[
https://issues.jboss.org/browse/TEIID-2564?page=com.atlassian.jira.plugin...
]
Steven Hawkins resolved TEIID-2564.
-----------------------------------
Resolution: Done
The initial fix (in 7.7.1) was not generally correct as the corrected symbol map needed to
reference the subquery expression rather than the derived grouping column.
Correlated subqueries throws TEIID30328
---------------------------------------
Key: TEIID-2564
URL:
https://issues.jboss.org/browse/TEIID-2564
Project: Teiid
Issue Type: Bug
Components: Query Engine
Affects Versions: 8.4
Reporter: Warren Gibson
Assignee: Steven Hawkins
Fix For: 8.4.1, 8.5
Attachments: server.log
Non-pushed correlated subqueries that use a correlation variable that is a grouping
column where the subquery is turned into a semi join (either through the MJ hint or
costing) will fail to find the correlation values.
For example the following against Netezza (which does not support correlated
subqueries):
SELECT A.INTKEY, A.STRINGNUM FROM BQT1.SMALLA AS A WHERE CONVERT(LONGNUM,
STRING) = 8 GROUP BY A.INTKEY, A.STRINGNUM HAVING A.STRINGNUM = (SELECT
MAX(B.STRINGNUM) FROM BQT1.SMALLA AS B WHERE A.INTKEY = B.INTKEY
throws "TEIID30328 Unable to evaluate A.IntKey: No value was available"
If the subplan is not converted to a join or the correlated column is not a grouping
expression, then there is no issue.
--
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