matplotlib-users Mailing List for matplotlib (Page 11)

Hi,

I'm running matplotlib v 0.99.0.

I can create a 3D scatterplot, like this:

import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

fig = plt.figure()
ax = Axes3D(fig)
ax.scatter( ... )

so far so good.

Now, say I want to add a 2D contourplot (a "slice" of the 3D scatter  
contoured).
Is there some way how I can specify along which coordinates the  
contourplot is plotted, or is it always plotted on the (x,y) plane?

How I have tried this is by
from matplotlib.mlab import griddata

fi = griddata( ... )
ax.contourf(x,z,fi, ... )
# even though I specify z, it plots the contourf along the y axis.


Thank you all for any pointers,
Cheers,
Claus

Hi,

Is there a way to somehow squeeze two datasets with two different color
maps into one pcolormesh? Say one dataset occupies one triangle (e. g.
numpy.triu(dataset1, 1)) and the second dataset a different area (e. g.
numpy.tril(dataset2, -1)) and then paste the two datasets in one
pcolormesh (e.g. pylab.pcolormesh(numpy.triu(dataset1, 1) +
numpy.tril(dataset2, -1)) ). This will obviously share one colormap. But
does anybody know a trick to use two different ones?

TIA
Christian

PS Forgot to thank Eric Firing for his tip last time I asked. Thank you
Eric: You solved my problem!

It's actually just a nice part of the python language, setattr(a,
> 'foo', 'bar') is the same code as a.foo = bar.  When working with
> NetCDF, the only time I've needed to use setarr (or getattr) is when
> the name of the attribute I want isn't a valid python identifier (like
> if an attribute has a - in it, e.g. 'unit-type').
>
>
Slick! Another great reason for python. Thanks for the rapid response.

On Wed, Nov 11, 2009 at 9:26 AM, John <was...@gm...> wrote:
> No problem.. just wanted to make sure it would work and is 'pythonic'.
>
> I guess the biggest concern is that for others using the netcdf file, say
> from matlab or somewhere else, will be able to access the attributes.
>
> It seemed so easy to do it this way, I was curious why the tutorials suggest
> using setattr(), and I wonder if perhaps what I'm setting is not 'global' or
> perhaps wouldn't match some ISO or otherwise NetCDF standard??

It's actually just a nice part of the python language, setattr(a,
'foo', 'bar') is the same code as a.foo = bar.  When working with
NetCDF, the only time I've needed to use setarr (or getattr) is when
the name of the attribute I want isn't a valid python identifier (like
if an attribute has a - in it, e.g. 'unit-type').

Ryan

-- 
Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma

No problem.. just wanted to make sure it would work and is 'pythonic'.

I guess the biggest concern is that for others using the netcdf file, say
from matlab or somewhere else, will be able to access the attributes.

It seemed so easy to do it this way, I was curious why the tutorials suggest
using setattr(), and I wonder if perhaps what I'm setting is not 'global' or
perhaps wouldn't match some ISO or otherwise NetCDF standard??

Thanks!

On Wed, Nov 11, 2009 at 4:03 PM, Ryan May <rm...@gm...> wrote:

> On Wed, Nov 11, 2009 at 7:31 AM, John [H2O] <was...@gm...> wrote:
> >
> > Hello, I'm trying to write a function to create a 'vanilla' NetCDF file
> to
> > which I can add data.
> >
> > So far I have created the following, which is designed to set up a netcdf
> > file for the addition of global or at least lat/lon datasets. My question
> > has to do with attribute setting. Is it valid to do what I'm doing below
> > "nco.attribute = attributevalue" or do you have to use the setattr
> function
> > as outlined in some tutorials?
> >
> > Thanks!
> >
> > def default_netcdf(nco_filename,
> >                   lon0=-179.5,lat0=-89.5,
> >                   nx=720,ny=360,
> >                   dx=0.5,dy=0.5):
> >    """ add default attributes and dimensions to the nc file """
> >    nco = NetCDFFile(nco_filename,'w')
> >    nco.author = "Some One"
> >    nco.createdate = dt.datetime.now().ctime()
> >    nco.contact = "som...@py..."
> >    nco.Conventions = "CF-1.4"
> >
> >    nco.createDimension('lon',nx)
> >    nco.createDimension('lat',ny)
> >    nco.createVariable('lat','d',('lat',))
> >    nco.createVariable('lon','d',('lon',))
> >    lon = np.arange(lon0,lon0+(nx*dx),dx)
> >    lat = np.arange(lat0,lat0+(ny*dy),dy)
> >    nco.variables['lat'][:] = lat
> >    nco.variables['lon'][:] = lon
> >    nco.createVariable('data','d',('lon','lat'))
> >
> >    return nco
>
> That should work fine for settings attributes.  Are you having a
> problem? If so, can you paste the actual traceback?
>
> Ryan
>
> --
> Ryan May
> Graduate Research Assistant
> School of Meteorology
> University of Oklahoma
>



-- 
Configuration
``````````````````````````
Plone 2.5.3-final,
CMF-1.6.4,
Zope (Zope 2.9.7-final, python 2.4.4, linux2),
Five 1.4.1,
Python 2.4.4 (#1, Jul 3 2007, 22:58:17) [GCC 4.1.1 20070105 (Red Hat
4.1.1-51)],
PIL 1.1.6
Mailman 2.1.9
Postfix 2.4.5
Procmail v3.22 2001/09/10

On Wed, Nov 11, 2009 at 7:31 AM, John [H2O] <was...@gm...> wrote:
>
> Hello, I'm trying to write a function to create a 'vanilla' NetCDF file to
> which I can add data.
>
> So far I have created the following, which is designed to set up a netcdf
> file for the addition of global or at least lat/lon datasets. My question
> has to do with attribute setting. Is it valid to do what I'm doing below
> "nco.attribute = attributevalue" or do you have to use the setattr function
> as outlined in some tutorials?
>
> Thanks!
>
> def default_netcdf(nco_filename,
>                   lon0=-179.5,lat0=-89.5,
>                   nx=720,ny=360,
>                   dx=0.5,dy=0.5):
>    """ add default attributes and dimensions to the nc file """
>    nco = NetCDFFile(nco_filename,'w')
>    nco.author = "Some One"
>    nco.createdate = dt.datetime.now().ctime()
>    nco.contact = "som...@py..."
>    nco.Conventions = "CF-1.4"
>
>    nco.createDimension('lon',nx)
>    nco.createDimension('lat',ny)
>    nco.createVariable('lat','d',('lat',))
>    nco.createVariable('lon','d',('lon',))
>    lon = np.arange(lon0,lon0+(nx*dx),dx)
>    lat = np.arange(lat0,lat0+(ny*dy),dy)
>    nco.variables['lat'][:] = lat
>    nco.variables['lon'][:] = lon
>    nco.createVariable('data','d',('lon','lat'))
>
>    return nco

That should work fine for settings attributes.  Are you having a
problem? If so, can you paste the actual traceback?

Ryan

-- 
Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma

Hi all,

Previously I was a user of gnuplot but have been giving matplotlib a  
try.  One thing I've run in to right away is that matplotlib appears  
to be significantly slower.

A script to produce a dozen plots was taking me ~1 second with  
gnuplot, and now takes me ~18 seconds with matplotlib.

I'm curious if anyone knows how to speed things up.  To figure out  
what is taking most of the time, I've used cProfile and pstats and  
below is the top 15 functions taking the most time.
   (note: "plotStackedJobsVsTime" is my function that uses matplotlib.)

----------------------------------------
Wed Nov 11 08:43:52 2009    fooprof

          3069840 function calls (3009711 primitive calls) in 17.658  
CPU seconds

    Ordered by: cumulative time
    List reduced from 873 to 15 due to restriction <15>

    ncalls  tottime  percall  cumtime  percall filename:lineno(function)
         1    0.000    0.000   17.658   17.658 <string>:1(<module>)
         1    0.007    0.007   17.658   17.658 /cms/cmsprod/bin/ 
prodJobMonitorPlots_matplotlib.py:151(main)
        13    0.620    0.048   17.647    1.357 /cms/cmsprod/bin/ 
prodJobMonitorPlots_matplotlib.py:54(plotStackedJobsVsTime)
        13    0.000    0.000   10.087    0.776 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/figure.py:959 
(savefig)
        13    0.028    0.002   10.086    0.776 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/ 
backend_bases.py:1372(print_figure)
        13    0.028    0.002    9.847    0.757 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/backends/ 
backend_agg.py:357(print_png)
        13    0.000    0.000    9.243    0.711 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/backends/ 
backend_agg.py:307(draw)
   3206/13    0.040    0.000    9.222    0.709 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/artist.py:44 
(draw_wrapper)
        13    0.001    0.000    9.222    0.709 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/figure.py:729 
(draw)
        13    0.005    0.000    9.200    0.708 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/axes.py:1632 
(draw)
        26    0.029    0.001    5.624    0.216 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/axis.py:727 
(draw)
1533/1325    0.078    0.000    3.663    0.003 /afs/hep.wisc.edu/cms/sw/ 
python/x86_64/2.5.4/lib/python2.5/site-packages/matplotlib-0.99.1.1_r0- 
py2.5-linux-x86_64.egg/matplotlib/axis.py:61(__init__)
       811    0.035    0.000    3.057    0.004 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/axis.py:181 
(draw)
        13    0.001    0.000    2.619    0.201 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/legend.py:342 
(draw)
  1065/961    0.012    0.000    2.543    0.003 /afs/hep.wisc.edu/cms/ 
sw/python/x86_64/2.5.4/lib/python2.5/site-packages/ 
matplotlib-0.99.1.1_r0-py2.5-linux-x86_64.egg/matplotlib/axis.py:1207 
(_get_tick)
----------------------------------------


My script, for the curious, is at
   http://www.hep.wisc.edu/cms/comp/routerqMonitor/prodJobMonitorPlots_matplotlib.py
and produces these plots:
   http://www.hep.wisc.edu/cms/comp/routerqMonitor/index.html


Any hints at what I can do to speed up my script?  Or is it out of my  
hands because it's all in matplotlib?

Thanks for any help,
Mike

Hello, I'm trying to write a function to create a 'vanilla' NetCDF file to
which I can add data.

So far I have created the following, which is designed to set up a netcdf
file for the addition of global or at least lat/lon datasets. My question
has to do with attribute setting. Is it valid to do what I'm doing below
"nco.attribute = attributevalue" or do you have to use the setattr function
as outlined in some tutorials?

Thanks!

def default_netcdf(nco_filename,
                   lon0=-179.5,lat0=-89.5,
                   nx=720,ny=360,
                   dx=0.5,dy=0.5):
    """ add default attributes and dimensions to the nc file """
    nco = NetCDFFile(nco_filename,'w')
    nco.author = "Some One"
    nco.createdate = dt.datetime.now().ctime()
    nco.contact = "som...@py..."
    nco.Conventions = "CF-1.4"
    
    nco.createDimension('lon',nx)
    nco.createDimension('lat',ny)
    nco.createVariable('lat','d',('lat',))
    nco.createVariable('lon','d',('lon',))
    lon = np.arange(lon0,lon0+(nx*dx),dx)
    lat = np.arange(lat0,lat0+(ny*dy),dy)
    nco.variables['lat'][:] = lat
    nco.variables['lon'][:] = lon
    nco.createVariable('data','d',('lon','lat'))
    
    return nco

-- 
View this message in context: http://old.nabble.com/function-to-create-netcdf-file-tp26301238p26301238.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

On Tue, 10 Nov 2009 13:13 -0500, "George Bonev"
<geo...@gm...> wrote:

  Hello,
  I am plotting a fig in Basemap, but I don't need the extra
  while space on the sides of the figure when I use savfig(). I
  tried axes([0,0,1,1]) but that just gave me a blank white
  screen with some ticks on the sides. Is there another way to
  do it with Basemap? Please help!
  Thank you,
  George B.


George:  If you want to preserve the aspect ratio of the map, you
need to calculate it ahead of time and make sure that you create
a figure with the same aspect ratio and define an axes with no
white space.  For example, for a global cylindrical equidistant
projection, the aspect ration should be 2:1, so you would create
a figure like this

from mpl_toolkits.basemap import Basemap
import matplotlib.pyplot as plt
import numpy as np
fig = plt.figure(figsize=(10,5))
ax = plt.axes([0,0,1,1])
m = Basemap(projection='cyl',lon_0=180)
m.drawcoastlines()
m.fillcontinents()
m.drawparallels(np.arange(-60,61,30))
m.drawmeridians(np.arange(60,301,60))
plt.savefig('nowhite.png')
plt.show()

Note that you need to create the axes instance before the Basemap
instance.

If you don't care about the aspect ratio of the map being
correct, you can set fix_aspect='False', create a map of any size
you like, and use an axes with no white space. For example,

from mpl_toolkits.basemap import Basemap
import matplotlib.pyplot as plt
import numpy as np
fig = plt.figure(figsize=(8,8))
ax = plt.axes([0,0,1,1])
m = Basemap(projection='cyl',lon_0=180,fix_aspect=False)
m.drawcoastlines()
m.fillcontinents()
m.drawparallels(np.arange(-60,61,30))
m.drawmeridians(np.arange(60,301,60))
plt.savefig('nowhite2.png')
plt.show()

If you don't set fix_aspect=False in this example, you'll get a
map with a 2:1 aspect ratio and lots of white space on the top
and bottom of the plot.

-Jeff
------
Jeffrey S. Whitaker
Phone: (303)497-6313
FAX: (303)497-6449
NOAA/OAR/CDC  R/CDC1         
325 Broadway, Boulder, CO, USA 80305-3328                             
http://www.cdc.noaa.gov/people/jeffrey.s.whitaker

Hi all, 
I am a very new user to both python and matplotlib so please beware that my
questions might be a real no-brainer. Here goes: 
I have a 3D array of data that I isolated from a much larger dataset. The
first two columns are the x and y positions while the third is the intensity
at the position (xi, yi). I would like to produce a colour map of the
intensity as a function of x and y positions. Numerous tries with pcolor
have got me nowhere. I constantly get an error message saying "Too many
indices". 

Is my use of pcolor correct? If not, what should I use?

Any help would be greatly appreciated. 

Thanks a lot! :-D

-- 
View this message in context: http://old.nabble.com/Noob-confusion%3A-Pcolor-tp26295820p26295820.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

I just upgraded to 0.99.1 and it's working perfectly now.  I was previously
hacking _png.so with install_name_tool, and that's not necessary anymore as
well.

Thank you to Emanuele and everyone else...much appreciated!

BZ

On Tue, Nov 10, 2009 at 1:38 PM, Emanuele Santos
<ema...@gm...>wrote:

> On Tue, Nov 10, 2009 at 11:57 AM, John Hunter <jd...@gm...> wrote:
>
>> On Tue, Nov 10, 2009 at 1:26 PM, Emanuele Santos
>> <ema...@gm...> wrote:
>> > Hi, Chris
>> >
>> > I think it is the official one (0.98.5).
>> > I didn't try the latest version but I can check if this is still true.
>>
>> Could you please -- there were some bugs in linking on the OSX
>> binaries that were fixed between 0.98.5 and 0.99.1 so I'd like to
>> confirm that this was fixes as well
>>
>>
> I just checked and it seems to be fixed in 0.99.1.
>
> -- Emanuele.
>
>
>

I
> On Tue, Nov 10, 2009 at 11:57 AM, John Hunter <jd...@gm...>  
> wrote:
> On Tue, Nov 10, 2009 at 1:26 PM, Emanuele Santos
> <ema...@gm...> wrote:
> > Hi, Chris
> >
> > I think it is the official one (0.98.5).
> > I didn't try the latest version but I can check if this is still  
> true.
>
> Could you please -- there were some bugs in linking on the OSX
> binaries that were fixed between 0.98.5 and 0.99.1 so I'd like to
> confirm that this was fixes as well
>

I just checked and it seems to be fixed in 0.99.1.

-- Emanuele.


> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

FWIW, I'm using the official 0.98.5.

Looking at the app bundle, all of the matplotlib .so files are linking
against:

@executable_path/../Frameworks/libgcc_s.1.dylib

...*except* matplotlib/backends/_macosx.so which is linking against:

/usr/lib/libgcc_s.1.dylib

So, I didn't see any looking in /usr/local/lib.

BZ


On Tue, Nov 10, 2009 at 11:57 AM, John Hunter <jd...@gm...> wrote:

> On Tue, Nov 10, 2009 at 1:26 PM, Emanuele Santos
> <ema...@gm...> wrote:
> > Hi, Chris
> >
> > I think it is the official one (0.98.5).
> > I didn't try the latest version but I can check if this is still true.
>
> Could you please -- there were some bugs in linking on the OSX
> binaries that were fixed between 0.98.5 and 0.99.1 so I'd like to
> confirm that this was fixes as well
>
>

On Tue, Nov 10, 2009 at 1:26 PM, Emanuele Santos
<ema...@gm...> wrote:
> Hi, Chris
>
> I think it is the official one (0.98.5).
> I didn't try the latest version but I can check if this is still true.

Could you please -- there were some bugs in linking on the OSX
binaries that were fixed between 0.98.5 and 0.99.1 so I'd like to
confirm that this was fixes as well

Hi, Chris

I think it is the official one (0.98.5).
I didn't try the latest version but I can check if this is still true.

-- Emanuele.

On Nov 10, 2009, at 11:14 AM, Christopher Barker wrote:

> Emanuele Santos wrote:
>> It should load /usr/lib/libgcc_s.1.dylib. The problem is that the  
>> binary
>> package of matplotlib for mac is linked
>> against /usr/local/lib/libgcc_s.1.dylib.
>
> ouch -- it shouldn't be. Did you build it yourself or is that the
> "official" one?
>
> Also, it seems that Py2app really should include libs that are in
> /usr/local with the bundle, so that may be a py2app bug/feature.
>
> Boy, this is a pain!
>
> oh well,
>
> -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...
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

-- Emanuele.

Hello all,

I'm having trouble setting the aspect ratio of an Axes3D to equal. Is it
possible ?

Here is example illustrating the problem:

from matplotlib.pylab import *
from mpl_toolkits.mplot3d import Axes3D
from  numpy import *
n = 100
r = 3
t = linspace(0,2*pi, n)
ax = Axes3D(figure())
ax.set_aspect('equal')
ax.plot(xs=cos(t), ys=sin(t), zs=0)
ax.plot(xs=[0]*n, ys=r*sin(t), zs=r*cos(t))
show()

As a workaround, I currently plot a big cube around my data.
I would be very glad if someone had a cleaner solution.

Thanks by advance

Regards, Sebastien

Hello,

I am plotting a fig in Basemap, but I don't need the extra while space on
the sides of the figure when I use savfig(). I tried axes([0,0,1,1]) but
that just gave me a blank white screen with some ticks on the sides. Is
there another way to do it with Basemap? Please help!

Thank you,

George B.

Emanuele Santos wrote:
> It should load /usr/lib/libgcc_s.1.dylib. The problem is that the binary 
> package of matplotlib for mac is linked 
> against /usr/local/lib/libgcc_s.1.dylib.

ouch -- it shouldn't be. Did you build it yourself or is that the 
"official" one?

Also, it seems that Py2app really should include libs that are in 
/usr/local with the bundle, so that may be a py2app bug/feature.

Boy, this is a pain!

oh well,

-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...

It should load /usr/lib/libgcc_s.1.dylib. The problem is that the  
binary package of matplotlib for mac is linked against /usr/local/lib/ 
libgcc_s.1.dylib.

I tried using install_name_tool and change this on the bundle and it  
does the trick.
$ cd Resources/lib/python2.5
install_name_tool -change @executable_path/../Frameworks/libgcc_s. 
1.dylib /usr/lib/libgcc_s.1.dylib ft2font.so

-- Emanuele.

On Nov 10, 2009, at 9:43 AM, Brian Zambrano wrote:

> Following those steps, my app started just fine on Snow Leopard.   
> Thank you Emanuele!
>
> I would, however, like to understand this problem a bit more so I  
> can fix the build.  I'd like this app to run out of the box.
>
> Looking at ft2font.so:
>
> otool -L Resources/lib/python2.5/matplotlib/ft2font.so
> Resources/lib/python2.5/matplotlib/ft2font.so:
>         @executable_path/../Frameworks/libfreetype.6.dylib  
> (compatibility version 10.0.0, current version 10.12.0)
>         /usr/lib/libz.1.dylib (compatibility version 1.0.0, current  
> version 1.2.3)
>         /usr/lib/libstdc++.6.dylib (compatibility version 7.0.0,  
> current version 7.4.0)
>         /usr/lib/libSystem.B.dylib (compatibility version 1.0.0,  
> current version 111.1.3)
>         @executable_path/../Frameworks/libgcc_s.1.dylib  
> (compatibility version 1.0.0, current version 1.0.0)
>
> For matplotlib to load successfully, which libgcc should be loaded?
>
> Thanks again,
> BZ
>
>
>
> On Mon, Nov 9, 2009 at 11:45 PM, Emanuele Santos <ema...@gm... 
> > wrote:
> We saw the same problem with VisTrails (www.vistrails.org).
>
> It seems that python.org's python can't build the fonts in  
> ~/.matplotlib on snow leopard.
> We noticed that importing matplotlib.axis using Apple's python will  
> do that.
>
> So this is the work-around we found for vistrails (bundled with python.org 
>  2.5)
>
> In a terminal on snow leopard:
>
> $ rm -rf ~/.matplotlib
> $ cd /Applications/Vistrails.app/Contents/Resources/lib/python2.5
> $ DYLD_LIBRARY_PATH=/usr/lib  /System/Library/Frameworks/ 
> Python.framework/Versions/2.5/bin/python
>
> >>> import matplotlib.axis
>
> Now running the app again works.
>
> The DYLD_LIBRARY_PATH is necessary because of problems we saw with  
> different versions of libgcc_s.1.dylib present in the system and the  
> wrong one was being loaded.
>
> I don't think it's a py2app bug because I can reproduce it by just  
> running python.org's python and importing matplotlib.axis.
>
> I guess it is a bug in matplotlib.
>
> -- Emanuele.
>
>
> On Nov 9, 2009, at 11:46 PM, Brian Zambrano wrote:
>
>> I'm getting my Py2app build running and think I've worked around  
>> some issues, but another major one has come up.  My application is  
>> being built on OS X 10.5.8, where matplotlib is, oviously,  
>> installed.  On another 10.5.X machine without any of the app's  
>> dependencies, my compiled app runs just fine.  Today, I tried  
>> getting it running on a new Mac with Snow Leopard, 10.6, and get  
>> the following bus error.
>>
>> From the crash report, it looks like these are relevant:
>>
>> Exception Type:  EXC_BAD_ACCESS (SIGBUS)
>> Exception Codes: KERN_PROTECTION_FAILURE at 0x0000000000000000
>> Crashed Thread:  0  Dispatch queue: com.apple.main-thread
>>
>> Thread 0 Crashed:  Dispatch queue: com.apple.main-thread
>> 0   ???                                 0000000000 0 + 0
>> 1   libSystem.B.dylib                   0x91176aa8  
>> _Unwind_GetLanguageSpecificData + 24
>> 2   libstdc++.6.dylib                   0x90057d86  
>> __gxx_personality_v0 + 120
>> 3   libgcc_s.1.dylib                    0x03801476  
>> _Unwind_Backtrace + 278
>> 4   libgcc_s.1.dylib                    0x03801890 _Unwind_Resume +  
>> 112
>> 5   ft2font.so                          0x038cd3b1 FT2Font::FT2Font 
>> (std::string) + 3737
>> 6   ft2font.so                          0x038cd4df  
>> ft2font_module::new_ft2font(Py::Tuple const&) + 291
>> 7   ft2font.so                          0x038d787f  
>> Py::ExtensionModule<ft2font_module>::invoke_method_varargs 
>> (std::string const&, Py::Tuple const&) + 261
>> 8   ft2font.so                          0x038df637  
>> method_varargs_call_handler + 301
>>
>> I'm not really sure where to go from here.  I've read a few other  
>> things which hint at Snow Leopard being a bit of a headache: http://bit.ly/2Z2Cil
>>
>> Has anyone run this setup before?  Suggestions?
>>
>> BZ
>>
>> ------------------------------------------------------------------------------
>> Let Crystal Reports handle the reporting - Free Crystal Reports  
>> 2008 30-Day
>> trial. Simplify your report design, integration and deployment -  
>> and focus on
>> what you do best, core application coding. Discover what's new with
>> Crystal Reports now.  http://p.sf.net/sfu/bobj-july_______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

-- Emanuele.

Following those steps, my app started just fine on Snow Leopard.  Thank you
Emanuele!

I would, however, like to understand this problem a bit more so I can fix
the build.  I'd like this app to run out of the box.

Looking at ft2font.so:

otool -L Resources/lib/python2.5/matplotlib/ft2font.so
Resources/lib/python2.5/matplotlib/ft2font.so:
        @executable_path/../Frameworks/libfreetype.6.dylib (compatibility
version 10.0.0, current version 10.12.0)
        /usr/lib/libz.1.dylib (compatibility version 1.0.0, current version
1.2.3)
        /usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current
version 7.4.0)
        /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current
version 111.1.3)
        @executable_path/../Frameworks/libgcc_s.1.dylib (compatibility
version 1.0.0, current version 1.0.0)

For matplotlib to load successfully, which libgcc should be loaded?

Thanks again,
BZ



On Mon, Nov 9, 2009 at 11:45 PM, Emanuele Santos
<ema...@gm...>wrote:

> We saw the same problem with VisTrails (www.vistrails.org).
>
> It seems that python.org's python can't build the fonts in ~/.matplotlib
> on snow leopard.
> We noticed that importing matplotlib.axis using Apple's python will do
> that.
>
> So this is the work-around we found for vistrails (bundled with python.org2.5)
>
> In a terminal on snow leopard:
>
> $ rm -rf ~/.matplotlib
> $ cd /Applications/Vistrails.app/Contents/Resources/lib/python2.5
> $ DYLD_LIBRARY_PATH=/usr/lib
>  /System/Library/Frameworks/Python.framework/Versions/2.5/bin/python
>
> >>> import matplotlib.axis
>
> Now running the app again works.
>
> The DYLD_LIBRARY_PATH is necessary because of problems we saw with
> different versions of libgcc_s.1.dylib present in the system and the wrong
> one was being loaded.
>
> I don't think it's a py2app bug because I can reproduce it by just running
> python.org's python and importing matplotlib.axis.
>
> I guess it is a bug in matplotlib.
>
> -- Emanuele.
>
>
> On Nov 9, 2009, at 11:46 PM, Brian Zambrano wrote:
>
> I'm getting my Py2app build running and think I've worked around some
> issues, but another major one has come up.  My application is being built on
> OS X 10.5.8, where matplotlib is, oviously, installed.  On another 10.5.X
> machine without any of the app's dependencies, my compiled app runs just
> fine.  Today, I tried getting it running on a new Mac with Snow Leopard,
> 10.6, and get the following bus error.
>
> From the crash report, it looks like these are relevant:
>
> Exception Type:  EXC_BAD_ACCESS (SIGBUS)
> Exception Codes: KERN_PROTECTION_FAILURE at 0x0000000000000000
> Crashed Thread:  0  Dispatch queue: com.apple.main-thread
>
> Thread 0 Crashed:  Dispatch queue: com.apple.main-thread
> 0   ???                                 0000000000 0 + 0
> 1   libSystem.B.dylib                   0x91176aa8
> _Unwind_GetLanguageSpecificData + 24
> 2   libstdc++.6.dylib                   0x90057d86 __gxx_personality_v0 +
> 120
> 3   libgcc_s.1.dylib                    0x03801476 _Unwind_Backtrace + 278
> 4   libgcc_s.1.dylib                    0x03801890 _Unwind_Resume + 112
> 5   ft2font.so                          0x038cd3b1
> FT2Font::FT2Font(std::string) + 3737
> 6   ft2font.so                          0x038cd4df
> ft2font_module::new_ft2font(Py::Tuple const&) + 291
> 7   ft2font.so                          0x038d787f
> Py::ExtensionModule<ft2font_module>::invoke_method_varargs(std::string
> const&, Py::Tuple const&) + 261
> 8   ft2font.so                          0x038df637
> method_varargs_call_handler + 301
>
> I'm not really sure where to go from here.  I've read a few other things
> which hint at Snow Leopard being a bit of a headache: http://bit.ly/2Z2Cil
>
> Has anyone run this setup before?  Suggestions?
>
> BZ
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.
> http://p.sf.net/sfu/bobj-july_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

Hi folks,
I am using matplotlib on debian unstable with Gtk (or gtkagg sometimes)
backend. Since the last update I have some troubles with the stop mouse
event in the ginput function:

$ python
Python 2.5.4 (r254:67916, Sep 26 2009, 08:19:36) 
[GCC 4.3.4] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import matplotlib.pyplot as plt
>>> plt.plot([1,2.,34.])
/usr/lib/pymodules/python2.5/matplotlib/backends/backend_gtk.py:621: DeprecationWarning: Use the new widget gtk.Tooltip
  self.tooltips = gtk.Tooltips()
[<matplotlib.lines.Line2D object at 0xa091f4c>]
>>> plt.ginput(n=0)
/usr/lib/pymodules/python2.5/matplotlib/backend_bases.py:1624: DeprecationWarning: Using default event loop until function specific to this GUI is implemented
  warnings.warn(str,DeprecationWarning)
Traceback (most recent call last):
  File "/usr/lib/pymodules/python2.5/matplotlib/backends/backend_gtk.py", line 208, in button_press_event
    FigureCanvasBase.button_press_event(self, x, y, event.button, guiEvent=event)
  File "/usr/lib/pymodules/python2.5/matplotlib/backend_bases.py", line 1196, in button_press_event
    self.callbacks.process(s, mouseevent)
  File "/usr/lib/pymodules/python2.5/matplotlib/cbook.py", line 163, in process
    func(*args, **kwargs)
  File "/usr/lib/pymodules/python2.5/matplotlib/blocking_input.py", line 46, in on_event
    self.post_event()
  File "/usr/lib/pymodules/python2.5/matplotlib/blocking_input.py", line 142, in post_event
    self.mouse_event()
  File "/usr/lib/pymodules/python2.5/matplotlib/blocking_input.py", line 153, in mouse_event
    self.mouse_event_stop(event)
  File "/usr/lib/pymodules/python2.5/matplotlib/blocking_input.py", line 196, in mouse_event_stop
    self.fig.canvas.stop_event_loop(event)
TypeError: stop_event_loop() takes exactly 1 argument (2 given)
[(0.44261065266316579, 6.9299095607235142), (1.3628407101775444, 16.765180878552972)]

In fact, I mentioned the argument n=0 to have an arbitrary of input
points, but when I press enter or middle-click to end the acquisition,
the exception raises. After the 30s default timeout, the selected points
are output.

Is it a bug related to the changes in 7591 ?

-- 
Fabrice Silva
Laboratory of Mechanics and Acoustics - CNRS
31 chemin Joseph Aiguier, 13402 Marseille, France.

We saw the same problem with VisTrails (www.vistrails.org).

It seems that python.org's python can't build the fonts in  
~/.matplotlib on snow leopard.
We noticed that importing matplotlib.axis using Apple's python will do  
that.

So this is the work-around we found for vistrails (bundled with python.org 
  2.5)

In a terminal on snow leopard:

$ rm -rf ~/.matplotlib
$ cd /Applications/Vistrails.app/Contents/Resources/lib/python2.5
$ DYLD_LIBRARY_PATH=/usr/lib  /System/Library/Frameworks/ 
Python.framework/Versions/2.5/bin/python

 >>> import matplotlib.axis

Now running the app again works.

The DYLD_LIBRARY_PATH is necessary because of problems we saw with  
different versions of libgcc_s.1.dylib present in the system and the  
wrong one was being loaded.

I don't think it's a py2app bug because I can reproduce it by just  
running python.org's python and importing matplotlib.axis.

I guess it is a bug in matplotlib.

-- Emanuele.


On Nov 9, 2009, at 11:46 PM, Brian Zambrano wrote:

> I'm getting my Py2app build running and think I've worked around  
> some issues, but another major one has come up.  My application is  
> being built on OS X 10.5.8, where matplotlib is, oviously,  
> installed.  On another 10.5.X machine without any of the app's  
> dependencies, my compiled app runs just fine.  Today, I tried  
> getting it running on a new Mac with Snow Leopard, 10.6, and get the  
> following bus error.
>
> From the crash report, it looks like these are relevant:
>
> Exception Type:  EXC_BAD_ACCESS (SIGBUS)
> Exception Codes: KERN_PROTECTION_FAILURE at 0x0000000000000000
> Crashed Thread:  0  Dispatch queue: com.apple.main-thread
>
> Thread 0 Crashed:  Dispatch queue: com.apple.main-thread
> 0   ???                                 0000000000 0 + 0
> 1   libSystem.B.dylib                   0x91176aa8  
> _Unwind_GetLanguageSpecificData + 24
> 2   libstdc++.6.dylib                   0x90057d86  
> __gxx_personality_v0 + 120
> 3   libgcc_s.1.dylib                    0x03801476 _Unwind_Backtrace  
> + 278
> 4   libgcc_s.1.dylib                    0x03801890 _Unwind_Resume +  
> 112
> 5   ft2font.so                          0x038cd3b1 FT2Font::FT2Font 
> (std::string) + 3737
> 6   ft2font.so                          0x038cd4df  
> ft2font_module::new_ft2font(Py::Tuple const&) + 291
> 7   ft2font.so                          0x038d787f  
> Py::ExtensionModule<ft2font_module>::invoke_method_varargs 
> (std::string const&, Py::Tuple const&) + 261
> 8   ft2font.so                          0x038df637  
> method_varargs_call_handler + 301
>
> I'm not really sure where to go from here.  I've read a few other  
> things which hint at Snow Leopard being a bit of a headache: http://bit.ly/2Z2Cil
>
> Has anyone run this setup before?  Suggestions?
>
> BZ
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

I'm getting my Py2app build running and think I've worked around some
issues, but another major one has come up.  My application is being built on
OS X 10.5.8, where matplotlib is, oviously, installed.  On another 10.5.X
machine without any of the app's dependencies, my compiled app runs just
fine.  Today, I tried getting it running on a new Mac with Snow Leopard,
10.6, and get the following bus error.

>From the crash report, it looks like these are relevant:

Exception Type:  EXC_BAD_ACCESS (SIGBUS)
Exception Codes: KERN_PROTECTION_FAILURE at 0x0000000000000000
Crashed Thread:  0  Dispatch queue: com.apple.main-thread

Thread 0 Crashed:  Dispatch queue: com.apple.main-thread
0   ???                                 0000000000 0 + 0
1   libSystem.B.dylib                   0x91176aa8
_Unwind_GetLanguageSpecificData + 24
2   libstdc++.6.dylib                   0x90057d86 __gxx_personality_v0 +
120
3   libgcc_s.1.dylib                    0x03801476 _Unwind_Backtrace + 278
4   libgcc_s.1.dylib                    0x03801890 _Unwind_Resume + 112
5   ft2font.so                          0x038cd3b1
FT2Font::FT2Font(std::string) + 3737
6   ft2font.so                          0x038cd4df
ft2font_module::new_ft2font(Py::Tuple const&) + 291
7   ft2font.so                          0x038d787f
Py::ExtensionModule<ft2font_module>::invoke_method_varargs(std::string
const&, Py::Tuple const&) + 261
8   ft2font.so                          0x038df637
method_varargs_call_handler + 301

I'm not really sure where to go from here.  I've read a few other things
which hint at Snow Leopard being a bit of a headache: http://bit.ly/2Z2Cil

Has anyone run this setup before?  Suggestions?

BZ

On Mon, Nov 2, 2009 at 11:51 PM, Jae-Joon Lee <lee...@gm...> wrote:

> On Mon, Nov 2, 2009 at 10:52 PM, David Sanders <dps...@gm...>
> wrote:
> > from pylab import *
> >
> > ion()
> >
> > N = 1000
> > pos = zeros((N,2))
> >
> > figure(figsize=(8,8))
> > points, = plot(pos[:,0], pos[:,1], ',')
> > axis([-20,20,-20,20])
> >
> > for t in range(1000):
> >
> >     pos += uniform(-1,1,N*2).reshape(N,2)
> >     points.set_data(pos[:,0].copy(), pos[:,1].copy())
> >     draw()
>
> The Line2D object keeps the input data as a cache and only update it
> (recache) if the new data is different than the cached one.
> The problem in this particular case is that the cache is actually a
> *pos* itself. And modifying the pos in place, actually modify the
> cache in the Line2D object. Thus, set_data sees that the given data is
> identical to the cached one, and skip the recaching.
> I'm not sure what is the best approach here, and I defer the fix (or
> not) to others.
> Meanwhile, you can force the recaching with recache method. i.e., call
> points.recache() after set_data. You don't need to make a copy also.
> As a matter of fact, I think it will give you a best performance (but
> not tested) if you directly update the cached data and do not call
> set_data.
> Note that in this particular case, pos == cache, so you actually don't
> need to call get_data, but this is not a general case.
>
> posx, posy = points.get_data(orig=True)
>
> for t in range(100):
>    dx, dy = uniform(-1,1,N*2).reshape(2, N) # note the change in the shape
>    posx += dx
>    posy += dy
>    points.recache()
>    draw()
>
>
Dear JJ,

Many thanks for your answer -- "points.recache()" is exactly what I was
looking for to make my animations work.

It seems to me that this must be a reasonably common question, but I could
not find it in the documentation.
Perhaps it could be added to the animation cookbook?

Thanks and best wishes,
David.

PS: Apologies for the late reply -- I was travelling with difficult internet
access.

Greetings,
I would much like to plot single point in existing figure. If the
point is outside of current axes limit, I would like the axes not to
resize.

Is there an easy way? Currently, to add a point x,y, I
scatter([x],[y]) then xlim and ylim to make sure the axes did not
rescale.

Pål