dolfin team mailing list archive

["Garth N. Wells" <gnw20@xxxxxxxxx>]

Ilmar Wilbers wrote:
> There have been problems with the demo pde/dg/advection_diffusion/python 
> on the mac vuildbot for some time now. Instant could not compile this 
> demo, and the reason is found in 
> /site-packages/dolfin/compile_functions.py, line 199:
> dolfin_include_dir = includes[0] # FIXME: is this safe?
>
> So this really is not safe, as pkg-config dolfin --cflags shows that the 
> first include directory is 
> /Users/ilmarw/local/petsc-2.3.3-p11/bmake/darwin9.2.0-c-debug, hence 
> instant recieves this directiry and its subfolder dolfin/swig, and will 
> not be able to locate dolfin.i
>
> Any ideas on how to pick the correct include folder?

Not really a fix, but we shouldn't need the file functions2D.h any more 
because Kristian has implemented a Function in SpecialFunctions.h to 
take care of the outflow facets.

We are though running into the issue that dolfin::OutflowFacet(...) 
takes a dolfin::Form in its constructor. I recall some issues around 
dolfin::Form versus ufc::form in the pyDOLFIN interface. Was there a 
resolution?

Garth

> ilmar
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev