Re: changesets yet to be pushed

[Ben Abbott]

On Jan 24, 2009, at 3:22 AM, Jaroslav Hajek wrote:

> On Sat, Jan 24, 2009 at 1:48 AM, Ben Abbott <address@hidden> wrote:
> > On Jan 2, 2009, at 8:06 PM, Ben Abbott wrote:
> >
> > > I have four outstanding changesets. They are below.
> > >
> > > (1,2) changesets for tests in eigs.cc and svds.m. These make some  
> > > tests
> > > optional depending upon whether or not ARPACK is present. JWE had  
> > > already
> > > handled many such tests for eigs.cc, but a few were missed.
> > >
> > > http://www.nabble.com/Re%3A-test-failures-for-chol.cc--changesets-for-eigs.cc-and-svds.m--p21171064.html
> > > changeset-eigs.patch
> > > changeset-svds.patch
> > >
> > > (3) Changeset for savepath: respect cmd-line and env paths
> > >
> > > http://www.nabble.com/Re%3A--changeset--%7E-.octaverc-overwrites-OCTAVE_PATH%21%21-p21170416.html
> > > changeset-octave_path.patch
> > >
> > > (4) freqz produces incorrect output
> > > http://www-old.cae.wisc.edu/pipermail/bug-octave/2008-December/007493.html
> > > changeset-freqz.patch
> > >
> > > Ben
> >
> > These changesets have yet to be pushed. I've built with them today,  
> > so there
> > should be no problem applying them ... provided there is no  
> > objection to the
> > implementation or intent.
> >
> > I'm can't vouch for the freqz change, and I recall JWE had some  
> > concern, so
> > if someone with an interest in signal processing is inclined to  
> > take a look,
> > it would be appreciated.
> >
> > Ben
>
> Hi Ben,
>
> I guess everyone is OK with these changes (at least 1,2,3). Given that
> your contributions are fine and frequent enough, perhaps you could
> give Mercurial a shot and try to set up direct access to Savannah for
> yourself? I think you're in the Octave group now.
>
> regards

I do have access to Savannah.

With regards to (1/2), I have no hesitation that these can be pushed.

Regarding (3), jwe had asked me to create a static function that  
returns the value for command_line_path so as to ensure it is not  
modifiable by users. As best I can tell, I have done that. I assume  
there's been no comment because I got it right !/?

I'll push (3) as well.

 Ben