feat(#32): make ScoringConfig injectable in TemplateDetector#36
Merged
feat(#32): make ScoringConfig injectable in TemplateDetector#36
Conversation
- Replace global DETECTOR_WEIGHTS / MIN_CONFIDENCE_THRESHOLD refs with self._scoring.weight_for() / self._scoring.min_confidence_threshold - Extract _run_scoring() private helper (scoring loop) from detect_template() - Add get_detection_explanation() -> DetectionExplanation for diagnostic introspection - Export ScoringConfig and DetectionExplanation from templates/__init__.py - Add 4 new tests: threshold boundary pass/fail, weight ordering, tie-break reason
…tractionResult - PDFTableExtractor.extract() now returns ExtractionResult (not 3-tuple) - Credit card early-return path returns ExtractionResult with warnings entry - Normal return path wraps dicts_to_transactions(rows) in ExtractionResult - extraction_facade.extract_tables_from_pdf() return annotation -> ExtractionResult - All 6 extraction-layer test files updated: field access replaces tuple unpacking - ExtractionOrchestrator.extract_from_pdf() unpacks ExtractionResult, re-packs as tuple for upstream callers (Rule 3 - blocking fix) - All mock return values in test_excluded_files_logging, test_processor, test_processor_refactored_methods, test_extraction_orchestrator updated to ExtractionResult - 1360 tests pass, 9 skipped (all expected)
…ct_from_pdf() - Update 4 tuple-unpack assertion sites to use ExtractionResult field access - Add assertIsInstance(result, ExtractionResult) to key assertions - Tests now expect extract_from_pdf() to return ExtractionResult (not tuple)
- Add ExtractionResult import from bankstatements_core.domain - Change return annotation from tuple[list, int, str | None] to ExtractionResult - Remove tuple unpacking; pass ExtractionResult through unchanged - Use result.iban for IBAN logging - Update docstring Returns section
- Add ExtractionResult and transactions_to_dicts imports - Change return annotation from tuple[list[dict], int, dict[str,str]] to list[ExtractionResult] - Replace tuple unpacking with result.page_count/iban/transactions field access - Exclusion logic preserved using ExtractionResult fields - filter_service.apply_all_filters() receives transactions_to_dicts(result.transactions) - Append excluded results to list (page_count preserved for callers) - Update processor._process_all_pdfs() to adapt list[ExtractionResult] -> 3-tuple - Update test_processor test mock to use list[ExtractionResult] return value
- _process_all_pdfs() now returns list[ExtractionResult] directly (no adapter) - run() loops over results, aggregates all_rows/pages_read/pdf_ibans via transactions_to_dicts() - Added ExtractionResult and transactions_to_dicts imports at module level - test_processor_refactored_methods.py updated: tuple-unpack assertions replaced with indexed list access
- group_words_by_y: 8 tests for Y-grouping with tolerance param - assign_words_to_columns: 11 tests for strict/relaxed x-position assignment - calculate_column_coverage: 7 tests for coverage fraction calculation
…ions - group_words_by_y: groups words by round(top, 0); tolerance param accepted for API compat - assign_words_to_columns: relaxed/strict rightmost column boundary check - calculate_column_coverage: returns fraction of columns with non-empty data across rows - Fix test rounding: round(100.7, 0) is 101.0 in Python banker rounding
- boundary_detector.py: import group_words_by_y/assign_words_to_columns/calculate_column_coverage; add pre-filter before group_words_by_y; delete _group_words_by_y, _build_row_from_words, _calculate_column_coverage - row_builder.py: import assign_words_to_columns/group_words_by_y; replace inline loops with word_utils calls; remove _rightmost_column/_column_names attrs - header_detection.py: import group_words_by_y; delete _group_words_by_y_coordinate private method - page_validation.py: import calculate_column_coverage as _calculate_column_coverage_impl; make calculate_column_coverage a thin wrapper - content_density.py: import assign_words_to_columns; replace inline column-assignment loop
…rd_utils - test_boundary_detector.py: delete test_group_words_by_y, test_group_words_filters_above_table_top, test_build_row_from_words, test_calculate_column_coverage_full, test_calculate_column_coverage_partial, test_calculate_column_coverage_empty - test_header_detection.py: delete test_group_words_by_y_coordinate - All deleted tests now covered by test_word_utils.py TestGroupWordsByY, TestAssignWordsToColumns, TestCalculateColumnCoverage
longieirl
force-pushed
the
feat/32-scoring-config-injectable
branch
from
0604547 to
69d25c1
Compare
This was referenced Mar 25, 2026
Closed
longieirl
pushed a commit
that referenced
this pull request
Mar 26, 2026
Introduces ExtractionScoringConfig (injectable frozen dataclass) holding penalty weights for extraction anomalies, following the ScoringConfig pattern established in PR #36. RowPostProcessor now applies penalties and stamps confidence_score on each transaction row dict: - DATE_PROPAGATED: -0.1 (date inferred from prior row or filename) - MISSING_BALANCE: -0.2 (balance field absent or empty) Score is clamped to [0.0, 1.0]. Rows with no anomalies keep 1.0. PDFTableExtractor accepts an optional scoring_config and threads it through to RowPostProcessor; defaults to ExtractionScoringConfig.default(). Adds CODE_MISSING_BALANCE warning code to extraction_warning.py. Exports ExtractionScoringConfig from domain/models/__init__.py. 19 new tests cover: full-confidence, date-propagated, missing-balance, both penalties, clamping, injectable config, and non-transaction rows.
9 tasks
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
2 participants
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.
Pull Request
Summary
Changes
Type
Testing
Checklist