[Gmsh] STL importer/ auto boundary/ open cascade

Christophe Geuzaine geuzaine at gmail.com
Thu Jan 11 12:27:42 CET 2007


Alfonso Medina wrote:
> Chris, I just spent a long time getting to learn perl (since my last 
> posting almost). I'm wondering, If I write something in perl, how can it 
> make its way into C++?   I'm in the middle of writing an STL file 
> importer to .GEO. ITs just that when I import anything on to GMSH, it 
> only saves the points. I'm doing it so that I can set up Open flower 
> problems easily.
> 
> My idea is to have a CAD program like Pro/e or a 3-d modeling program 
> like Blender, produce an ASCII stl output file for a model. At the 
> moment, I can just import points. This week probably (probably not 
> also), I'll be having it make lines, line loops and planes.
> 
> Then, I'll have another script that does the following:  I open the 
> newly imported .GEO, then I make two virtual surfaces. One inlet one 
> outlet, it could be more, but just to say.  The script then takes the 
> rest of the un used planes and makes them virtual walls.  And thats it. 
> Probably also I'll add some kind of variable set to each import.
> 
> But that is very specific to openflower. Now I'm reading about the open 
> cascade thing. Does that mean you will interpret STEP/IGES style  with 
> no triangles in the way? That would be soo awesome.
> 

Yes, Gmsh 2.0 will be able to read STEP and IGES files and mesh them
(but not modify them). We plan to release it (at least a beta) later 
this month.




> Alfonso
> 
> 
> 
> Christophe Trophime wrote:
>> On Mon, 2006-10-09 at 19:21 -0400, Christophe Geuzaine wrote:
>>  
>>> Christophe Trophime wrote:
>>>    
>>>> Hi,
>>>> trying to create a mesh from a "template" geo file I append to notice
>>>> that I cannot create mesh file larger than 2 Gb. As I run gmsh 
>>>> (1.65) on
>>>> Linux FC5 I think that this limit is no longer due to the system 
>>>> (kernel
>>>> 2.6.xx). Is this limitation really specific of gmsh?
>>>>
>>>>       
>>> No. Do you run into this limit with a version you compiled on that 
>>> machine, or with the stock pre-compiled version from the web site?
>>>
>>>     
>>
>> I use the source version of 1.65 and "limit" do not report limiation on
>> the file size.
>>
>>  
>>>> I also wish to make gmsh able to write some other mesh file format
>>>> (especially for freefem++ and some cfd codes like Dolfyn, gerris, 
>>>> ...) .
>>>> Is there a simple way to tell 2D and 3D meshes when it comes to write
>>>> the mesh data to file (the only way I find so far is to look for
>>>> volumes).       
>>> That's reasonable: Gmsh does not distinguish between 1D, 2D or 3D 
>>> meshes.
>>>
>>>    
>>>> I also need to write the vertices nodes list as follows :
>>>>
>>>> x y bc_id
>>>>
>>>> How can I simply retrieve the boundary id (bc_id) for the vertices?
>>>>       
>>> You'll have to write a new output routine. I would strongly encourage 
>>> you to code it in the new code base if you actually do it: the whole 
>>> output layer from 1.65 has been replaced in the upcoming Gmsh 2.0.
>>>
>>>
>>>     
>>
>> I have manage to produce BAMG mesh for freefem++ and it should also work
>> for dolfyn. The only problem in this case is that you have to generate a
>> file for the vertices, one for the "cells" and one for the boundaries.
>>
>>
>>  
>>>    
>>>> Best regards
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------ 
>>>>
>>>>
>>>> _______________________________________________
>>>> gmsh mailing list
>>>> gmsh at geuz.org
>>>> http://www.geuz.org/mailman/listinfo/gmsh
>>>>       
>>>     
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> gmsh mailing list
>>> gmsh at geuz.org
>>> http://www.geuz.org/mailman/listinfo/gmsh
>>>     
> 

-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.case.edu/artsci/math/geuzaine