matplotlib-users Mailing List for matplotlib

On Aug 5, 2013, at 4:43 PM, Matthew Brett <mat...@be...> wrote:

> Hi,
> 
> On Sun, Aug 4, 2013 at 7:26 PM, Scott Lasley <sl...@sp...> wrote:
>> 
>> On Aug 4, 2013, at 4:47 PM, Matthew Brett <mat...@be...> wrote:
>> 
>>> Hi,
>>> 
>>> Congratulations on the new release.
>>> 
>>> I just tried installing in a fresh installation of Python 2.7 from Python.org
>>> 
>>> Am I right in thinking the recommended method is:
>>> 
>>>   pip install matplotlib
>>> 
>>> ?  I did this, and then:
>>> 
>>>>>> import matplotlib.pyplot as plt
>>> Traceback (most recent call last):
>>> File "<stdin>", line 1, in <module>
>>> File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/pyplot.py",
>>> line 24, in <module>
>>>   import matplotlib.colorbar
>>> File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/colorbar.py",
>>> line 29, in <module>
>>>   import matplotlib.collections as collections
>>> File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/collections.py",
>>> line 23, in <module>
>>>   import matplotlib.backend_bases as backend_bases
>>> File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/backend_bases.py",
>>> line 50, in <module>
>>>   import matplotlib.textpath as textpath
>>> File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/textpath.py",
>>> line 14, in <module>
>>>   from matplotlib.mathtext import MathTextParser
>>> File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/mathtext.py",
>>> line 62, in <module>
>>>   import matplotlib._png as _png
>>> ImportError: dlopen(/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/_png.so,
>>> 2): Library not loaded: /usr/X11/lib/libpng12.0.dylib
>>> Referenced from:
>>> /Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/_png.so
>>> Reason: Incompatible library version: _png.so requires version
>>> 42.0.0 or later, but libpng12.0.dylib provides version 36.0.0
>>> 
>>> There were no previous versions of matplotlib installed.  This on OSX 10.6:
>>> 
>>> Python 2.7.5  just installed from python.org
>>> X11 installed from the Apple installation disks for 10.6 before `pip
>>> install matplotlib`
>>> No homebrew etc
>>> 
>>> Any hints about how I should debug this further?
>>> 
>>> Thanks a lot for any help,
>>> 
>>> Matthew
>> 
>> I was able to install numpy 1.7.1 and matplotlib 1.30 under OS X 10.6.8 with X11 version XQuartz 2.3.6 (xorg-server 1.4.2-apple56) and python 2.7.5 64-bit from python.org using pip 1.4.
>> 
>> Have you tried running Software Update?  If you installed X11 after updating to 10.6.8 you can run the combo updater ( http://support.apple.com/kb/DL1399 ) to get the latest X11.   Alternately, you could try installing XQuartz from http://xquartz.macosforge.org/landing/
> 
> Yes, the system was up to date before I installed matplotlib.   Just
> to check my failing memory, I checked for updates (none), reinstalled
> and got the same answer.
> 
> The combo updater looks like it's a fast track to the same outcome as
> Software update - is it something different?

I don't remember if Software Update detects updates to X11.  What version of X11 are you running?  The version on the Snow Leopard install disks, 2.3.4, is older than version 2.3.6 installed by the combo updater.  As far as I know there is no standalone X11 2.3.6 updater for Snow Leopard.  You'll have to re-install the security updates that came out after the combo updater was released if you use it, which is a bit of a pain.  According to otool -L, the libpng12.0.dylib installed with X11 2.3.6 is version 45, compatibility version 45, so it should work with matplotlib.

If you don't want to run the combo updater because of possible side effects or install XQuartz you can download libpng from http://sourceforge.net/projects/libpng/files/libpng15/1.5.17/, unpack it and do the standard ./configure, make, sudo make install to install it in /usr/local/lib.  setup.py should find it there.

hth,
Scott

> If Matplotlib is only expected to work for the Quartz installs - is it
> worth trying to detect that and warn / error at install time?
> 
> Cheers,
> 
> Matthew

On 2013/08/05 10:43 AM, Matthew Brett wrote:
> If Matplotlib is only expected to work for the Quartz installs - is it
> worth trying to detect that and warn / error at install time?

Logically, XQuartz should not be needed at all; mpl on the mac does not 
require X11.  It looks like XQuartz is an accidental dependency, as a 
way to get access to a suitable libpng.

A link from the libpng web site led to this:
http://ethan.tira-thompson.com/Mac_OS_X_Ports.html
which might be a viable alternative.

Eric

Hi,

On Sun, Aug 4, 2013 at 7:26 PM, Scott Lasley <sl...@sp...> wrote:
>
> On Aug 4, 2013, at 4:47 PM, Matthew Brett <mat...@be...> wrote:
>
>> Hi,
>>
>> Congratulations on the new release.
>>
>> I just tried installing in a fresh installation of Python 2.7 from Python.org
>>
>> Am I right in thinking the recommended method is:
>>
>>    pip install matplotlib
>>
>> ?  I did this, and then:
>>
>>>>> import matplotlib.pyplot as plt
>> Traceback (most recent call last):
>>  File "<stdin>", line 1, in <module>
>>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/pyplot.py",
>> line 24, in <module>
>>    import matplotlib.colorbar
>>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/colorbar.py",
>> line 29, in <module>
>>    import matplotlib.collections as collections
>>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/collections.py",
>> line 23, in <module>
>>    import matplotlib.backend_bases as backend_bases
>>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/backend_bases.py",
>> line 50, in <module>
>>    import matplotlib.textpath as textpath
>>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/textpath.py",
>> line 14, in <module>
>>    from matplotlib.mathtext import MathTextParser
>>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/mathtext.py",
>> line 62, in <module>
>>    import matplotlib._png as _png
>> ImportError: dlopen(/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/_png.so,
>> 2): Library not loaded: /usr/X11/lib/libpng12.0.dylib
>>  Referenced from:
>> /Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/_png.so
>>  Reason: Incompatible library version: _png.so requires version
>> 42.0.0 or later, but libpng12.0.dylib provides version 36.0.0
>>
>> There were no previous versions of matplotlib installed.  This on OSX 10.6:
>>
>> Python 2.7.5  just installed from python.org
>> X11 installed from the Apple installation disks for 10.6 before `pip
>> install matplotlib`
>> No homebrew etc
>>
>> Any hints about how I should debug this further?
>>
>> Thanks a lot for any help,
>>
>> Matthew
>
> I was able to install numpy 1.7.1 and matplotlib 1.30 under OS X 10.6.8 with X11 version XQuartz 2.3.6 (xorg-server 1.4.2-apple56) and python 2.7.5 64-bit from python.org using pip 1.4.
>
> Have you tried running Software Update?  If you installed X11 after updating to 10.6.8 you can run the combo updater ( http://support.apple.com/kb/DL1399 ) to get the latest X11.   Alternately, you could try installing XQuartz from http://xquartz.macosforge.org/landing/

Yes, the system was up to date before I installed matplotlib.   Just
to check my failing memory, I checked for updates (none), reinstalled
and got the same answer.

The combo updater looks like it's a fast track to the same outcome as
Software update - is it something different?

If Matplotlib is only expected to work for the Quartz installs - is it
worth trying to detect that and warn / error at install time?

Cheers,

Matthew

The problem is that a 0-length dash or space is undefined.  In Agg, it 
causes an infinite loop (presumably because the line cursor never 
moves).  Saving it to a PDF file and opening it in Acrobat Reader 
reveals a blank page (presumably because it's doing something smarter, 
but also basically throwing up its hands).  In SVG, you get a solid 
line, which may or may not be the right behavior.

Given that a value of 0 doesn't make much sense anyway, I thought it 
best to just disallow it.  Jeffrey: Do you have a good need for this?

Here's the original PR:

https://github.com/matplotlib/matplotlib/pull/1999

Mike

On 08/05/2013 01:36 PM, Benjamin Root wrote:
> @mdboom, from git blame, this looks to be specifically introduced by 
> you via |7e7b5320 
> <https://github.com/matplotlib/matplotlib/commit/7e7b532057c08541489203697987a924e56a7aeb> 
> on May 15th, and you even added some tests for handling path 
> clipping.  Perhaps the choice of "<=" should have been just "<"?|
>
>
> ------------------------------------------------------------------------------
> Get your SQL database under version control now!
> Version control is standard for application code, but databases havent
> caught up. So what steps can you take to put your SQL databases under
> version control? Why should you start doing it? Read more to find out.
> http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

@mdboom, from git blame, this looks to be specifically introduced by you
via 7e7b5320<https://github.com/matplotlib/matplotlib/commit/7e7b532057c08541489203697987a924e56a7aeb>on
May 15th, and you even added some tests for handling path clipping.
Perhaps the choice of "<=" should have been just "<"?

docutils is the library that supports the format (restructuredtext) that 
these docstrings are written in.  It *may* (I haven't looked) contain 
functionality to render as clean plain text.

Mike

On 08/05/2013 09:57 AM, federico vaggi wrote:
> Hi,
>
> SciPy (and NumPy) docstrings are written with a special kind of mark up:
>
> For example, the docstring for the russellrao distance function looks 
> like this:
>
> '\n    Computes the Russell-Rao dissimilarity between two boolean 1-D 
> arrays.\n\n    The Russell-Rao dissimilarity between two boolean 1-D 
> arrays, `u` and\n    `v`, is defined as\n\n    .. math::\n\n     
>  \\frac{n - c_{TT}}\n {n}\n\n    where :math:`c_{ij}` is the number of 
> occurrences of\n    :math:`\\mathtt{u[k]} = i` and 
> :math:`\\mathtt{v[k]} = j` for\n    :math:`k < n`.\n\n    Parameters\n 
>  ----------\n    u : (N,) array_like, bool\n        Input array.\n   
>  v : (N,) array_like, bool\n        Input array.\n\n    Returns\n   
>  -------\n    russellrao : double\n        The Russell-Rao 
> dissimilarity between vectors `u` and `v`.\n\n    '
>
> What's the most efficient way to turn this into a format where you can 
> format it nicely as a matplotlib text object?
>
> I tried:
>
>     fig = plt.figure()
>
> ax = fig.add_subplot(111)
>
> props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)
>
> textstr = dist_fcn.__doc__
>
> textstr = textstr.replace('math:',' ')
>
> textstr = textstr.replace('`', '$')
>
> textstr = textstr.replace('\n\n where', '$\n\n where')
>
> ax.text(0.05, 0.95, textstr, transform=ax.transAxes, fontsize=14,
>
> verticalalignment='top', bbox=props)
>
>
> Which does an 'ok' job, at best, since fractions aren't converted 
> properly.  Is there a way to do it nicely short of using some 
> horrendous regular expressions?
>
>
> Federico
>
>
>
> ------------------------------------------------------------------------------
> Get your SQL database under version control now!
> Version control is standard for application code, but databases havent
> caught up. So what steps can you take to put your SQL databases under
> version control? Why should you start doing it? Read more to find out.
> http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

On 08/03/2013 07:50 AM, Rita wrote:
> Same problem in Linux also. Here is what I did to fix it: Remove the 
> freetype/fontconfig rpm from my local install (yum remove) and then 
> place the proper PKG_CONFIG_PATH to point to my remote 
> freetype/fontconfig.

By remote, you mean self-built, rather than from a package?

> The problem is there is a bug with setupext.py. We ought to prepend 
> PKG_CONFIG_PATH in the gcc compile statement.  I hope this helps.

Can you elaborate?  The setupext.py just calls whatever pkg-config is 
first on the PATH, which should then in turn obey PKG_CONFIG_PATH.  If 
the user needs a custom PKG_CONFIG_PATH, it is generally the 
resposibility of the user to set it correctly -- and matplotlib's build 
system should (and does) use it.  Or maybe I'm just misunderstanding 
what you're suggesting.

Cheers,
Mike

>
>
>
>
>
> On Fri, Aug 2, 2013 at 6:53 AM, Andrew Jaffe <a.h...@gm... 
> <mailto:a.h...@gm...>> wrote:
>
>     Hi,
>
>
>     On 01/08/2013 19:06, Michael Droettboom wrote:
>     > On behalf of a veritable army of super coders, I'm pleased to
>     announce
>     > the release of matplotlib 1.3.0.
>
>     Two issues on OSX 10.8.4. I had been previously using the dmg
>     installer.
>     Lacking that, I tried easy-install and pip install, both of which gave
>     me the following problems:
>
>       - I needed to set CC=clang
>       - When attempting to load matplotlib, I got the following error:
>
>     /Volumes/Data/Users/jaffe/Library/Python/2.7/lib/python/site-packages/matplotlib/font_manager.py
>     in <module>()
>           51 import matplotlib
>           52 from matplotlib import afm
>     ---> 53 from matplotlib import ft2font
>           54 from matplotlib import rcParams, get_cachedir
>           55 from matplotlib.cbook import is_string_like
>
>     ImportError:
>     dlopen(/Volumes/Data/Users/jaffe/Library/Python/2.7/lib/python/site-packages/matplotlib/ft2font.so,
>     2): Symbol not found: _FT_Attach_File
>        Referenced from:
>     /Volumes/Data/Users/jaffe/Library/Python/2.7/lib/python/site-packages/matplotlib/ft2font.so
>        Expected in: flat namespace
>       in
>     /Volumes/Data/Users/jaffe/Library/Python/2.7/lib/python/site-packages/matplotlib/ft2font.so
>
>
>     This is a freetype problem, probably an incompatible version
>     somewhere.
>     Ideas?
>
>     Andrew
>
>
>
>     ------------------------------------------------------------------------------
>     Get your SQL database under version control now!
>     Version control is standard for application code, but databases havent
>     caught up. So what steps can you take to put your SQL databases under
>     version control? Why should you start doing it? Read more to find out.
>     http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
>     _______________________________________________
>     Matplotlib-users mailing list
>     Mat...@li...
>     <mailto:Mat...@li...>
>     https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>
>
> -- 
> --- Get your facts first, then you can distort them as you please.--
>
>
> ------------------------------------------------------------------------------
> Get your SQL database under version control now!
> Version control is standard for application code, but databases havent
> caught up. So what steps can you take to put your SQL databases under
> version control? Why should you start doing it? Read more to find out.
> http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hi,

SciPy (and NumPy) docstrings are written with a special kind of mark up:

For example, the docstring for the russellrao distance function looks like
this:

'\n    Computes the Russell-Rao dissimilarity between two boolean 1-D
arrays.\n\n    The Russell-Rao dissimilarity between two boolean 1-D
arrays, `u` and\n    `v`, is defined as\n\n    .. math::\n\n      \\frac{n
- c_{TT}}\n           {n}\n\n    where :math:`c_{ij}` is the number of
occurrences of\n    :math:`\\mathtt{u[k]} = i` and :math:`\\mathtt{v[k]} =
j` for\n    :math:`k < n`.\n\n    Parameters\n    ----------\n    u : (N,)
array_like, bool\n        Input array.\n    v : (N,) array_like, bool\n
   Input array.\n\n    Returns\n    -------\n    russellrao : double\n
   The Russell-Rao dissimilarity between vectors `u` and `v`.\n\n    '

What's the most efficient way to turn this into a format where you can
format it nicely as a matplotlib text object?

I tried:

    fig = plt.figure()

ax = fig.add_subplot(111)

props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)

textstr = dist_fcn.__doc__

textstr = textstr.replace('math:',' ')

textstr = textstr.replace('`', '$')

textstr = textstr.replace('\n\n where', '$\n\n where')

ax.text(0.05, 0.95, textstr, transform=ax.transAxes, fontsize=14,

verticalalignment='top', bbox=props)


Which does an 'ok' job, at best, since fractions aren't converted properly.
 Is there a way to do it nicely short of using some horrendous regular
expressions?


Federico

Hi Chao,

I think I could make it work somehow but I thought that such a functionality
would be really nice to have in general. 

Tyrax



--
View this message in context: http://matplotlib.1069221.n5.nabble.com/Change-column-number-within-legend-tp41704p41709.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

On Aug 4, 2013, at 4:47 PM, Matthew Brett <mat...@be...> wrote:

> Hi,
> 
> Congratulations on the new release.
> 
> I just tried installing in a fresh installation of Python 2.7 from Python.org
> 
> Am I right in thinking the recommended method is:
> 
>    pip install matplotlib
> 
> ?  I did this, and then:
> 
>>>> import matplotlib.pyplot as plt
> Traceback (most recent call last):
>  File "<stdin>", line 1, in <module>
>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/pyplot.py",
> line 24, in <module>
>    import matplotlib.colorbar
>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/colorbar.py",
> line 29, in <module>
>    import matplotlib.collections as collections
>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/collections.py",
> line 23, in <module>
>    import matplotlib.backend_bases as backend_bases
>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/backend_bases.py",
> line 50, in <module>
>    import matplotlib.textpath as textpath
>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/textpath.py",
> line 14, in <module>
>    from matplotlib.mathtext import MathTextParser
>  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/mathtext.py",
> line 62, in <module>
>    import matplotlib._png as _png
> ImportError: dlopen(/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/_png.so,
> 2): Library not loaded: /usr/X11/lib/libpng12.0.dylib
>  Referenced from:
> /Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/_png.so
>  Reason: Incompatible library version: _png.so requires version
> 42.0.0 or later, but libpng12.0.dylib provides version 36.0.0
> 
> There were no previous versions of matplotlib installed.  This on OSX 10.6:
> 
> Python 2.7.5  just installed from python.org
> X11 installed from the Apple installation disks for 10.6 before `pip
> install matplotlib`
> No homebrew etc
> 
> Any hints about how I should debug this further?
> 
> Thanks a lot for any help,
> 
> Matthew

I was able to install numpy 1.7.1 and matplotlib 1.30 under OS X 10.6.8 with X11 version XQuartz 2.3.6 (xorg-server 1.4.2-apple56) and python 2.7.5 64-bit from python.org using pip 1.4.

Have you tried running Software Update?  If you installed X11 after updating to 10.6.8 you can run the combo updater ( http://support.apple.com/kb/DL1399 ) to get the latest X11.   Alternately, you could try installing XQuartz from http://xquartz.macosforge.org/landing/ 

hth,
Scott