[Gmsh] Bug report and a question about Gmsh

Christophe Geuzaine cgeuzaine at ulg.ac.be
Thu Feb 8 11:25:57 CET 2007


Gael de Bressy de Guast wrote:
> Hello,
>  
> First of all, I'd like to thank you for developping a tool as Gmsh. I 
> just recently discovered it and I must admit that the program is very 
> easy to use, performs well in all kind of situation and gives very 
> satisfying results when it comes to meshes.
>  
> Until now, I was using Gmsh to generate 2D meshes from a geometry I 
> defined using the built-in CAD tool. I'm now trying to import a more 
> complex 3D geometry in order to build a volumic tetrahedra mesh. This 
> means that I cannot create the geometry myself and import models from 
> other CAD tools, such as SolidWorks.
>  
> Alas, it seems that I have encountered a bug ! While opening a .STP 
> file, Gmsh 2.00 sent me back to the desktop with this error message (in 
> gmsh-errors) :
> ------------------------------------------------------------------------
> /Info    : Reading 'C:/Gaël/Mailleurs non structurés/Exemples de 
> Maillages/Cube avec Tore inclu dedans.STEP'
> Fatal   : Segmentation violation (invalid memory reference)/
> ------------------------------------------------------------------------
>  

Could you send the file?


>  
> Then, when importing an IGES file, there is no volume defined, which 
> implies that no 3D mesh can be created, only 2D elements. When I try to 
> modify the geometry in order to add a volume (elementary 
> entities>add>new>volume) by selecting a surface loop, the program returns :
> ------------------------------------------------------------------------
> /Error   : Unknown surface 8
> Error   : Unknown surface 8
> Info    : Surface Loop(1) = {8};/
> ------------------------------------------------------------------------
> 
> This implies that the renumbering of elementary entities from an 
> imported file is not stable, or not written in the program 
> memory.  Obviously, there is an algorithm that translates imported 
> features into Gmsh own language. This can be seen when passing the mouse 
> over the geometry, the dialog bar returns the name of the feature, its 
> number and the list of features that compose it.

This feature is not included yet: at the moment you can only import 
external models, not modify them.

We plan to add simple modifications (like adding a volume) in the next 
release.


>  
> Knowing this, is it possible to easily translate an .igs file into a 
> .geo file without the 'merge' command ?
>  

Not that I know of...


> I'll keep you informed if I ever come across another bug or if I have 
> another comment. Thanks again for spending your time developping this 
> wonderful tool and I hope my comments will help you improve it even further.
>  
> Sincerely, 
>  
> Gaël de BRESSY
>  
> SERGI Research Department
> 186, avenue du Gal de GAULLE
> P.O. Box 90 - 78260 ACHERES - FRANCE
> Tel: 33 - 1 39 22 37 18 / Fax: 33 - 1 39 22 48 55
> =========================================
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> 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.montefiore.ulg.ac.be/~geuzaine