octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Dolfin mesh conversion functions [was: Re: Welcome GSoC students!]


From: Marco Vassallo
Subject: RE: Dolfin mesh conversion functions [was: Re: Welcome GSoC students!]
Date: Fri, 7 Jun 2013 19:25:48 +0200

> Subject: Dolfin mesh conversion functions [was: Re: Welcome GSoC students!]
> From: address@hidden
> Date: Wed, 5 Jun 2013 07:11:46 +0200
> To: address@hidden
> CC: address@hidden
>
>
> On 5 Jun 2013, at 06:54, c. <address@hidden> wrote:
>
> >
> > On 5 Jun 2013, at 02:17, Marco Vassallo <address@hidden> wrote:
> >
> >> Ok, I create a /devel directory and add the code
> > OK.
> >
> > As for the code congratulattions these are already quite useful functions!
> >
> > a few quick comments:
> >
> > 1: the name of the functions is completely misleading,
> > it sounds like if they were for parsing generic xml files,
> > which is not the case. Also, msh has a "namespace" convention
> > to start all its functions by "msh".
> >
> > 2: please follow Octave code formatting guidelines:
> > http://www.gnu.org/software/octave/doc/interpreter/C_002b_002b-Sources.html#C_002b_002b-Sources
> >
> > 3: you should chech for errors in input data types with the "error_state" macro:
> >
> > std::string mesh_to_read = args(0).string_value();
> > if (! error_state)
> > {
> > // do stuff
> > }
> > return retva;
> >
> > 4: it would be helpful to add a makefile, you could grab an example from some ither package to start with
> >
> >> Thanks
> >> Marco
> >
>
> wow, it happened yet one more time … I again sent the message before it was complete.
>
> 5: demo code should be prepended with "%!demo", also adding at least a test would be useful
>
> 6: the next step could be to implement "msh2m_refine" and "msh3m_refine" using dolfin's mesh refinement.
>
I have implemented  msh2m_refine and msh3m_refine but for the moment they only work with uniform refinement.(and I'm still working for being able to keep the label on the side vertices of the refined mesh)
I was thinking about other possibilities for the refinement.. maybe a boolean vector of size equal to the number of cells which says if a cell has to be refined or not? or maybe something related to the label of the border ?

I have also seen that there is a plot option under dolfin.. we could integrate it with something like msh3p_mesh ??

Marco
> c.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]