jingz-db commented on code in PR #48838:
URL: https://github.com/apache/spark/pull/48838#discussion_r1861301227
##########
python/pyspark/sql/tests/pandas/test_pandas_transform_with_state.py:
##########
@@ -858,33 +913,30 @@ def init(self, handle: StatefulProcessorHandle) -> None:
self.handle = handle
self.max_state = handle.getValueState("max_state", state_schema)
- def handleInputRows(
- self, key, rows, timer_values, expired_timer_info
- ) -> Iterator[pd.DataFrame]:
- if expired_timer_info.is_valid():
- self.max_state.clear()
- self.handle.deleteTimer(expired_timer_info.get_expiry_time_in_ms())
- str_key = f"{str(key[0])}-expired"
- yield pd.DataFrame(
- {"id": (str_key,), "timestamp":
str(expired_timer_info.get_expiry_time_in_ms())}
- )
+ def handleExpiredTimer(self, key, timer_values, expired_timer_info) ->
Iterator[pd.DataFrame]:
+ self.max_state.clear()
+ self.handle.deleteTimer(expired_timer_info.get_expiry_time_in_ms())
+ str_key = f"{str(key[0])}-expired"
+ yield pd.DataFrame(
+ {"id": (str_key,), "timestamp":
str(expired_timer_info.get_expiry_time_in_ms())}
+ )
- else:
- timestamp_list = []
- for pdf in rows:
- # int64 will represent timestamp in nanosecond, restore to
second
- timestamp_list.extend((pdf["eventTime"].astype("int64") //
10**9).tolist())
+ def handleInputRows(self, key, rows, timer_values) ->
Iterator[pd.DataFrame]:
+ timestamp_list = []
+ for pdf in rows:
+ # int64 will represent timestamp in nanosecond, restore to second
+ timestamp_list.extend((pdf["eventTime"].astype("int64") //
10**9).tolist())
- if self.max_state.exists():
- cur_max = int(self.max_state.get()[0])
- else:
- cur_max = 0
- max_event_time = str(max(cur_max, max(timestamp_list)))
+ if self.max_state.exists():
+ cur_max = int(self.max_state.get()[0])
+ else:
+ cur_max = 0
+ max_event_time = str(max(cur_max, max(timestamp_list)))
- self.max_state.update((max_event_time,))
-
self.handle.registerTimer(timer_values.get_current_watermark_in_ms())
+ self.max_state.update((max_event_time,))
+ self.handle.registerTimer(timer_values.get_current_watermark_in_ms() +
1)
Review Comment:
I modified this to current batch timestamp + 1 for testing with more common
use cases as registering with current batch timestamp is not a very common use
case.
--
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: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
