[ 
https://issues.apache.org/jira/browse/BEAM-10697?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17179223#comment-17179223
 ] 

Daniel Oliveira commented on BEAM-10697:
----------------------------------------

Would this count as a regression for the 2.24.0 release? And how likely is it 
to happen in real user pipelines? Unfortunately 2.24.0 is still supposed to 
support Python 2.7, so this bug might need to get fixed regardless if it would 
impact users.

I suppose another alternative is to just add this as a known issue and suggest 
python 2.7 users stick with 2.23.0.

> Python Precommit failing due to type check failure.
> ---------------------------------------------------
>
>                 Key: BEAM-10697
>                 URL: https://issues.apache.org/jira/browse/BEAM-10697
>             Project: Beam
>          Issue Type: Bug
>          Components: sdk-py-core, test-failures
>            Reporter: Daniel Oliveira
>            Assignee: Udi Meiri
>            Priority: P0
>             Fix For: 2.24.0
>
>          Time Spent: 40m
>  Remaining Estimate: 0h
>
> Test suite: https://ci-beam.apache.org/job/beam_PreCommit_Python_Cron/
> First failure: https://ci-beam.apache.org/job/beam_PreCommit_Python_Cron/3110/
> Looks like it's failing for about a day now.
> Error:
> {noformat}
> self = <apache_beam.transforms.ptransform_test.PTransformTypeCheckTestCase 
> testMethod=test_pardo_properly_type_checks_using_type_hint_decorators>
>     def test_pardo_properly_type_checks_using_type_hint_decorators(self):
>       @with_input_types(a=str)
>       @with_output_types(typing.List[str])
>       def to_all_upper_case(a):
>         return [a.upper()]
>     
>       # If this type-checks than no error should be raised.
>       d = (
>           self.p
>           | 'T' >> beam.Create(['t', 'e', 's', 't']).with_output_types(str)
>           | 'Case' >> beam.FlatMap(to_all_upper_case))
>       assert_that(d, equal_to(['T', 'E', 'S', 'T']))
>       self.p.run()
>     
>       # Output type should have been recognized as 'str' rather than 
> List[str] to
>       # do the flatten part of FlatMap.
> >     self.assertEqual(str, d.element_type)
> E     AssertionError: <type 'str'> != Any
> apache_beam/transforms/ptransform_test.py:1182: AssertionError
> {noformat}



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to