[
https://issues.apache.org/jira/browse/DERBY-6004?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13510048#comment-13510048
]
Rick Hillegas commented on DERBY-6004:
--------------------------------------
Hi Dag,
I went through exactly the same reasoning, starting out with ORDERED BY and
ending up with ORDER BY to avoid the extra keyword.
I think that collations introduce extra complexity but I would lump this case
together with the more general problem of an ordered table function which has a
bug and returns a row order which violates its contract. Here the table
function declares that it returns rows in the database's collation order, but
in fact it doesn't. I see that as a bug in the table function. I think the
collation case would be handled by the solution to the general problem: have
Derby raise an exception when it detects an out-of-sequence row.
Thanks,
-Rick
> Make it possible for table functions to declare that they return sorted
> results and make the optimizer exploit this information.
> --------------------------------------------------------------------------------------------------------------------------------
>
> Key: DERBY-6004
> URL: https://issues.apache.org/jira/browse/DERBY-6004
> Project: Derby
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 10.10.0.0
> Reporter: Rick Hillegas
>
> A derby-dev discussion has highlighted the need for table functions to return
> sorted results which the optimizer can exploit. See
> http://old.nabble.com/Have-Derby-Network-Server-having-an-out-of-memory-%28PermGen%29-to34704415.html#a34706722
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
