You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@drill.apache.org by "Jason Altekruse (JIRA)" <ji...@apache.org> on 2014/02/20 04:02:21 UTC

[jira] [Commented] (DRILL-375) VarBinary hash function computing hash for entire buffer

    [ https://issues.apache.org/jira/browse/DRILL-375?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13906535#comment-13906535 ] 

Jason Altekruse commented on DRILL-375:
---------------------------------------

patch is saved with .html extension. I doubt git will care, but when trying to view the patch in the browser it removes all of the whitespace and smashes all of the text together.

> VarBinary hash function computing hash for entire buffer
> --------------------------------------------------------
>
>                 Key: DRILL-375
>                 URL: https://issues.apache.org/jira/browse/DRILL-375
>             Project: Apache Drill
>          Issue Type: Bug
>            Reporter: Steven Phillips
>            Assignee: Steven Phillips
>         Attachments: DRILL-375
>
>
> the hash function for VarBinary data type should be computing the hash function for a particular record. But it is currently computing for the entire buffer, which is all of the data for the ValueVector.
> the fix is to set the starting and ending indices for the particular value.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)