db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Myrna van Lunteren (JIRA)" <j...@apache.org>
Subject [jira] Resolved: (DERBY-3831) RuntimeStatisticsParser can not distinguish names of table or index well and truly
Date Thu, 14 Aug 2008 00:26:44 GMT

     [ https://issues.apache.org/jira/browse/DERBY-3831?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Myrna van Lunteren resolved DERBY-3831.
---------------------------------------

       Resolution: Fixed
    Fix Version/s: 10.5.0.0
       Derby Info:   (was: [Patch Available])

This was a good find, thx...
However, when I ran the tests that use the modified methods in RuntimeStatisticsParser:
    ...lang.UpdateStatisticsTest
    ...lang.PredicatePushdownTest and the newly added
    ...lang.OptimizerOverridesTest
I got a failure in PredicatePushdownTest on the usedIndexRowToBaseRow method. The runtimestatistics
output actually never shows a space in that line after the tablename, but a : (colon).
I modified that method to look for a colon after the table name and committed the change with
revision 685733.


> RuntimeStatisticsParser can not distinguish names of table or index well and truly
> ----------------------------------------------------------------------------------
>
>                 Key: DERBY-3831
>                 URL: https://issues.apache.org/jira/browse/DERBY-3831
>             Project: Derby
>          Issue Type: Bug
>          Components: Tools
>            Reporter: Junjie Peng
>            Assignee: Junjie Peng
>             Fix For: 10.5.0.0
>
>         Attachments: derby-3831-1.patch, derby-3831-1.stat
>
>
> RuntimeStatisticsParser can not distinguish names of table or index well and truly.
> i.e. there are two indexes t1_c1c2 and t1_c1 on table t1. Execute "select * from t1 --DeRbY-pRoPeRtIeS
index = t1_c1c2"  on t1. Then,RuntimeStatisticsParser.usedSpecificIndexForIndexScan("T1","T1_C1C2")
will return true, while RuntimeStatisticsParser.usedSpecificIndexForIndexScan("T1","T1_C1")
will return true too. It's not correct.
> The similar errors also exist on usedIndexScan(), usedIndexRowToBaseRow(), usedDistinctScan()
and usedTableScan() in the class of  RuntimeStatisticsParser.

-- 
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