[JBoss JIRA] Created: (TEIID-894) Query inexplicably failing Query Testing only with MySQL
by Paul Nittel (JIRA)
Query inexplicably failing Query Testing only with MySQL
--------------------------------------------------------
Key: TEIID-894
URL: https://jira.jboss.org/jira/browse/TEIID-894
Project: Teiid
Issue Type: Bug
Components: Query Engine
Affects Versions: 7.0
Environment: Fedora 10, Teiid 7.0.0 M1
Reporter: Paul Nittel
Assignee: Steven Hawkins
Series 9000, #7 fails with MySql and passes with all the other sources.
Original Query
SELECT BQT1.SmallA.IntKey, BQT2.SmallB.FloatNum FROM BQT1.SmallA, BQT2.SmallB WHERE BQT1.SmallA.IntKey = BQT2.SmallB.FloatNum AND BQT1.SmallA.IntKey >= 0 AND BQT2.SmallB.IntKey >= 0 ORDER BY BQT1.SmallA.IntKey;
It's supposed to return 25 rows, but returns 0. Let's ignore the "ORDER BY" since it just adds to the code (and doesn't factor into the issue since it works the same with or without).
If I remove the "AND BQT2.SmallB.IntKey >= 0", it works! I get 25 rows.
SELECT BQT1.SmallA.IntKey, BQT2.SmallB.FloatNum FROM BQT1.SmallA, BQT2.SmallB WHERE BQT1.SmallA.IntKey = BQT2.SmallB.FloatNum AND BQT1.SmallA.IntKey >= 0;
If I switch it so the first condition is removed and the second remains, it FAILS
SELECT BQT1.SmallA.IntKey, BQT2.SmallB.FloatNum FROM BQT1.SmallA, BQT2.SmallB WHERE BQT1.SmallA.IntKey = BQT2.SmallB.FloatNum AND BQT2.SmallB.IntKey >= 0;
What doesn't it like about "BQT2.SmallB >= 0" ?? I ran this query--and the one with only "BQT1.SmallA >= 0"--with OPTION DEBUG.
Stuff is attached.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 6 months
[JBoss JIRA] Resolved: (TEIID-152) New SQL directive OPTION NOJOINPUSHDOWN that tells query planner to not push down any joins
by Steve Hawkins (JIRA)
[ https://jira.jboss.org/browse/TEIID-152?page=com.atlassian.jira.plugin.sy... ]
Steve Hawkins resolved TEIID-152.
---------------------------------
Fix Version/s: (was: 8.x)
Resolution: Won't Fix
With the other improvements to capabilities there's shouldn't be a need to work this issue.
> New SQL directive OPTION NOJOINPUSHDOWN that tells query planner to not push down any joins
> -------------------------------------------------------------------------------------------
>
> Key: TEIID-152
> URL: https://jira.jboss.org/browse/TEIID-152
> Project: Teiid
> Issue Type: Feature Request
> Components: Query Engine
> Affects Versions: 8.x
> Reporter: Greg Haber
> Priority: Minor
>
> This capability has been requested by a customer(see I-T ticket 217562). They want us to add in the following functionality:
> -add an OPTION directive (OPTION NOJOINPUSHDOWN) to MetaMatrix SQL that tells the query planner whether or not to attempt to push down joins for a particular query. This would be a general purpose MetaMatrix SQL change. The default would be for our query planner to act as it does today (do join pushdown as allowed by connector capabilities), and the behavior if this OPTION directive was used would be to instead not do any join pushdown for any part of the user query, even if allowed by connector capabilities.
> What this would allow them to do, if JBEDSP-348 is also implemented, is to leverage the limited join capability that SFDC has via its "relationship query" functionality. Because SFDC doesn't have full join capabilities, this would allow them to first try to submit a query against a VDB that uses SFDC as a source and see if the part of the join that the query planner pushes down to SFDC is something that SFDC can actually do. If not, the SFDC connector would throw an error, and the customer would resubmit their query with this OPTION NOJOINPUSHDOWN option.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 6 months
[JBoss JIRA] Resolved: (TEIID-188) Improve resolution of parameter references used in compare criteria
by Steve Hawkins (JIRA)
[ https://jira.jboss.org/browse/TEIID-188?page=com.atlassian.jira.plugin.sy... ]
Steve Hawkins resolved TEIID-188.
---------------------------------
Fix Version/s: (was: 8.x)
Resolution: Duplicate Issue
Duplicate of TEIID-235
> Improve resolution of parameter references used in compare criteria
> -------------------------------------------------------------------
>
> Key: TEIID-188
> URL: https://jira.jboss.org/browse/TEIID-188
> Project: Teiid
> Issue Type: Feature Request
> Components: Query Engine
> Affects Versions: 8.x
> Reporter: Mark Drilling
> Assignee: Steven Hawkins
> Priority: Minor
> Attachments: GregComments.txt
>
> Original Estimate: 1 day
> Remaining Estimate: 1 day
>
> Entering this JIRA from Issue 185605 in issue tracker. Greg Haber reports problem using Penrose off the shelf application. Example query as follows
> select distinct petOwnersFromMM.owner, petOwnersFromMM.name from petVirtualDB.gregtest_virtual.gregtest.pet petOwnersFromMM where lower(petOwnersFromMM.owner) = lower(?) order by petOwnersFromMM.owner
> He is getting "com.metamatrix.jdbc.MMSQLException: The function 'lower(?)' has more than one possible signature." since the lower function is ambiguous.
> Workaround is to use "lower(convert(?, string))" in the sql. There is a statement in the release notes for a future enhancement that we could use the type info from the other side of the compare criteria to determine what the type should be.
> I will attach Gregs further comments for reference.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 6 months
[JBoss JIRA] Updated: (TEIID-168) Refine caching behavior
by Steve Hawkins (JIRA)
[ https://jira.jboss.org/browse/TEIID-168?page=com.atlassian.jira.plugin.sy... ]
Steve Hawkins updated TEIID-168:
--------------------------------
Fix Version/s: 7.1
(was: 7.0)
Labels: performance cache (was: performance)
Moving remaining work with relation and procedure caching to 7.1. ResultSet caching and proper buffering of cache results was handled in 7.0.
> Refine caching behavior
> -----------------------
>
> Key: TEIID-168
> URL: https://jira.jboss.org/browse/TEIID-168
> Project: Teiid
> Issue Type: Feature Request
> Components: JDBC Connector, Query Engine
> Affects Versions: 7.0
> Reporter: Steven Hawkins
> Assignee: Steven Hawkins
> Fix For: 7.1
>
>
> Connector resultset caching should be replaced with relation and procedure caching. This would initially be backed by the buffermanager, based upon hints/costing, and be utilized by the planner rather than the resolver. It would be good if we also detected the level of determinism.
> Query resultset caching should also be backed by the buffermanager.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 6 months