From 443c7a48dbb7d57ce9ea5d9548cc32402cd39060 Mon Sep 17 00:00:00 2001
From: Marcel <marcel@familienarchiv>
Date: Mon, 25 May 2026 20:46:42 +0200
Subject: [PATCH] fix(normalizer): don't convert plausible typo years as Excel
 serials

---
 tools/import-normalizer/persons_tree.py       | 44 ++++++++++++++-----
 .../tests/test_persons_tree.py                |  7 +++
 2 files changed, 40 insertions(+), 11 deletions(-)

diff --git a/tools/import-normalizer/persons_tree.py b/tools/import-normalizer/persons_tree.py
index a52787a6..68e77ffb 100644
--- a/tools/import-normalizer/persons_tree.py
+++ b/tools/import-normalizer/persons_tree.py
@@ -11,14 +11,26 @@ import dates
 from persons import _strip_accents
 
 
+_MIN_YEAR = 1700
+_MAX_YEAR = 2100
+# Threshold: if parse_date parses a pure-digit string as a year outside [_MIN_YEAR, _MAX_YEAR],
+# but the year is a plausible typo (1000-3000), don't try serial conversion.
+# Years outside this range (e.g., 7568) are implausible and should try serial conversion.
+_PLAUSIBLE_TYPO_MIN = 1000
+_PLAUSIBLE_TYPO_MAX = 3000
+
+
 def _parse_year(raw: str | None) -> int | None:
     """Extract a birth/death year from an Excel cell string.
 
-    Handles four cases:
-    1. ISO string (openpyxl date-formatted cell) → parse_date()
-    2. Numeric string that is an Excel serial (1-80000) → timedelta conversion
-    3. Any other string → parse_date()
-    4. Unresolvable → None
+    Handles three cases:
+    1. ISO / German / text string parseable by parse_date() → extract year if in range
+    2. Pure-integer string (out-of-range or unparseable) → try Excel serial conversion
+       (unless it's a plausible typo year, e.g., "1023" for "1923")
+    3. Mixed-format or unresolvable → None
+
+    Serial conversion only fires for pure-digit strings and implausible years,
+    preventing typo years like "1023" from being mis-converted as serials.
     """
     if raw is None:
         return None
@@ -26,21 +38,31 @@ def _parse_year(raw: str | None) -> int | None:
     if not s:
         return None
 
+    # Check if it's a pure-digit string (candidate for serial conversion)
+    is_pure_digit = re.fullmatch(r"\d+", s) is not None
+
     # Try parse_date first (handles ISO, DD.MM.YYYY, year-only, month+year, etc.)
     result = dates.parse_date(s)
     if result.iso:
         year = int(result.iso[:4])
-        # Reject years outside 1700-2100 (same guard as expand_year())
-        if 1700 <= year <= 2100:
+        if _MIN_YEAR <= year <= _MAX_YEAR:
             return year
+        # Year is out of range. Only try serial conversion if it's an implausible year.
+        # Plausible typos (e.g., 1023 for 1923) should not be converted as serials.
+        if is_pure_digit and not (_PLAUSIBLE_TYPO_MIN <= year <= _PLAUSIBLE_TYPO_MAX):
+            n = int(s)
+            if 1 <= n <= 80_000:
+                d = datetime.date(1899, 12, 30) + datetime.timedelta(days=n)
+                if _MIN_YEAR <= d.year <= _MAX_YEAR:
+                    return d.year
+        return None
 
-    # If it's a pure integer string, try Excel serial conversion.
-    # parse_date() may parse large serials like "7568" as year 7568 or other edge cases.
-    if re.fullmatch(r"\d+", s):
+    # parse_date() found nothing. Try serial conversion only for pure-digit strings.
+    if is_pure_digit:
         n = int(s)
         if 1 <= n <= 80_000:
             d = datetime.date(1899, 12, 30) + datetime.timedelta(days=n)
-            if 1700 <= d.year <= 2100:
+            if _MIN_YEAR <= d.year <= _MAX_YEAR:
                 return d.year
 
     return None
diff --git a/tools/import-normalizer/tests/test_persons_tree.py b/tools/import-normalizer/tests/test_persons_tree.py
index 30c602ff..0419ebcd 100644
--- a/tools/import-normalizer/tests/test_persons_tree.py
+++ b/tools/import-normalizer/tests/test_persons_tree.py
@@ -52,3 +52,10 @@ def test_parse_year_unresolvable_truncated():
 def test_parse_year_typo_year():
     # "4.3.1023" — year 1023 outside 1500-2100 guard — returns None
     assert persons_tree._parse_year("4.3.1023") is None
+
+
+def test_parse_year_bare_out_of_range_year_is_none():
+    # "1023" is a plausible typo for "1923" but is NOT an Excel serial.
+    # parse_date("1023") parses it as year 1023 (out of 1700-2100 guard).
+    # The serial branch must NOT re-interpret it as a serial.
+    assert persons_tree._parse_year("1023") is None