On Thu, Jun 30, 2011 at 08:59:26PM +0200, Marie E. Rognes wrote: > On 06/30/11 16:24, nore...@launchpad.net wrote: > >------------------------------------------------------------ > >revno: 5994 > >committer: Martin Sandve Alnæs<marti...@simula.no> > >branch nick: work > >timestamp: Thu 2011-06-30 16:09:15 +0200 > >message: > > Fix bug in assemble_system (not passing subdomains to Form), > > and make subdomains handling in Form more robust > > (only calling set_foo_domains for the foo_domains that are provided). > > Subdomains that are passed as arguments will override those > > found in the form. > > Martin: Could you provide an example of the behaviour that you > wanted to change and what you've changed it to? > > I think we should keep as much as possible regarding the logic of > what overrides what on the C++ side.
In particular in relation to this blueprint which is not yet implemented as far as I know: https://blueprints.launchpad.net/dolfin/+spec/subdomains-in-system-assembler The logic of subdomains attached to forms, subdomains attached to meshes and subdomains passed as arguments to the assembler is handled in Assembler.cpp (we have 3 different ways to specify subdomains...). I don't think this has yet been added to SystemAssembler.cpp. -- Anders _______________________________________________ Mailing list: https://launchpad.net/~dolfin Post to : dolfin@lists.launchpad.net Unsubscribe : https://launchpad.net/~dolfin More help : https://help.launchpad.net/ListHelp
