[petsc-dev] Could generatefortranstubs.py be lazier about modification times?

Satish Balay balay at mcs.anl.gov
Mon Oct 4 16:26:29 CDT 2010


On Mon, 4 Oct 2010, Barry Smith wrote:

> 
> On Oct 4, 2010, at 4:19 PM, Jed Brown wrote:
> 
> > On Mon, Oct 4, 2010 at 23:15, Barry Smith <bsmith at mcs.anl.gov> wrote:
> >> Yuck, not really. First it runs bfort on everything in a directory then FixFile() on each generated file.
> > 
> > That's what I thought.  Is it always safe to look at ../fname.c?
> 
>   Beats me. You are on your own :-(


Hmm - if there is a file 'xxx.c' - then bfort generates ftn-auto/xxxf.c

so if looping over generated files - one can use this reverse maping
to identify the correct source mapping

Satish



More information about the petsc-dev mailing list