[
https://issues.apache.org/jira/browse/TRAFODION-2843?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16289915#comment-16289915
]
ASF GitHub Bot commented on TRAFODION-2843:
-------------------------------------------
GitHub user DaveBirdsall opened a pull request:
https://github.com/apache/incubator-trafodion/pull/1340
[TRAFODION-2843] Fix outer join on aggregate bug
Formerly, the code was allowing an uncovered aggregate expression to be
generated as a before join predicate. This lead to a Generator assert. The test
for uncovered expressions in Join::preCodeGen has been tightened up so this
does not happen.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/DaveBirdsall/incubator-trafodion Trafodion2843
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/incubator-trafodion/pull/1340.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #1340
----
commit ccb17c28e7fa74b5c58dbf37a02239316184bdd5
Author: Dave Birdsall <dbirds...@apache.org>
Date: 2017-12-13T21:04:44Z
[TRAFODION-2843] Fix outer join on aggregate bug
----
> Internal error (or core) on full outer join on an aggregate
> -----------------------------------------------------------
>
> Key: TRAFODION-2843
> URL: https://issues.apache.org/jira/browse/TRAFODION-2843
> Project: Apache Trafodion
> Issue Type: Bug
> Components: sql-cmp
> Affects Versions: 2.3-incubating
> Environment: All
> Reporter: David Wayne Birdsall
> Assignee: David Wayne Birdsall
>
> The following sqlci script demonstrates the problem. The "prepare x1" fails
> with an error 7000 (generator internal error) on release builds, and cores on
> debug builds.
> ?section setup
> drop schema if exists mytest cascade;
> create schema if not exists mytest;
> set schema mytest;
> Create table D03s
> (
> pk int not null not droppable primary key
> , val01 int
> , val02 int
> );
> Create table F01s
> (
> pk int not null not droppable primary key
> , fk_d01 int not null
> , fk_d02 int not null
> , fk_d03 int not null
> , fk_d04 int not null
> , fk_d05 int not null
> , fk_d06 int not null
> , fk_d07 int not null
> , fk_d08 int not null
> , fk_d09 int not null
> , fk_d10 int not null
> , val01 int
> , val02 int
> , val01_d01 int
> , val02_d01 int
> , val01_d02 int
> , val02_d02 int
> , val01_d03 int
> , val02_d03 int
> ) salt using 8 partitions;
> insert into D03s
> select c1+c2*10+c3*100, c1, c1+c2*10
> from (values(1)) T
> transpose 0,1 as c1
> transpose 0,1 as c2
> transpose 0,1 as c3;
> insert with no rollback into F01s
> select c1+c2*10+c3*100+c4*1000+c5*10000+c6*100000
> ,c1
> ,c1+c2*10
> ,c1+c2*10+c3*100
> ,c1
> ,c1+c2*10
> ,c1+c2*10+c3*100
> ,c1
> ,c1+c2*10
> ,c1+c2*10+c3*100
> ,c1
> ,c1+c2*10
> ,mod(c1+c2*100+c3*100,200)
> ,mod(c1,3)
> ,mod(c1,6)
> ,mod(c1+c2*10,5)
> ,c1
> ,c1
> ,c1+c2*10
> from (values(1)) T
> transpose 0,1 as c1
> transpose 0,1 as c2
> transpose 0,1 as c3
> transpose 0,1 as c4
> transpose 0 as c5
> transpose 0 as c6
> ;
> ?section testit
> set schema mytest;
> prepare x1 from
> select F01s.val01, TD03.val01
> From F01s
> full outer join
> (select D03s.val01,count(D03s.pk)
> from D03s
> group by D03s.val01) as TD03(val01,pk)
> on (TD03.pk=F01s.fk_d03
> AND TD03.pk>0 );
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
