[petsc-dev] ugliness due to missing lapack routines

[Jed Brown]

On Thu, Feb 7, 2013 at 11:48 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:

> This is not a problem, because I will be implementing in Haskell a system
> to manipulate Python code. Thus managing the python code that manages the C
> code will become a far easier task :-)
>

Where does m4 fit in?

And hopefully Perl. Anything that uses _both_ m4 and another language to
manipulate a third language is bound to be good.


>
>    You are perfectly happy using a really crappy system for manipulating C
> code (CPP) but fear that a better system would be impossible to get right?
>  What if I proposed just one tweak to CPP to make PETSc source code better,
> would you consider that?
>

I would consider it based on the value of that tweak, acknowledging that
changing CPP in any way presents a severe workflow contortion.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20130207/134b5485/attachment.html>