[Insight-users] WrapITK Problem

Gaëtan Lehmann gaetan.lehmann at jouy.inra.fr
Wed Oct 25 18:05:13 EDT 2006


Le Wed, 25 Oct 2006 16:37:53 +0200, Gheorghe Postelnicu  
<gheorghe.postelnicu at gmail.com> a écrit:

> No, I have one I downloaded about 1 week ago. I noticed that the build
> time (from scratch) was huge with WrapITK enabled - it took about 8
> hours. Is this normal?

It seem a bit long, but depending on the option you have chosen and your  
host, it can be as long as that, and even longer

>
> Also, I assume that updating from the CVS and then building would take
> less time. Correct?

Well, it depends of the changes: some of them are forcing WrapITK to  
rebuild entirely. That's the case today, with some changes in  
FlatStructuringElement

>
> Finally, I had a question yesterday about adding new classes to the
> Python wrapper. I meant one of the existing ITK classes, such as the
> LBFGSBOptimizer. It was not hard for me to manually add that line in
> the wrap_ITKOptimizers.cmake file, but is there an "ITK way" of
> preserving that modification through an update? Sorry, probably this
> question is a little silly, I know....

Send your changes on the list, I'll be pleased to review them and put them  
in ITK :-)
You can also use an external project to do that.

Gaetan

>
> On 10/25/06, Gaetan Lehmann <gaetan.lehmann at jouy.inra.fr> wrote:
>>
>> Hi,
>>
>> On Wed, 25 Oct 2006 15:54:21 +0200, Gheorghe Postelnicu
>> <gheorghe.postelnicu at gmail.com> wrote:
>>
>> > Hi,
>> >
>> > I have 2 question regarding the Python WrapITK CVS distribution:
>> >
>> > 1. when I use the syntax
>> >
>> > from itk import * in a python script, I get the message:
>> >
>> > Python 2.3.4 (#1, Mar 10 2006, 06:12:09)
>> > [GCC 3.4.5 20051201 (Red Hat 3.4.5-2)] on linux2
>> > Type "help", "copyright", "credits" or "license" for more information.
>> >>>> from itk import *
>> > Traceback (most recent call last):
>> >   File "<stdin>", line 1, in ?
>> >   File "/local/lib/InsightToolkit/WrapITK/Python/itkLazy.py", line 18,
>> > in __getattribute__
>> >     itkBase.LoadModule(module, namespace)
>> >   File "/local/lib/InsightToolkit/WrapITK/Python/itkBase.py", line 85,
>> > in LoadModule
>> >     if not swigModuleName in sys.modules: module =
>> > loader.load(swigModuleName)
>> >   File "/local/lib/InsightToolkit/WrapITK/Python/itkBase.py", line  
>> 192,
>> > in load
>> >     return imp.load_module(name, fp, pathname, description)
>> >   File  
>> "/local/lib/InsightToolkit/WrapITK/lib/IntensityFiltersPython.py",
>> > line 1, in ?
>> >     tionImageFilterINS_5ImageItLj2EEES2_E13CreateAnotherEv
>> > NameError: name
>> > 'tionImageFilterINS_5ImageItLj2EEES2_E13CreateAnotherEv' is not
>> > defined
>>
>> It works fine for me.
>> Are you using the last cvs version ?
>>
>> >
>> > 2. It seems the Python examples in the ITK directories follow some
>> > slightly different syntax, namely they use import InsightToolkit.
>> > However, the install of the WrapITK produces the itk namespace. Is
>> > this normal?
>>
>> Yes, that's normal.
>> The itk namespace has been chosen to be more consistent with c++, and
>> because it is shorter.
>> Note that with the old wrappers, you had to do:
>>
>>    from InsighToolkit import *
>>    median = itkMedianImageFilter...
>>
>> and with the new:
>>
>>    import itk
>>    median = itk.MedianImageFilter...
>>
>> This last form has a huge advantage compared to the "from itk import *"
>> you are using: it use the lazy loading, and should save lots of the
>> loading time
>>
>> Gaetan
>>
>>
>> --
>> 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
>>
>
>



-- 
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


More information about the Insight-users mailing list