I have a weird problem with Unicode strings. I am using windows 7 with
Python 2.7.2 64 bit, and matplotlib 1.2.1 (package from Unofficial Windows
Binaries at Archived: Python Extension Packages for Windows - Christoph Gohlke ).
When i run:
-*- coding: utf-8 -*-
import matplotlib.pyplot as plt
fig = plt.figure( )
ax = fig.add_subplot(111)
ax.hist([1,2,3,4,5,6,7,8,9,10], bins=range(1,11) )
plt.ylabel(u'ä')
fig.savefig('test.eps')
I get a crash:
pythonw -u "bugtest.py"
Exit code: -1073740777
but if i modify my code to read:
-*- coding: utf-8 -*-
import matplotlib.pyplot as plt
fig = plt.figure( )
ax = fig.add_subplot(111)
ax.hist([1,2,3,4,5,6,7,8,9,10], bins=range(1,11) )
plt.ylabel(u'äł')
fig.savefig('test.eps')
Then no longer crashes. Why is it that "a umlaut" alone crashes
but when the string contains a "l stroke" character the Unicode string
behaves fine. In both cases it works fine if I show the plot but not when
I write eps either in code or when saving.