[Insight-users] ITK current CVS build error (CableSwig/Python wrapping)

[Charl P. Botha]

Hello Hans,

On Wed, 2003-11-26 at 03:04, Hans J. Johnson wrote:
> Thanks for the update.  You are correct, and I've committed those
> changes to the CVS source tree.

Great, I see that you've removed them from the CMakeLists.txt.  Wouldn't
it be prudent to remove them from the "MAKE_ONLY_FLOAT_TYPES" variable
in BasicFilters/MakeConsistentWrappedClasses.sh as well, just in case
somebody decides to run that script again or is mislead by the classes
that it lists?

Thanks,
Charl

> 
> On Tue, 2003-11-25 at 16:40, Charl P. Botha wrote:
> > Brad King wrote:
> > > I think the std::fill one is due to a missing #include <algorithm> in that
> > > source file.  Please try adding it.  However, I also noticed that I was
> > > missing std::endl and std::ends from the header wrappers, so please update
> > > GCC_XML again too.
> > 
> > Thanks again!  Your changes and the #include <algorithm> line fixes my 
> > build on GCC 2.95.
> > 
> > One more thing (and this is more meant for Hans I think): it seems 
> > itkCurvatureFlowImageFilter and itkMinMaxCurvatureFlowImageFilter are 
> > being wrapped in BasicFilters by the MakeConsistentWrappedClasses.sh 
> > script and ALSO manually in Algorithms.  The Algorithms wrapping (only 
> > F2 and F3) is overwriting the BasicFilters one causing havoc during an 
> > "import InsightToolkit" (fatal).
> > 
> > Can I remove these two from the MakeConsistentWrappedClasses.sh script 
> > (as they actually seem to belong under Algorithms) or is there some 
> > greater plan?
> > 
> > Thanks again,
> > Charl
-- 
charl p. botha http://cpbotha . net/ http://visualisation . tudelft . nl/