MongoDB: Handle too large $date.$numberLong values gracefully

CHANGES.md

@@ -3,6 +3,7 @@
 ## Unreleased
 - DynamoDB: Change CrateDB data model to use (`pk`, `data`, `aux`) columns
   Attention: This is a breaking change.
+- MongoDB: Handle too large `$date.$numberLong` values gracefully
 
 ## 2024/09/26 v0.0.19
 - DynamoDB CDC: Fix `MODIFY` operation by propagating `NewImage` fully

src/commons_codec/transform/mongodb.py

@@ -123,7 +123,11 @@ def decode_extended_json(self, value: t.Dict[str, t.Any]) -> t.Any:
         type_ = next(iter(value))  # Get key of first item in dictionary.
         is_date_numberlong = type_ == "$date" and "$numberLong" in value["$date"]
         if is_date_numberlong:
-            out = dt.datetime.fromtimestamp(int(value["$date"]["$numberLong"]) / 1000, tz=dt.timezone.utc)
+            try:
+                out = dt.datetime.fromtimestamp(int(value["$date"]["$numberLong"]) / 1000, tz=dt.timezone.utc)
+            except ValueError as ex:
+                logger.error(f"Decoding legacy timestamp failed: {ex}. value={value}")
+                out = 0
         else:
             out = object_hook(value)
 

tests/transform/mongodb/data.py

@@ -58,7 +58,10 @@
         "code_bytes": {"$code": "ab\u0000ab\u0000"},
         "code_scope": {"$code": "abab", "$scope": {"x": {"$numberInt": "42"}}},
         "date_iso8601": {"$date": "2015-09-23T10:32:42.33Z"},
-        "date_numberlong": {"$date": {"$numberLong": "1356351330000"}},
+        "date_numberlong_valid": {"$date": {"$numberLong": "1356351330000"}},
+        "date_numberlong_invalid": {
+            "$date": {"$numberLong": "-9223372036854775808"}
+        },  # year -292275055 is out of range
         "dbref": {
             "$id": {"$oid": "56027fcae4b09385a85f9344"},
             "$ref": "foo",
@@ -169,7 +172,8 @@
             },
         },
         "date_iso8601": 1443004362000,
-        "date_numberlong": 1356351330000,
+        "date_numberlong_valid": 1356351330000,
+        "date_numberlong_invalid": 0,
         "dbref": {
             "$id": "56027fcae4b09385a85f9344",
             "$ref": "foo",