tools/import-normalizer/persons.py

@@ -1,6 +1,7 @@
 """Person register parsing, name splitting, alias resolution."""
 import re
 import unicodedata
+from collections import Counter
 from dataclasses import dataclass, field
 
 import config
@@ -31,7 +32,7 @@ class Person:
     maiden_name: str = ""
     title: str = ""
     nickname: str = ""
-    extra_given_names: list = field(default_factory=list)
+    extra_given_names: list[str] = field(default_factory=list)
     birth_date: str | None = None
     birth_date_raw: str = ""
     birth_place: str = ""
@@ -41,7 +42,7 @@ class Person:
     spouse: str = ""
     generation: str = ""
     notes: str = ""
-    aliases: list = field(default_factory=list)
+    aliases: list[str] = field(default_factory=list)
     provisional: bool = False
 
 
@@ -77,12 +78,12 @@ def parse_register(rows: list[dict]) -> list[Person]:
             spouse=spouse_raw, generation=(r.get("generation") or "").strip(),
             notes=(r.get("notes") or "").strip(), provisional=False,
         ))
-    # De-duplicate colliding ids with numeric suffix
+    # De-duplicate colliding ids: every member of a colliding group gets a numeric suffix
-    seen = {}
+    # (-1, -2, …) so no id is left as an ambiguous "base". Unique ids are untouched.
+    counts = Counter(p.person_id for p in people)
+    seen: dict[str, int] = {}
     for p in people:
-        if p.person_id in seen:
+        if counts[p.person_id] > 1:
-            seen[p.person_id] += 1
+            seen[p.person_id] = seen.get(p.person_id, 0) + 1
             p.person_id = f"{p.person_id}-{seen[p.person_id]}"
-        else:
-            seen[p.person_id] = 1
     return people

tools/import-normalizer/tests/test_persons.py

@@ -27,3 +27,13 @@ def test_parse_register_basic():
     assert p2.maiden_name == "Cram"
     assert p2.spouse == "Ludwig Bohrmann"
     assert p2.provisional is False
+
+def test_parse_register_dedups_colliding_ids():
+    # Two people with the same first+last name: BOTH get a numeric suffix (no ambiguous base id).
+    people = persons.parse_register([
+        {"last_name": "Cram", "first_name": "Hans"},
+        {"last_name": "Cram", "first_name": "Hans"},
+    ])
+    ids = [p.person_id for p in people]
+    assert ids == ["cram-hans-1", "cram-hans-2"]
+    assert len(set(ids)) == 2