What version of matplotlib are you referring to? master has a similar cycle, but it isn't problematic (and the line number you refer to doesn't match). Can you attach a traceback if you are getting one?

I'm using 1.1, downloaded a few days ago
there is no traceback to speak of as the call to rcParams is encapsulated
in a try/except clause.

I've commented that clause to obtain as follows

c:\code\trunk\OPL\opl\test>plottemporal.py
Traceback (most recent call last):
File "C:\code\trunk\OPL\opl\test\plottemporal.py", line 11, in
from opl.test.plot import iTestPlot, iTestShownDrawBase, testPlotFactory
File "c:\code\trunk\OPL\opl\test\plot.py", line 4, in
from opl.draw.canvas import SimpleCanvas
File "c:\code\trunk\OPL\opl\draw__init__.py", line 139, in
import matplotlib.cbook as cbook
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib__init__.py",
line 133, in
from matplotlib.rcsetup import (defaultParams,
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib\rcsetup.py",
line 19, in
from matplotlib.colors import is_color_like
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib\colors.py",
line 54, in
import matplotlib.cbook as cbook
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib\cbook.py",
line 34, in
matplotlib.rcParams['axes.formatter.use_locale']).strip()
AttributeError: 'module' object has no attribute 'rcParams'

On Wed, Nov 16, 2011 at 8:25 PM, Michael Droettboom <
[email protected]

wrote:

What version of matplotlib are you referring to? master has a similar
cycle, but it isn't problematic (and the line number you refer to doesn't
match). Can you attach a traceback if you are getting one?

---

Reply to this email directly or view it on GitHub:
#593 (comment)

My apologies, I realize I read your mail too quickly:
You say the cycle is not a problem. I believe it is, although a try
except clause does hide this.

The bug is that matplotlib.rcParams raises an AttributeError. The exception
is caught but the result is that the module variable "preferedencoding" is
not initialized correctly. In turn, this later results in UnicodeError
when, for exemple, one tries to plot dates.

Regards,

On Wed, Nov 16, 2011 at 10:01 PM, Pol d'Avezac [email protected]:

I'm using 1.1, downloaded a few days ago
there is no traceback to speak of as the call to rcParams is encapsulated
in a try/except clause.

I've commented that clause to obtain as follows

c:\code\trunk\OPL\opl\test>plottemporal.py
Traceback (most recent call last):
File "C:\code\trunk\OPL\opl\test\plottemporal.py", line 11, in
from opl.test.plot import iTestPlot, iTestShownDrawBase, testPlotFactory
File "c:\code\trunk\OPL\opl\test\plot.py", line 4, in
from opl.draw.canvas import SimpleCanvas
File "c:\code\trunk\OPL\opl\draw__init__.py", line 139, in
import matplotlib.cbook as cbook
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib__init__.py",
line 133, in
from matplotlib.rcsetup import (defaultParams,
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib\rcsetup.py",
line 19, in
from matplotlib.colors import is_color_like
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib\colors.py",
line 54, in
import matplotlib.cbook as cbook
File
"C:\code\environment\python\v27\lib\site-packages\matplotlib\cbook.py",
line 34, in
matplotlib.rcParams['axes.formatter.use_locale']).strip()
AttributeError: 'module' object has no attribute 'rcParams'

On Wed, Nov 16, 2011 at 8:25 PM, Michael Droettboom <
[email protected]

wrote:

What version of matplotlib are you referring to? master has a similar
cycle, but it isn't problematic (and the line number you refer to doesn't
match). Can you attach a traceback if you are getting one?

---

Reply to this email directly or view it on GitHub:
#593 (comment)

Thanks for clarifying. I'll have a fix for this shortly.

@poldavezac: When you get a chance, can you confirm this patch fixes the issue for you?

Yes it works

Thanks,

On Thu, Nov 17, 2011 at 5:46 PM, Michael Droettboom <
[email protected]

wrote:

@poldavezac: When you get a chance, can you confirm this patch fixes the
issue for you?

---

Reply to this email directly or view it on GitHub:
#593 (comment)

@mdboom: can you merge it into the master branch? currently it gives a merge conflict.

@leejjoon: Done