[Soc-2009-dev] Reports + Midterms + Blendernation

[Ken Hughes]

Be aware that, in trunk at least, the scanfill code has some problems.  
I found cases where collinear vertices resulted in incorrect fills.

Ken

Sebastian wrote:
> I must admit that i was too lazy to search the existing code base ;) 
> scanfill is the better solution.
>
> Sebastian
>
>
>
> joe wrote:
>   
>> Eh please don't use GLU.  We have our own tesselation library, no need
>> for a glu dependancy.
>>
>> Jow
>>
>> On Sun, Jun 28, 2009 at 1:47 PM, Sebastian<sebastian at opencollada.org> wrote:
>>     
>>> Hi Arystan,
>>>
>>> those samples rely on a GLU implementation. You might find the sources
>>> of those within the reference implementation by SGI or the MESA3D libraries.
>>>
>>> The code has not been changed during the last nine years and i have no
>>> experience with it.
>>>
>>> http://oss.sgi.com/cgi-bin/cvsweb.cgi/projects/ogl-sample/main/gfx/lib/glu/libtess/
>>> http://cgit.freedesktop.org/mesa/mesa/tree/src/glu/sgi/libtess
>>> http://cgit.freedesktop.org/mesa/mesa/tree/src/glu/mesa
>>>
>>> It is probably sufficient for the seldom case where COLLADA contains
>>> complex polygons. You probably get triangulated data most of the time...
>>> Use it ;)
>>>
>>> Sebastian
>>>
>>>
>>> Arystan Dyussenov wrote:
>>>       
>>>> Hi, Sebastian,
>>>>
>>>> What do you think about this approach:
>>>> http://www.songho.ca/opengl/gl_tessellation.html?
>>>>
>>>> It seems optimal to me. Since some parts of Blender are already using
>>>> GLU, it doesn't create additional dependencies.
>>>>
>>>> Arystan
>>>>
>>>> On Fri, Jun 26, 2009 at 10:46 PM, Sebastian <sebastian at opencollada.org
>>>> <mailto:sebastian at opencollada.org>> wrote:
>>>>
>>>>     Chingiz,
>>>>
>>>>      > Does anyone have an idea how to import polygons with more than 4
>>>>      > vertices? Currently the importer simply ignores such polygons.
>>>>
>>>>     That's a non-trivial task especially when your polygon includes holes,
>>>>     too ;)
>>>>
>>>>     http://www.vterrain.org/Implementation/Libs/triangulate.html lists some
>>>>     of the available algorithms. Internally we are using FIST, an
>>>>     alternative would be to figure out the corresponding parts from
>>>>     OpenCascade.
>>>>
>>>>     Sebastian
>>>>     _______________________________________________
>>>>     Soc-2009-dev mailing list
>>>>     Soc-2009-dev at blender.org <mailto:Soc-2009-dev at blender.org>
>>>>     http://lists.blender.org/mailman/listinfo/soc-2009-dev
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------
>>>>
>>>> _______________________________________________
>>>> Soc-2009-dev mailing list
>>>> Soc-2009-dev at blender.org
>>>> http://lists.blender.org/mailman/listinfo/soc-2009-dev
>>>>         
>>> _______________________________________________
>>> Soc-2009-dev mailing list
>>> Soc-2009-dev at blender.org
>>> http://lists.blender.org/mailman/listinfo/soc-2009-dev
>>>
>>>       
>> _______________________________________________
>> Soc-2009-dev mailing list
>> Soc-2009-dev at blender.org
>> http://lists.blender.org/mailman/listinfo/soc-2009-dev
>>     
> _______________________________________________
> Soc-2009-dev mailing list
> Soc-2009-dev at blender.org
> http://lists.blender.org/mailman/listinfo/soc-2009-dev
>