[Insight-developers] Library dependency Oddness, IO for testing modules
Bradley Lowekamp
blowekamp at mail.nih.gov
Thu Feb 16 13:09:28 EST 2012
Hello,
Thanks for the explanation of how the test modules depend on the IO modes, I have been wondering about that.
Unfortunately, this issue has to do with a mis-configuration of the tiff IO library for big endian systems. I believe I have addressed it here:
http://review.source.kitware.com/#/c/4293/
Brad
On Feb 16, 2012, at 1:02 PM, Brad King wrote:
> On 2/15/2012 3:10 PM, Bradley Lowekamp wrote:
>> Could not create IO object for file .../HeadMRVolumeMaximumProjection0.tif
>> Tried to create one of the following:
> [snip]
>> TIFFImageIO
>
> If it tried TIFFImageIO then it is not a registration problem.
> The CreateTestDriver macro handles registration of all main IO types:
>
> Modules/Core/TestKernel/CreateTestDriver.cmake
> Modules/Core/TestKernel/include/itkTestDriverIncludeRequiredIOFactories.h
>
> The tests for each module are built in a pseudo-module for the module's
> "test" directory. It is to this test module that TEST_DEPENDS applies.
> They all depend on TestKernel which depends on all the IO types:
>
> $ cat Modules/Core/TestKernel/itk-module.cmake
> ...
> itk_module(ITKTestKernel
> DEPENDS
> ...
> ITKIOTIFF
>
> If it did not then the test drivers would not link.
>
> I do not think it is a problem with module dependencies or IO registration.
>
> Can you verify the input file existence and type and reproduce the failure
> when running the test by hand in the dashboard build tree?
>
> -Brad K
========================================================
Bradley Lowekamp
Medical Science and Computing for
Office of High Performance Computing and Communications
National Library of Medicine
blowekamp at mail.nih.gov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-developers/attachments/20120216/f458e5bc/attachment.htm>
More information about the Insight-developers
mailing list