fix(targets): Safely skip parsing record field as date-time if it is missing in schema

singer_sdk/sinks/core.py

@@ -365,12 +365,15 @@ def _parse_timestamps_in_record(
             schema: TODO
             treatment: TODO
         """
-        for key in record:
+        for key, value in record.items():
+            if key not in schema["properties"]:
+                self.logger.debug("No schema for record field '%s'", key)
+                continue
             datelike_type = get_datelike_property_type(schema["properties"][key])
             if datelike_type:
-                date_val = record[key]
+                date_val = value
                 try:
-                    if record[key] is not None:
+                    if value is not None:
                         date_val = parser.parse(date_val)
                 except parser.ParserError as ex:
                     date_val = handle_invalid_timestamp_in_record(

tests/core/sinks/test_validation.py

@@ -0,0 +1,44 @@
+from __future__ import annotations
+
+import datetime
+
+from tests.conftest import BatchSinkMock, TargetMock
+
+
+def test_validate_record():
+    target = TargetMock()
+    sink = BatchSinkMock(
+        target,
+        "users",
+        {
+            "type": "object",
+            "properties": {
+                "id": {"type": "integer"},
+                "created_at": {"type": "string", "format": "date-time"},
+            },
+        },
+        ["id"],
+    )
+
+    record_message = {
+        "type": "RECORD",
+        "stream": "users",
+        "record": {
+            "id": 1,
+            "created_at": "2021-01-01T00:00:00+00:00",
+            "missing_datetime": "2021-01-01T00:00:00+00:00",
+        },
+        "time_extracted": "2021-01-01T00:00:00+00:00",
+        "version": 100,
+    }
+    record = record_message["record"]
+    updated_record = sink._validate_and_parse(record)
+    assert updated_record["created_at"] == datetime.datetime(
+        2021,
+        1,
+        1,
+        0,
+        0,
+        tzinfo=datetime.timezone.utc,
+    )
+    assert updated_record["missing_datetime"] == "2021-01-01T00:00:00+00:00"