matplotlib-users Mailing List for matplotlib

The new Figure Options button in the PyQt plot toolbar is a really nice
feature that allows the user to change the line style for curves in the
plot.  Unfortunately, the legend does not get updated, so the legend becomes
out of sync with the plot - when a user changes a line from black to red,
the legend will still show the line as black, for instance.  Forcing the
legend to redraw itself via a draw() operation does not solve the problem -
it redraws, but still with the old color.

The issue seems to be that there doesn't appear to be a way to have a legend
update what its artists should look like based on their latest appearances.
I've developed a fix for this that seems to work, but wanted to see if
anyone had something better, or if people think I should submit this as a
potential fix.  I'm not thrilled with it (it will fail if there are multiple
legends on the figure, for instance), but it does work for simple plots and
figures with multiple subplots.

The fix is in the apply_callback function inside the the
matplotlib.backends.qt4_editor.figureoptions.figure_edit function.  I've
included the whole function below, with the fix contained within the
designated comments.

Comments?  Is there an easier, better way to do this?

Dave

    def apply_callback(data):
        """This function will be called to apply changes"""
        if has_curve:
            general, curves = data
        else:
            general, = data

        # Set / General
        title, xmin, xmax, xlabel, xscale, ymin, ymax, ylabel, yscale =
general
        axes.set_xscale(xscale)
        axes.set_yscale(yscale)
        axes.set_title(title)
        axes.set_xlim(xmin, xmax)
        axes.set_xlabel(xlabel)
        axes.set_ylim(ymin, ymax)
        axes.set_ylabel(ylabel)

        if has_curve:
            # Set / Curves
            for index, curve in enumerate(curves):
                line = linedict[curvelabels[index]]
                label, linestyle, linewidth, color, \
                    marker, markersize, markerfacecolor, markeredgecolor =
curve
                line.set_label(label)
                line.set_linestyle(linestyle)
                line.set_linewidth(linewidth)
                line.set_color(color)
                if marker is not 'none':
                    line.set_marker(marker)
                    line.set_markersize(markersize)
                    line.set_markerfacecolor(markerfacecolor)
                    line.set_markeredgecolor(markeredgecolor)

### Fix begins here
            # Recalculate the color and style of legend artists - note, this
            # does not work for multiple legends on the same figure.
            figure = axes.get_figure()
            all_axes = figure.get_axes()
            for legend in figure.legends:
                # Need to gather all the legend items from all the subplots
                all_handles = []
                all_labels = []
                for one_axes in all_axes:
                    handles, labels = one_axes.get_legend_handles_labels()
                    all_handles.extend(handles)
                    all_labels.extend(labels)
                # Location and title are altered by the _init_legend_box
                # function, so preserve the current values
                loc = legend._loc
                title = legend.get_title().get_text()
                if title == 'None':
                    title = None
                # This function recalculates colors and styles
                legend._init_legend_box(all_handles, all_labels)
                legend._loc = loc
                legend.set_title(title)
### Fix ends here

        # Redraw
        figure = axes.get_figure()
        figure.canvas.draw()

Hi,

The following code displays a figure with both a red and a blue arrow 
(generated by annotate). But the pdf that is saved only shows the red 
one. It seems that the line ax1.set_xscale('log') is causing the 
trouble, because if it is commented out, the blue arrow reappears in the 
pdf. Any hints?

import numpy as np
from matplotlib import pyplot as pp

fig = pp.figure()
ax2 = fig.add_subplot(111)
ax1 = ax2.twinx()
ax1.set_xlim(xmin=100,xmax=10000)
ax1.set_ylim(ymin=-60, ymax=5)
ax1.set_xscale('log')
ax2.set_xlim(xmin=100,xmax=10000)
ax2.set_ylim(ymin=-5,ymax=35)
ax2.set_xscale('log')

ax1.annotate("", (10000,-30), (2000,-30),
              arrowprops=dict(arrowstyle="-|>",fc="b", ec="b"))
ax2.annotate("", (100, 5), (2000, 5),
              arrowprops=dict(arrowstyle="-|>",fc="r", ec="r"))
pp.savefig('test_arrow.pdf')
pp.show()

-- 
Christopher Brown, Ph.D.
Associate Research Professor
Department of Speech and Hearing Science
Arizona State University
http://pal.asu.edu

On Thu, Jul 14, 2011 at 3:57 PM, T. Tofus von Blisstein
<tuf...@go...> wrote:
> Hi,
>
> how can I invert the colors of axes/background from black/white to white/black?
>
> thanks... I have been googling for a while...

If you want to do it for all your plots, you can mess with all of the
'color*' settings in the .matplotlibrc file:
   http://matplotlib.sourceforge.net/users/customizing.html#the-matplotlibrc-file

There are plenty of colors to change (font, line, axes, figure, patch,
etc), so just grep through there. You'll probably want to change the
color_cycle list.

To do the same thing at runtime, play around with matplotlib.rc(),
passing in the ones you'd change from above:
   http://matplotlib.sourceforge.net/api/matplotlib_configuration_api.html#matplotlib.rc

Then you can reset it using rcdefaults():
   http://matplotlib.sourceforge.net/api/matplotlib_configuration_api.html#matplotlib.rcdefaults

If you don't like those approaches, you can specify the color for
about everything using keyword arguments as you create the figure,
axes, etc (using the object-oriented API), or call plot(), subplot(),
etc (using the interactive pylab API). Try help(figure) and
help(colors) for some examples.

Hope that helps,
    Justin

hi, is it possible to restrict a contour plot to a
country? if i grid my data to projection coordinates
and then make a contour plot i can draw the country
borders on top, but since the data plotted outside
the country is a gridding artifact i would rather
not plot it.

to make it more clear what i mean:
http://rokuko.net/basicmap.png

any ideas? can one make a single country transparent?

best regards, yoshi

my problem was fixed with this

var = float(sys.argv[1])

thanks!



On 14 July 2011 16:24, Benjamin Root <ben...@ou...> wrote:
> On Thursday, July 14, 2011, Pau <vim...@go...> wrote:
>> Hi,
>>
>> I have a very naive question ... howmdo I pass an arg to a matplotlib
>> script? I need to run MyScript 34 12 67 25 for the script to use those
>> numbers to multiply quantities for producing a plot... I am using
>> sysarg[1], 2, 3 but is seems not to work.
>>
>> thanks
>>
>
> Could you attach your script so that we can see if there is a mistake
> of some sort.
>
> Ben Root
>

Hi,

how can I invert the colors of axes/background from black/white to white/black?

thanks... I have been googling for a while...

T.

On Wed, Jul 13, 2011 at 6:49 PM, Benjamin Root <ben...@ou...> wrote:
> On Wednesday, July 13, 2011, Justin McCann <jn...@gm...> wrote:
>> $ ipython -pylab
>> # ====
>> from matplotlib.collections import LineCollection
>> f = figure()
>> plot()
>> ax = gca()
>> vec = numpy.random.random((10,3))
>> segs = []
>> for i in range(0, len(vec)):
>>     x1 = vec[i][0]
>>     #x1 = 3000000
>>     x2 = vec[i][1]
>>     #x2 = 4000000
>>     y1 = y2 = vec[i][2]
>>     segs.extend( [[(x1,y1),(x2,y2)]] )
>>
>> line_segments = LineCollection(segs, linewidth=3, alpha=0.3, colors =
>> 'r', linestyle = 'solid')
>> ax.add_collection(line_segments)
>> ax.set_xlim(0,1)
>> ax.set_ylim(0,1)
>> show() # hmmmm... nothing yet
>> draw() # force a draw; now it works
>>
...
> Just an observation (I haven't tested anything)... But what is up with
> the call to plot()?  It might be causing issues with the autoscaler.
> Any line collections created without other plotting functions is going
> to need the axes limits set.
>
> Ben Root
>

Yeah, that was weird. :)

I added that when I was first messing around, and the axes didn't show
up even when I called show(). If you do the draw() at the end, then
you don't need to call plot(), and then you also don't need to mess
with the view limits (set_{x,y}lim).

I guess the moral of the story is, "if you don't explicitly plot() [or
a variant], you must explicitly draw()."

===
from matplotlib.collections import LineCollection
f = figure()
ax = gca()
vec = numpy.random.random((10,3))
segs = []
for i in range(0, len(vec)):
   x1 = vec[i][0]
   x2 = vec[i][1]
   y1 = y2 = vec[i][2]
   segs.extend( [[(x1,y1),(x2,y2)]] )

line_segments = LineCollection(segs, linewidth=3, alpha=0.3,
colors='r', linestyle='solid')
ax.add_collection(line_segments)
draw() # force a draw; now it works
====

On Thursday, July 14, 2011, Pau <vim...@go...> wrote:
> Hi,
>
> I have a very naive question ... howmdo I pass an arg to a matplotlib
> script? I need to run MyScript 34 12 67 25 for the script to use those
> numbers to multiply quantities for producing a plot... I am using
> sysarg[1], 2, 3 but is seems not to work.
>
> thanks
>

Could you attach your script so that we can see if there is a mistake
of some sort.

Ben Root

Sorry, it doesn't work.
ax.set_ylabel("foo") 
doesn't change anything (before "ax.yaxis.cla()").

And f.clf(keep_observers=False)  still locks the process.

But maybe I used wrong terms. Here is the displays ; left before the
clearing ; right, after the clearing ; the plot is correctly renewed, but
the ticks labels are printed over the previous ones.
http://old.nabble.com/file/p32060202/Capture-6.png 

Thank you.
JD
-- 
View this message in context: http://old.nabble.com/Clearing-the-axis-in-a-figure-embedded-in-Tkinter-tp32054762p32060202.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Sorry, it doesn't work.
ax.set_ylabel("foo") 
doesn't change anything (before "ax.yaxis.cla()").

And f.clf(keep_observers=False)  still locks the process.

But maybe I used wrong terms. Here is the displays ; left before the
clearing ; right, after the clearing ; the plot is correctly renewed, but
the ticks labels are printed over the previous ones.
http://old.nabble.com/file/p32060191/Capture-6.png Capture-6.png 

Thank you.
JD
-- 
View this message in context: http://old.nabble.com/Clearing-the-axis-in-a-figure-embedded-in-Tkinter-tp32054762p32060191.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Hi,

I have a very naive question ... howmdo I pass an arg to a matplotlib
script? I need to run MyScript 34 12 67 25 for the script to use those
numbers to multiply quantities for producing a plot... I am using
sysarg[1], 2, 3 but is seems not to work.

thanks

On 07/13/2011 09:18 PM, Armagan Tarim wrote:
> Hi All,
> I have tried the sample PySide code given at
> http://www.scipy.org/Cookbook/Matplotlib/PySide
> the below excerpt is from my "Python Shell" for this sample code, which
> gives this error,

Github master works with pyside.  It is expected to emerge as a release 
within a few weeks.

Eric

> "ImportError: Warning: formlayout requires PyQt4 >v4.3"
> It seems that it is looking for PyQt; but the whole point of using
> PySide is to avoid using PyQt.
> Any help is appreciated.
> -- Armagan
> Python 2.7.2 (default, Jun 12 2011, 15:08:59) [MSC v.1500 32 bit
> (Intel)] on win32
> Type "copyright", "credits" or "license()" for more information.
>  >>> import sys
>  >>> import matplotlib
>  >>> matplotlib.use('Qt4Agg')
>  >>> import pylab
> Traceback (most recent call last):
> File "<pyshell#3>", line 1, in <module>
> import pylab
> File "C:\Python27\lib\site-packages\pylab.py", line 1, in <module>
> from matplotlib.pylab import *
> File "C:\Python27\lib\site-packages\matplotlib\pylab.py", line 263, in
> <module>
> from matplotlib.pyplot import *
> File "C:\Python27\lib\site-packages\matplotlib\pyplot.py", line 95, in
> <module>
> new_figure_manager, draw_if_interactive, show = pylab_setup()
> File "C:\Python27\lib\site-packages\matplotlib\backends\__init__.py",
> line 25, in pylab_setup
> globals(),locals(),[backend_name])
> File
> "C:\Python27\lib\site-packages\matplotlib\backends\backend_qt4agg.py",
> line 12, in <module>
> from backend_qt4 import QtCore, QtGui, FigureManagerQT, FigureCanvasQT,\
> File "C:\Python27\lib\site-packages\matplotlib\backends\backend_qt4.py",
> line 18, in <module>
> import matplotlib.backends.qt4_editor.figureoptions as figureoptions
> File
> "C:\Python27\lib\site-packages\matplotlib\backends\qt4_editor\figureoptions.py",
> line 11, in <module>
> import matplotlib.backends.qt4_editor.formlayout as formlayout
> File
> "C:\Python27\lib\site-packages\matplotlib\backends\qt4_editor\formlayout.py",
> line 51, in <module>
> raise ImportError, "Warning: formlayout requires PyQt4 >v4.3"
> ImportError: Warning: formlayout requires PyQt4 >v4.3
>
> -------------------------------------------------
>
>
>
> ------------------------------------------------------------------------------
> AppSumo Presents a FREE Video for the SourceForge Community by Eric
> Ries, the creator of the Lean Startup Methodology on "Lean Startup
> Secrets Revealed." This video shows you how to validate your ideas,
> optimize your ideas and identify your business strategy.
> http://p.sf.net/sfu/appsumosfdev2dev
>
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hi All,

I have tried the sample PySide code given at 
http://www.scipy.org/Cookbook/Matplotlib/PySide

the below excerpt is from my "Python Shell" for this sample code, which gives 
this error,

"ImportError: Warning: formlayout requires PyQt4 >v4.3" 

It seems that it is looking for PyQt; but the whole point of using PySide is to 
avoid using PyQt. 


Any help is appreciated.


-- Armagan


Python 2.7.2 (default, Jun 12 2011, 15:08:59) [MSC v.1500 32 bit (Intel)] on 
win32
Type "copyright", "credits" or "license()" for more information.
>>> import sys
>>> import matplotlib
>>> matplotlib.use('Qt4Agg')
>>> import pylab
Traceback (most recent call last):
  File "<pyshell#3>", line 1, in <module>
    import pylab
  File "C:\Python27\lib\site-packages\pylab.py", line 1, in <module>
    from matplotlib.pylab import *
  File "C:\Python27\lib\site-packages\matplotlib\pylab.py", line 263, in 
<module>
    from matplotlib.pyplot import *
  File "C:\Python27\lib\site-packages\matplotlib\pyplot.py", line 95, in 
<module>
    new_figure_manager, draw_if_interactive, show = pylab_setup()
  File "C:\Python27\lib\site-packages\matplotlib\backends\__init__.py", line 25, 
in pylab_setup
    globals(),locals(),[backend_name])
  File "C:\Python27\lib\site-packages\matplotlib\backends\backend_qt4agg.py", 
line 12, in <module>
    from backend_qt4 import QtCore, QtGui, FigureManagerQT, FigureCanvasQT,\
  File "C:\Python27\lib\site-packages\matplotlib\backends\backend_qt4.py", line 
18, in <module>
    import matplotlib.backends.qt4_editor.figureoptions as figureoptions
  File 
"C:\Python27\lib\site-packages\matplotlib\backends\qt4_editor\figureoptions.py", 
line 11, in <module>
    import matplotlib.backends.qt4_editor.formlayout as formlayout
  File 
"C:\Python27\lib\site-packages\matplotlib\backends\qt4_editor\formlayout.py", 
line 51, in <module>
    raise ImportError, "Warning: formlayout requires PyQt4 >v4.3"
ImportError: Warning: formlayout requires PyQt4 >v4.3 
 
-------------------------------------------------

Hi,

I am trying to plot some arrays with missing data, noted as -9999.9. I have
tried setting these values to NaN and using numpy masked arrays. Neither
produces the correct plot when using box plot. Any suggestions?

e.g.

import matplotlib.pyplot as plt
import numpy as np

# works fine
y = np.array([568., 576., 436.])
plt.boxplot(y, notch=0, sym='+', vert=1, whis=1.5)
plt.show()

# same array, but with some missing data
x = np.array([-9999.9, 568., -9999.9, -9999.9, 576., -9999.9, 436.])

# mark missing data
xx = np.where(x < -9000.0, np.nan, x)

# try and plot that, nope.
plt.boxplot(xx, notch=0, sym='+', vert=1, whis=1.5)
plt.show()

# mask the array, plot that?
xxx = np.ma.array(xx, mask=np.isnan(xx))
plt.boxplot(xxx, notch=0, sym='+', vert=1, whis=1.5)
plt.show()

thanks 

Martin
-- 
View this message in context: http://old.nabble.com/Box-plot-and-missing-data-%28NaN-Masked%29-tp32058883p32058883.html
Sent from the matplotlib - users mailing list archive at Nabble.com.