HeartSaVioR commented on code in PR #48838:
URL: https://github.com/apache/spark/pull/48838#discussion_r1857706752
##########
python/pyspark/sql/pandas/serializers.py:
##########
@@ -1201,11 +1201,89 @@ def generate_data_batches(batches):
def dump_stream(self, iterator, stream):
"""
- Read through an iterator of (iterator of pandas DataFrame), serialize
them to Arrow
- RecordBatches, and write batches to stream.
+ Read through chained return results from a single partition of
handleInputRows.
Review Comment:
Any reason we can't do this in `if key is None:` in `transformWithStateUDF`
and `transformWithStateWithInitStateUDF`?
This was my suggestion and I believe you can just do retrieve expired timers
and timestamps, and call handleExpiredTimer() with these information, and done.
I don't think this complication is necessary - if we can't do this in `key is
None` in some reason, I suspect fixing that would be much easier.
--
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]
