[
https://issues.apache.org/jira/browse/DERBY-2710?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12499692
]
Mamta A. Satoor commented on DERBY-2710:
----------------------------------------
The collation rule for c1 like c2 should be same as c1 = c2 as per SQL
specification.
As per SQL spec, Section 8.5 <like predicate>, Syntax Rules 3d),
"The collation used for <like predicate> is determined by applying Subclause
9.13, "Collation determination", with operands CVE, PC, and (if specified) EC."
As per SQL spec, Section 8.2 <comparison predicate>, General Rules 3a),
"Let CS be the collation as determined by Subclause 9.13, "Collation
determination", for the declared types of the two character strings."
Please let me know if you have any further questions in this area.
> disable dynamic like optimization for collated databases
> --------------------------------------------------------
>
> Key: DERBY-2710
> URL: https://issues.apache.org/jira/browse/DERBY-2710
> Project: Derby
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 10.3.0.0
> Reporter: Mike Matrigali
> Assignee: Mike Matrigali
>
> I think Code needs to be added to
> opensource/java/engine/org/apache/derby/impl/sql/compile/LikeEscapeOperatorNode.java
> to disable like optimization for databases with non-default collation. There
> is left over code in this routine that says it disables it for national char
> by adding a cast to one of the two sides which I assume then stops some other
> code from considering the
> optimization.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
