[Gmsh] Boundary Conditions

Christophe Geuzaine geuzaine at acm.caltech.edu
Tue Jun 17 01:12:51 CEST 2003


Jonas Forssell wrote:
> Hello,
>  
> Congratulations to a very nice piece of software! A tip of the hat also
> to the guy who have written the docs. Very nice!

Thanks!

>  
> I have written a solver http://impact.sourceforge.net which I am
> concidering interfacing with gmsh.
> However, the physical option within gmsh does not seem to allow me to
> define thicknesses, number of integration points, material etc for the
> different elements. It just allows me to group them.
>  
> Also, I cannot find that boundary conditions can be defined in gmsh.
>  
> How can I make this data available in gmsh? Do I need an intermediate
> step to modify the .msh into an indata file for my solver?
>  

Yes, you need to define these things in your code, as Gmsh's philosophy
is to be as solver-agnostic as possible.

Note that you don't have to modify the .msh file: just define all the
info you need by referring to the physical numbers. And if you don't
like manipulating the numbers directly (I know I don't), just define a
name for all your physical entities in a small text file that you
include in both your .geo file and in your own code (see e.g.
http://www.geuz.org/pipermail/gmsh/2001/000493.html).

Cheers,

Christophe

-- 
Christophe A. Geuzaine
Applied and Computational Mathematics, Caltech
geuzaine at acm.caltech.edu - http://geuz.org