jedarden/pdftract
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
527 commits 1 branch 0 tags 7.5 GiB
Commit graph

3 commits

Author SHA1 Message Date
jedarden
e6bf3dd290 feat(pdftract-3s2i): implement Phase 5.5.2 validation filter 
Implement per-word validation filter for assisted-OCR BrokenVector path.

Changes:
- Add SpanSource::OcrAssisted variant to hybrid.rs
- Add Span::ocr_assisted() helper method
- Implement validate_ocr_with_position_hints() in ocr.rs
  - 5pt distance threshold for position validation
  - 0.4 confidence cap for rejected words
  - Linear scan for nearest-neighbor lookup
- Add unit tests for validation filter

Closes: pdftract-3s2i

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
 2026-05-24 04:57:17 -04:00
jedarden
9365bb404c test(pdftract-2zw): add reproducibility gate perturbation test 
Adds test_reproducibility_gate_with_perturbation which verifies that the
reproducibility check correctly detects when classification results differ.
This test intentionally perturbs a confidence value and asserts that the
reproducibility gate fails with a clear diff message.

Acceptance criteria for pdftract-2zw:
- Reproducibility gate fails on intentional perturbation

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
 2026-05-23 15:04:05 -04:00
jedarden
9215892f95 feat(pdftract-2zw): page classification fixtures + integration tests + reproducibility gate 
Implement page classification test fixtures, integration tests, and
reproducibility CI gate for Phase 5.1.5.

Fixtures (4 total, 3.6 KB):
- vector_pure: Pure text PDF (born-digital)
- scanned_single: Image-only PDF (scanned)
- brokenvector_pdfa: Invisible text + image
- hybrid_header_body: Text header + scanned body

Integration tests (crates/pdftract-core/tests/page_classification.rs):
- test_page_classification_fixtures: Validates classification correctness
- test_page_classification_reproducibility: CI gate for byte-identical JSON
- test_fixture_files_exist_and_size: Infrastructure validation
- test_expected_json_validity: JSON schema validation

Acceptance criteria:
- ✅ 4 fixtures present in tests/fixtures/page_class/
- ✅ cargo test page_classification passes (4/4 tests)
- ✅ Reproducibility gate fails on perturbation
- ✅ Fixtures total < 1 MB (3.6 KB)

Refs: pdftract-2zw, plan.md lines 1840-1844

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
 2026-05-23 15:04:05 -04:00