What you're trying to do just can't work. The look_at method don't just 
modify the Z axis of the coordsyst but the whole coordsyst orientation. 
So does the look_at_x method. I don't know how you want to set your 
solid orientation but calling a look_at method after an other look_at 
method is not the proper way.

Perhaps you want to set the absolute orientation of your solid and not 
only relative rotation from its current orientation, is that it ? You 
can do that by first using the set_identity method that reset your 
coordsyst position and rotation and after that rotate or turn methods.

Souvarine.



Pablo Angulo wrote:
>   Hello, everybody. I'm a new soya user, and I was surprised because it 
> was so easy to develop a simple 3D app. I had a kind of respect for the 
> field, but the soya tutorials made the learning curve almost flat. Up to 
> this moment, whenever I had a problem, I found that the authors had 
> already though about it and offered a good solution. The embedded 
> CoordSyst really make most tasks simple. And there is probably one 
> 'soya' solution to the problem I have now, but I couldn't find it.
>
>   I understand soya does not use matrices, and there is one thing I 
> don't know how to do without them. I need to set the full orientation of 
> an object in 3D space. I can set the position, and I can use the 
> 'look_at' method to set the z-vector of the CoordSyst to the one I need, 
> but I also need to set the x-vector, and I found that if I set the 
> z-vector first with the look_at method and then the x-vector with the 
> look_at_x method, the second call sends the z-vector to an arbitrary 
> position.
>
>  In [2]: soya.CoordSyst.look_at_x?
> Type:           method_descriptor
> Base Class:     <type 'method_descriptor'>
> String Form:    <method 'look_at_x' of '_soya.CoordSyst' objects>
> Namespace:      Interactive
> Docstring:
>     CoordSyst.look_at_x(target)
>    
>     Rotate so that this object's positive-X points at target.
>
>
>   So what is the soya way to set all the position, the z-vector and the 
> x-vector (provided, of course, that the two vectors are orthogonal)?
>
>   Thank you very much
> Pablo
>
> _______________________________________________
> Soya-user mailing list
> Soya-user@gna.org
> https://mail.gna.org/listinfo/soya-user
>
>   

_______________________________________________
Soya-user mailing list
Soya-user@gna.org
https://mail.gna.org/listinfo/soya-user

Reply via email to