[
https://issues.apache.org/jira/browse/CASSANDRA-5443?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13641701#comment-13641701
]
Sylvain Lebresne commented on CASSANDRA-5443:
---------------------------------------------
bq. With regard to the result, it would be consistent with SQL to return a
count of affected logical rows; not a boolean.
I suppose that would work, but it feels to me that returning whether the
condition did apply is more "direct" that the number of affected rows.
I understand that the motivation is to mimick SQL more, but then we would need
to return the number of affected rows for normal updates too (otherwise that
would feel inconsistent to me). Also, my understanding (that is possibly
broken) is that in SQL the number of affected rows is returned as some metadata
of the query (typically, in JDBC, as the result value of executeUpdate()), not
as a result set as we would do here, so would we be really fully consistent
with SQL anyway?
Anyway, not that I'm really against the idea, just wondering.
> Add CAS CQL support
> -------------------
>
> Key: CASSANDRA-5443
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5443
> Project: Cassandra
> Issue Type: Sub-task
> Components: API, Core
> Reporter: Jonathan Ellis
> Assignee: Sylvain Lebresne
> Fix For: 2.0
>
>
--
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
