db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "A B (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DERBY-3061) Wrong results from query with two conjuncts
Date Fri, 07 Sep 2007 15:57:30 GMT

    [ https://issues.apache.org/jira/browse/DERBY-3061?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12525757
] 

A B commented on DERBY-3061:
----------------------------

I ran the repro with svn # 541822 ant it fails with an ASSERT failure:

shared.common.sanity.AssertFailure: ASSERT FAILED All multi-probing result sets are expected
to have a single key that is both the start key AND the stop key, but that is not the case.

    at shared.common.sanity.SanityManager.ASSERT(SanityManager.java(Compiled Code))
    at impl.sql.execute.MultiProbeTableScanResultSet.<init>(MultiProbeTableScanResultSet.java:158)
    ...

But that ASSERT was removed as part of changes for DERBY-2740, hence no failure in 10.3.1.4
nor latest runk.  It looks like that fix was intended to allow different start/stop keys to
work (hence the assertion was removed), but apparently it does not do the correct thing after
all.

> Wrong results from query with two conjuncts
> -------------------------------------------
>
>                 Key: DERBY-3061
>                 URL: https://issues.apache.org/jira/browse/DERBY-3061
>             Project: Derby
>          Issue Type: Bug
>          Components: SQL
>    Affects Versions: 10.3.1.4
>            Reporter: Rick Hillegas
>            Priority: Critical
>
> Tim Dudgeon, on the user list, reports that the following query returns no results in
10.3.1.4 but works correctly in 10.2. I have verified that the query returns no results in
the mainline as well. If you eliminate either of the the conjuncts, then the query returns
the correct results:
> SELECT MYTABLE.MY_ID
>  FROM MYTABLE
>  WHERE MYTABLE.MY_ID < 100 AND MYTABLE.MY_ID IN (
> 2,15,19,20,21,48,49
> )
> Here is a more complete script which demonstrates the problem:
> drop table mytable;
> create table mytable ( id int primary key );
> insert into mytable ( id )
> values
> ( 0 ), ( 1 ), ( 2 ), ( 3 ), ( 4 ), ( 5 ), ( 6 ), ( 7 ), ( 8 ), ( 9 );
> insert into mytable select id + 10 from mytable;
> insert into mytable select id + 20 from mytable;
> insert into mytable select id + 40 from mytable;
> insert into mytable select id + 100 from mytable;
> select mytable.id
> from mytable
> where mytable.id < 100;
> select mytable.id
> from mytable
> where mytable.id in ( 2, 15, 19, 20, 21, 48, 49 );
> select mytable.id
> from mytable
> where mytable.id < 100
> and mytable.id in ( 2, 15, 19, 20, 21, 48, 49 );
>  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message