2010/6/2 Philippe Miron <[email protected]>: > I try to define the Line Loops in a different order but still getting node > number in the clockwise direction.. > I think I'll just do a postprocessing script to fix all triangles and edges > elements.
Really? I tried the following two geometry files for a simple triangular Plane Surface, with the characteristic length set large enough so that the (MEDIT) mesh ends up with a single Triangle element. Both have the Points numbered anticlockwise: 1, 2, 3. In positive.geo I join them anticlockwise, in negative.geo clockwise, into Lines, and I get the expected correct orientation in the MEDIT: positive.mesh has its triangle 2 3 1 and negative 2 1 3. I run these with, e.g., $ gmsh -2 -format mesh -string "Mesh.SaveElementTagType=2;" positive.geo
triangle.geo
Description: Binary data
positive.geo
Description: Binary data
negative.geo
Description: Binary data
positive.mesh
Description: Mesh model
negative.mesh
Description: Mesh model
_______________________________________________ gmsh mailing list [email protected] http://www.geuz.org/mailman/listinfo/gmsh
