You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@beam.apache.org by "Yu Feng (Jira)" <ji...@apache.org> on 2022/03/05 04:15:00 UTC

[jira] [Updated] (BEAM-14052) Input type annotation of MapTuple and FlatMapTuple are ignored

     [ https://issues.apache.org/jira/browse/BEAM-14052?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yu Feng updated BEAM-14052:
---------------------------

+ [~yeandy] 

> Input type annotation of MapTuple and FlatMapTuple are ignored
> --------------------------------------------------------------
>
>                 Key: BEAM-14052
>                 URL: https://issues.apache.org/jira/browse/BEAM-14052
>             Project: Beam
>          Issue Type: Bug
>          Components: sdk-py-core
>            Reporter: Yu Feng
>            Priority: P2
>
> This is a tracking bug for the remaining issue of type annotation support in MapTuple and FlatMapTuple described in PR #16351.
> Consider 2 cases:
>  - MapTuple(fn(a: int, b: int, *args))
>  - MapTuple(fn(a: int, b: int, args: Tuple[Any, ...]))
> We unpack (1, 2, 3, 4) to a=1, b=1, args=(3, 4) in first case,  but fail in second case.
> My concern is that the beam representation of these 2 signatures are *likely* the same:
> input_types[0] = (int, int, Tuple[Any, ...])
> input_types[1] = {}
>  



--
This message was sent by Atlassian Jira
(v8.20.1#820001)