[SciPy-dev] Re[2]: [SciPy-user] Re: plot_func

Alan G Isaac aisaac at american.edu
Fri Aug 6 11:14:12 CDT 2004

On Thu, 05 Aug 2004, Travis Oliphant apparently wrote:
> I'm not sure who is using pyPlot (it was something of a stop-gap) and I
> would highly recommend using matplotlib at this point.  There may be a
> case for including it in the Enthon distribution but Enthought would
> have to comment on that.  I'm still using xplt heavily (on Linux mostly)
> and so I know it works there.

> Changes to SciPy are always welcome.  If you just have a few you can
> post bug reports at
> http://www.scipy.net/roundup/scipy/index   --- you might want to post on
> scipy-dev as well if it's a serious bug.

> General complaints are also welcome and will be attended to as soon as a
> developer gets time.  If you would like to become a developer, don't
> hesitate to ask about getting CVS access so you can make changes directly.

1. Code snippets
To gplt/pyPlot.py please add:
   def replot(self,cmd=''):

To gplt/interface.py please add:
def replot(cmd=''):

This will give general access to the gnuplot command line.
(I would gladly make these changes myself, if I am given
CVS access.)

2. Offer/query:
If SciPy has a commitment to maintaining gplt's pyPlot module,
and *if* there is at least one good reason not to promote
Gnuplot.py instead, I can spend a little time on pyPlot.
It sounds to me, however, that SciPy is currently promoting
a shift to matplotlib and that interest in gplt has died,
so that users interested in gnuplot should shift to Gnuplot.py.
Please let me know.

3. Possibility:
If SciPy is no longer interested in pyPlot, perhaps
it should nevertheless retain its interest in gnuplot.
(I am personally a strong gnuplot admirer.)
In which case, perhaps the gplt interface should be
to Gnuplot.py instead of pyPlot.py.  If so, I would consider
working on this.  (However, I haven't tried out Gnuplot.py

Thank you,
Alan Isaac

More information about the Scipy-dev mailing list