[Gmsh] Mapped meshing

Christophe Geuzaine cgeuzaine at ulg.ac.be
Wed Apr 23 10:26:33 CEST 2008


Atanas Pavlov wrote:
> Yes, I had not understand the concept of Transfinite grids, it is 
> exactly what I need. Thank you very much.
> I have a few other questions, if someone happens to have an answer to 
> them :)
> 
>   1) I have not been able to find an Offset command in Gmsh (generate 
> curve which is offset from another curve by a given distance). Any 
> suggestions?
> 
>   2) I am considering integrating Gmsh in a simple application, which 
> generates a certain geometrical configuration and meshes it. I have not 
> been able to find any example how to do that, my attempts to compile 
> anything against libGmsh.a (generated by make install-lib) have been 
> fruitless ( I cannot check the archive content with nm either, it seems 
> to be archive made of other archives and I do not know how to compile 
> against it). When I try to compile anything also against the full set of 
> libraries created in the lib build directory, I get a lot of error 
> messages about unresolved references. Is there any example anywhere, 
> which shows how to use Gmsh API to create and mesh geometries?


cf. utils/misc/driver.cpp in the source distribution




> 
> Thanks a lot,
> Atanas
> 
> 2008/4/21 Christophe Geuzaine <cgeuzaine at ulg.ac.be 
> <mailto:cgeuzaine at ulg.ac.be>>:
> 
>     Atanas Pavlov wrote:
> 
>         Hello,
> 
>         I am trying to create a mesh for an axisymmetric geometry, which
>         is relatively simple, defined by a few points, resp curves
>         connecting them. I would like, however, to perform mapped
>         meshing where I partition the curves, then use these partitions
>         on 4-side surface regions to partition the regions, and then
>         revolve-extrude the mesh around the symmetry axis. As far as I
>         can see, the first step is currently not possible in GMSH,
>         because for surface mesh it can only generate triangle mesh (or
>         extrude 1D mesh, which however is not possible in this case). Is
>         there any workaround? Is mapped meshing planned to be
>         implemented in GMSH, and for simple cases like that, can it be a
>         short term feasible task?
> 
> 
>     Could you use a Transfinite grid like this?
> 
>     Point(1) = {0,0,0,0.1};
>     Point(2) = {1,0,0,0.1};
>     Point(3) = {0,1,0,0.1};
>     Point(4) = {3,0,0,0.1};
>     Point(5) = {3,3,0,0.1};
>     Point(6) = {0,3,0,0.1};
>     Line(1) = {3,6};
>     Line(2) = {6,5};
>     Line(3) = {5,4};
>     Line(4) = {4,2};
>     Circle(5) = {2,1,3};
>     Line Loop(6) = {3,4,5,1,2};
>     Plane Surface(7) = {6};
>     Transfinite Line {1} = 20 Using Progression 1.2;
>     Transfinite Line {4} = 20 Using Progression 1./1.2;
>     Transfinite Line {3,2} = 10 Using Progression 1;
>     Transfinite Line {5} = 19 Using Progression 1;
>     Transfinite Surface {7} = {3,6,4,2} Alternated;
>     Recombine Surface {7};
> 
> 
> 
>         Cheers,
>         Atanas
> 
> 
> 
>         ------------------------------------------------------------------------
> 
>         _______________________________________________
>         gmsh mailing list
>         gmsh at geuz.org <mailto:gmsh at geuz.org>
>         http://www.geuz.org/mailman/listinfo/gmsh
> 
> 
> 
>     -- 
>     Prof. Christophe Geuzaine
>     University of Liege, Electrical Engineering and Computer Science
>     http://www.montefiore.ulg.ac.be/~geuzaine
>     <http://www.montefiore.ulg.ac.be/%7Egeuzaine>
> 
> 
> 
> 
> -- 
> Atanas Pavlov
> Luitfriedstr. 18, München 80995


-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine