[Cdash] ITK appears in GDCM cdash

[David Cole]

How about crc32 concat length concat some substring from the data ...
?  Still easy to compute, but less likely to collide.


On Wednesday, September 2, 2009, Julien Jomier
<julien.jomier at kitware.com> wrote:
> Mathieu,
>
> This was a good one. So we found a limitation of the crc32 algorithm. In that case the two tests: 'itkBinaryDilateImageFilterTest-100-150-1' and the missing 'TestDumper', even if they are different, are returning the same crc32 value. So CDash thought they were the same and assign the previously stored (itkBinaryDilateImageFilter) to GDCM instead of TestDumper.
>
> It is not surprising but I was hoping that we could avoid this type of collision even with a descent size database (currently storing 12,417,508 tests on http://cdash.org).
>
> One option is to use md5 or a more complex algorithm, but then we will need to do a string matching in SQL and I'd like to avoid this if we can for speed reasons.
>
> I'm open to suggestions,
> Julien
>
> Mathieu Malaterre wrote:
>
> Another ITK test on GDCM dashboard:
>
> http://www.cdash.org/CDash/viewTest.php?onlyfailed&buildid=415610
>
> itkBinaryDilateImageFilterTest-100-150-1 is not part of GDCM
>
> HTH
>
> On Tue, Sep 1, 2009 at 2:34 PM, Mathieu
> Malaterre<mathieu.malaterre at gmail.com> wrote:
>
> Long time I did not see one:
>
> http://www.cdash.org/CDash/viewBuildError.php?buildid=415126
>
> This should be a gdcm build ...
>
> --
> Mathieu
>
>
>
>
>
>
> _______________________________________________
> Cdash mailing list
> Cdash at public.kitware.com
> http://public.kitware.com/cgi-bin/mailman/listinfo/cdash
>