[Gnuplot-py-users] (no subject)

[<e.pasca@nk...>]

Dear all, 

I have a strange behaviuor when I try to plot. Unfortunately I couldn't yet discover something reproducible in the error. 

Here it is the gnuplot output

put data ('e' ends) > 3095.52929688 5558.18603516 input data ('e' ends) > 3109.60009766 5553.88037109 input data ('e' ends) > 3123.6706543 5549.58398438171 554186 178182 660198 24 input data ('e' ends) > 4352383.282863.721483.361153.9 189236 792242 271257 8201
79552555.243225.264085.196755.427425.189185.160855.6819.13 127423 664428 221434 825440 458455 016551 687567 2935725.534805.845565.5 input data ('e' ends) > 393446 input data ('e' ends) > 85.9 input data ('e' ends) >  

what does it mean? It seems that the numbers got mixed up, isn't it?

I'm plotting two numpy.arrays as

Gnuplot.Data(X,Y, with_="lines", title="some title")

Now at some point I could fix this problem by plotting "lines" instead of "linespoints". But indeed the problem is not that.

Thanks

Edo

Edoardo Pasca, Ph.D
The Netherlands Cancer Institute -- Antoni van Leeuwenhoek Hospital
Radiotherapy Department
Plesmanlaan 121
1066 CX Amsterdam
tel: 0031 (0)20 512 1759 

Re: [Gnuplot-py-users] export output from gnuplot to python

[Krzysztof Berniak <krzysztof.berniak@gm...>]

Hi,

Thanks for your answers.
I don't have more time to play with gnuplot so I decided to use only
python's tools
Finally I use scipy to fittnig my data.

This is my code
def main():
    z,f = numbers_col()
    fitfunc = lambda v, z: v[0]*exp(z*v[1])
    errfunc = lambda v, z,f: fitfunc(v,z) - f
    v0 = [10., 0.005]
    v1,success = optimize.leastsq(errfunc, v0[:], args =
(z,f),full_output=True)

best regars,
Cris

2011/11/14 Michael Haggerty <mhagger@...>

> On 11/12/2011 05:43 PM, Krzysztof Berniak wrote:
> > I'm writing script in python, which fitting exponencial curve to data (
> > f(x) = a*exp(x*b).
> > To this problem, I use gnuplot in my script. Gnuplot display parameters
> > ( a +/- delta a; b +/- delta b)
> > How Can I use/save this parameters in python variables in next steps of
> > my script,
> >
> > def main():
> >    ...
> >
> >
> >     plot = Gnuplot.Gnuplot()
> >
> >     plot('f1(x) = a1*exp(b1*x)')
> >     plot('a1 = 300; b1 = 0.005;')
> >     plot('fit f1(x) "data.txt" using 1:2 via a1, b1')
> >
> >     print "first parameter", a1
> >     print "second parameter", b1   # is it feasible ? Or there is
> > another way to see the results ( parameter a1 and b1) of gnuplot by
> python
> >
> >     #plot('set terminal postscript')
> >     #plot('set output "output.ps <http://output.ps>"';)
>
> The communication from Gnuplot.py to gnuplot is currently
> one-directional only.  Therefore, there is no way for Gnuplot.py to see
> the output that gnuplot writes to the terminal.
>
> So without rewriting Gnuplot.py to change that fact (which would be a
> big job), the easiest thing would be to have gnuplot write the results
> of the fit to a file (see "help update"), then have Python read that
> file.  If you want more information about the fit, you can read the
> logfile that gnuplot's "fit" command generates (see "help fit", "help
> set fit").
>
> Michael
>
> --
> Michael Haggerty
> mhagger@...
> http://softwareswirl.blogspot.com/
>

Re: [Gnuplot-py-users] tics labels

[Lorenzo Mentaschi <lorenzo.mentaschi@ya...>]

Hallo Michael,
even if it is not frequently maintained, I find gnuploy.py still very useful, and, I think, it's clear and well written. So, thank you for this project!

About set_tics method, I think your way is far more general than mine, and agree with your approach.

lorenzo

--- Sab 19/11/11, Michael Haggerty <mhagger@...> ha scritto:

> Da: Michael Haggerty <mhagger@...>
> Oggetto: Re: [Gnuplot-py-users] tics labels
> A: "Lorenzo Mentaschi" <lorenzo.mentaschi@...>
> Cc: gnuplot-py-users@...
> Data: Sabato 19 novembre 2011, 14:36
> On 10/09/2011 12:03 PM, Lorenzo
> Mentaschi wrote:
> > in order to make gnuplot-py deal with xtics labels of
> string type
> > (without building the gnuplot command in user code), I
> added this method
> > to Gnuplot class:
> > 
> > 
> > ######
> >     def set_tics_labels(self,
> labels, axis = 'x'):
> >         cmd = 'set
> %stics (' % (axis,)
> >         for indx, label
> in zip(range(len(labels)), labels):
> >           cmd +=
> '"%s" %i,' % (label, indx,)
> >         cmd =
> cmd.strip(',') + ")"
> >         self(cmd) 
>   
> > ######
> > 
> > 
> > This seems to work for me. Is there any smarter way to
> do this?
> 
> Thanks for your submission.
> 
> If I understand correctly, your function takes a list of
> label strings
> as argument and places the 0th label in the list at axis
> position 0.0,
> the 1st label at axis position 1.0, etc.  Is that
> correct?
> 
> Because this seems quite special-purpose.  The gnuplot
> "set *tics"
> commands are much more flexible.  They allow labels to
> be placed at
> arbitrary positions and allow the user to specify which tic
> marks are
> major vs. minor.  If we are going to build a
> set_tics_labels() command
> into Gnuplot, then I think it should provide access to more
> of gnuplot's
> functionality.
> 
> What do you think of the attached patch?
> 
> Michael
> 
> -- 
> Michael Haggerty
> mhagger@...
> http://softwareswirl.blogspot.com/
> 

Re: [Gnuplot-py-users] can't set date format, is that a bug?

[Michael Haggerty <mhagger@al...>]

On 11/19/2011 02:55 PM, Michael Haggerty wrote:
> On 11/18/2009 11:15 AM, zhangwei02 wrote:

Oops, that's embarrassing.  Somehow I had my email sorted strangely and
I thought that this was a brand new message that I was responding to.  I
didn't notice that it was 1 day *plus 2 years* old.  My response is
probably too late to be helpful :-(

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

Re: [Gnuplot-py-users] can't set date format, is that a bug?

[Michael Haggerty <mhagger@al...>]

On 11/18/2009 11:15 AM, zhangwei02 wrote:
> [...]
> gnuplot: unable to open display ''
> gnuplot: X11 aborted.
>          line 0: Bad format character

Why do you think that your problem is related to setting a date format?
 Can you get your script to plot *anything*?

It looks to me like gnuplot is trying to plot something to an X11 window
(even though your goal seems to be only to output a PNG file).  But it
fails, perhaps because your DISPLAY variable is not set or because you
are running this script in a non-X11 environment.

I'm not sure why gnuplot is trying to output to an X11 window.  You
don't seem to generate any output before resetting the terminal type to
png.  Try loading the same commands (the ones output by Gnuplot.py's
debug=1 mode) directly into gnuplot in the same environment where the
script is run, to see whether it gives the same error.  If so, try
running the same commands but without the "set terminal x11" line and
see if the error goes away.  If it goes away, then you might be able to
work around the problem by setting your "default_term" to something
other than x11 using something like

    import Gnuplot
    Gnuplot.gp.GnuplotOpts.default_term = 'dumb'

before you create the Gnuplot object.

Please let us know what you find out.

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

Re: [Gnuplot-py-users] tics labels

[Michael Haggerty <mhagger@al...>]

On 10/09/2011 12:03 PM, Lorenzo Mentaschi wrote:
> in order to make gnuplot-py deal with xtics labels of string type
> (without building the gnuplot command in user code), I added this method
> to Gnuplot class:
> 
> 
> ######
>     def set_tics_labels(self, labels, axis = 'x'):
>         cmd = 'set %stics (' % (axis,)
>         for indx, label in zip(range(len(labels)), labels):
>           cmd += '"%s" %i,' % (label, indx,)
>         cmd = cmd.strip(',') + ")"
>         self(cmd)    
> ######
> 
> 
> This seems to work for me. Is there any smarter way to do this?

Thanks for your submission.

If I understand correctly, your function takes a list of label strings
as argument and places the 0th label in the list at axis position 0.0,
the 1st label at axis position 1.0, etc.  Is that correct?

Because this seems quite special-purpose.  The gnuplot "set *tics"
commands are much more flexible.  They allow labels to be placed at
arbitrary positions and allow the user to specify which tic marks are
major vs. minor.  If we are going to build a set_tics_labels() command
into Gnuplot, then I think it should provide access to more of gnuplot's
functionality.

What do you think of the attached patch?

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

Re: [Gnuplot-py-users] PlotItem: fs option

[Michael Haggerty <mhagger@al...>]

On 10/09/2011 03:31 PM, Lorenzo Mentaschi wrote:
> Hi all, I modified PlotItem class adding the option "fs", meaning
> "fill style", an option useful when plotting bar diagrams or
> histograms (added in gnuplot 4). Theoretically it should be possible
> to set this option also by "with boxes fs <pattern>", but this seem
> to lose box width settings ...)
> 
> I'm sending you the modified PlotItem.py module, and a sample script
> to make clear when this might come useful (in the example the method
> I described in my previous mail is also used).
> 
> Do you think what I did is right? Or is there a better approach?

Sorry for the delay in responding.  (Gnuplot.py is not really being
maintained anymore.)

Your suggestion looks fine.  I just committed it to Gnuplot.py as r317.

Thanks!
Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

Re: [Gnuplot-py-users] newbie w/ stockcharts with time in x

[Michael Haggerty <mhagger@al...>]

On 10/07/2011 02:45 PM, Charles Gagnon wrote:
> I had a gnoplot script I am trying to convert into python using
> gnuplot-py. I have a list of historical prices like this:
> 
> 2010-11-24,19.33,19.55,19.32,19.46,63880400,19.2
> 
> My basic gnuplot call were:
> 
> set grid nopolar
> set grid xtics nomxtics ytics nomytics noztics nomztics nox2tics
> nomx2tics noy2tics nomy2tics nocbtics nomcbtics
> set grid layerdefault   linetype 0 linewidth 1.000,  linetype 0 linewidth 1.000
> 
> # Chart Title
> set title "CSCO 1yr"
> 
> set xdata time
> set timefmt "%Y-%m-%d"
> set format x "%m/%y"
> 
> set lmargin  9
> set rmargin  2
> set datafile separator ","
> 
> plot 'data/CSCO_365.csv' using 1:2:3:4:5 notitle with financebars lt 8
> 
> Instead of doing it through a csv, I was trying to plot from the list
> directly. So I took out the datafile seperator but I am having a hard
> time porting that to python. Most of my options were just set with a
> 
> g('set lmargin 9')
> 
> line. But I'm not sure how to handle my plot statement. I can pass my
> list to g.plot() but it lacks all the other definitions. What am I
> missing?

Gnuplot.py only has support for plotting numbers, not dates.  You would
probably have to either derive your own PlotItem class or teach the
existing PlotItem classes how to work with dates.

One alternative that might work: if gnuplot can parse some numerical
format as dates (e.g., unix timestamp long values), then maybe you could
convert the dates to numbers then pass them to gnuplot as numbers.
However, you might lose too much precision, as Gnuplot.py would probably
want to convert the numbers to doubles or floats.

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

[Gnuplot-py-users] R: export output from gnuplot to python

[Lorenzo Mentaschi <lorenzo.mentaschi@ya...>]

Hi, I'm not an expert of scipy, but you might perform the fit using scipy, and then pass all to gnuplot.

http://www.scipy.org/Cookbook/FittingData

lorenzo

--- Sab 12/11/11, Krzysztof Berniak <krzysztof.berniak@...> ha scritto:

Da: Krzysztof Berniak <krzysztof.berniak@...>
Oggetto: [Gnuplot-py-users] export output from gnuplot to python
A: gnuplot-py-users@...
Data: Sabato 12 novembre 2011, 17:43

Hi all,
I'm writing script in python, which fitting exponencial curve to data ( f(x) = a*exp(x*b).To this problem, I use gnuplot in my script. Gnuplot display parameters ( a +/- delta a; b +/- delta b)
How Can I use/save this parameters in python variables in next steps of my script, 
def main():   ...            plot = Gnuplot.Gnuplot()
        plot('f1(x) = a1*exp(b1*x)')    plot('a1 = 300; b1 = 0.005;')    plot('fit f1(x) "data.txt" using 1:2 via a1, b1')        print "first parameter", a1
    print "second parameter", b1   # is it feasible ? Or there is another way to see the results ( parameter a1 and b1) of gnuplot by python  
    #plot('set terminal postscript')    #plot('set output "output.ps"')   
regards and please help,
Cristopher


-----Segue allegato-----

------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
-----Segue allegato-----

_______________________________________________
Gnuplot-py-users mailing list
Gnuplot-py-users@...
https://lists.sourceforge.net/lists/listinfo/gnuplot-py-users

Re: [Gnuplot-py-users] export output from gnuplot to python

[Michael Haggerty <mhagger@al...>]

On 11/12/2011 05:43 PM, Krzysztof Berniak wrote:
> I'm writing script in python, which fitting exponencial curve to data (
> f(x) = a*exp(x*b).
> To this problem, I use gnuplot in my script. Gnuplot display parameters
> ( a +/- delta a; b +/- delta b)
> How Can I use/save this parameters in python variables in next steps of
> my script, 
> 
> def main():
>    ...
>     
>     
>     plot = Gnuplot.Gnuplot()
>     
>     plot('f1(x) = a1*exp(b1*x)')
>     plot('a1 = 300; b1 = 0.005;')
>     plot('fit f1(x) "data.txt" using 1:2 via a1, b1')
>     
>     print "first parameter", a1
>     print "second parameter", b1   # is it feasible ? Or there is
> another way to see the results ( parameter a1 and b1) of gnuplot by python 
>  
>     #plot('set terminal postscript')
>     #plot('set output "output.ps <http://output.ps>"';)

The communication from Gnuplot.py to gnuplot is currently
one-directional only.  Therefore, there is no way for Gnuplot.py to see
the output that gnuplot writes to the terminal.

So without rewriting Gnuplot.py to change that fact (which would be a
big job), the easiest thing would be to have gnuplot write the results
of the fit to a file (see "help update"), then have Python read that
file.  If you want more information about the fit, you can read the
logfile that gnuplot's "fit" command generates (see "help fit", "help
set fit").

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

[Gnuplot-py-users] export output from gnuplot to python

[Krzysztof Berniak <krzysztof.berniak@gm...>]

Hi all,

I'm writing script in python, which fitting exponencial curve to data (
f(x) = a*exp(x*b).
To this problem, I use gnuplot in my script. Gnuplot display parameters ( a
+/- delta a; b +/- delta b)
How Can I use/save this parameters in python variables in next steps of my
script,

def main():
   ...


    plot = Gnuplot.Gnuplot()

    plot('f1(x) = a1*exp(b1*x)')
    plot('a1 = 300; b1 = 0.005;')
    plot('fit f1(x) "data.txt" using 1:2 via a1, b1')

    print "first parameter", a1
    print "second parameter", b1   # is it feasible ? Or there is another
way to see the results ( parameter a1 and b1) of gnuplot by python

    #plot('set terminal postscript')
    #plot('set output "output.ps"')


regards and please help,
Cristopher

[Gnuplot-py-users] PlotItem: fs option

[Lorenzo Mentaschi <lorenzo.mentaschi@ya...>]

Hi all,
I modified PlotItem class adding the option "fs", meaning "fill style", an option useful when plotting bar diagrams or histograms (added in gnuplot 4). Theoretically it should be possible to set this option also by "with boxes fs <pattern>", but this seem to lose box width settings ...)

I'm sending you the modified PlotItem.py module, and a sample script to make clear when this might come useful (in the example the method I described in my previous mail is also used).

Do you think what I did is right? Or is there a better approach?

Thank you, lorenzo

[Gnuplot-py-users] tics labels

[Lorenzo Mentaschi <lorenzo.mentaschi@ya...>]

Hi all,
in order to make gnuplot-py deal with xtics labels of string type (without building the gnuplot command in user code), I added this method to Gnuplot class:


######
    def set_tics_labels(self, labels, axis = 'x'):
        cmd = 'set %stics (' % (axis,)
        for indx, label in zip(range(len(labels)), labels):
          cmd += '"%s" %i,' % (label, indx,)
        cmd = cmd.strip(',') + ")"
        self(cmd)    
######


This seems to work for me. Is there any smarter way to do this?

best regards, lorenzo

[Gnuplot-py-users] newbie w/ stockcharts with time in x

[Charles Gagnon <charlesg@un...>]

I had a gnoplot script I am trying to convert into python using
gnuplot-py. I have a list of historical prices like this:

2010-11-24,19.33,19.55,19.32,19.46,63880400,19.2

My basic gnuplot call were:

set grid nopolar
set grid xtics nomxtics ytics nomytics noztics nomztics nox2tics
nomx2tics noy2tics nomy2tics nocbtics nomcbtics
set grid layerdefault   linetype 0 linewidth 1.000,  linetype 0 linewidth 1.000

# Chart Title
set title "CSCO 1yr"

set xdata time
set timefmt "%Y-%m-%d"
set format x "%m/%y"

set lmargin  9
set rmargin  2
set datafile separator ","

plot 'data/CSCO_365.csv' using 1:2:3:4:5 notitle with financebars lt 8

Instead of doing it through a csv, I was trying to plot from the list
directly. So I took out the datafile seperator but I am having a hard
time porting that to python. Most of my options were just set with a

g('set lmargin 9')

line. But I'm not sure how to handle my plot statement. I can pass my
list to g.plot() but it lacks all the other definitions. What am I
missing?

-- 
Charles Gagnon
charlesg at unixrealm.com

Re: [Gnuplot-py-users] How to contribute termdefs?

[Michael Haggerty <mhagger@al...>]

On 09/29/2011 09:18 PM, Greg van Anders wrote:
> I've just started using gnuplot-py and have been finding it very useful.
> 
> Version 1.8 doesn't seem to have the epslatex terminal option, which
> is the one I use most often in ordinary gnuplot. I added an epslatex
> terminal option and I was wondering how to contribute it as I thought
> others might find it useful.

Best would be to submit a patch to the mailing list.  Let me know if you
need more information about how to create a patch.

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

[Gnuplot-py-users] How to contribute termdefs?

[Greg van Anders <gregvananders@gm...>]

Hello,

I've just started using gnuplot-py and have been finding it very useful.

Version 1.8 doesn't seem to have the epslatex terminal option, which
is the one I use most often in ordinary gnuplot. I added an epslatex
terminal option and I was wondering how to contribute it as I thought
others might find it useful.

Thanks!
Greg

Re: [Gnuplot-py-users] Gnuplot.py, FIFOs and threading

[Ken Smith <k.w.smith@qu...>]

Many thanks Michael

Indeed I just needed to close my Gnuplot() objects.  Not a single  
error (in ~10,000 plots) after I did that.

Cheers,

Ken


On 22 Aug 2011, at 09:22, Michael Haggerty wrote:

> On 08/18/2011 05:31 PM, Ken Smith wrote:
>> Apologies in advance if there's an FAQ, but I can't find it...  (Feel
>> free to point it out to me...)
>
> There is a FAQ.txt in the Gnuplot.py distribution.  The first FAQ
> question is closely related to your problem.
>
>> The problem is that (presumably because of threading) Gnuplot.py is
>> sometimes not completing its work before I attempt to convert the
>> plot.  Hence ImageMagic complains about missing postscript files - or
>> the postscript files are incomplete when conversion starts.
>>
>> Is there a way to turn threading/FIFOs OFF?  Or is there a way I can
>> use "join" to force my python program to WAIT for the plot to be
>> completed before attempting to convert it?
>
> The problem is not so much the multithreading in Gnuplot.py (which
> anyway should be turned off if your turn off the use of FIFOs) but
> rather the fact that a separate process (the gnuplot program) is doing
> the real work.  I believe that if you close() the Gnuplot object
> explicitly then it will wait for gnuplot to finish (though of course  
> you
> will have to create a new Gnuplot object if you want to do more work
> afterwards):
>
> g = Gnuplot.Gnuplot()
> # Use g to plot what you want
> g.close()
> # At this point I think that all of the gnuplot output should be
> # done.
>
> Please try this out and tell us if it works.
>
> Michael
>
> -- 
> Michael Haggerty
> mhagger@...
> http://softwareswirl.blogspot.com/

Re: [Gnuplot-py-users] Multiple Datasets Gnuplot

[Michael Haggerty <mhagger@al...>]

On 08/23/2011 04:33 AM, Sears Merritt wrote:
> I am trying plot many datasets (1500) in a single plot and am running
> into an I/O error due to too many open files.
> 
> The code is as follows:
> 
> for row in dataSet:
> plots.append(Gnuplot.PlotItems.Data(eval(row[2])))
> g.plot(*plots)
> g.hardcopy(filename='output.ps')
> 
> Has anyone found a workaround for this issue?

You might want to put multiple datasets in a single file (e.g.,
separated by blank lines).  This means that they will all be plotted in
the same style, but with 1500 datasets I doubt that the datasets are
very distinct anyway.

If you insist on having so many distinct datasets, they you have to
increase the number of file descriptors allowed to your process.  See,
for example, [1] or [2].

Michael

[1]
http://stackoverflow.com/questions/880557/linux-socket-accept-too-many-open-files
[2]
http://www.cyberciti.biz/faq/linux-increase-the-maximum-number-of-open-files/

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

[Gnuplot-py-users] Multiple Datasets Gnuplot

[Sears Merritt <sears.merritt@gm...>]

I am trying plot many datasets (1500) in a single plot and am running into an I/O error due to too many open files.

The code is as follows:

for row in dataSet:
plots.append(Gnuplot.PlotItems.Data(eval(row[2])))
g.plot(*plots)
g.hardcopy(filename='output.ps')

Has anyone found a workaround for this issue?

Thanks in advance 

Re: [Gnuplot-py-users] Gnuplot.py, FIFOs and threading

[Michael Haggerty <mhagger@al...>]

On 08/18/2011 05:31 PM, Ken Smith wrote:
> Apologies in advance if there's an FAQ, but I can't find it...  (Feel  
> free to point it out to me...)

There is a FAQ.txt in the Gnuplot.py distribution.  The first FAQ
question is closely related to your problem.

> The problem is that (presumably because of threading) Gnuplot.py is  
> sometimes not completing its work before I attempt to convert the  
> plot.  Hence ImageMagic complains about missing postscript files - or  
> the postscript files are incomplete when conversion starts.
> 
> Is there a way to turn threading/FIFOs OFF?  Or is there a way I can  
> use "join" to force my python program to WAIT for the plot to be  
> completed before attempting to convert it?

The problem is not so much the multithreading in Gnuplot.py (which
anyway should be turned off if your turn off the use of FIFOs) but
rather the fact that a separate process (the gnuplot program) is doing
the real work.  I believe that if you close() the Gnuplot object
explicitly then it will wait for gnuplot to finish (though of course you
will have to create a new Gnuplot object if you want to do more work
afterwards):

g = Gnuplot.Gnuplot()
# Use g to plot what you want
g.close()
# At this point I think that all of the gnuplot output should be
# done.

Please try this out and tell us if it works.

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

[Gnuplot-py-users] Gnuplot.py, FIFOs and threading

[Ken Smith <k.w.smith@qu...>]

Folks

Apologies in advance if there's an FAQ, but I can't find it...  (Feel  
free to point it out to me...)  There also doesn't seem to be a simple  
way to search the Sourceforge archives for this group...

I've been using Gnuplot.py for a while now, and I keep running into  
the same problem.  I'm creating postscript plots, but at the end of my  
script I'd like to convert them into JPEGs using ImageMagick (convert).

The problem is that (presumably because of threading) Gnuplot.py is  
sometimes not completing its work before I attempt to convert the  
plot.  Hence ImageMagic complains about missing postscript files - or  
the postscript files are incomplete when conversion starts.

Is there a way to turn threading/FIFOs OFF?  Or is there a way I can  
use "join" to force my python program to WAIT for the plot to be  
completed before attempting to convert it?

I've tried setting the following after my Gnuplot import:

Gnuplot.GnuplotOpts.prefer_fifo_data = 0
Gnuplot.GnuplotOpts.support_fifo = 0
Gnuplot.GnuplotOpts.prefer_inline_data = 1

but to no avail.  The only thing I'm doing at the moment is checking  
that the postscript file exists - but this is just a fudge and doesn't  
prevent other issues if the file is incomplete.  Below are example  
errors...

Ken



Example errors:

MISSING ps file

convert: unable to open image `/tmp/tempPSPlots/ 
1095652401023547700_lc.ps':  @
error/blob.c/OpenBlob/2584.
convert: missing an image filename
`/psdb/images/ps1ss/lightcurves/1095652401023547700_lc.jpeg' @
error/convert.c/ConvertImageCommand/2949.


INCOMPLETE ps file (presumably that's what causes ghostscript to barf)

Error: /syntaxerror in -file-
Operand stack:
    MLshow
Execution stack:
    %interp_exit   .runexec2   --nostringval--   --nostringval--
    %--nostringval--   2   %stopped_push   --nostringval--   -- 
nostringval--
    %--nostringval--   false   1   %stopped_push   1894   1   3    
%oparray_pop
    %1893   1   3   %oparray_pop   1877   1   3   %oparray_pop    
1771   1   3
    %%oparray_pop   --nostringval--   %errorexec_pop   .runexec2
    %--nostringval--   --nostringval--   --nostringval--   2    
%stopped_push
Dictionary stack:
    --dict:1158/1684(ro)(G)--   --dict:0/20(G)--   --dict:80/200(L)--
--dict:176/256(L)--
Current allocation mode is local
Last OS error: 2
GPL Ghostscript  9.00: Unrecoverable error, exit code 1

Re: [Gnuplot-py-users] keep plot open after program close?

[Michael Haggerty <mhagger@al...>]

On 07/17/2011 05:03 AM, tony wrote:
> I'm using gnuplot-py to monitor data from a serial port, and it works 
> fine while the main python program is running. But I would like to keep 
> the plot open after python closes - is that possible? I can obviously 
> reopen the data file and reconstruct the plot, but it would be useful to 
> keep the original plot window open.

g = Gnuplot.Gnuplot(persist=True) , if your platform supports it.

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/

[Gnuplot-py-users] keep plot open after program close?

[tony <t.vassallo@in...>]

Michael
I'm using gnuplot-py to monitor data from a serial port, and it works 
fine while the main python program is running. But I would like to keep 
the plot open after python closes - is that possible? I can obviously 
reopen the data file and reconstruct the plot, but it would be useful to 
keep the original plot window open.
Thanks
Tony

[Gnuplot-py-users] jpeg or png output

[alain dimier <alain.dimier@gm...>]

Hi,
Using gnuplot.py in an interactive manner, I would like to generate in the
mean time a png or jpeg output. Is there a way to do that while keeping the
interactive window?

Regards

Alain
--

Re: [Gnuplot-py-users] labels into plot

[Michael Haggerty <mhagger@al...>]

On 07/03/2011 11:54 AM, tony wrote:
> Thanks Michael
> I can now place a label on the graph, but what I really want to do is 
> have the label be a variable that is updated every few seconds.
> I've tried to include this in
>     g('set label 1 sprintf("%f",inc) at 2,2')
> where inc is the variable but in a While loop I keep getting an 
> undefined variable error, even though the variable is defined in the 
> line above. I've seen this done in other scripts, but not necessarily in 
> a while loop - is that the problem?

It sounds like you are defining the variable "inc" in Python but then
expecting gnuplot to know how to put it into a string.  Either you have
to define the variable in gnuplot or you have to do the string
interpolation in Python.

You might try the "debug" option to the Gnuplot constructor to see what
commands it is passing to gnuplot.  That might make the situation clearer.

Michael

-- 
Michael Haggerty
mhagger@...
http://softwareswirl.blogspot.com/