hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gopal V (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-9684) Incorrect disk range computation in ORC because of optional stream kind
Date Fri, 13 Feb 2015 20:04:12 GMT

    [ https://issues.apache.org/jira/browse/HIVE-9684?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14320674#comment-14320674
] 

Gopal V commented on HIVE-9684:
-------------------------------

LGTM +1.

This needs the extra condition because unknown enum fields default to the first field (PRESENT).


> Incorrect disk range computation in ORC because of optional stream kind
> -----------------------------------------------------------------------
>
>                 Key: HIVE-9684
>                 URL: https://issues.apache.org/jira/browse/HIVE-9684
>             Project: Hive
>          Issue Type: Bug
>          Components: File Formats
>    Affects Versions: 1.0.0, 1.1.0, 1.0.1
>            Reporter: Prasanth Jayachandran
>            Assignee: Prasanth Jayachandran
>            Priority: Critical
>         Attachments: HIVE-9684.1.patch, HIVE-9684.branch-1.0.patch, HIVE-9684.branch-1.1.patch
>
>
> HIVE-9593 changed all required fields in ORC protobuf message to optional field. But
DiskRange computation and stream creation code assumes existence of stream kind everywhere.
This leads to incorrect calculation of diskranges resulting in out of range exceptions. The
proper fix is to check if stream kind exists using stream.hasKind() before adding the stream
to disk range computation.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message