Dear Yao Koutsawa,

It can be extended of course to geometric transformations. I added the
function also (in bgeot_geometric_trans.h).

Of course, when the fem/integration method/geometric transformation  is
of general interest, it is more interesting to add it in the
corresponding files of getfem and submit a patch. In particular, the
"QT_Q2_INCOMPLETE(P)" geometric transformation may be interesting for
other users of Getfem.

Yves.



Le 14/11/2012 09:58, Epiphane Y. K. a écrit :
> Dear Yves,
>
> I am wondering if this feature can be extended to the geometric
> transformations. Actually I am looking for the way to use
> isoparametric Quad 8 and Hexa 20 serendipity elements. The FEMs
> "FEM_Q2_INCOMPLETE(P)"  are well defined but I would like to define a
> convex with a geometric transformation like "QT_Q2_INCOMPLETE(P)" . 
> In short words how can I define for example Quad 8 or  Hexa 20 convex?
>
> Best regards,
> Yao Koutsawa
>
>
>     ------------------------------------------------------------------------
>     *De :* Yves Renard <[email protected]>
>     *À :* Torquil Macdonald Sørensen <[email protected]>
>     *Cc :* [email protected]
>     *Envoyé le :* Mercredi 14 novembre 2012 8h55
>     *Objet :* Re: [Getfem-users] Add integration method
>
>
>     Dear Torquil,
>
>     You are right, a function equivalent to "add_fem_name" do not
>     exist for
>     integration methods. Of course, since the two systems for fems and
>     integration method derive from the same mechanism (dal_naming_system),
>     the same function can be easily added to getfem_integration.cc. It
>     should be as follow:
>
>     void add_integration_name(std::string name,
>                 dal::naming_system<integration_method>::pfunction f) {
>         dal::singleton<im_naming_system>::instance().add_suffix(name, f);
>       }
>
>
>     I add it to the svn version of getfem.
>
>
>     Yves.
>
>
>     Le 13/11/2012 21:30, Torquil Macdonald Sørensen a écrit :
>     > Hi!
>     >
>     > I have created a custom mesh_fem, and I made it available in my
>     > program using the code
>     >
>     > getfem::add_fem_name("CUSTOM", custom_fem);
>     >
>     > where custom_fem is a function that returns a getfem::pfem,
>     similarly
>     > to how it is done for the built-in mesh_fems in the Getfem
>     source code.
>     >
>     > I select this mesh_fem using the syntax
>     >
>     > mf.set_finite_element(getfem::fem_descriptor("FEM_CUSTOM"));
>     >
>     > However, now I would like to create my own integration method
>     > IM_CUSTOM to go with FEM_CUSTOM. My goal is to define an integration
>     > method that is exact on FEM_CUSTOM.
>     >
>     > Is there a nice way to do this without recompiling the Getfem source
>     > code? I didn't find any functions analogous to "add_fem_name",
>     but for
>     > integration methods rather than mesh_fems.
>     >
>     > Best regards
>     > Torquil Sørensen
>     >
>     >
>     > _______________________________________________
>     > Getfem-users mailing list
>     > [email protected] <mailto:[email protected]>
>     > https://mail.gna.org/listinfo/getfem-users
>
>
>     -- 
>
>       Yves Renard ([email protected]
>     <mailto:[email protected]>)      tel : (33) 04.72.43.87.08
>       Pole de Mathematiques, INSA-Lyon            fax : (33)
>     04.72.43.85.29
>       20, rue Albert Einstein
>       69621 Villeurbanne Cedex, FRANCE
>       http://math.univ-lyon1.fr/~renard
>     <http://math.univ-lyon1.fr/%7Erenard>
>
>     ---------
>
>
>     _______________________________________________
>     Getfem-users mailing list
>     [email protected] <mailto:[email protected]>
>     https://mail.gna.org/listinfo/getfem-users
>
>


-- 

  Yves Renard ([email protected])       tel : (33) 04.72.43.87.08
  Pole de Mathematiques, INSA-Lyon             fax : (33) 04.72.43.85.29
  20, rue Albert Einstein
  69621 Villeurbanne Cedex, FRANCE
  http://math.univ-lyon1.fr/~renard

---------

_______________________________________________
Getfem-users mailing list
[email protected]
https://mail.gna.org/listinfo/getfem-users

Reply via email to