On Nov 29, 2010, at 4:17 PM, Matthew Knepley wrote:

> On Mon, Nov 29, 2010 at 4:09 PM, Jed Brown <jed at 59a2.org> wrote:
> On Mon, Nov 29, 2010 at 23:04, Barry Smith <bsmith at mcs.anl.gov> wrote:
> Changing the name means introducing a new concept that doesn't exist 
> elsewhere and boy do I hate having tons of concepts in PETSc.
> 
> Agree. 
> 
>    Plus few sane people would use this routine so changing its meaning won't 
> effect many end users.
> 
> Disagree.  I think it's pretty common (I know at least five people, including 
> myself, who have done this independently) to create two or more DAs that are 
> compatible (2D matches with 3D, or cell-centered and one node-centered) and 
> the only robust way to do these things is to pass lx,ly,lz along to the next 
> Create (perhaps with some modifications).
> 
> This does scream out for an API extension, DADuplicate(), with the same kind 
> of semantics as VecDuplicate().

   Doesn't work because the new DA needed is fundamentally a different type of 
DA then the original, it is not a duplicate.

   barry

> 
>    Matt
>  
> Jed
> 
> 
> 
> -- 
> What most experimenters take for granted before they begin their experiments 
> is infinitely more interesting than any results to which their experiments 
> lead.
> -- Norbert Wiener


Reply via email to