[Gmsh] surface normals -- 2

[Umut Tabak]

Dear all and developers,

I have experimented in gmsh to see what kind of solution I can find to 
my problem of 'finding the surface normals' in gmsh, one gentle poster 
suggested that they can be extracted from the stl info of the 
triangulated surface information.

There is a question from my side in gmsh, for these kinds of 'normal' 
finding trick. Attached is a simple geometry extrusion example to find 
the surface normals of a side, there is a strange issue when extruding 
the geometry with some layers, the normals of the surface is not found 
right when you do a negative extrusion in the -z direction, without 
using the layers information, the normals are pointing in the wrong 
direction.

Actually by using the Layers, the normals start to point in the 
direction of the extrusion in the negative extrusion. In the positive 
extrusion, the normals are pointing in the right direction, out of the 
object considered. I added the attached files some comments, you can 
check them quickly. Does Layers doing sth different we should know 
practically?

P.S. negExt2.stl shows the extrusion without the layers in -z direction.

Best,
Umut
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: boxAttalaTria.geo
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20091214/52f2a24b/attachment.geo>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: modelAttala.geo
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20091214/52f2a24b/attachment-0001.geo>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: negExt1.stl
Type: application/vnd.ms-pki.stl
Size: 10935 bytes
Desc: not available
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20091214/52f2a24b/attachment.stl>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: negExt2.stl
Type: application/vnd.ms-pki.stl
Size: 10860 bytes
Desc: not available
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20091214/52f2a24b/attachment-0001.stl>