matplotlib-devel Mailing List for matplotlib

Hi everyone!
I'm trying to build matplotlib 1.1.0 using Python-2.7.3 on CentOS 6.4. I
constantly keep getting the following error:
/usr/bin/ld: /usr/local/lib/libpython2.7.a(parsetok.o): relocation
R_X86_64_32 against `.rodata.str1.1' can not be used when making a shared
object; recompile with -fPIC
/usr/local/lib/libpython2.7.a: could not read symbols: Bad value
collect2: ld returned 1 exit status
error: command 'g++' failed with exit status 1
Can anyone please advice on what is going on and how I can resolve the
issue? Thanks a lot.
Regards,
Ali
--
View this message in context: http://matplotlib.1069221.n5.nabble.com/Error-in-building-Matplotlib-tp41968.html
Sent from the matplotlib - devel mailing list archive at Nabble.com.

On Mon, Sep 2, 2013 at 12:27 AM, Filipe Saraiva <ma...@fi...>wrote:
> Hello,
>
> First, thanks for this great library.
>
> My name is Filipe Saraiva, I am developing a python backend for Cantor,
> the KDE mathematical software. More infos can be read in
> http://blog.filipesaraiva.info/?tag=gsoc2013-python-backend (in
> portuguese and english).
>
> Currently I have a problem when I try import pyplot in Cantor. I am using
> Python 2.7.5 and matplotlib 1.3.0. The error is below:
>
> import matplotlib.pyplot as plt
> Traceback (most recent call last):
> File "<string>", line 1, in <module>
> File "/usr/lib64/python2.7/site-packages/matplotlib/pyplot.py", line 98,
> in <module>
> _backend_mod, new_figure_manager, draw_if_interactive, _show =
> pylab_setup()
> File
> "/usr/lib64/python2.7/site-packages/matplotlib/backends/__init__.py", line
> 25, in pylab_setup
> globals(),locals(),[backend_name])
> File
> "/usr/lib64/python2.7/site-packages/matplotlib/backends/backend_qt4agg.py",
> line 19, in <module>
> _decref = ctypes.pythonapi.Py_DecRef
> File "/usr/lib64/python2.7/ctypes/__init__.py", line 378, in __getattr__
> func = self.__getitem__(name)
> File "/usr/lib64/python2.7/ctypes/__init__.py", line 383, in __getitem__
> func = self._FuncPtr((name_or_ordinal, self))
> AttributeError: kde/bin/cantor: undefined symbol: Py_DecRef
>
> Well, anyone have any idea about how can I fix it?
>
> Thank you,
>
>
My only guess is that there is some sort of linking/build error. Perhaps
the python-qt4 library was built and linked against a different python on
your system?
Cheers!
Ben Root

Sorry if this is a double post; used the wrong email in the first reply and
got at least one failure notice...
This is terminal IPython and IPython Notebook. (I noticed it first in the
IPython Notebook). The backends are the same in both Python and IPython
(and IPython Notebook). I have not changed any of the default IPython
settings.
I'm wondering if this is a path issue of some sorts...namely IPython is
picking up an old version of geos that I cannot seem to find. At least this
is the angle I'm currently trying to work on. I should add that I've remove
all my manually installed modules and "started over". I built Numpy,
Matplotlib, Basemap, and then IPython (in that order), and still have the
same segfault issue.
Patrick
On Tue, Sep 3, 2013 at 1:15 PM, MinRK <ben...@gm...> wrote:
> This is terminal IPython?
>
> Is there a chance that the matplotlib backend is different in each case?
> Have you enabled matplotlib eventloop integration in IPython (otherwise, it
> will block when you draw a plot).
>
>
> On Tue, Sep 3, 2013 at 9:36 AM, Patrick Marsh <pat...@gm...>wrote:
>
>> Hi, All,
>>
>> I'm not sure what is going on, but the following fails in IPython
>> (terminal and notebook) but works just fine in regular Python.
>>
>> I'm using OS X 10.8.4; git master for git master for Matplotlib
>> (a091f6d), IPython (9f92804), and Basemap (1d7664c); and geos version 3.4.2
>> (built by Homebrew).
>>
>> The following script will work from a Python prompt but fails (segfaults)
>> from the IPython prompt. After doing some digging it appears that the
>> segfault occurs when accessing some of the methods associated with the
>> generated geos module (_geoslib.so). I've rolled back to previous versions
>> of everything and I still have the same issues.
>>
>> What I cannot figure out is why this would work with a pure Python
>> interpreter, but fail in the IPython console.
>>
>>
>> =====
>> import matplotlib.pyplot as plt
>> from mpl_toolkits.basemap import Basemap
>>
>> m = Basemap() # This is where the segfault occurs
>>
>> m.drawcoastlines()
>> plt.show()
>> =====
>>
>>
>> I originally posted this on the IPython Users list and got no response,
>> so I thought I would try here before posting a bug on both IPython and
>> Basemap's Github issue trackers.
>>
>>
>> Thanks for any help or ideas on fixing this issue, or at least on how to
>> track down this issue...
>>
>>
>> Patrick
>>
>> _______________________________________________
>> IPython-dev mailing list
>> IPy...@sc...
>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>
>>
>
> _______________________________________________
> IPython-dev mailing list
> IPy...@sc...
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>

This is terminal IPython?
Is there a chance that the matplotlib backend is different in each case?
Have you enabled matplotlib eventloop integration in IPython (otherwise, it
will block when you draw a plot).
On Tue, Sep 3, 2013 at 9:36 AM, Patrick Marsh <pat...@gm...>wrote:
> Hi, All,
>
> I'm not sure what is going on, but the following fails in IPython
> (terminal and notebook) but works just fine in regular Python.
>
> I'm using OS X 10.8.4; git master for git master for Matplotlib (a091f6d),
> IPython (9f92804), and Basemap (1d7664c); and geos version 3.4.2 (built by
> Homebrew).
>
> The following script will work from a Python prompt but fails (segfaults)
> from the IPython prompt. After doing some digging it appears that the
> segfault occurs when accessing some of the methods associated with the
> generated geos module (_geoslib.so). I've rolled back to previous versions
> of everything and I still have the same issues.
>
> What I cannot figure out is why this would work with a pure Python
> interpreter, but fail in the IPython console.
>
>
> =====
> import matplotlib.pyplot as plt
> from mpl_toolkits.basemap import Basemap
>
> m = Basemap() # This is where the segfault occurs
>
> m.drawcoastlines()
> plt.show()
> =====
>
>
> I originally posted this on the IPython Users list and got no response, so
> I thought I would try here before posting a bug on both IPython and
> Basemap's Github issue trackers.
>
>
> Thanks for any help or ideas on fixing this issue, or at least on how to
> track down this issue...
>
>
> Patrick
>
> _______________________________________________
> IPython-dev mailing list
> IPy...@sc...
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>

Hi, All,
I'm not sure what is going on, but the following fails in IPython (terminal
and notebook) but works just fine in regular Python.
I'm using OS X 10.8.4; git master for git master for Matplotlib (a091f6d),
IPython (9f92804), and Basemap (1d7664c); and geos version 3.4.2 (built by
Homebrew).
The following script will work from a Python prompt but fails (segfaults)
from the IPython prompt. After doing some digging it appears that the
segfault occurs when accessing some of the methods associated with the
generated geos module (_geoslib.so). I've rolled back to previous versions
of everything and I still have the same issues.
What I cannot figure out is why this would work with a pure Python
interpreter, but fail in the IPython console.
=====
import matplotlib.pyplot as plt
from mpl_toolkits.basemap import Basemap
m = Basemap() # This is where the segfault occurs
m.drawcoastlines()
plt.show()
=====
I originally posted this on the IPython Users list and got no response, so
I thought I would try here before posting a bug on both IPython and
Basemap's Github issue trackers.
Thanks for any help or ideas on fixing this issue, or at least on how to
track down this issue...
Patrick