版本升级后在matplotlib中没有绘图

Question

I just updated matplotlib to 1.1.0 on a server running ubuntu 10.04 LTS in order to play better with pandas. Pandas was converting my index according the functionality of a different version of matplotlib. I installed on one server using "easyinstall -U matplotlib" and "pip install -U matplotlib" on the other.

I cannot plot in any of my previous working scripts or in ipython. show() has stopped working in pylab. Could someone point me in the direction of what may be broken?

I took the following test script "simple_plot.py" from matplotlib's site and tested it on various servers after first deleting my config directory.

from pylab import *
plot([1,2,3])
show()

simple_plot produces a plot on all servers that are running on versions <= 0.99 but has no output on version 1.1.

here is the debug output on one server that does not work:

$HOME=/home/michael
CONFIGDIR=/home/michael/.matplotlib
matplotlib data path /usr/local/lib/python2.6/dist-packages/matplotlib-1.1.0-py2.6-linux-x86_64.egg/matplotlib/mpl-data
loaded rc file /usr/local/lib/python2.6/dist-packages/matplotlib-1.1.0-py2.6-linux-x86_64.egg/matplotlib/mpl-data/matplotlibrc
matplotlib version 1.1.0
verbose.level debug
interactive is False
platform is linux2
Using fontManager instance from /home/michael/.matplotlib/fontList.cache
backend agg version v2.2
python version:2.6.5
findfont: Matching :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=medium to Bitstream Vera Sans (/usr/local/lib/python2.6/dist-packages/matplotlib-1.1.0-py2.6-linux- x86_64.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf) with score of 0.000000

Answer 1

You are using the Agg backend, which does not show the figure on the display. This is the default for matplotlib. You need to change your backend in your matplotlib configuration file (usually ~/.matplotlib/matplotlibrc). Look for the part:

backend      : Agg

And replace 'Agg' with one of: GTKAgg, Qt4Agg, TkAgg, WXAgg. You may not have all of these (or any!) installed in your system, so try one that works.

If you don't have the file ~/.matplotlib/matplotlibrc, then copy it from your main configuration:

cp /usr/local/lib/python2.6/dist-packages/matplotlib-1.1.0-py2.6-linux-x86_64.egg/matplotlib/mpl-data/matplotlibrc ~/.matplotlib/matplotlibrc