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

tools/import-normalizer/persons_tree.py

@@ -74,3 +74,57 @@ def _parse_generation(raw: str | None) -> int | None:
         return None
     m = re.search(r"\d+", str(raw))
     return int(m.group()) if m else None
+
+
+_GEO_SUFFIXES = {"aachen", "mex", "mexiko", "sen", "jun", "jr"}
+
+
+def _norm_tree(s: str) -> str:
+    """Normalize a name string for tree matching.
+
+    - Strip surrounding quotes, remove parenthetical substrings
+    - Diacritic → ASCII (ä→ae etc.), lowercase, dots → spaces
+    - Remove known geographic/honorific suffix tokens
+    - Collapse whitespace
+    """
+    s = (s or "").strip().strip("\"'")
+    s = re.sub(r"\([^)]*\)", "", s)
+    s = _strip_accents(s).lower().replace(".", " ")
+    tokens = [t for t in s.split() if t and t not in _GEO_SUFFIXES]
+    return " ".join(tokens).strip("., ")
+
+
+def _build_index(persons: list[dict]) -> dict[str, list[str]]:
+    """Build a name → [rowId, …] lookup index with four keys per person."""
+    index: dict[str, list[str]] = {}
+
+    def _add(key: str, row_id: str) -> None:
+        if key:
+            index.setdefault(key, []).append(row_id)
+
+    for p in persons:
+        row_id = p["rowId"]
+        first = p.get("firstName") or ""
+        last = p.get("lastName") or ""
+        maiden = p.get("maidenName") or ""
+
+        _add(_norm_tree(f"{first} {last}"), row_id)
+        _add(_norm_tree(f"{last} {first}"), row_id)
+        if maiden:
+            _add(_norm_tree(f"{first} {maiden}"), row_id)
+        _add(_norm_tree(last), row_id)
+
+    return index
+
+
+def _resolve_one(raw: str, index: dict[str, list[str]]) -> tuple[str | None, str | None]:
+    """Return (row_id, None) on unique match, (None, reason) otherwise."""
+    key = _norm_tree(raw)
+    if not key:
+        return None, "empty"
+    hits = index.get(key, [])
+    if len(hits) == 1:
+        return hits[0], None
+    if len(hits) == 0:
+        return None, "not_found"
+    return None, "ambiguous"