[ODE] get triangle index of a trimesh

[Ettore Pancini]

I've tried dGeomTriMeshSetCallback and it works to an extent

just some notes:
* there was a bug in the trimesh-sphere collider. solved by Geoff Carlton
(http://q12.org/pipermail/ode/2005-March/015563.html). it seems is not
applied to unstable, but i can confirm it works
* ode will callback many times for the same collision. rather then telling
you
the exact triangle that collides, you'll have a bunch of potentially
colliding tris.
or at least, this is how i understand it

there was some talk in the past about adding an Index param in the
collision callback. the meaning could be depending on the geom type:
it could be the side of the box, or the triangle index for a trimesh.
i think is not a very general solution, but i could be very useful

cheers


----- Original Message ----- 
From: "Jon Watte (ODE)" <hplus-ode at mindcontrol.org>
To: <adastra at inorbit.com>
Cc: "ODE mailing_list" <ode at q12.org>
Sent: Thursday, March 09, 2006 18:30
Subject: Re: [ODE] get triangle index of a trimesh


>
> In my copy of the manual, on page 66, section 10.7.6, it talks about the
> callbacks that will tell you which triangles are hit during collision;
> it's an old version, so it may have moved around. Search for
> dTriCallback and dTriRayCallback to locate it.
>
> If you have read this part, and experimented with the code, and it still
> won't let you do what you want, please post a description of what you
> have tried, and why it didn't work for you.
>
> Cheers,
>
> / h+
>
>
> adastra at inorbit.com wrote:
> > I had the same problem with my indices. I just used my visualization
> > software (VTK) to work it out.
> > I used GetCell() and then GetPoints() for each cell and made an array
> > with that data.
> >
> > Is there a way to do this using ODE without wasting processing time on
> > finding the same index so many times?
> >
> > Jemma
>
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode