Spaces:
Sleeping
Sleeping
| from __future__ import annotations | |
| import sys | |
| import os | |
| import pytest | |
| sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..")) | |
| from comment_target_resolver import ( | |
| _build_char_map, | |
| _normalize_for_match, | |
| _find_all, | |
| _resolve_target_in_paragraph, | |
| resolve_single_target, | |
| ResolveResult, | |
| ) | |
| from models import ParagraphText | |
| class TestCharMap: | |
| def test_basic(self): | |
| cm = _build_char_map("Hello World") | |
| assert cm.normalized == "HelloWorld" | |
| assert cm.orig_offsets == [0, 1, 2, 3, 4, 6, 7, 8, 9, 10] | |
| def test_nbsp_stripped(self): | |
| cm = _build_char_map("Hello\u00a0World") | |
| assert cm.normalized == "HelloWorld" | |
| def test_nbsp_offsets(self): | |
| cm = _build_char_map("A\u00a0B") | |
| assert cm.normalized == "AB" | |
| assert cm.orig_offsets == [0, 2] | |
| def test_nbsp(self): | |
| cm = _build_char_map("Hello\u00a0World") | |
| assert cm.normalized == "HelloWorld" | |
| def test_smart_quotes(self): | |
| cm = _build_char_map("\u201cHello\u201d") | |
| assert cm.normalized == '"Hello"' | |
| def test_em_dash(self): | |
| cm = _build_char_map("A\u2014B") | |
| assert cm.normalized == "A-B" | |
| class TestNormalize: | |
| def test_missing_spaces(self): | |
| norm = _normalize_for_match("the Sellershall") | |
| assert norm == "theSellershall" | |
| def test_nbsp_normalized(self): | |
| norm = _normalize_for_match("Hello\u00a0World") | |
| assert norm == "HelloWorld" | |
| def test_spaces_stripped(self): | |
| norm = _normalize_for_match("The Seller shall") | |
| assert norm == "TheSellershall" | |
| class TestFindAll: | |
| def test_single(self): | |
| assert _find_all("abc", "b") == [1] | |
| def test_multiple(self): | |
| assert _find_all("ababa", "a") == [0, 2, 4] | |
| def test_none(self): | |
| assert _find_all("abc", "z") == [] | |
| class TestResolveTargetInParagraph: | |
| def test_exact_match(self): | |
| result = _resolve_target_in_paragraph( | |
| "The Seller shall deliver the goods promptly.", | |
| "shall deliver the goods", | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "exact" | |
| assert result.confidence == 1.0 | |
| assert result.target_text == "shall deliver the goods" | |
| def test_exact_match_with_anchors(self): | |
| result = _resolve_target_in_paragraph( | |
| "The Seller shall deliver the goods promptly.", | |
| "shall deliver the goods", | |
| "Seller ", | |
| " promptly", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "exact" | |
| assert result.was_resolved is False | |
| assert result.before_anchor == "Seller " | |
| assert result.after_anchor == " promptly" | |
| def test_missing_spaces(self): | |
| result = _resolve_target_in_paragraph( | |
| "The Seller shall not be liable for any loss of revenue.", | |
| "the Sellershall not be liable for anyloss of revenue", | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "whitespace_normalized" | |
| assert "Seller shall not be liable for any loss of revenue" in result.target_text | |
| def test_missing_spaces_around_parens(self): | |
| result = _resolve_target_in_paragraph( | |
| "ABC Construction Ltd (The Buyer)", | |
| "ABC Construction Ltd(TheBuyer)", | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "whitespace_normalized" | |
| assert result.target_text == "ABC Construction Ltd (The Buyer)" | |
| def test_missing_space_after_punctuation(self): | |
| result = _resolve_target_in_paragraph( | |
| "from the factory. This warranty extends only to", | |
| "factory.This warrantyextendsonly to", | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "whitespace_normalized" | |
| assert "factory. This warranty extends only" in result.target_text | |
| def test_nbsp_vs_space(self): | |
| result = _resolve_target_in_paragraph( | |
| "The\u00a0materials remain the sole property of the Seller", | |
| "The materials remain the sole property ofthe Seller", | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "whitespace_normalized" | |
| def test_smart_quotes(self): | |
| result = _resolve_target_in_paragraph( | |
| 'He said \u201cno liability\u201d for damages.', | |
| 'He said "no liability" for damages', | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy == "whitespace_normalized" | |
| def test_em_dash(self): | |
| result = _resolve_target_in_paragraph( | |
| "Payment \u2014 if delayed \u2014 incurs interest.", | |
| "Payment - if delayed - incurs interest.", | |
| "", "", | |
| ) | |
| assert result is not None | |
| def test_duplicate_with_anchors(self): | |
| para = "The Seller shall deliver. The Seller shall deliver promptly." | |
| result = _resolve_target_in_paragraph( | |
| para, | |
| "The Sellershall deliver", | |
| "The Seller", "promptly", | |
| ) | |
| assert result is not None | |
| assert result.strategy in ("anchor_normalized", "whitespace_normalized") | |
| def test_no_match_returns_none(self): | |
| result = _resolve_target_in_paragraph( | |
| "Completely unrelated text here.", | |
| "The Seller shall not be liable", | |
| "", "", | |
| ) | |
| assert result is None | |
| def test_sentence_fallback(self): | |
| para = "The Seller reserves all rights under this clause. The Buyer shall pay within 30 days. Force majeure does not apply." | |
| result = _resolve_target_in_paragraph( | |
| para, | |
| "The Sellerreserves all rightsunder this clause", | |
| "", "", | |
| ) | |
| assert result is not None | |
| assert result.strategy in ("whitespace_normalized", "sentence_fallback") | |
| assert "The Seller reserves all rights under this clause" in result.target_text | |
| def test_sentence_fallback_low_confidence_fails(self): | |
| para = "The Seller reserves all rights. The Buyer shall pay. Force majeure does not apply." | |
| result = _resolve_target_in_paragraph( | |
| para, | |
| "completely different and unrelated gibberish text", | |
| "", "", | |
| ) | |
| assert result is None | |
| class TestResolveSingleTarget: | |
| def test_with_paragraph_list(self): | |
| paragraphs = [ | |
| ParagraphText(id="P_0001", text="The Seller shall deliver the goods."), | |
| ParagraphText(id="P_0002", text="The Buyer shall pay the price."), | |
| ] | |
| result = resolve_single_target( | |
| paragraphs, "P_0001", | |
| "the Sellershall deliver the goods", | |
| "", "", | |
| ) | |
| assert result.strategy == "whitespace_normalized" | |
| assert result.paragraph_id == "P_0001" | |
| assert "The Seller shall deliver the goods" in result.target_text | |
| def test_missing_paragraph(self): | |
| paragraphs = [ | |
| ParagraphText(id="P_0001", text="Some text."), | |
| ] | |
| result = resolve_single_target( | |
| paragraphs, "P_9999", | |
| "Some text", | |
| "", "", | |
| ) | |
| assert result.strategy == "not_resolved" | |
| assert result.confidence == 0.0 | |
| def test_paragraph_fallback(self): | |
| paragraphs = [ | |
| ParagraphText(id="P_0001", text="Short paragraph here."), | |
| ] | |
| result = resolve_single_target( | |
| paragraphs, "P_0001", | |
| "totally unrelated", | |
| "", "", | |
| allow_paragraph_fallback=True, | |
| ) | |
| assert result.strategy == "paragraph_fallback" | |
| def test_paragraph_fallback_disabled(self): | |
| paragraphs = [ | |
| ParagraphText(id="P_0001", text="Short paragraph here."), | |
| ] | |
| result = resolve_single_target( | |
| paragraphs, "P_0001", | |
| "totally unrelated", | |
| "", "", | |
| allow_paragraph_fallback=False, | |
| ) | |
| assert result.strategy == "not_resolved" | |
| assert result.confidence == 0.0 | |
| if __name__ == "__main__": | |
| pytest.main([__file__, "-v"]) | |