beam-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jozef Vilcek (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (BEAM-5180) Broken FileResultCoder via parseSchema change
Date Tue, 21 Aug 2018 10:55:00 GMT

    [ https://issues.apache.org/jira/browse/BEAM-5180?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16587285#comment-16587285
] 

Jozef Vilcek commented on BEAM-5180:
------------------------------------

For reference, this is maybe a root cause which this change in scheme parsing triggered
https://issues.apache.org/jira/browse/BEAM-2277?focusedCommentId=16587202&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#comment-16587202

> Broken FileResultCoder via parseSchema change
> ---------------------------------------------
>
>                 Key: BEAM-5180
>                 URL: https://issues.apache.org/jira/browse/BEAM-5180
>             Project: Beam
>          Issue Type: Bug
>          Components: sdk-java-core
>    Affects Versions: 2.6.0
>            Reporter: Jozef Vilcek
>            Assignee: Kenneth Knowles
>            Priority: Blocker
>          Time Spent: 1h
>  Remaining Estimate: 0h
>
> Recently this commit
> [https://github.com/apache/beam/commit/3fff58c21f94415f3397e185377e36d3df662384]
> introduced more strict schema parsing which is breaking the contract between _FileResultCoder_ and
_FileSystems.matchNewResource()_.
> Coder takes _ResourceId_ and serialize it via `_toString_` methods and then relies on
filesystem being able to parse it back again. Having strict _scheme://_ breaks this at least
for Hadoop filesystem which use _URI_ for _ResourceId_ and produce _toString()_ in form
of `_hdfs:/some/path_`
> I guess the _ResourceIdCoder_ is suffering the same problem.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message