db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mike Matrigali <mikem_...@sbcglobal.net>
Subject Re: [jira] Commented: (DERBY-2656) Run suites.All against a collated database
Date Wed, 11 Jul 2007 22:10:30 GMT
Adding some targeted tests to be run every run into collation sounds 
good.  I agree the metadata tests are a good candidate.

It would also be nice if we could make running all the tests against
collation repeatable.  I agree that many are likely not adding value
so would really not want it part of the standard derbyall - but does seem
like picking and choosing might miss something and sometimes it is just
easier to throw hardware at it.  At first I didn't think we would be
able to easily do it as I thought different collations might correctly
return different results for a number of tests and thus provide lots
of "expected" diffs - thus expecting the exercise to be a one off.  But
the initial run showed that at least the current tests with the 
collation chosen worked.

Also how hard would it be to checkin something that would allow someone 
to run all or almost all junit tests against a collated test.  Once this
work was in then maybe those with the hardware could figure out a way
to run this more than once if it doesn't make sense to run every day.
Now that we have a good set of junit tests maybe 10.4 is a good time to
produce different configurations that easily run all the tests under
different interesting configurations.  Let's leverage all those junit
conversions.

Daniel John Debrunner (JIRA) wrote:
>     [ https://issues.apache.org/jira/browse/DERBY-2656?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12511843
] 
> 
> Daniel John Debrunner commented on DERBY-2656:
> ----------------------------------------------
> 
> Or a bug was introduced after you ran the tests, or a new test case was added ...
> 
> Rather than a one-off (two-off) task, I wonder if we could make this repeatable by adding
more existing tests into the CollationTest.
> Adding the DatabaseMetaDataTest is a good choice because meta data is where we know there
might be issues given the different collations.
> Are there any other potential areas that could be covered by existing tests?
> 
> E.g. testing of the math funtions (sin cos etc.) is very unlikely to be affected by collation
so little benefit in adding that test case but what about others?
> 
>> Run suites.All against a collated database
>> ------------------------------------------
>>
>>                 Key: DERBY-2656
>>                 URL: https://issues.apache.org/jira/browse/DERBY-2656
>>             Project: Derby
>>          Issue Type: Task
>>          Components: Test
>>    Affects Versions: 10.3.0.0
>>            Reporter: Kathey Marsden
>>            Assignee: Kathey Marsden
>>            Priority: Minor
>>             Fix For: 10.3.0.0
>>
>>         Attachments: DERBY-2656_diff.txt
>>
>>
>> It would be good to leverage the existing tests to exercise the collation code. 
Investigate running all the tests against a collated database.
>> Kathey
> 

Mime
View raw message