dolfin team mailing list archive

[Johan Hake <johan.hake@xxxxxxxxx>]

On Friday January 21 2011 01:16:24 Anders Logg wrote:
> B1;2600;0cOn Thu, Jan 20, 2011 at 04:13:19PM -0800, Johan Hake wrote:
> > Hello!
> > 
> > For awhile I have wanted to change the naming of cells, vertices, aso to
> > itercells, itervertices, as it would be more Pythonic.
> 
> Is 'iterfoo' common in Python? I haven't seen it before. So to me
> 
>   for cell in cells(mesh):
>       ...
> 
> still looks very neat.

Compare the values and itervalues method in a dict. I would expect cell(mesh) 
to return a list of cells or somthing. 

Maybe we can add:

  mesh.itercells, mesh.iterentity(2), aso

then we are talking pythonic. We could keep cells, vertices aso, for 
compatability.

> The C++ iterators are also don't follow the regular C++ style but I
> think
> 
>   for (CellIterator cell(mesh); !cell.end(); ++cell)
> 
> looks better than
> 
>   for (<something>::iterator cell = mesh.<something>.begin(); cell !=
> mesh.<something>.end(); ++cell)
> 
> so we're not necessarily following the standard convention always if
> we think we can do better.

sure. But it has turned out that Python is a pretty good standard. :)

> > I also wonder if we should change the naming of
> > 
> >   MeshFunction.values
> > 
> > to
> > 
> >   MeshFunction.array.
> 
> That sounds good.

Ok will do change.

Johan


 
> --
> Anders
> 
> > I think cells is just to general. What you say?
> > 
> > This will be consistent with the vector, and matrix interface.
> > 
> > Johan
> > 
> > _______________________________________________
> > Mailing list: https://launchpad.net/~dolfin
> > Post to     : dolfin@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~dolfin
> > More help   : https://help.launchpad.net/ListHelp