[
https://issues.apache.org/jira/browse/CASSANDRA-4327?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13408309#comment-13408309
]
Jonathan Ellis commented on CASSANDRA-4327:
-------------------------------------------
I think we should still disallow ordering w/ no key restriction, instead of
allowing but ignoring.
buildCompositeValue looks expensive, can we compare w/o that copy?
Nit: prefer isEmpty to size == 0.
> Sorting results when using IN()
> --------------------------------
>
> Key: CASSANDRA-4327
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4327
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.1.1
> Reporter: Stephen Powis
> Assignee: Pavel Yaskevich
> Priority: Minor
> Labels: cql3, patch, sort
> Fix For: 1.1.3
>
> Attachments: CASSANDRA-4327.patch, trunk-4327.txt
>
>
> Using the following test schema:
> CREATE TABLE test (
> my_id varchar,
> time_id uuid,
> value int,
> PRIMARY KEY (my_id, time_id)
> );
> When you issue a CQL3 query like:
> select * from test where my_id in('key1', 'key2') order by time_id;
> You receive the error:
> "Ordering is only supported if the first part of the PRIMARY KEY is
> restricted by an Equal"
> I'm including a patch I put together after spending an hour or two poking
> thru the code base that sorts the results for these types of queries. I'm
> hoping someone with a deeper understanding of Cassandra's code base can take
> a look at it, clean it up or use it as a starting place, and include it in an
> upcoming release.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
