[
https://issues.apache.org/jira/browse/BEAM-8655?focusedWorklogId=344384&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-344384
]
ASF GitHub Bot logged work on BEAM-8655:
----------------------------------------
Author: ASF GitHub Bot
Created on: 15/Nov/19 16:22
Start Date: 15/Nov/19 16:22
Worklog Time Spent: 10m
Work Description: robertwb commented on pull request #10096: [BEAM-8655]
Strengthen trigger transcript tests.
URL: https://github.com/apache/beam/pull/10096#discussion_r346678764
##########
File path: sdks/python/apache_beam/transforms/trigger_test.py
##########
@@ -859,13 +892,92 @@ class
WeakTestStreamTranscriptTest(BaseTestStreamTranscriptTest):
allow_out_of_order = True
+class BatchTranscriptTest(TranscriptTest):
+
+ def _execute(
+ self, window_fn, trigger_fn, accumulation_mode, timestamp_combiner,
+ transcript, spec):
+ if timestamp_combiner == TimestampCombiner.OUTPUT_AT_EARLIEST_TRANSFORMED:
+ self.skipTest(
+ 'Non-fnapi timestamp combiner: %s' % spec.get('timestamp_combiner'))
+
+ if accumulation_mode != AccumulationMode.ACCUMULATING:
Review comment:
In discarding mode it's non-trivial to compute what the final frames should
be when the trigger is only fired once. (One could try to take the union of
values, but the timestamps would also be different in some cases as well.)
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 344384)
Time Spent: 1h (was: 50m)
> Strengthen trigger transcript tests
> -----------------------------------
>
> Key: BEAM-8655
> URL: https://issues.apache.org/jira/browse/BEAM-8655
> Project: Beam
> Issue Type: Bug
> Components: testing
> Reporter: Robert Bradshaw
> Priority: Major
> Time Spent: 1h
> Remaining Estimate: 0h
>
> We should run them with multiple keys, using combiners as well as
> group-by-key, and when possible in batch mode.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
