You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@openjpa.apache.org by "Pinaki Poddar (JIRA)" <ji...@apache.org> on 2010/04/26 20:04:33 UTC

[jira] Commented: (OPENJPA-1440) Allow COUNT(*) instead of COUNT(t0.id) for simple queries as an option

    [ https://issues.apache.org/jira/browse/OPENJPA-1440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12861018#action_12861018 ] 

Pinaki Poddar commented on OPENJPA-1440:
----------------------------------------

Please correct the header on the TestCase attached with the patch.

> Allow COUNT(*) instead of COUNT(t0.id) for simple queries as an option
> ----------------------------------------------------------------------
>
>                 Key: OPENJPA-1440
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-1440
>             Project: OpenJPA
>          Issue Type: Improvement
>            Reporter: Pinaki Poddar
>            Assignee: Pinaki Poddar
>             Fix For: 2.0.0-beta
>
>         Attachments: OPENJPA-1440-1.2.x.patch.txt
>
>
> SQL generated for "select p for PObject p" is SELECT COUNT(t0.id) FROM POBJECT t0". This is valid and according to JPQL spec which does not accept COUNT(*).
> However, JDBC Drivers for legacy databases may not always support COUNT(columnName) as reported in [1].
> OpenJPA should support such cases with a boolean option DBDictionary.useWildCardForCount.
> [1] http://n2.nabble.com/COUNT-t-to-COUNT-tc4176827.html#a4176827

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.