[Gmsh] Fwd: Flat Quad mesh with Delaunay for quads

[Michel Cassagnes]

Dear all,

The routine Delaunay for quads produces flat quad mesh (two sides aligned).
It would be great if a test was added to prevent these flat mesh.
Thank you
Best regards,

Michel Cassagnes
---------- Forwarded message ----------
From: Michel Cassagnes <michel.cassagnes at noos.fr>
Date: 2017-11-26 21:57 GMT+01:00
Subject: Flat Quad mesh with Delaunay for quads
To: gmsh at onelab.info


Dear all,
I get a bad quad mesh when I use Delaunay for quads (still labeled as
experimental) with Recombine all triangulatr meshes and Standard algorithm.
Several quads have two aligned edges : see attached images, one image with
triangular meshes, before recombine and another image where we can see the
flat quad mesh.
The recombine algorithm already leaves some triangle in the mesh because
all cannot be quad or with bad shape. So gmsh works very very well and mesh
is very nice (see attached, global view). It just remains the problem that
it accepts flat meshes and so doesn't allow to use it for calculations.

Thank you.and best regards,

Michel Cassagnes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://onelab.info/pipermail/gmsh/attachments/20171130/31007316/attachment.html>