matplotlib-users Mailing List for matplotlib

Michael Droettboom wrote:
> All that said, you could experiment with rendering to a vector format 
> (e.g. Pdf or Svg) and then have wxPython render it directly with each 
> window resize. 

If you did want to go that route, someone wrote a SVG renderer for 
wxPython -- search the wxpython-users list to find it.

I do question why you want this -- MPLs current behavior generally makes 
the most sense -- you generally want the same font sizes and line 
thicknesses in a GUI app regardless of how large your Window is.

Unless you really need it to look that same as it would on paper, for 
instance.

-Chris




-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chr...@no...

Jurgen,

I generally lurk on the list, but I'll speak up now since I have 
basically done what you are talking about.  I integrated MPL (TkAgg) 
into an existing application that resizes (PyRAF).

The path I chose isn't particularly pretty, but it does the job and 
sounds like it might be flexible enough for what you are doing.  I 
created an adapter class which wraps the MPL calls.  The adapter is what 
the existing application calls to make plots instead of calling MPL 
directly.  The adapter also keeps a normalized (0-1) copy of all plot 
data arrays (and any lines and other items).  When the canvas is 
resized, I go through all my lines (Line2D), patches (Rectangle), and 
text, and scale each based on the new canvas dimensions.  Then the new 
items are inserted directly into the Figure object, replacing the old 
values.  This resizes everything so that it is correct when redrawn.  I 
was initially concerned about performance, but the resizings are very 
smooth (thanks matplotlib!).  You could also enforce an aspect ratio 
this way.

I had to write this adapter not because of the resizing issue, but due 
to the constraints of the existing application with which I was 
working.  It did however make resizing easy.  As long as your plotted 
items are relatively simple (lines, patches, text), this works very 
well.  Using more complicated items like legends would not work as well 
unless you either allowed them to stay a static size or manually 
composed them of simpler objects.  In general, this idea is probably 
more work than John's DPI suggestions.

Chris

Doh!  Why didn't I think of that?

Mike

John Hunter wrote:
> On Jan 14, 2008 3:55 PM, Michael Droettboom <md...@st...> wrote:
> 
>> I think what you're asking for would require a pretty major overhaul of
>> matplotlib.  The fact that all the text etc. remains the same size is a
> 
> matplotlib is designed to scale in the way Jurgen requests with DPI
> (eg fonts, line thicknesses and the like scale with DPI).
> 
> In [65]: fig = figure()
> 
> In [66]: plot([1,2,3])
> Out[66]: [<matplotlib.lines.Line2D instance at 0xbf965ec>]
> 
> In [67]: fig.dpi.set(200)
> 
> In [68]: fig.canvas.draw()
> 
> with a little work, one could hook into the resize mechanism to
> increase the dpi to create the desired effect w/o a major overhaul.
> By default what happens is the width and height in inches are changed
> with a resize but the dpi is held constant.  One could trick
> matplotlib by computing a new width, height in inches, and a new dpi
> so that the resized canvas width in pixels is the requested size based
> on the resize event but the dpi is increased to create the microcal
> effect
> 
> JDH

-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

On Jan 14, 2008 3:55 PM, Michael Droettboom <md...@st...> wrote:

> I think what you're asking for would require a pretty major overhaul of
> matplotlib.  The fact that all the text etc. remains the same size is a

matplotlib is designed to scale in the way Jurgen requests with DPI
(eg fonts, line thicknesses and the like scale with DPI).

In [65]: fig = figure()

In [66]: plot([1,2,3])
Out[66]: [<matplotlib.lines.Line2D instance at 0xbf965ec>]

In [67]: fig.dpi.set(200)

In [68]: fig.canvas.draw()

with a little work, one could hook into the resize mechanism to
increase the dpi to create the desired effect w/o a major overhaul.
By default what happens is the width and height in inches are changed
with a resize but the dpi is held constant.  One could trick
matplotlib by computing a new width, height in inches, and a new dpi
so that the resized canvas width in pixels is the requested size based
on the resize event but the dpi is increased to create the microcal
effect

JDH

DaFudl wrote:
> I integrated mpl into my wxPython application for plotting simple lines.
> I would like my canvas to behave differently when the containing window is
> being resized.
> At the moment all text, all lines, all markers, the legend... remain in the
> same state (same size and thickness) regardless whether the plot is small or
> big. As a result the plot looks fine at one size only.
> I need all elements to be resized with the window just like it happens in
> Microcal's scientific plotting software "origin". Everything should resize
> as a "whole".

I think what you're asking for would require a pretty major overhaul of 
matplotlib.  The fact that all the text etc. remains the same size is a 
deliberate design decision, and actually requires a lot more work to 
make fast enough, since a lot more must be recalculated with each 
redraw, rather than just multiplying everything by a scale factor.

All that said, you could experiment with rendering to a vector format 
(e.g. Pdf or Svg) and then have wxPython render it directly with each 
window resize.  (This would be much the same way that resizing a PDF 
generated by matplotlib in Acrobat Reader scales everything as a whole).

Alternatively, you could try using a backend (e.g. Cairo) that allows 
you to push a global affine transformation onto the transformation 
stack.  (At present, the Agg backends don't have that functionality -- 
but it could be added, especially since the recent overhaul of 
transformations in general in matplotlib.)  That way, you could tell the 
underlying rendering engine (below matplotlib) to scale everything up 
and down for any subsequent commands it receives from matplotlib.

But, anyway you slice it, I think you're looking at getting your hands 
dirty in some matplotlib internals.  I'm happy to help if you want to 
work through it, but I don't think I would have time myself.

> Also, it would be great to keep the aspect ratio of the plot when resizing
> and zooming. I could not figure out how to do that. Using methods like
> apply_aspect and set_adjustable is not enough.

That seems to be easy enough to do in the window resize callback -- just 
programmatically force-resize the window to the desired aspect ratio.

Cheers,
Mike

-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

I integrated mpl into my wxPython application for plotting simple lines.
I would like my canvas to behave differently when the containing window is
being resized.
At the moment all text, all lines, all markers, the legend... remain in the
same state (same size and thickness) regardless whether the plot is small or
big. As a result the plot looks fine at one size only.
I need all elements to be resized with the window just like it happens in
Microcal's scientific plotting software "origin". Everything should resize
as a "whole".
Also, it would be great to keep the aspect ratio of the plot when resizing
and zooming. I could not figure out how to do that. Using methods like
apply_aspect and set_adjustable is not enough.

Is there a way to make those things work?

Thanks,
Jurgen
-- 
View this message in context: http://www.nabble.com/resizing-behaviour-like-in-Mircrocal-Origin-tp14814996p14814996.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

>

I have some code that produces a series of graphs of data over time.   
For the most part it works well, but for certain combinations of  
dates the two plots do not line up, due to the width of the bars in  
the first subplot.  What is the best way for force my second  
subplot's x axis to be identical to that of the first subplot?

Thanks!

Here is minimal code reproducing the problem:

import matplotlib
from datetime import *
import time
from pylab import *


def toOrd(strDate):
	return datetime(*time.strptime(strDate,"%m/%d/%y")[0:5]).toordinal()

def main():
	
	startDate = toOrd("05/24/05")
	endDate = toOrd("11/21/06")
	
	dates = range(startDate,endDate,7)
	figure(1)
	
	axis = subplot(211,axisbelow=True)
	plot_date(dates,dates,visible = False)
	axis.bar(dates,dates, width=3.0)
	
	axis = subplot(212,axisbelow=True)
	plot_date(dates,dates) #,'-',color = priceColor)
	
	show()

	

if __name__ == '__main__':
	main()

j. vickroy wrote:
> Hello, I am a first-time user of matplotlib.
>
> When trying to run the plotmap.py example, the following traceback is 
> generated:
>
>  >plotmap.py
> Traceback (most recent call last):
>  File "C:\Documents and Settings\jim.vickroy\My 
> Documents\Projects\High-latitudes D-region\plotmap.py", line 13, in 
> <module>
>    topoin = load('etopo20data.gz')
>  File "C:\Python25\Lib\site-packages\matplotlib\mlab.py", line 1252, in 
> load
>    fh = cbook.to_filehandle(fname)
>  File "C:\Python25\Lib\site-packages\matplotlib\cbook.py", line 236, in 
> to_filehandle
>    fh = gzip.open(fname, flag)
>  File "C:\Python25\lib\gzip.py", line 49, in open
>    return GzipFile(filename, mode, compresslevel)
>  File "C:\Python25\lib\gzip.py", line 95, in __init__
>    fileobj = self.myfileobj = __builtin__.open(filename, mode or 'rb')
> IOError: [Errno 2] No such file or directory: 'etopo20data.gz'
>
> How do I obtain this file and where should it reside?
>
> I have downloaded proj446_win32_bin.zip and unzipped it, but I do not 
> see any such file.
>
> I have also attempted to search this user-group archive for references 
> to "etopo20data.gz" but repeatedly get an "Unable to connect to Search 
> Server" error so I do not know what I'm doing incorrectly.
>
> I have installed:
> - matplotlib-0.91.2.win32-py2.5.exe
> - basemap-0.9.9.1.win32-py2.5.exe
> - httplib2-0.4.0.zip
> for use with Python 2.5.1.
>
> I apologize if this has been previously addressed.
>
> Thanks,
> -- jv
>   

It's in the examples directory of the basemap source tarball 
(basemap-0.9.9.1.tar.gz).  The examples directory is not included with 
the binary installers.

-Jeff

-- 
Jeffrey S. Whitaker         Phone  : (303)497-6313
Meteorologist               FAX    : (303)497-6449
NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
325 Broadway                Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg

Hello, I am a first-time user of matplotlib.

When trying to run the plotmap.py example, the following traceback is 
generated:

 >plotmap.py
Traceback (most recent call last):
 File "C:\Documents and Settings\jim.vickroy\My 
Documents\Projects\High-latitudes D-region\plotmap.py", line 13, in 
<module>
   topoin = load('etopo20data.gz')
 File "C:\Python25\Lib\site-packages\matplotlib\mlab.py", line 1252, in 
load
   fh = cbook.to_filehandle(fname)
 File "C:\Python25\Lib\site-packages\matplotlib\cbook.py", line 236, in 
to_filehandle
   fh = gzip.open(fname, flag)
 File "C:\Python25\lib\gzip.py", line 49, in open
   return GzipFile(filename, mode, compresslevel)
 File "C:\Python25\lib\gzip.py", line 95, in __init__
   fileobj = self.myfileobj = __builtin__.open(filename, mode or 'rb')
IOError: [Errno 2] No such file or directory: 'etopo20data.gz'

How do I obtain this file and where should it reside?

I have downloaded proj446_win32_bin.zip and unzipped it, but I do not 
see any such file.

I have also attempted to search this user-group archive for references 
to "etopo20data.gz" but repeatedly get an "Unable to connect to Search 
Server" error so I do not know what I'm doing incorrectly.

I have installed:
- matplotlib-0.91.2.win32-py2.5.exe
- basemap-0.9.9.1.win32-py2.5.exe
- httplib2-0.4.0.zip
for use with Python 2.5.1.

I apologize if this has been previously addressed.

Thanks,
-- jv

Hi folks,

I have encountered a problem using colorbar and set_under with a color 
map that has only 32 colors.  The downward arrow is not filled with the 
correct color.  If 64 colors are used, the arrow is filled correctly.  I 
actually want 8 colors.

The test code is essentially image_masked.py from the 0.90.0 examples.

if I use: palette = cm.get_cmap("gray", 64)
the plot is OK.
if I use: palette = cm.get_cmap("gray", 32)
it is not OK.

In both cases, the upper arrow is the correct color.

I am using matplotlib version 0.90.0.


Here is the complete code:
#!/usr/bin/env python
'''imshow with masked array input and out-of-range colors.

'''

from pylab import *
import matplotlib.numerix.ma as ma
import matplotlib.colors as colors

delta = 0.025
x = y = arange(-3.0, 3.0, delta)
X, Y = meshgrid(x, y)
Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0)
Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1)
Z = 10 * (Z2-Z1)  # difference of Gaussians

# Set up a colormap:
#palette = cm.gray	# OK
palette = cm.get_cmap("gray",32)	# Not OK
#palette = cm.get_cmap("gray",64)	# OK
palette.set_over('r', 1.0)
palette.set_under('g', 1.0)
palette.set_bad('b', 1.0)
# Alternatively, we could use
# palette.set_bad(alpha = 0.0)
# to make the bad region transparent.  This is the default.
# If you comment out all the palette.set* lines, you will see
# all the defaults; under and over will be colored with the
# first and last colors in the palette, respectively.
Zm = ma.masked_where(Z > 1.2, Z)

# By setting vmin and vmax in the norm, we establish the
# range to which the regular palette color scale is applied.
# Anything above that range is colored based on palette.set_over, etc.

im = imshow(Zm, interpolation='bilinear',
     cmap=palette,
     norm = colors.Normalize(vmin = -1.0, vmax = 1.0, clip = False),
     origin='lower', extent=[-3,3,-3,3])
title('Green=low, Red=high, Blue=bad')
# extend choices are "both", "min", "max" and "neither"
colorbar(im, extend='both', shrink=0.8)
show()
-- 
Ms. Carol A. Leger
SRI International			Phone: (650) 859-4114
333 Ravenswood Avenue G-273
Menlo Park, CA 94025                    e-mail: le...@sr...

   Hi,

I need to make some Weibull analysis and I wanted to make it with numpy and
scipy.
Theses analysis  are  based on  Weibull plots :
   - on X, time values are reported according to a log scale
   - on Y, probability (F) is reported according to a  special scale, which
lets Weibull random variables fit on a line.
So the pair (t, F) should be plotted on ( ln(t),  ln( -ln(1-F)) )

You can find an example of weibull plot here :
http://www.itl.nist.gov/div898/handbook/eda/section3/weibplot.htm

Here is my way to  make weibull plot with matplotlib :

[code]

from numpy import *
from matplotlib.ticker import FuncFormatter
import pylab as p

# I'm used to  the ln notation for the natural log
from numpy import log as ln

# Paramters
beta = 5.2
eta = 12

# Genrate 10 numbers following a Weibull distribution
x = eta *random.weibull(beta, size=10)
F = 1 - exp( -(x/eta)**beta )

# Estimate Weibull parameters
lnX = ln(x)
lnF = ln( -ln(1-F) )
a, b = polyfit(lnF, lnX, 1)
beta0 = 1/a
eta0  = exp(b)

# ideal line
F0 = array([1e-3, 1-1e-3])
x0 = eta0 * (-ln(1-F0))**(1/beta0)
lnF0 = ln(-ln(1-F0))


# Weibull plot
p.figure()
ax = p.subplot(111)
p.semilogx(x, lnF, "bs")
p.plot(x0, lnF0, 'r-', label="beta= %5G\neta = %.5G" % (beta0, eta0) )
p.grid()
p.xlabel('x')
p.ylabel('Cumulative Distribution Function')
p.legend(loc='lower right')

# ticks
def weibull_CDF(y, pos):
    return "%G %%" % (100*(1-exp(-exp(y))))

formatter = FuncFormatter(weibull_CDF)
ax.yaxis.set_major_formatter(formatter)

yt_F = array([ 0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
           0.6, 0.7, 0.8, 0.9, 0.95, 0.99])
yt_lnF = ln( -ln(1-yt_F))
p.yticks(yt_lnF)

p.show()

[/code]

This works, but I 'm pretty sure this could be improved :

1)  Is there a way to use matplotlib.transforms to handle the convertion
between F and y=ln(-ln(1-F)) ?

2) I would like to have vertical lines for major and minor ticks. How can I
do that ?

3) In this case, the log notation for the X axis is not pertinent. How could
I switch to a more common floating point notation ?

Any hints ?

Thanks

-- 
LB

On Monday 14 January 2008 02:27:28 am Tom Johnson wrote:
> I was looking at rc_params() and saw
>
>     if not os.path.exists(fname):
>         message = 'could not find rc file; returning defaults'
>         ret =  dict([ (key, tup[0]) for key, tup in defaultParams.items()])
>         warnings.warn(message)
>         return ret
>
> Is this correct?  It seems that it returns a regular dictionary rather
> than an instance of RcParams.

You have a sharp eye. It is incorrect, but it turns out that block of code is 
never executed (unless the user deletes the default rc file from 
site-packages/matplotlib/mpl-data). But I changed it in the trunk to return 
an RcParams object. Thanks for the report.

Darren 

I've committed these changes to matplotlib SVN, and it should make it 
into the next release of 0.91.x, if we decide to make one.

You may be right that Qt's SVG support is iffy (I don't know enough 
about the spec to be sure, so I'm not conceding that), but either way it 
doesn't bother me to make changes that help SVG work in more places. 
It's like all those pragmatic web developers who have to make things 
work with IE...

Thanks for your help.  I'll try to get the Qt SVG demo installed here so 
I can test with that the next time our SVG code changes.

Cheers,
Mike

Christiaan Putter wrote:
> Hi Michael,
> 
> Sorry for getting back to you only now, and thanks for the help.
> 
> If you have Qt installed there should be an sample Svg viewer app in the 
> examples somewhere.  If you'd like I can send it to you somehow, just 
> let me know.
> 
> I tested the new .svg you sent me it's rendering the same in Firefox and 
> Qt.  So it works.
> 
> Once again thanks for your help and let me know if I can send you 
> something to help with testing in Qt.  Personally I think Qt's support 
> for svg is still a bit iffy.
> 
> Have a nice,
> 
> Christiaan
> 
> On 04/01/2008, *Michael Droettboom* <md...@st... 
> <mailto:md...@st...>> wrote:
> 
>     Can you please test the attached SVG file?
> 
>     Cheers,
>     Mike
> 
>     Michael Droettboom wrote:
>      > Between 0.90 and 0.91, the SVG backend was changed to store the glyph
>      > outlines of the characters in the SVG file itself.  (This is on by
>      > default, but can be turned off with the rc parameter
>      > svg.embed_char_paths).  This helps make the SVG files much more
>      > portable, as the need to install the math fonts has long been a
>     FAQ on
>      > this list.  I've been doing all my testing with Firefox and
>     Inkscape.
>      > Is there a simple Qt-based SVG viewer I could add to my testing
>     regimen?
>      >
>      > I won't pretend to be an expert on the SVG spec, but it does say
>     this:
>      >
>      > < http://www.w3.org/TR/SVG/struct.html#Head>
>      >
>      > ====
>      >
>      > To provide some SVG user agents with an opportunity to implement
>      > efficient implementations in streaming environments, creators of SVG
>      > content are encouraged to place all elements which are targets of
>     local
>      > URI references within a 'defs' element which is a direct child of
>     one of
>      > the ancestors of the referencing element. For example:
>      >
>      > <?xml version="1.0" standalone="no"?>
>      > <!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
>      > " http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
>      > <svg width="8cm" height="3cm"
>      >       xmlns="http://www.w3.org/2000/svg">
>      >    <desc>Local URI references within ancestor's 'defs'
>     element.</desc>
>      >    <defs>
>      >      <linearGradient id="Gradient01">
>      >        <stop offset="20%" stop-color="#39F" />
>      >        <stop offset="90%" stop-color="#F3F" />
>      >      </linearGradient>
>      >    </defs>
>      >    <rect x="1cm" y="1cm" width="6cm" height="1cm"
>      >          fill="url(#Gradient01)"  />
>      >    <!-- Show outline of canvas using 'rect' element -->
>      >    <rect x=".01cm" y=".01cm" width="7.98cm" height=" 2.98cm"
>      >          fill="none" stroke="blue" stroke-width=".02cm" />
>      > </svg>
>      >
>      > View this example as SVG (SVG-enabled browsers only)
>      >
>      > In the document above, the linear gradient is defined within a
>     'defs'
>      > element which is the direct child of the 'svg' element, which in
>     turn is
>      > an ancestor of the 'rect' element which references the linear
>     gradient.
>      > Thus, the above document conforms to the guideline.
>      >
>      > ====
>      >
>      > So we are complying to that part of the spec.  The spec doesn't
>     seem to
>      > say anything about whether the defs must appear before or after their
>      > use -- but maybe I just can't find the relevant paragraph.
>      >
>      > In any case, this should be easy enough to fix on matplotlib's
>     end, and
>      > certainly won't break compliance with the spec.  I'll have a
>     look, and
>      > may come back to you for help with testing with Qt if you don't
>     mind.
>      >
>      > Cheers,
>      > Mike
>      >
>      > Christiaan Putter wrote:
>      >> Hi there,
>      >>
>      >> Just an update regarding the svg problem I was having:
>      >>
>      >> I simply went back to 0.90 and that's working now.
>      >>
>      >> Would still be nice to know if the svg output from matplotlib
>     complies
>      >> with the standard or whether it's Qt that's messing things up.
>      >>
>      >> Merry x-mass!
>      >>
>      >> cputter
>      >>
>      >>
>      >>
>      >> On 21/12/2007, *Christiaan Putter* <cep...@go...
>     <mailto:cep...@go...>
>      >> <mailto: cep...@go...
>     <mailto:cep...@go...>>> wrote:
>      >>
>      >>     Hi guys and girls,
>      >>
>      >>     Quick question regarding matplotlib's svg backend...
>      >>
>      >>     I've embeded pyhton into c++ and Qt's (4.3.3) svg support is
>     having
>      >>     some problems with .svg files created by matplotlib.  Text isn't
>      >>     showing up.  Firefox displays the same .svg file correctly
>     though...
>      >>
>      >>     The problem:
>      >>
>      >>     It seems some text stuff is stored in a section called defs
>     at the
>      >>     end of the file with stuff linking to this earlier in the
>     file.  Qt
>      >>     doesn't like that and only displays the normal plot stuff
>     (lines,
>      >>     etc.) but not the labels and other text.
>      >>
>      >>     More exact:
>      >>
>      >>     <use xlink:href="#c_7" .....
>      >>
>      >>     references
>      >>
>      >>     <path id="c_7" d="M10.6875 .....
>      >>
>      >>     at the end of the file.
>      >>
>      >>
>      >>
>      >>     When I simply cut and paste the defs section to the
>     beginning of the
>      >>     file it solves the problem.
>      >>
>      >>     What does the svg standard say about this?  I assume Qt's
>      >>     implementation is defect...
>      >>
>      >>     Has anybody else encountered this problem?  I'll send an
>     e-mail to
>      >>     Qt too and ask them about it.
>      >>
>      >>     Hope you're all having a nice day.
>      >>
>      >>     Regards,
>      >>     cputter
>      >>
>      >>
>      >>
>      >>    
>     -------------------------------------------------------------------------
> 
>      >>     This SF.net email is sponsored by: Microsoft
>      >>     Defy all challenges. Microsoft(R) Visual Studio 2005.
>      >>     http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>     <http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/>
>      >>     _______________________________________________
>      >>     Matplotlib-users mailing list
>      >>     Mat...@li...
>     <mailto:Mat...@li...>
>      >>     <mailto:Mat...@li...
>     <mailto:Mat...@li...>>
>      >>     https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>      >>     <https://lists.sourceforge.net/lists/listinfo/matplotlib-users >
>      >>
>      >>
>      >>
>      >>
>     ------------------------------------------------------------------------
>      >>
>      >>
>     -------------------------------------------------------------------------
> 
>      >> This SF.net email is sponsored by: Microsoft
>      >> Defy all challenges. Microsoft(R) Visual Studio 2005.
>      >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>     <http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/>
>      >>
>      >>
>      >>
>     ------------------------------------------------------------------------
>      >>
>      >> _______________________________________________
>      >> Matplotlib-users mailing list
>      >> Mat...@li...
>     <mailto:Mat...@li...>
>      >> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>     <https://lists.sourceforge.net/lists/listinfo/matplotlib-users>
>      >
> 
>     --
>     Michael Droettboom
>     Science Software Branch
>     Operations and Engineering Division
>     Space Telescope Science Institute
>     Operated by AURA for NASA
> 
> 

-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

On Thu, 2008-01-10 at 14:10 +0100, Henry Proudhon wrote:
> Hi Matplotlib users,
> 
> I'm experiencing some problems when closing a figure (I'm using
> matplotlib 0.91.0).
> the function is working correctly when invoked inside the module where
> it is defined:
> 
> from pylab import *
> 
> def test():
>     t = arange(0.0, 1.0+0.01, 0.01)
>     s = cos(2*2*pi*t)
>     plot(t, s, '-', lw=2)
>     show()
>     print 'after show'
> 
> if __name__ == '__main__':
>     print test()
> 
> now if I call test() from another module, the figure closes but show()
> does not return ('after show' does not get printed) so the execution is
> stuck.
> 
> Can somebody shed some light here, I'm really missing something...
> 
> Thanks a lot
> Henry

In case somebody has the same problem, I worked around this by embedding
a matplotlib figure in a gtk window but it's a lot heavier.

Henry

I was looking at rc_params() and saw

    if not os.path.exists(fname):
        message = 'could not find rc file; returning defaults'
        ret =  dict([ (key, tup[0]) for key, tup in defaultParams.items()])
        warnings.warn(message)
        return ret

Is this correct?  It seems that it returns a regular dictionary rather
than an instance of RcParams.

~tj

On some OSX systems, pylab is failing on importing pylab because of  
lack of GTK. This appears to be a bug; if the system doesnt have gtk,  
it should simply not import that backend to begin with.

 >>> from matplotlib import pylab
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Library/Python/2.5/site-packages/matplotlib-0.98pre-py2.5- 
macosx-10.5-i386.egg/matplotlib/pylab.py", line 292, in <module>
    from matplotlib.pyplot import *
  File "/Library/Python/2.5/site-packages/matplotlib-0.98pre-py2.5- 
macosx-10.5-i386.egg/matplotlib/pyplot.py", line 39, in <module>
    new_figure_manager, draw_if_interactive, show = pylab_setup()
  File "/Library/Python/2.5/site-packages/matplotlib-0.98pre-py2.5- 
macosx-10.5-i386.egg/matplotlib/backends/__init__.py", line 24, in  
pylab_setup
    globals(),locals(),[backend_name])
  File "/Library/Python/2.5/site-packages/matplotlib-0.98pre-py2.5- 
macosx-10.5-i386.egg/matplotlib/backends/backend_gtkagg.py", line 10,  
in <module>
    from matplotlib.backends.backend_gtk import gtk, FigureManagerGTK,  
FigureCanvasGTK,\
  File "/Library/Python/2.5/site-packages/matplotlib-0.98pre-py2.5- 
macosx-10.5-i386.egg/matplotlib/backends/backend_gtk.py", line 6, in  
<module>
    import gobject
ImportError: No module named gobject
 >>>

OK, I managed to build with the setupext.py list empty as you  
specified. However, for some reason, the resulting matplotlib egg  
*still* looks for dynamic libraries in /usr/local:

ImportError: dlopen(/Library/Python/2.5/site-packages/ 
matplotlib-0.98pre-py2.5-macosx-10.5-i386.egg/matplotlib/_path.so, 2):  
Library not loaded: /usr/local/lib/libpng12.0.dylib
  Referenced from: /Library/Python/2.5/site-packages/ 
matplotlib-0.98pre-py2.5-macosx-10.5-i386.egg/matplotlib/_path.so
  Reason: image not found

This is pretty bizarre.

On 12/01/2008, at 12:42 PM, Charlie Moad wrote:

> I just had to rebuild the OSX binary because a similar reason.  I
> forgot to change the basedir['darwin'] entry in setupext.py.  Around
> line 50.....
>
> Change:
>   'darwin' : ['/sw/lib/freetype2', '/sw/lib/freetype219', '/usr/ 
> local',
>               '/usr', '/sw', '/usr/X11R6'],
> To:
>   'darwin' : [],
>
> This will ensure you don't pick up shared libraries over the static
> ones since they are preferred by gcc.