cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-5626) Support empty IN queries
Date Wed, 24 Jul 2013 09:51:49 GMT


Sylvain Lebresne commented on CASSANDRA-5626:

On the patch:
* The "simplification" in getKeyBounds is not completely correct.  If we do a 2ndary index
query with a EQ on the partition key (which is supported), we will still go through getKeyBounds
even though we don't use the token function (more precisely, this patch breaks the
dtest) We can still simplify things a bit, because we don't need the current generality, but
I think I'd rather open a separate ticket for that.
* SelectStatement.getSliceCommands() was assuming that names filters were never null, which
is not the case anymore with this patch. Concretely, I've pushed a test for this in dtests
( and with the attached patch there's a NPE thrown by this test.
* Why remove the break when checking if we have at least one indexed EQ clause?
* Nit: in getIndexExpressions, let's add an assert that we're not in the IN case, if only
for documenting that we only use the first element on purpose.

> Support empty IN queries
> ------------------------
>                 Key: CASSANDRA-5626
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 1.2.0
>            Reporter: Alexander Solovyev
>            Assignee: Aleksey Yeschenko
>            Priority: Minor
>              Labels: cql3
>         Attachments: 5626.txt
> It would be nice to have support of empty IN queries. 
> Example: "SELECT a FROM t WHERE aKey IN ()". 
> One of the reasons is to have such support in DataStax Java Driver (see discussion here:

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:

View raw message