matplotlib-users Mailing List for matplotlib (Page 2)

Hi David,

The Mac OS X backend does not use Agg, so I am not surprised that this script won't work with Mac OS X backend. The script probably also won't work with other non-Agg backends such as GtkCairo.

--Michiel.

--- On Mon, 2/15/10, David Arnold <dwa...@su...> wrote:

> From: David Arnold <dwa...@su...>
> Subject: [Matplotlib-users] agg_buffer_to_array.py
> To: mat...@li...
> Date: Monday, February 15, 2010, 2:12 PM
> All,
> This error:The
> debugged program raised the exception unhandled AttributeError
> "'FigureCanvasMac' object
> has no attribute 'buffer_rgba'"
> File: /Users/darnold/Documents/temp/Matplotlib/PylabExamples/agg_buffer_to_array.py,
> Line: 16is raised
> by the following script on my Macbook.
> #
> agg_buffer_to_array.py import
> matplotlibmatplotlib.use('macosx')from
> pylab import figure, showimport numpy as
> np
> # make an agg figurefig =
> figure()ax =
> fig.add_subplot(111)ax.plot([1,2,3])ax.set_title('a
> simple figure')fig.canvas.draw()
> # grab rhe pixel buffer and dumpy it into a
> numpy arraybuf =
> fig.canvas.buffer_rgba(0,0)l, b, w, h =
> fig.bbox.boundsX = np.fromstring(buf,
> np.uint8)X.shape = h,w,4
> # now display the array X as an Axes in a new
> figurefig2 = figure()ax2 =
> fig2.add_subplot(111,
> frameon=False)ax2.imshow(X)show()
> This is captured from:  http://matplotlib.sourceforge.net/examples/pylab_examples/agg_buffer_to_array.html
> With:
> matplotlib.use('Agg')
> Nothing happens at all. With --verbose-helpful,
> yields the following:
> $HOME=/Users/darnoldCONFIGDIR=/Users/darnold/.matplotlibmatplotlib
> data path
> /Library/Frameworks/Python.framework/Versions/6.0.0/lib/python2.6/site-packages/matplotlib/mpl-dataloaded
> rc file
> /Users/darnold/.matplotlib/matplotlibrcmatplotlib
> version 0.99.1.1verbose.level
> helpfulinteractive is Falseunits is
> Falseplatform is darwinUsing
> fontManager instance from
> /Users/darnold/.matplotlib/fontList.cachebackend
> agg version v2.2findfont: Matching
> :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=medium
> to Bitstream Vera Sans
> (/Library/Frameworks/Python.framework/Versions/4.3.0/lib/python2.5/site-packages/matplotlib-0.98.5.2n2-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf)
> with score of 0.000000findfont: Matching
> :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=large
> to Bitstream Vera Sans
> (/Library/Frameworks/Python.framework/Versions/4.3.0/lib/python2.5/site-packages/matplotlib-0.98.5.2n2-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf)
> with score of 0.000000
> Both
> matplotlib.use('tkagg')
> and
> matplotlib.use('wxagg')
> work as they should.
> David.
> -----Inline Attachment Follows-----
> 
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features
> such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris
> 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev
> -----Inline Attachment Follows-----
> 
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 


      

Thanks! That was at least part of the problem.
I've committed a bug fix to the svn repository.

Thanks gain,

--Michiel.

--- On Mon, 2/15/10, John Hunter <jd...@gm...> wrote:

> From: John Hunter <jd...@gm...>
> Subject: Re: [Matplotlib-users] Enter Figure on Macs
> To: "Michiel de Hoon" <mjl...@ya...>
> Cc: "David Arnold" <dwa...@su...>, mat...@li...
> Date: Monday, February 15, 2010, 8:07 AM
> On Mon, Feb 15, 2010 at 6:50 AM,
> Michiel de Hoon <mjl...@ya...>
> wrote:
> > I almost have a solution for this for the Mac OS X
> backend. I am stuck though at what I should pass to
> enter_notify_event and leave_notify_event for the guiEvent:
> >
> >    def leave_notify_event(self, guiEvent=None):
> >        """
> >        Backend derived classes should call this
> function when leaving
> >        canvas
> >
> >        *guiEvent*
> >            the native UI event that generated
> the mpl event
> >
> >        """
> >
> > What are the requirements for guiEvent? If I call
> leave_notify_event without guiEvent, so guiEvent = None,
> then the example gives me the following error:
> 
> we don't make any assumptions about what kind of object the
> gui event
> is.  We provide the GUI event because sometimes when
> using a specific
> backend, the user wants to drill into the GUI native event
> (eg a
> button press event) but we don't use it anywhere in the mpl
> frontend
> because this would break the abstraction.  So if you
> have some event
> that is being fired at the UI level on figure enter, pass
> that in.
> 
> It looks like you may be having a problem because the
> leave_notify_event is getting called more than once, or is
> called for
> a figure that has not been entered.  Check the logic
> in
> backend_bases.FigureCanvasBase.leave_notify_event
> 
> 
>     def leave_notify_event(self, guiEvent=None):
>         """
>         Backend derived classes should
> call this function when leaving
>         canvas
> 
>         *guiEvent*
>             the native UI
> event that generated the mpl event
> 
>         """
>        
> self.callbacks.process('figure_leave_event',
> LocationEvent.lastevent)
>         LocationEvent.lastevent = None
> 
> 
> It looks like your figure_leave_event is being triggered
> with
> LocationEvent.lastevent = None (so it is not a problem with
> your
> guiEvent).  This could happen if a leave event was
> processed *before*
> and enter event (which sets the lastevent), or if a leave
> event was
> processed twice.
> 
> Hopefully this will help you drill down into the source of
> the problem
> 
> JDH
> 


      

Works pretty well.
I've now implemented something like

========================== *snip* ==========================
def find_associated_colorbar( obj ):
      for child in obj.get_children():
              try:
                      cbar = child.colorbar
              except AttributeError:
                      continue
              if not cbar == None: # really necessary?
                      # if fetch was successful, cbar contains
                      # ( reference to colorbar, reference to axis
containing colorbar )
                      return cbar[0]
      return None
========================== *snip* ==========================

How did you find out about the colormap attribute? Was that by taking
a good guess in looking at the source code, or are the public
attributes of a class documented?

Cheers,
Nico



On Tue, Feb 16, 2010 at 1:18 PM, Jae-Joon Lee <lee...@gm...> wrote:
> set_colorbar sets colorbar attribute. So I guess you can just check if
> Mappable.colorbar is None or not.
> Mappable.colorbar, when set, should be a tuple whose first item is an
> image for colorbar and the second item is an colorbar axes.
>
> Regards,
>
> -JJ
>
> On Tue, Feb 16, 2010 at 5:26 AM, Nico Schlömer <nic...@gm...> wrote:
>> Alright, so I dug the sources a bit and found the snippet
>>
>> ========================== *snip* ==========================
>>        cb = cbar.Colorbar(cax, mappable, **kw)
>>
>>        def on_changed(m):
>>            #print 'calling on changed', m.get_cmap().name
>>            cb.set_cmap(m.get_cmap())
>>            cb.set_clim(m.get_clim())
>>            cb.update_bruteforce(m)
>>
>>        self.cbid = mappable.callbacksSM.connect('changed', on_changed)
>>        mappable.set_colorbar(cb, cax)
>> ========================== *snap* ==========================
>>
>> I guess what happens is that a Colorbar is created, and a callback
>> function registered which adapts this very color bar whenever there is
>> a change in color maps/limits.
>> Well. -- I reckon that means that at the moment there's no way to tell
>> if a ScalarMappable has a color bar associated or not. :/ -- At least
>> I don't see how it would be possible to dig up on_changed( ) from the
>> list of callbacks and extract cb from it.
>>
>> Aaand everybody: "Fea-ture request, fea-ture request!"
>> get_colorbar() for ScalarMappables :)
>>
>> Cheers,
>> Nico
>>
>>
>>
>>
>> On Mon, Feb 15, 2010 at 7:58 PM, Nico Schlömer <nic...@gm...> wrote:
>>>> As far as I can see, it is the other way around, i.e., mappables
>>>> (e.g., images) know about the colorbar they are connected.
>>>
>>> Well yeah, that'd be even better. I'll check out the API. -- Hints
>>> would still be appreciated of course.
>>>
>>> --Nico
>>>
>>
>

 Thanks Ernest, I had no idea that the DateFormatter was going to be treated
as latex as well.

However, escaping the \ with another \ did not worked.
I tried:

majorF = DateFormatter("\\n \\n %b")

How should I escape the \n ?


> majorF = DateFormatter("\n \n %b") # problem
>                          ^^  ^^
> A common mistake.
> You forgot to escape the "\" characters.
>
> Bye.
>
> Ernest
>

set_colorbar sets colorbar attribute. So I guess you can just check if
Mappable.colorbar is None or not.
Mappable.colorbar, when set, should be a tuple whose first item is an
image for colorbar and the second item is an colorbar axes.

Regards,

-JJ

On Tue, Feb 16, 2010 at 5:26 AM, Nico Schlömer <nic...@gm...> wrote:
> Alright, so I dug the sources a bit and found the snippet
>
> ========================== *snip* ==========================
>        cb = cbar.Colorbar(cax, mappable, **kw)
>
>        def on_changed(m):
>            #print 'calling on changed', m.get_cmap().name
>            cb.set_cmap(m.get_cmap())
>            cb.set_clim(m.get_clim())
>            cb.update_bruteforce(m)
>
>        self.cbid = mappable.callbacksSM.connect('changed', on_changed)
>        mappable.set_colorbar(cb, cax)
> ========================== *snap* ==========================
>
> I guess what happens is that a Colorbar is created, and a callback
> function registered which adapts this very color bar whenever there is
> a change in color maps/limits.
> Well. -- I reckon that means that at the moment there's no way to tell
> if a ScalarMappable has a color bar associated or not. :/ -- At least
> I don't see how it would be possible to dig up on_changed( ) from the
> list of callbacks and extract cb from it.
>
> Aaand everybody: "Fea-ture request, fea-ture request!"
> get_colorbar() for ScalarMappables :)
>
> Cheers,
> Nico
>
>
>
>
> On Mon, Feb 15, 2010 at 7:58 PM, Nico Schlömer <nic...@gm...> wrote:
>>> As far as I can see, it is the other way around, i.e., mappables
>>> (e.g., images) know about the colorbar they are connected.
>>
>> Well yeah, that'd be even better. I'll check out the API. -- Hints
>> would still be appreciated of course.
>>
>> --Nico
>>
>

15/02/10 @ 19:22 (-0500), thus spake Filipe Pires Alvarenga Fernandes:
> Hello list,
> 
> If I use DateFormatter with latex and lines breaks like this
> >>> DateFormatter("\n \n %b") I get an latex error:
> 
> http://pastebin.com/m5b186ded
> 
> Although, if I do not use the line breaks,
> >>> DateFormatter("%b")
> The problem disappears.
> 
> Below is a script that reproduces what I'm talking about:
> 
> #Example:
> from pylab            import *
> from matplotlib       import rcParams
> rcParams['text.usetex'] = True
> fig = figure()
> ax  = fig.add_subplot(111)
> fig.subplots_adjust(bottom=0.2)
> t = arange(100,110,.1)
> u = sin(t)
> v = cos(t)
> quiver([t],[[0]*len(t)],u,v)
> major  = DayLocator([10,04])
> minor  = DayLocator()
> majorF = DateFormatter("\n \n %b") # problem
                         ^^  ^^
A common mistake.
You forgot to escape the "\" characters.

Bye.

Ernest

Alright, so I dug the sources a bit and found the snippet

========================== *snip* ==========================
        cb = cbar.Colorbar(cax, mappable, **kw)

        def on_changed(m):
            #print 'calling on changed', m.get_cmap().name
            cb.set_cmap(m.get_cmap())
            cb.set_clim(m.get_clim())
            cb.update_bruteforce(m)

        self.cbid = mappable.callbacksSM.connect('changed', on_changed)
        mappable.set_colorbar(cb, cax)
========================== *snap* ==========================

I guess what happens is that a Colorbar is created, and a callback
function registered which adapts this very color bar whenever there is
a change in color maps/limits.
Well. -- I reckon that means that at the moment there's no way to tell
if a ScalarMappable has a color bar associated or not. :/ -- At least
I don't see how it would be possible to dig up on_changed( ) from the
list of callbacks and extract cb from it.

Aaand everybody: "Fea-ture request, fea-ture request!"
get_colorbar() for ScalarMappables :)

Cheers,
Nico




On Mon, Feb 15, 2010 at 7:58 PM, Nico Schlömer <nic...@gm...> wrote:
>> As far as I can see, it is the other way around, i.e., mappables
>> (e.g., images) know about the colorbar they are connected.
>
> Well yeah, that'd be even better. I'll check out the API. -- Hints
> would still be appreciated of course.
>
> --Nico
>

When I come back tonight I will try to fix the errors for you.

If you are interested in contributing to matplotlib check out this link:

http://matplotlib.sourceforge.net/faq/howto_faq.html#contributing-howto

The PDF is not the "original source", it's a product of the sphinx 
documentation system I think so contributing happens in simple plain text 
files. You can send a patch then.

Regards,
Philipp

Hi,

I'm trying to grid irregularly spaced data, such that the convex hull
of the data is not rectangular.  Specifically, all my data lies in an
equilateral triangle inside the unit circle.  I found:

     http://www.scipy.org/Cookbook/Matplotlib/Gridding_irregularly_spaced_data

and tried the suggested technique.  For my grid, I made a square of
the min and max of my data.  However, it had problems:

...
  File "/home/guest/.local/lib/python2.6/site-packages/matplotlib/delaunay/triangulate.py",
line 125, in _compute_convex_hull
    hull.append(edges.pop(hull[-1]))
KeyError: 0


Should I expect matplotlib.mlab.griddata to work with a dataset like
this?  I know that I can use hexbin, but it'd be really nice to see
contours explicitly.

<PH...@Ge...> wrote:

> ...maybe dividing the markers up into 2, 3, or 4 sections would be  
> useful too.
> ...

There's a gallery example doing that in general, making pie-charts out  
of the markers:

http://matplotlib.sourceforge.net/examples/api/scatter_piecharts.html

although I think my demo of it shows off its data-representation better:

# Piechart markers from matplotlib gallery, thanks to Manuel Metz for  
the original example
# CPHLewis, 2010.

import math
import numpy as np
import matplotlib.pyplot as plt

def payoff(x,y):
     return x*y*400

def outcome_xwins(x,y):
     return x/(x+y)

def outcome_ywins(x,y):
     return y/(x+y)

x_cases = [ .25, .5, .75]
y_cases = [ .33, .5, .66]
outcomes = [('x wins', outcome_xwins, 'blue'),
             ('y wins', outcome_ywins, 'green')] #the name,  
calculation, and plotting color for categories of outcome

fig = plt.figure()
ax = fig.add_subplot(111)
ax.set_title('Small multiples: pie charts calculated based on (x,y)')
legend_once = True
#At each point in the plot we calculate everything about the outcomes.
for x in x_cases:
     for y in y_cases:
         size = payoff(x,y)
         start_at = 0
         for result in outcomes:
             result_share = result[1](x,y)
             xpt = [0] + np.cos(np.linspace(2*math.pi*start_at,  
2*math.pi*(result_share+start_at), 10)).tolist()
             ypt = [0] + np.sin(np.linspace(2*math.pi*start_at,  
2*math.pi*(result_share+start_at), 10)).tolist()
             xypt = zip(xpt, ypt)
             ax.scatter([x],[y], marker = (xypt, 0), s = size,  
facecolor = result[2], label=result[0])
             start_at = start_at + result_share
         if legend_once: ax.legend() #don't know why this isn't  
picking up the labels.
         legend_once = False
plt.show()

On Mon, Feb 15, 2010 at 5:22 PM, John Hunter <jd...@gm...> wrote:
> Very nice and thorough work.  I think this should be included, but
> I'll wait to hear from other developers before committing.  Could you
> confirm that the unit tests pass?
>
>>>> import matplotlib
>>>> matplotlib.test()
>

Confirmed on rev 8133:

Ran 124 tests in 341.585s

FAILED (KNOWNFAIL=2, errors=2)

and the errors were something to do with hexbin extents and the figimage method.


> I think the markerangle would also be a useful contribution, though it
> would render some of the markers redundant (eg triangle left, right,
> etc, would all just be triangles with different angles...)
>

That was a concern I had as well, but I suppose > ^ v < (etc) could
just be considered shortcuts to particular angles.  Presumably, we
would not be removing them.  Correct?  Also, is the standard to have
the angle specified in degrees?  So what is more useful:  markerangle
or markerdeg?

The other difference is that when one specifies fillstyle='left', then
it would only apply to the marker at 0 degrees.  Whereas, marker='v',
fillstyle='left', markerangle=0  would correspond to marker='^',
fillstyle='right', markerangle=180   (or something like that).

On Mon, Feb 15, 2010 at 5:59 PM, T J <tj...@gm...> wrote:
> On Sun, Feb 14, 2010 at 2:49 PM, T J <tj...@gm...> wrote:
>> I ran across:
>>
>>    http://old.nabble.com/half-filled-markers-td24003576.html
>>
>> The name "fillstyle" can give the wrong impression about what is being
>> filled.  For example, see the comment here:
>>
>>   http://www.mail-archive.com/mat...@li.../msg13074.html
>>
>> It's probably too late, but would "markerfillstyle" be a better name for this?
>>
>> Also, the current implementation fills half of the marker with the
>> markerfacecolor and doesn't fill the marker at all for the other half.
>>  I think a neat (and simple) feature would be for users to specify two
>> colors.  Perhaps 'markerfacecolor2'.   The change to the code is
>> minimal, but the functionality it brings is quite flexible.
>> markerfacecolor2 can default to 'none' to maintain current
>> functionality.
>>
>> Should I file a ticket for this?



>>
>
> I went ahead and implemented this.  The user can now specify
> 'markercoloralt'.  In the process, I finished the "half-marker" code
> for all remaining filled_markers.   The diff is attached, which also
> includes a fix for bug #560720:
>
>    https://sourceforge.net/tracker/?func=detail&aid=2952236&group_id=80706&atid=560720
>
> Please comment.  If there are no comments in a few days, I'll file a ticket.
>
> A picture demonstrating this functionality is attached.  Demos for all
> filled markers can be obtained here:
>
>    http://www.filedropper.com/filledmarkers
>
> Note, the diff contains the script used to generate the pictures. More
> generally, I wonder if there should be a 'markerangle' keyword. I can
> probably push through and implement this, but I'd like to hear what
> people think about it.

Very nice and thorough work.  I think this should be included, but
I'll wait to hear from other developers before committing.  Could you
confirm that the unit tests pass?

>>> import matplotlib
>>> matplotlib.test()

I think the markerangle would also be a useful contribution, though it
would render some of the markers redundant (eg triangle left, right,
etc, would all just be triangles with different angles...)

Many thanks,
JDH

Hello list,

If I use DateFormatter with latex and lines breaks like this
>>> DateFormatter("\n \n %b") I get an latex error:

http://pastebin.com/m5b186ded

Although, if I do not use the line breaks,
>>> DateFormatter("%b")
The problem disappears.

Below is a script that reproduces what I'm talking about:

#Example:
from pylab            import *
from matplotlib       import rcParams
rcParams['text.usetex'] = True
fig = figure()
ax  = fig.add_subplot(111)
fig.subplots_adjust(bottom=0.2)
t = arange(100,110,.1)
u = sin(t)
v = cos(t)
quiver([t],[[0]*len(t)],u,v)
major  = DayLocator([10,04])
minor  = DayLocator()
majorF = DateFormatter("\n \n %b") # problem
#majorF = DateFormatter("%b") # no problem
minorF = DateFormatter('%d')
ax.xaxis.set_major_locator(major)
ax.xaxis.set_minor_locator(minor)
ax.xaxis.set_major_formatter(majorF)
ax.xaxis.set_minor_formatter(minorF)
gca().xaxis_date()
title(r"$\alpha \beta \gamma \delta$")
show()


I'm using the latest svn version with qt4 as backend.

Thanks, Filipe

I'm curious if anyone knows a good way to embed pydot
(http://code.google.com/p/pydot/) graphs (or really, any
graphviz-style graphs) inside matplotlib somehow.  I could easily
write out a png or something from pydot and then imshow it, but that
seems very kludgy.  Is there some way to load svg or other vector data
into matplotlib to be shown inside a figure?

TJ,

I think this current implementation and adding a marker rotation key word would be wonderful. When I get stuck doing GIS work, I end up using split markers very often. It's quite useful.

Now's probably not the time, but maybe dividing the markers up into 2, 3, or 4 sections would be useful too. 
But perhaps at that point it'd be easier to create half and quarter markers and then rotate accordingly as the user makes multiple calls to plot. I dunno...just thinking aloud.

This is a great contribution. Thanks a lot!
-paul

> -----Original Message-----
> From: T J [mailto:tj...@gm...]
> Sent: Monday, February 15, 2010 4:00 PM
> To: Matplotlib Users
> Subject: Re: [Matplotlib-users] half-filled markers, two-colors
> 
> On Sun, Feb 14, 2010 at 2:49 PM, T J <tj...@gm...> wrote:
> > I ran across:
> >
> >    http://old.nabble.com/half-filled-markers-td24003576.html
> >
> > The name "fillstyle" can give the wrong impression about what is being
> > filled.  For example, see the comment here:
> >
> >   http://www.mail-archive.com/matplotlib-
> us...@li.../msg13074.html
> >
> > It's probably too late, but would "markerfillstyle" be a better name
> for this?
> >
> > Also, the current implementation fills half of the marker with the
> > markerfacecolor and doesn't fill the marker at all for the other half.
> >  I think a neat (and simple) feature would be for users to specify two
> > colors.  Perhaps 'markerfacecolor2'.   The change to the code is
> > minimal, but the functionality it brings is quite flexible.
> > markerfacecolor2 can default to 'none' to maintain current
> > functionality.
> >
> > Should I file a ticket for this?
> >
> 
> I went ahead and implemented this.  The user can now specify
> 'markercoloralt'.  In the process, I finished the "half-marker" code
> for all remaining filled_markers.   The diff is attached, which also
> includes a fix for bug #560720:
> 
> 
> https://sourceforge.net/tracker/?func=detail&aid=2952236&group_id=80706&a
> tid=560720
> 
> Please comment.  If there are no comments in a few days, I'll file a
> ticket.
> 
> A picture demonstrating this functionality is attached.  Demos for all
> filled markers can be obtained here:
> 
>     http://www.filedropper.com/filledmarkers
> 
> Note, the diff contains the script used to generate the pictures. More
> generally, I wonder if there should be a 'markerangle' keyword. I can
> probably push through and implement this, but I'd like to hear what
> people think about it.