yade-dev team mailing list archive

[Vincent Richefeu <richefeu@xxxxxxxxx>]

Hi,

There is some mistakes in 'basicVTKWitter' (I will rename this class!). I will correct this on Wednesday when I'll be back from Sophia Antipolis. 

As you can see in the attached file, there is very flat tetrahedra at the boundaries of the triangulation. Is there a way to avoid that whit CGAL?

Vincent

Le 16 janv. 2010 à 16:28, Bruno Chareyre a écrit :

> Hello
> 
> Vaclav, I wrote a function returning python::dict as discussed before. Check TesselationWrapper.cpp::394.
> Problem : including <yade/lib-pyutil/numpy.hpp> somewhere is breaking compilation (namespace conflict in numpy-boost, see below).
> Do you have any idea how to fix that?
> 
> Vincent, could you try loading the attached file and see if you get a correct display? I thought everything was ok last time we checked, but it was not (non access to the scalar value, and the display of matrix components is bugged somehow, with strange scaling).
> 
> Thanks.
> 
> Bruno
> 
> 
> C /home/bruno/YADE/build-last-opt/plugins0.cpp
> In file included from /home/bruno/YADE/build-last-opt/include/yade-last/yade/lib-pyutil/numpy.hpp:3,
> from /home/bruno/YADE/yade-bzr/pkg/dem/Engine/GlobalEngine/TesselationWrapper.cpp:10,
> from /home/bruno/YADE/build-last-opt/plugins0.cpp:12:
> /home/bruno/YADE/build-last-opt/include/yade-last/yade/lib-pyutil/numpy_boost.hpp: In constructor 'numpy_boost<T, NDims>::numpy_boost(PyObject*)':
> /home/bruno/YADE/build-last-opt/include/yade-last/yade/lib-pyutil/numpy_boost.hpp:146: error: reference to 'detail' is ambiguous
> /home/bruno/YADE/build-last-opt/include/yade-last/yade/lib-pyutil/numpy_boost.hpp:50: error: candidates are: namespace detail { }
> /usr/include/boost/type_traits/detail/cv_traits_impl.hpp:24: error: namespace boost::detail { }
> 
> 
> 
> Václav Šmilauer a écrit :
>>>  I see. I will generate vectors on request then, as for now the only array is for strains. Volumes and porosities are stored in the vertices directly.
>>> Would numpy apply if vertices are in a vector and the data is accessed using "vertex->info().something" (imposed by CGAL design)?
>>>    
>> I think you have to allocate new array anyways (it would probably be
>> possible without, using some magic, but I don't know how). Some of them
>> will get just copied values, other will be filled by
>> vertex->info().something. Does that make sense?
>> 
>> Vaclav
>> 
>> 
>> 
>> _______________________________________________
>> Mailing list: https://launchpad.net/~yade-dev
>> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
>> Unsubscribe : https://launchpad.net/~yade-dev
>> More help   : https://help.launchpad.net/ListHelp
>> 
>>  
> 
> 
> -- 
> _______________
> Bruno Chareyre
> Associate Professor
> Grenoble INP
> Lab. 3SR
> BP 53 - 38041, Grenoble cedex 9 - France
> Tél : 33 4 56 52 86 21
> Fax : 33 4 76 82 70 43
> ________________
> 
> <def.vtk.tar.gz>_______________________________________________
> Mailing list: https://launchpad.net/~yade-dev
> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~yade-dev
> More help   : https://help.launchpad.net/ListHelp