vinodkc commented on code in PR #54237:
URL: https://github.com/apache/spark/pull/54237#discussion_r2789920280
##########
python/pyspark/sql/tests/test_python_streaming_datasource.py:
##########
@@ -509,6 +510,60 @@ def check_batch(df, batch_id):
q.awaitTermination(timeout=30)
self.assertIsNone(q.exception(), "No exception has to be propagated.")
+ def test_simple_stream_reader_offset_did_not_advance_raises(self):
+ """Validate that returning end == start with non-empty data raises
STREAM_READER_OFFSET_DID_NOT_ADVANCE."""
+ from pyspark.sql.datasource_internal import _SimpleStreamReaderWrapper
+
+ class BuggySimpleStreamReader(SimpleDataSourceStreamReader):
+ def initialOffset(self):
+ return {"offset": 0}
+
+ def read(self, start: dict):
+ # Bug: return same offset as end despite returning data
+ start_idx = start["offset"]
+ it = iter([(i,) for i in range(start_idx, start_idx + 3)])
+ return (it, {"offset": start_idx})
+
+ def readBetweenOffsets(self, start: dict, end: dict):
+ return iter([])
+
+ def commit(self, end: dict):
+ pass
+
+ reader = BuggySimpleStreamReader()
+ wrapper = _SimpleStreamReaderWrapper(reader)
+ with self.assertRaises(PySparkException) as cm:
+ wrapper.latestOffset({"offset": 0}, ReadAllAvailable())
+ self.assertEqual(
+ cm.exception.getCondition(),
+ "STREAM_READER_OFFSET_DID_NOT_ADVANCE",
+ )
+
+ def
test_simple_stream_reader_empty_iterator_start_equals_end_allowed(self):
+ """When read() returns end == start with an empty iterator, no
exception and no cache entry."""
+ from pyspark.sql.datasource_internal import _SimpleStreamReaderWrapper
+
+ class EmptyBatchReader(SimpleDataSourceStreamReader):
+ def initialOffset(self):
+ return {"offset": 0}
+
+ def read(self, start: dict):
+ # Valid: same offset as end but empty iterator (no data)
+ return (iter([]), {"offset": start["offset"]})
Review Comment:
Done
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
