gnuplot-py-users Mailing List for Gnuplot.py

Well, it took a lot of digging to figure this one out. I won't bore you with
the details; the bottom line is that the gnuplot.py directory was misnamed.

On Windows 2000 it was

C:\Python25\Lib\site-packages\gnuplot

instead of 

C:\Python25\Lib\site-packages\Gnuplot

    jab--who has a MUCH cleaner workstation now

John Bollinger, CFA, CMT
www.BollingerBands.com

If you advance far enough, you arrive at the beginning.


      ____________________________________________________________________________________
Be a better friend, newshound, and 
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

BBands wrote:
> I am running the SVN version of Gnuplot.py, Python 2.5.2 and gnuplot 4.1.
> 
> import Gnuplot fails
> 
> import gnuplot succeeds
> 
> Since test.py and demo.py emply the Gnuplot syntax this seems a bit odd.
> 
> I must be missing something, but I am not sure what.

Yes, the Python module is called "Gnuplot", not "gnuplot".

My guess is that either you have some other unrelated gnuplot module
installed, or you are on an OS with a case-insensitive filesystem and
somehow the main directory got installed with a lowercase letter instead
of uppercase.

Michael

I am running the SVN version of Gnuplot.py, Python 2.5.2 and gnuplot 4.1.

import Gnuplot fails

import gnuplot succeeds

Since test.py and demo.py emply the Gnuplot syntax this seems a bit odd.

I must be missing something, but I am not sure what.

     jab 






John Bollinger, CFA, CMT
www.BollingerBands.com

If you advance far enough, you arrive at the beginning.


      ____________________________________________________________________________________
Be a better friend, newshound, and 
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

Mark Alford wrote:
> On Mon, 28 Apr 2008, John K. Parejko wrote:
>> The version of Gnuplot.py in SVN replaces Numeric with numpy.  It hasn't been 
>> turned into a release yet, but if you can download the latest subversion 
>> trunk (there should be information on how to do this from the sourceforge 
>> page), it should work just fine with numpy.
> 
> What is stopping us from turning the SVN version in to a release?
> I tested it on Fedora and it works fine.
> Is there anything else that I and others can do to expedite the process?

All that is missing is a volunteer who will follow through.

If you want to be that person, let me know what your sourceforge
username is and I'll give you the necessary permissions.  There's even a
document RELEASES.txt telling roughly what I had to do to make the last
release.

Michael

On Mon, 28 Apr 2008, John K. Parejko wrote:
> The version of Gnuplot.py in SVN replaces Numeric with numpy.  It hasn't been 
> turned into a release yet, but if you can download the latest subversion 
> trunk (there should be information on how to do this from the sourceforge 
> page), it should work just fine with numpy.

What is stopping us from turning the SVN version in to a release?
I tested it on Fedora and it works fine.
Is there anything else that I and others can do to expedite the process?

2008/4/29 Johan Ekh <ekh...@gm...>:

> Thanks Benny,
> I think I may have an old version of Gnuplot.py. When I use "with_" I get
> an error and when I use "with"
> I still get the same behavior as before.
>
> How do I download and install the most recent version?


obtain the version in svn, then run :
python setup.py  install

>
>
> /Johan
>
> 2008/4/29 Benny Malengier <ben...@gm...>:
>
> Please, have a look in test.py file in subversion.
> >
> > To plot this do:
> >
> > import Gnuplot
> > >
> > > x = linspace(0.0, 1.8, 18)
> > > y = 3 - 2*x + 2*sin(4*x)
> > > points2 = transpose(array([x,y]))
> >
> > --> no, not needed !
> >
> > >
> > >
> > > g = Gnuplot.Gnuplot(persist=1)
> > > d = Gnuplot.Data(points2, with='linespoints', title='points2')
> >
> > --> no, do:
> >  d = Gnuplot.Data(x, y, with_='linespoints', title='points2')
> >
> > Benny
> >
> >
>

Thanks Benny,
I think I may have an old version of Gnuplot.py. When I use "with_" I get an
error and when I use "with"
I still get the same behavior as before.

How do I download and install the most recent version?

/Johan

2008/4/29 Benny Malengier <ben...@gm...>:

> Please, have a look in test.py file in subversion.
>
> To plot this do:
>
> import Gnuplot
> >
> > x = linspace(0.0, 1.8, 18)
> > y = 3 - 2*x + 2*sin(4*x)
> > points2 = transpose(array([x,y]))
>
> --> no, not needed !
>
> >
> >
> > g = Gnuplot.Gnuplot(persist=1)
> > d = Gnuplot.Data(points2, with='linespoints', title='points2')
>
> --> no, do:
>  d = Gnuplot.Data(x, y, with_='linespoints', title='points2')
>
> Benny
>
>

Please, have a look in test.py file in subversion.

To plot this do:

import Gnuplot
>
> x = linspace(0.0, 1.8, 18)
> y = 3 - 2*x + 2*sin(4*x)
> points2 = transpose(array([x,y]))

--> no, not needed !

>
>
> g = Gnuplot.Gnuplot(persist=1)
> d = Gnuplot.Data(points2, with='linespoints', title='points2')

--> no, do:
 d = Gnuplot.Data(x, y, with_='linespoints', title='points2')

Benny

Hi all,
I'm trying to make my first plot using Gnuplot.py.
I'm using scipy and numpy and I have issued the following commands in my
script:


import Gnuplot

x = linspace(0.0, 1.8, 18)
y = 3 - 2*x + 2*sin(4*x)
points2 = transpose(array([x,y]))

g = Gnuplot.Gnuplot(persist=1)
d = Gnuplot.Data(points2, with='linespoints', title='points2')
g.plot(d)


However, the plot consists of only one point, even though the vectors x and
y look as I expect.
I also get the following message:

In [3]: Warning: empty x range [0:0], adjusting to [-1:1]
Warning: empty y range [3:3], adjusting to [2.97:3.03]
gnuplot_x11: Some character sets not available


Does anyone know what I'm doing wrong?

Best regards,
Johan

> So the question is: what should I do?
> gnuplot-py requires Numeric which requires python 2.4
>
> Isn't there a way to make gnuplot-py working with the new "num-py"  
> module and so with python 2.5?


The version of Gnuplot.py in SVN replaces Numeric with numpy.  It  
hasn't been turned into a release yet, but if you can download the  
latest subversion trunk (there should be information on how to do  
this from the sourceforge page), it should work just fine with numpy.

John