[Gmsh] bug (?) in gmsh-1.65

Christophe Geuzaine cag32 at case.edu
Thu Aug 24 02:42:21 CEST 2006


Cristian Tibirna wrote:
> Hello
> 
> I wanted to upgrade our lab to gmsh-1.65 today (from 1.60) and we fell on a 
> possible bug in the UNV exporting functionality.
> 
> The attached geo/unv pairs should demonstrate the bug:
> 
> exporting a 2D mesh in UNV format includes now elements of type 21 (beam) for 
> the border of geometry (see carre.geo/carre.unv)
> 
> The same happens in 3d. When exporting the tetra mesh generated from the 
> cube.geo (from the demos directory of gmsh-1.65) in UNV format both the tetra 
> elements and triangles for the geometric faces of the grid are included in 
> the exported file.
> 
> I must acknowledge that all this doesn't have the feel of a bug, but I have 
> fallen into it because trying to import with our in-house code the UNV meshes 
> generated by gmsh-1.65 doesn't give us the same result as with the same 
> meshes exported by gmsh-1.60.
> 

We changed the behaviour a while ago to export all the elements by 
default, so this looks normal to me.


> Thanks.
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh


-- 
Christophe Geuzaine
Assistant Professor, Case Western Reserve University, Mathematics
http://www.case.edu/artsci/math/geuzaine