"""Integration tests for cross-reference detection and resolution.""" import pytest from uuid import uuid4 from src.extraction.cross_ref_detector import ( CrossRefDetector, CrossReferenceMatch, CrossRefResolutionSummary, ) from src.extraction.models import CrossRefResolutionSummary as ModelCrossRefSummary class TestCrossRefDetectorIntegration: """Integration tests for CrossRefDetector end-to-end flows.""" def test_full_detection_flow(self): """Test complete detection flow from content to matches.""" detector = CrossRefDetector() # Content with multiple reference types content = { "description": "Main specification document", "notes": "See Sheet Configuration for settings", "reference": "Item 2024-4_0019 applies here", "link": "Ref: ABC-123", "source": "Page 01.2-5, Item 3", } matches = detector.detect_references(content) # Verify we detected multiple reference types ref_types = {m.reference_type for m in matches} assert "sheet_ref" in ref_types assert "item_ref" in ref_types # May or may not have code_ref depending on pattern overlap # Verify field names are captured field_names = {m.field_name for m in matches} assert len(field_names) > 1 def test_japanese_reference_detection(self): """Test detection of Japanese cross-reference patterns.""" detector = CrossRefDetector() content = { "説明": "このドキュメントは設定を説明します", "参考": "参照 シート「データ一覧」を確認してください", "項目リンク": "項目 ABC-001 を参照", } matches = detector.detect_references(content) # Should detect at least one Japanese reference assert len(matches) >= 1 # Verify Japanese patterns work jp_matches = [ m for m in matches if "参照" in m.reference_text or "項目" in m.reference_text ] assert len(jp_matches) >= 1 def test_combined_sheet_item_reference(self): """Test detection and parsing of combined sheet+item references.""" detector = CrossRefDetector() content = { "cross_ref": "See Sheet Mapping, Item 2024-001", } matches = detector.detect_references(content) # Find sheet_item_ref or separate matches has_combined = any(m.reference_type == "sheet_item_ref" for m in matches) has_sheet = any(m.reference_type == "sheet_ref" for m in matches) # Should detect either combined or individual patterns assert has_combined or has_sheet def test_no_false_positives_normal_text(self): """Test that normal text doesn't generate false positive matches.""" detector = CrossRefDetector() content = { "name": "John Smith", "address": "123 Main Street", "description": "This is a normal description without references", "status": "Active", "date": "2024-01-15", } matches = detector.detect_references(content) # Should have few or no matches sheet_refs = [m for m in matches if m.reference_type == "sheet_ref"] assert len(sheet_refs) == 0 def test_content_item_search_integration(self): """Test item code search across content dictionaries.""" detector = CrossRefDetector() # Test various content structures test_cases = [ ({"item_code": "ABC-123"}, "ABC-123", True), ({"Item": "ABC-123"}, "ABC-123", True), ({"description": "Linked to ABC-123"}, "ABC-123", True), ({"value": "XYZ-999"}, "ABC-123", False), ({}, "ABC-123", False), ] for content, item_code, expected in test_cases: result = detector._content_contains_item(content, item_code) assert result == expected, f"Failed for content={content}, item={item_code}" def test_summary_statistics(self): """Test that summary statistics are correctly calculated.""" doc_id = uuid4() summary = CrossRefResolutionSummary( document_id=doc_id, references_found=10, references_resolved=7, unresolved_count=3, unresolved_refs=["Ref A", "Ref B", "Ref C"], warnings=[], ) # Verify arithmetic assert summary.references_resolved + summary.unresolved_count == summary.references_found assert len(summary.unresolved_refs) == summary.unresolved_count # Verify dict conversion d = summary.to_dict() assert d["references_found"] == 10 assert d["references_resolved"] == 7 assert d["unresolved_count"] == 3 class TestCrossRefResolutionSummaryCompatibility: """Test compatibility between CrossRefResolutionSummary definitions.""" def test_model_summary_compatibility(self): """Test that models.py and cross_ref_detector.py definitions are compatible.""" doc_id = uuid4() # Create from extraction models model_summary = ModelCrossRefSummary( document_id=doc_id, references_found=5, references_resolved=3, unresolved_count=2, unresolved_refs=["A", "B"], warnings=["Warning 1"], ) # Both should have same to_dict structure d = model_summary.to_dict() assert "document_id" in d assert "references_found" in d assert "references_resolved" in d assert "unresolved_count" in d assert "unresolved_refs" in d assert "warnings" in d class TestCrossRefPatternCoverage: """Test coverage of various cross-reference patterns.""" def test_pattern_variations(self): """Test various pattern variations are detected.""" detector = CrossRefDetector() test_cases = [ # English sheet references ("See Sheet Data", "sheet_ref"), ("see sheet config", "sheet_ref"), ("SEE SHEET MAIN", "sheet_ref"), ("Refer to Sheet Settings", "sheet_ref"), # Item references ("Item ABC-123", "item_ref"), ("item xyz_001", "item_ref"), ("ITEM 2024-001", "item_ref"), # Code references ("Ref: ABC-123", "code_ref"), ("ref: xyz_001", "code_ref"), ("Reference: CODE-001", "code_ref"), ] for text, expected_type in test_cases: content = {"field": text} matches = detector.detect_references(content) # Find match of expected type type_matches = [m for m in matches if m.reference_type == expected_type] assert ( len(type_matches) >= 1 ), f"Expected {expected_type} for '{text}', got {[m.reference_type for m in matches]}" class TestExtractionSummaryIntegration: """Test integration with extraction pipeline summary.""" def test_extraction_summary_includes_cross_refs(self): """Test that ExtractionSummary includes cross-reference fields.""" from src.extraction.pipeline import ExtractionSummary doc_id = uuid4() summary = ExtractionSummary( document_id=doc_id, sheets_processed=3, tables_detected=5, records_extracted=100, duration_ms=5000, warnings=[], cross_references_found=10, cross_references_resolved=8, cross_references_unresolved=2, ) d = summary.to_dict() assert d["cross_references_found"] == 10 assert d["cross_references_resolved"] == 8 assert d["cross_references_unresolved"] == 2 def test_extraction_summary_defaults(self): """Test that cross-reference fields default to 0.""" from src.extraction.pipeline import ExtractionSummary doc_id = uuid4() summary = ExtractionSummary( document_id=doc_id, sheets_processed=3, tables_detected=5, records_extracted=100, duration_ms=5000, warnings=[], ) assert summary.cross_references_found == 0 assert summary.cross_references_resolved == 0 assert summary.cross_references_unresolved == 0