[Insight-developers] redefinition of HAVE_SYS_TYPES_H and HAVE_STDINT_H

[Bradley Lowekamp]

*sigh* This cmake configuration has been surprisingly tricky and error prone.

I have since patched itkConfigure.h.in with what should be a working version now.


1)

That is a very good find! Thank you! I would not be certain moving the CONFIGURE_FILE would be free from side effects, better I just move the fenv.h check.

2)

My concern still stands with the #ifundef guarding. If another config header has a define for HAVE_INCLUDE_H but is include after our header, we will have redefinitions and generate a compiler warning. I think adding the ITK_ prefix is safest.

http://www.itk.org/cgi-bin/viewcvs.cgi/itkConfigure.h.in?root=Insight&r1=1.36&r2=1.37&sortby=date



Brad

On Nov 23, 2009, at 1:25 PM, Tom Vercauteren wrote:

> Hi Brad,
> 
> I just tried to use some of your recent changes (to move on with the
> unicode filename patch) and found two issues.
> 
> 1) The availability of fenv.h is the last thing that is done in the
> main CMakeLists.h
> http://www.itk.org/cgi-bin/viewcvs.cgi/CMakeLists.txt?root=Insight&r1=1.355&r2=1.356&sortby=date
> However HAVE_FENV_H is used by itkConfigure.h.in
> http://www.itk.org/cgi-bin/viewcvs.cgi/itkConfigure.h.in?root=Insight&r1=1.35&r2=1.36&sortby=date
> 
> This means that when starting from a clean build tree
>  CONFIGURE_FILE(${ITK_SOURCE_DIR}/itkConfigure.h.in
>    ${ITK_BINARY_DIR}/itkConfigure.h)
> will be called before checking for HAVE_FENV_H and ITK_HAVE_FENV_H will be false
> 
> The fix seems easy:
>  CONFIGURE_FILE(${ITK_SOURCE_DIR}/itkConfigure.h.in
>    ${ITK_BINARY_DIR}/itkConfigure.h)
> should be the last thing done in the main CMakeLists.txt, right?
> 
> 
> 
> 2) It looks like the following patch is wrong
> http://www.itk.org/cgi-bin/viewcvs.cgi/itkConfigure.h.in?root=Insight&r1=1.35&r2=1.36&sortby=date
> 
> I don't get how this kind of code may work:
>  #if @HAVE_FENV_H@
>  #define ITK_HAVE_FENV_H
>  #endif /* @HAVE_FENV_H@ */
> 
> What I just tried (with a cvs version of cmake) is to add this right
> before the previous call to CONFIGURE_FILE
>  CHECK_INCLUDE_FILES("unistd.h" HAVE_UNISTD_H)
>  CHECK_INCLUDE_FILES("windows.h" HAVE_WINDOWS_H)
> and
>  #if @HAVE_UNISTD_H@
>  #define ITK_HAVE_UNISTD_H
>  #endif /* @HAVE_UNISTD_H@ */
> 
>  #ifdef @HAVE_WINDOWS_H@
>  #define ITK_HAVE_WINDOWS_H
>  #endif /* @HAVE_WINDOWS_H@ */
> 
> 
> Looking at the generated itkConfigure.h, I get (on linux)
>  #if 1
>  #define ITK_HAVE_UNISTD_H
>  #endif /* 1 */
> 
>  #if
>  #define ITK_HAVE_WINDOWS_H
>  #endif /*  */
> 
> This leads to a gcc error:
>  error: #if with no expression
> 
> 
> Reverting to cmakedefine works fine in this case.
> 
> I am not sure how to ideally achieve your goal but something like
>  #ifndef HAVE_UNISTD_H
>  #cmakedefine HAVE_UNISTD_H
>  #endif
>  #ifdef HAVE_UNISTD_H
>  #define ITK_HAVE_UNISTD_H
>  #endif
> seems to do the trick (even thought it is quite ugly).
> 
> 
> Regards,
> Tom
> 
> 
> On Mon, Nov 23, 2009 at 15:41, Bradley Lowekamp <blowekamp at mail.nih.gov> wrote:
>> The complication is that the CHECK_TYPE_SIZE macro implicitly searches for
>> some headers. So for this particular header there is no CHECK_INCLUDE_FILES
>> directly called by Insight/CMake.
>> I am trying my approach now, and hopefully will have a commit to remove
>> those pesky warning shortly.
>> Brad
>> On Nov 23, 2009, at 3:43 AM, Gaëtan Lehmann wrote:
>> 
>> Le 20 nov. 09 à 14:01, Bradley Lowekamp a écrit :
>> 
>> But if  itkConfigure.h is included before phython then it wont work, because
>> phython is not ifndef. I think what may be needed is similar to the
>> following:
>> 
>> #if @HAVE_STDINT_H@
>> 
>> #define ITK_HAVE_STDINT_H
>> 
>> #endif
>> 
>> I'd need to look at what the value of @HAVE_STDINT_H@ would be. By doing
>> this we can use the standard naming inside CMAKE, and convert it to our own
>> in the ITK code.
>> 
>> 
>> Why not using directly the ITK_ prefix in cmake?
>> The chance to have a collision in that case seems quite low.
>> 
>> Am I missing something?
>> 
>> Gaëtan
>> 
>> --
>> Gaëtan Lehmann
>> Biologie du Développement et de la Reproduction
>> INRA de Jouy-en-Josas (France)
>> tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
>> http://voxel.jouy.inra.fr  http://www.itk.org
>> http://www.mandriva.org  http://www.bepo.fr
>> 
>> 
>> ========================================================
>> 
>> Bradley Lowekamp
>> 
>> Lockheed Martin Contractor for
>> 
>> Office of High Performance Computing and Communications
>> 
>> National Library of Medicine
>> 
>> blowekamp at mail.nih.gov
>> 
>> 
>> _______________________________________________
>> Powered by www.kitware.com
>> 
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>> 
>> Kitware offers ITK Training Courses, for more information visit:
>> http://kitware.com/products/protraining.html
>> 
>> Please keep messages on-topic and check the ITK FAQ at:
>> http://www.itk.org/Wiki/ITK_FAQ
>> 
>> Follow this link to subscribe/unsubscribe:
>> http://www.itk.org/mailman/listinfo/insight-developers
>> 
>> 

========================================================
Bradley Lowekamp  
Lockheed Martin Contractor 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/mailman/private/insight-developers/attachments/20091123/eb68763a/attachment.htm>