gnuplot-py-users Mailing List for Gnuplot.py

Mark Alford wrote:
> I have been using gnuplot-py-1.7 for many years.
> I would like to help testing/development of gnuplot-py-1.8 under Fedora 
> Core Linux. But I don't know how to get the latest version. I have never 
> used subversion.
> 
> Could someone tell me a simple series of steps for downloading
> the latest version of the code?

Your help is much appreciated.  It's been a long time since Gnuplot.py
has had any serious maintenance.

To check out the current trunk version, type:

svn co https://gnuplot-py.svn.sourceforge.net/svnroot/gnuplot-py/trunk
gnuplot-py

This will create a subdirectory called "gnuplot-py" under your current
directory containing the current cvs2svn source code.

Once you have made some changes, you can create a patch by typing "svn
diff" in the gnuplot-py directory.  Please test any changes against the
test suite to make sure you haven't broken anything.

Michael

I have been using gnuplot-py-1.7 for many years.
I would like to help testing/development of gnuplot-py-1.8 under Fedora 
Core Linux. But I don't know how to get the latest version. I have never 
used subversion.

Could someone tell me a simple series of steps for downloading
the latest version of the code?

Thank you

Mark

It seems that there is a bug in Gnuplot/utils.py write_array for the
case of 2 dimensional array.
I use gnuplot.py 1.7 with python 2.5 on FreeBSD.

        for point in set[1:]:
            f.write(nest_sep + nest_prefix)
            f.write(fmt % tuple(point.tolist()))

I think that write_array is called with array as a parameter, not a
regular list and it seems that slice syntax is not exactly applicable to
arrays in the same way as it is for lists.
It seems that no iterations actually happened and only data for set[0]
was written to a file.

I had to change the quoted code to the following:
        for point in set.tolist()[1:]:
            f.write(nest_sep + nest_prefix)
            f.write(fmt % tuple(point))


The problem was doing stuff like the following:
x = Numeric.arange(10) / 10.
y = Numeric.sin(x)
g = Gnuplot.Gnuplot()
g.plot(Gnuplot.Data(x,y))

Only one point was plotted.

P.S. I think that this link pretty well describes things for python 2.5
and Numeric (array):
http://projects.scipy.org/pipermail/numpy-discussion/2006-November/024428.html

-- 
Andriy Gapon

Mustafa Burak Boz wrote:
> I want to plot a second histogram on the top the first one. Is there a
> way to do that with gnuplot.py ?

Sorry, I don't think gnuplot even supported histograms when Gnuplot.py
was written, and right now nobody is actively maintaining Gnuplot.py.
(Volunteers always welcome!)

Michael

Hi there,
I want to plot a second histogram on the top the first one. Is there a  
way to do that with gnuplot.py ?

here is the script that I use to generate the first histogram:

     g=Gnuplot.Gnuplot(persist=1)

     max_x=find_max_x(x_y)
     max_y=find_max_y(x_y)

     g("set style data histograms ")
     g("set style fill solid 1.0 border -1")

     g("set xrange [0:"+str(max_x)+"]")
     g("set yrange [0:"+str(max_y+10)+"]")

     graph=Gnuplot.Data(x_y,with="boxes 3")
     g.plot(graph)
     g.hardcopy(of1+".ps",enhanced=1,color=1)

In the Gnuplot manual, it is described like this:

	newhistogram  { "title" } { <linestyle> }. For example
	set style histogram cluster
	plot newhistogram "Set A", ’a’ using 1, ’’ using 2, ’’ using 3, \
	newhistogram "Set B", ’b’ using 1, ’’ using 2, ’’ using 3

How can I do this in gnuplot.py and does g.hardcopy method give me the  
graph of 2 hostograms ?

Thank in advance...
mbb