[petsc-users] petsc4py: Passing command line switches

[Florian Lindner]

Am Mittwoch, 25. März 2015, 10:17:45 schrieb Matthew Knepley:
> On Wed, Mar 25, 2015 at 10:13 AM, Florian Lindner <mailinglists at xgm.de>
> wrote:
> 
> > Hello,
> >
> > I'm using the petsc4py. It works fine after some hurdling with missing
> > documentation. (were not able to find API doc anywhere, I generated it
> > myself now).
> >
> > I try to use the switches like -ksp_monitor like that:
> >
> > import petsc4py
> > from petsc4py import PETSc
> >
> > petsc4py.init(sys.argv)
> >
> > but no reaction when calling my script like that:
> >
> > python prbf.py -ksp_monitor
> >
> > What am I doing wrong? I also tried calling init before importing PETSc
> > but no changes.
> >
> 
> Did you remember to call setFromOptions() on the KSP? We do not configure
> anything from
> the command line unless you tell us to. That way, people can do things
> solely from the API
> is they want.

Double fail from my said.

I works when called like that:

import petsc4py
petsc4py.init(sys.argv)
from petsc4py import PETSc

Importing PETSc probably calls init and subsequent calls have no effect.
 
> How did you generate the documentation?

Second fail. I tried to generate it yesterday after I cloned the git and was unable to generate myself. make html does not brought much sphinx doc. Then I learned that the tarball contains an epydoc. Is that available online somewhere?

The online repo is a bit confusing, it contains an index.rst https://bitbucket.org/petsc/petsc4py/src/9dce29cab0b03f4d75fdebab087575b9635bd206/docs/index.rst?at=master
but this file seems to be competely dysfunctional. It looks like links in the document but nothing clickable (I can't even mark test), the Blame/Embed/Raw also seem dysfunctional.

Thanks,
Florian

> 
>   Thanks,
> 
>     Matt
> 
> 
> > Thanks,
> > Florian
> >
> 
> 
> 
>