db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Henri van de Scheur <Henri.Vandesch...@Sun.COM>
Subject Re: Code Coverage Results for 10.4.1.3
Date Thu, 08 May 2008 09:24:07 GMT
Hi again!

I do not understand these differences. I have instrumented the same 
jars, but not all. I did not include derbyTesting.jar and the 
derbyLocale*.jars. So I instrumented the 5 following jars: derby.jar, 
derbyclient.jar, derbynet.jar, derbyrun.jar and derbytools.jar.
I hope to find an explanation when I can compare your reports with mine.

Henri
Manjula Kutty wrote:
> I took the 10.4.1.3 <http://10.4.1.3> Release candidate jars.
> I used the sun jdks. Also is there a chance that I was giving all 
> permission to all the codebases in the derby_tests policy file?
>  
> -Manjula
>
>  
> On 5/7/08, *Henri van de Scheur* <Henri.Vandescheur@sun.com 
> <mailto:Henri.Vandescheur@sun.com>> wrote:
>
>     Hi Manjula!
>
>     I was able to fix the problem with linking the source-code, so now
>     the site I mentioned before shows correct and complete results.
>     To summarize and compare with your results:
>     OVERALL COVERAGE SUMMARY
>     --------------------------------------------------
>     all classes: 90%  (1264/1397)
>     Method : 75%  (16948/22513)
>     Block : 73%  (479495/660441)
>     Line : 71%  (100760/141054)
>
>     OVERALL STATS SUMMARY
>     total packages: 89
>     total executable files: 1280
>     total classes: 1397
>     total methods: 22513
>     total executable lines: 141054
>
>     So we see a couple of (general) differences:
>
>        1. In general you report higher numbers for the totals
>        2. In general I find higher numbers for coverage
>
>     Can this be explained by just the differences in our jvm's (as
>     different vendors) or should we investigate more?
>     For my numbers I did a little workaround regarding the
>     source-code: I used the source-code from the 10.4-branch and not
>     the official 10.4.1.3 <http://10.4.1.3/>
>
>     Thanks !
>
>     Henri
>
>
>     Manjula Kutty wrote:
>>     Hi Henri
>>      
>>     I ran suites.all. I used my derby_tests policy (Actually the one
>>     which gives all permission). I ran it with jdk142, 15 and 16. I
>>     will be linking the full report to the wiki.
>>      
>>     Thanks
>>     Manjula
>>
>>      
>>     On 5/6/08, *Henri van de Scheur* <Henri.Vandescheur@sun.com
>>     <mailto:Henri.Vandescheur@sun.com>> wrote:
>>
>>         Manjula Kutty wrote:
>>
>>             OVERALL COVERAGE SUMMARY
>>             --------------------------------------------------
>>             all classes: 87%  (1226/1404) Method : 74%  (16629/22556)
>>             Block : 69%  (486715/704201)
>>             Line : 69%  (101022.4/146764)
>>
>>             OVERALL STATS SUMMARY
>>             total packages: 90
>>             total executable files: 1286
>>             total classes: 1404
>>             total methods: 22556
>>             total executable lines: 146764
>>
>>
>>
>>             -- 
>>             Thanks,
>>             Manjula.
>>
>>         Hi Manjula!
>>
>>         What testsuite did you use and which jvm-versions?
>>         I used Emma to run with suitesAll on jvm1.4, jvm1.5 and
>>         jvm1.6 (both 32- and 64-mode). Results can be found here:
>>         http://dbtg.thresher.com/derby/test/10.4.1.3_RC/Emma.html
>>         Unfortunately I have had some problems linking the
>>         source-code to these reports, but I plan to do that within
>>         the next 24 hours (if svn allows me....).
>>
>>         Thanks,
>>
>>         Henri
>>
>
>
>
>
> -- 
> Thanks,
> Manjula. 


-- 

With regards,



Henri van de Scheur, Database Technology Group,
Sun Microsystems, Trondheim, Norway


Mime
View raw message