[Gmsh] 2D isotropic mesh algorithm : bug in node ordering ?

[Christophe Geuzaine]

Bioul Francois wrote:

> Mr Geuzaine, 
> 
> I use the 2D isotropic mesh algorithm (#1), and I have noted a bug (?)
> in the triangle node ordering : nodes describing triangles are numbered
> clockwise instead of counter-clockwise (conventionally).
> 
> In the attached file "bug.msh", the triangle #27 for example is defined
> by the nodes "5 1 22" instead of "5 22 1".
> 
> Remark : with the anisotropic algorithms, there are no problems.
> 
> Is it possible to correct this node ordering in the 2D isotropic
> algorithm?
> 

Yes, it's on our TODO list.

Meanwhile, you can simply specify the orientation using the 'Physical
Surface' command: just replace

Physical Surface (1) = {1};

with

Physical Surface (1) = {-1};

Christophe

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