efficient diacritic normalization in dropdown filtering @shamoon (#13347)
  • Fix: dedupe permission-visible documents when combined with multi-tag filters @stumpylog (#13345)
  • Fixhancement: pass LLM output language to chat if specified @shamoon (#13340)
  • Fix: ensure preview reload on live changes @shamoon (#13321)
  • Fix: clamp all out out-of-range possible fields @stumpylog (#13316)
  • Fix: add docstring to DocumentClassifierSchema for cleaner LLM tool description @stumpylog (#13315)
  • Fix: guard build_document_node against stale FK on deleted correspondent/doc type @stumpylog (#13318)
  • Fix: prevent search filter loss when closing document with Escape key @shamoon (#13317)
  • Fix: fix frontend permissions display for stats/system perms @shamoon (#13305)
Documentation
  • Documentation: fix PAPERLESS_AI_LLM_OUTPUT_LANGUAGE heading level @SuperSandro2000 (#13341)
Dependencies
  • Chore: resolve npm provenance issues with chokidar and semver @shamoon (#13323)
All App Changes
  • Fix: prevent pdfjs highlight scrolling from affecting the entire page @shamoon (#13355)
  • Fix: don't skip OCR/archive for tagged PDFs with no actual text @stumpylog (#13351)
  • Performance: more efficient diacritic normalization in dropdown filtering @shamoon (#13347)
  • Performance: prefetch notes and custom fields for LLM index text building @stumpylog (#13350)
  • Fix: dedupe permission-visible documents when combined with multi-tag filters @stumpylog (#13345)
  • Performance: Scope llm index updates to actually modified documents @stumpylog (#13322)
  • Fixhancement: pass LLM output language to chat if specified @shamoon (#13340)
  • Chore: resolve npm provenance issues with chokidar and semver @shamoon (#13323)
  • Fix: ensure preview reload on live changes @shamoon (#13321)
  • Fix: clamp all out out-of-range possible fields @stumpylog (#13316)
  • Fix: add docstring to DocumentClassifierSchema for cleaner LLM tool description @stumpylog (#13315)
  • Fix: guard build_document_node against stale FK on deleted correspondent/doc type @stumpylog (#13318)
  • Fix: prevent search filter loss when closing document with Escape key @shamoon (#13317)
  • Fix: fix frontend permissions display for stats/system perms @shamoon (#13305)
View original

Upgraded? How did it go?

Discussion