dolfin team mailing list archive

[Johan Hake <johan.hake@xxxxxxxxx>]

On Tuesday September 13 2011 10:00:35 Garth N. Wells wrote:
> On 13 September 2011 17:57, Anders Logg <logg@xxxxxxxxx> wrote:
> > There are currently 4 ways to specify subdomains:
> > 
> >  /// 1. By explicitly passing _MeshFunction_ (as pointers) to the
> >  ///    assemble functions
> >  ///
> >  /// 2. By assigning subdomain indicators specified by _MeshFunction_
> >  ///    to the _Form_ being assembled:
> >  ///
> >  ///    .. code-block:: c++
> >  ///
> >  ///        form.dx = cell_domains
> >  ///        form.ds = exterior_facet_domains
> >  ///        form.dS = interior_facet_domains
> >  ///
> >  /// 3. By _MeshFunction_ stored in _MeshData_ as
> >  ///
> >  ///    * "cell_domains"
> >  ///    * "exterior_facet_domains"
> >  ///    * "interior_facet_domains"
> >  ///
> >  /// 4. By specifying a _SubDomain_ which specifies the domain
> >  /// numbered as 0 (with the rest treated as domain number 1)
> > 
> > I'm going to add a 5th (!) which seems a bit silly, so I suggest
> > removing at least option (3). A warning will be added to the MeshData
> > class for use of those previously "reserved" names.
> > 
> > Option (3) will then be replaced by storage of boundary markers in the
> > new MeshDomains class.
> > 
> > Objections?
> 
> Sounds good (especially since I don't like 3 ;)).

++

Also is 4 used in any C++ demos? (The new measure!) We could easily add it to 
the Python interface if it was removed.

Johan

> Garth
> 
> > --
> > Anders
> > 
> > _______________________________________________
> > Mailing list: https://launchpad.net/~dolfin
> > Post to     : dolfin@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~dolfin
> > More help   : https://help.launchpad.net/ListHelp