[Gmsh] gmsh syntax

Christophe Geuzaine geuzaine at acm.caltech.edu
Wed Mar 12 18:15:11 CET 2003


walter steffe wrote:
> Thanks for your help. Your reference were useful but I have a few more 
> questions.
> 
> To better clarify my problem let us consider a solid made up of some boxes 
> that are connected at their faces. The intersection of the adjacient surfaces 
> of two connected boxes may be distinct from both faces because it is not 
> garanted that a face is included in the other.  So I have to compute the line 
> loop associated with this intersection and add it (as a hole) to the 
> definition of both surfaces. I know that the hole must be internal to the 
> face so I am thinking to assure this condition in any case by applying a 
> small perturbation on the exact hole (the effect on electromagnetic result 
> should be negligible).
> 
> I am planning to write the following functions:
> 
> -A function "box" which defines the geometrical and topological entities of 
> the isolated box (placed in its final position) and maps the local numeration 
> (defined inside the box) of these entities with the global numeration (using 
> newp and newreg). 
> 
> -A function "unite" which computes the hole (line loop) and adds it to the 
> interested faces.
> 
> The last function need access to the following data:
>  - the list of lines in a given line loop
>  - the extreme points of a given line.
>  - the coordinates of a given point.  
> Is it there a way to get this information using the gmsh language ? 

You would have to store the info you need when you create these entities 
(the "geo" language doesn't have any functions to retrieve data from the 
database once the entities are created).

> 
> In the computation of the faces intersection (and in many other cases) I would  
> also appreciate the possibility of calling an external code so that I could 
> reuse some existing code instead of rewriting it in the gmsh language. Is it 
> possible to do that ?
> 

There is no infrastructure for this at the moment. You'll have to add 
some C++ code directly in the sources.

Christophe

-- 
Christophe Geuzaine

Tel: (626) 395-4552    http://www.geuz.org
Fax: (626) 578-0124    mailto:geuzaine at acm.caltech.edu