[petsc-dev] preparing for PETSc release 3.0.1

[Matthew Knepley]

On Sat, Feb 27, 2010 at 5:41 PM, Satish Balay <balay at mcs.anl.gov> wrote:

> First issues:
>
> * help should work with or without PETSC_ARCH/PETSC_DIR set.
>

We would need

  1) a more sophisticated way to find BuildSystem

or

  2) to switch to optparse, which is missing features you like

PETSC_ARCH missing is definitely a bug. I will fix.

* Perhaps PETSC_DIR/PETSC_ARCH should be specifiable with command line
> option..
>

Yes, I agree

  Matt


> asterix:/home/balay/tmp/petsc-dev>./config/builder.py -h
> Traceback (most recent call last):
>  File "./config/builder.py", line 4, in <module>
>    sys.path.insert(0, os.path.join(os.environ['PETSC_DIR'], 'config'))
>  File "/usr/lib64/python2.6/UserDict.py", line 22, in __getitem__
>    raise KeyError(key)
> KeyError: 'PETSC_DIR'
> asterix:/home/balay/tmp/petsc-dev>export PETSC_DIR=$PWD
> asterix:/home/balay/tmp/petsc-dev>./config/builder.py -h
> Traceback (most recent call last):
>  File "./config/builder.py", line 269, in <module>
>    PETScMaker().buildAll()
>  File "./config/builder.py", line 15, in __init__
>    argDB.saveFilename = os.path.join(os.environ['PETSC_DIR'],
> os.environ['PETSC_ARCH'], 'conf', 'RDict.db')
>  File "/usr/lib64/python2.6/UserDict.py", line 22, in __getitem__
>    raise KeyError(key)
> KeyError: 'PETSC_ARCH'
> asterix:/home/balay/tmp/petsc-dev>export PETSC_ARCH=asterix64
> asterix:/home/balay/tmp/petsc-dev>./config/builder.py -h
> Script Help
> -----------
> Script:
>  --help    : Print this help message             current: 0
>  --h       : Print this help message             current: 1
> RepManager:
>  --rootDir : The root directory for this build   current:
> /home/balay/tmp/petsc-dev
>  --dryRun  : Only output what would be run       current: 1
>  --verbose : The verbosity level                 current: 1
> asterix:/home/balay/tmp/petsc-dev>
>
>
>
>
> On Sat, 27 Feb 2010, Matthew Knepley wrote:
>
> > 1) builder.py --help :)
> >
> > 2) You give it a root directory to start (default is PETSC_DIR) and it
> does
> > recursive make
> >
> > 3) The current target is libpetscfake.a so I never accidentally blow away
> my
> > install
> >
> >    Matt
> >
> > On Sat, Feb 27, 2010 at 4:44 AM, Satish Balay <balay at mcs.anl.gov> wrote:
> >
> > > Matt,
> > >
> > > Any instructions on how to use/test builder.py?
> > >
> > > Satish
> > >
> > > On Fri, 26 Feb 2010, Matthew Knepley wrote:
> > >
> > > > What do people think about builder.py for the release? I really
> really
> > > want
> > > > a version
> > > > I can use in China.
> > > >
> > > >    Matt
> > > >
> > > > On Thu, Feb 25, 2010 at 2:47 PM, Barry Smith <bsmith at mcs.anl.gov>
> wrote:
> > > >
> > > > >
> > > > >  I'd like to get the next PETSc release out very soon (so we can
> start
> > > > > ripping petsc-dev apart and making it way cooler :-). Please send
> in
> > > any
> > > > > outstanding bug reports and test your applications to make sure the
> > > code is
> > > > > in good shape.
> > > > >
> > > > >   Thanks
> > > > >
> > > > >    Barry
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > >
> > >
> >
> >
> >
>
>


-- 
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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20100228/1eaf0f1f/attachment.html>