Hello you all,
I hope I am not speaking about a subject that has already been discussed.
I am using the "*.mesh" format for my meshes because I find this format really easy to read. I have noticed that in this format the boundary edges are not tagged by the number of the physical entity they belong to, but by the number of the line they belong to.

Here is a small example :
I have generated a very awful mesh with the Carre.geo file attached by command
"gmsh-2.4.2 -2 -o Carre.mesh -format mesh Carre.geo"
and obtained the mesh described in Carre.mesh.
I got 12 edges :

Edges
12
1 2 1
2 3 2
3 4 3
4 1 4
5 9 6
9 6 6
6 10 7
10 7 7
7 11 8
11 8 8
8 12 9
12 5 9

that are tagged by the line they belong to and not by the physical tag I gave to those lines :

Physical Line(1) = {1,2,3,4};
Physical Line(2) = {6,7,8,9};

The same for the triangles that are tagged by the number of the surface they belong to (1) and not by the physical tag of this surface (0).

Is there any reason for that ? Maybe there exists a option that allows me to take those physical tags into account ?

Thanks,

Adam

// Variables
size_carre = 10.0;
char_carre = size_carre;
size_hole = 1.0;
char_hole = size_hole;

// Points
Point(1) = { size_carre,-size_carre,0,char_carre};
Point(2) = { size_carre, size_carre,0,char_carre};
Point(3) = {-size_carre, size_carre,0,char_carre};
Point(4) = {-size_carre,-size_carre,0,char_carre};

Point(5) = { size_hole,        0.0,0,char_hole};
Point(6) = {        0.0, size_hole,0,char_hole};
Point(7) = {-size_hole,        0.0,0,char_hole};
Point(8) = {        0.0,-size_hole,0,char_hole};

Point(9) = {        0.0,       0.0,0,char_hole};

// Lines and Line Loops
Line(1) = {1,2};
Line(2) = {2,3};
Line(3) = {3,4};
Line(4) = {4,1};
Line Loop(5) = {1,2,3,4};

Circle(6) = {5,9,6};
Circle(7) = {6,9,7};
Circle(8) = {7,9,8};
Circle(9) = {8,9,5};
Line Loop(10) = {6,7,8,9};

Physical Line(1) = {1,2,3,4};
Physical Line(2) = {6,7,8,9};

// Surface
Plane Surface(1) = {5,10};
Physical Surface(0) = {1};

Attachment: Carre.mesh
Description: Mesh model

_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh

Reply via email to