CGAL users discussion list

[wlee <>]

Hi Peter.

I understand what you mean. Maybe general transformation make
some vertices, which are on the same plane before the transformation,
not on the same plane after the transformation.

If I have some of line equations and calculate the intersections of the
line
with a plane using CGAL, are those intersection points are on the same
plane without rounding problem ?

Wonwoo.





Peter Hachenberger wrote:
> Hi,
>
> transformin Nef polyhedra can be quite complicated. That's why I tried
> to only allow a subset of all possible transformations. We did not test
> the code for general transformations. There are two things you should
> change. First, you should do two separate transformations: one
> translation and one rotation. Second, if it still does not work, you
> should use a rational rotation matrix for the rotation. The rotation
> that you perform will probably deform the polyhedron because of rounding
> issues. The code is only tested with rotations that do not deform the
> polyhedron. You can look up how to do it in demo/Nef_3/nef_3.C. There we
> handle rotations around all three coordinate axis. Search for the key
> word "rotx".
>
> Peter
>
> On Sat, 2007-02-17 at 00:24 +0900, Wonwoo Lee wrote:
>   
>> Hi. 
>>
>> I created a pyramid with Polyhedron_3 and converted to Nef_3
>> polyhedron. 
>>
>> Than, I appled a transformation, which consists of rotation and
>> translation by using Nef_polyhedron_3::Aff_transformation_3. 
>>
>> But after transformation. the Nef_3 polyhedron is not displayed in
>> OpenGL space. 
>>
>> If I apply only translation, it does not disappear. I think the
>> problem is in applying rotation. 
>>
>> What happens to my Nef_3 polyhedron ? 
>>
>> My code is something like below : 
>>
>> Nef_polyhedron_3 N ; 
>>
>> Nef_polyhedron_3::Aff_transformation_3 transf(1,0,0, 200, 
>> 0,1/sqrt(2),-1/sqrt(2), 200, 
>> 0,1/sqrt(2), 1/sqrt(2), 200) ; 
>>
>> N.transform(transf) ; 
>>
>>     
>
>
>
>
>