feat(normalizer): add name normalization + lookup index to persons_tree

tools/import-normalizer/tests/test_persons_tree.py

@@ -83,3 +83,91 @@ def test_parse_generation_empty():
 
 def test_parse_generation_none():
     assert persons_tree._parse_generation(None) is None
+
+
+def test_norm_tree_basic():
+    assert persons_tree._norm_tree("Werner Allemeyer") == "werner allemeyer"
+
+
+def test_norm_tree_diacritics():
+    assert persons_tree._norm_tree("Wöhler") == "woehler"
+
+
+def test_norm_tree_strips_parens():
+    assert persons_tree._norm_tree("Otto (Herbert)") == "otto"
+
+
+def test_norm_tree_strips_quotes():
+    assert persons_tree._norm_tree('"Tante Lolly"') == "tante lolly"
+
+
+def test_norm_tree_strips_geographic_suffix():
+    assert persons_tree._norm_tree("Walter Cram Aachen") == "walter cram"
+
+
+def test_norm_tree_strips_mexiko():
+    assert persons_tree._norm_tree("Hans Cram Mexiko") == "hans cram"
+
+
+def test_norm_tree_collapses_whitespace():
+    assert persons_tree._norm_tree("  Clara   de Gruyter  ") == "clara de gruyter"
+
+
+def test_build_index_forward_lookup():
+    persons = [{"rowId": "row_002", "firstName": "Werner", "lastName": "Allemeyer", "maidenName": None}]
+    idx = persons_tree._build_index(persons)
+    assert "werner allemeyer" in idx
+    assert idx["werner allemeyer"] == ["row_002"]
+
+
+def test_build_index_reversed_lookup():
+    persons = [{"rowId": "row_002", "firstName": "Werner", "lastName": "Allemeyer", "maidenName": None}]
+    idx = persons_tree._build_index(persons)
+    assert idx.get("allemeyer werner") == ["row_002"]
+
+
+def test_build_index_maiden_name_lookup():
+    persons = [{"rowId": "row_002", "firstName": "Elsgard", "lastName": "Allemeyer", "maidenName": "Wöhler"}]
+    idx = persons_tree._build_index(persons)
+    assert idx.get("elsgard woehler") == ["row_002"]
+
+
+def test_build_index_single_token_fallback():
+    persons = [{"rowId": "row_028", "firstName": "Herbert", "lastName": "Cram", "maidenName": None}]
+    idx = persons_tree._build_index(persons)
+    assert idx.get("cram") == ["row_028"]
+
+
+def test_build_index_ambiguous_single_token():
+    persons = [
+        {"rowId": "row_028", "firstName": "Herbert", "lastName": "Cram", "maidenName": None},
+        {"rowId": "row_019", "firstName": "Clara", "lastName": "Cram", "maidenName": None},
+    ]
+    idx = persons_tree._build_index(persons)
+    assert set(idx["cram"]) == {"row_028", "row_019"}
+
+
+def test_resolve_one_found():
+    persons = [{"rowId": "row_003", "firstName": "Werner", "lastName": "Allemeyer", "maidenName": None}]
+    idx = persons_tree._build_index(persons)
+    row_id, reason = persons_tree._resolve_one("Allemeyer Werner", idx)
+    assert row_id == "row_003"
+    assert reason is None
+
+
+def test_resolve_one_not_found():
+    idx = {}
+    row_id, reason = persons_tree._resolve_one("Nobody Unknown", idx)
+    assert row_id is None
+    assert reason == "not_found"
+
+
+def test_resolve_one_ambiguous():
+    persons = [
+        {"rowId": "row_028", "firstName": "Herbert", "lastName": "Cram", "maidenName": None},
+        {"rowId": "row_019", "firstName": "Clara", "lastName": "Cram", "maidenName": None},
+    ]
+    idx = persons_tree._build_index(persons)
+    row_id, reason = persons_tree._resolve_one("Cram", idx)
+    assert row_id is None
+    assert reason == "ambiguous"