[Matplotlib-users] What changes from matplotlib 1.3.1 to matplotlib 1.5.1 break the following code?

Thu Mar 24 13:05:21 EDT 2016

The zero might have been a position, which zero means "best". I should
point out that we are not very strict about keeping backwards compatibility
for automated legend placement. So, I hope your tests aren't depending on
the legend to always be in the exact same place in the figure.

Try specifying "loc=0" to keep the zero in-play in that call.

On Thu, Mar 24, 2016 at 1:02 PM, Benjamin Root <ben.v.root at gmail.com> wrote:

> try taking out the zero in the legend() call. The call signature does not
> support more than 2 positional arguments, I don't think. I don't know what
> that zero would have mean.
>
> Most likely, we tightened our error-checking since the v1.3.1 days.
>
>
> On Thu, Mar 24, 2016 at 12:52 PM, Jacob Barhak <jacob.barhak at gmail.com>
> wrote:
>
>> Sorry Paul,
>>
>> The 5 lines with DB in them should ignored. So please comment the 5 lines
>> after the line that things broke in - it is a file that MIST uses and is
>> not needed for this script. You may also make sure the is a Temp sub
>> directory below the execution path since the pdf filename should be stored
>> there.
>>
>> Yet after making those fixes, you will find out that the script works on
>> matplotlib 1.3.1 and breaks on matplotlib 1.5.1. with the following error:
>>
>> Traceback (most recent call last):
>>   File "PlotGeneration_Mod.py", line 21, in <module>
>>     TheLegend = HandleAxes.legend(LineHandles, LabelHandles, 0, prop =
>> HandleSma
>> llerFont, handlelength = 2.0/1.0)
>>   File "C:\Anaconda\lib\site-packages\matplotlib\axes\_axes.py", line
>> 536, in le
>> gend
>>     raise TypeError('Invalid arguments to legend.')
>> TypeError: Invalid arguments to legend.
>>
>>
>> Hopefully this behavior can be explained with changes made.
>>
>> I am unsure I wish to upgrade the code, I am considering tow solutions 1)
>> keeping the old matplotlib version, 2) making some code changes to match
>> the newer version.
>>
>> Your explanation will help me decide on the best solution.
>>
>>
>>                  Jacob
>>
>> On Thu, Mar 24, 2016 at 10:45 AM, Paul Hobson <pmhobson at gmail.com> wrote:
>>
>>> Can you trim this down to a reproducible example?
>>>
>>> It's hard to debug a "this doesn't work" question when I can't get past
>>> the imports ("import DataDef as DB")
>>>
>>> At the very least, could you tell use where the error is occurring and
>>> what the error message is?
>>> -paul
>>>
>>> On Thu, Mar 24, 2016 at 1:02 AM, Jacob Barhak <jacob.barhak at gmail.com>
>>> wrote:
>>>
>>>> Greetings,
>>>>
>>>> Are there specific new changes in matplotlib 1.5.1 that break previous
>>>> code?
>>>>
>>>> Below is a script that works nicely with matplotlib 1.3.1 and breaks on
>>>> matplotlib 1.5.1 that comes with the new anaconda on Windows.
>>>>
>>>> I am trying to decide if a downgrade of the matplotlib version is the
>>>> best solution or is there another simple fix that will make the code work
>>>> again?
>>>>
>>>> Note that this is code generated by the MIST system test code, so do
>>>> not look for logical output - the current goal is just to pass the test by
>>>> not breaking on any commands.
>>>>
>>>> I will appreciate your insight.
>>>>
>>>>               Jacob
>>>>
>>>>
>>>> ############# Code that Breaks with 1.5.1 ##############
>>>>
>>>> from __future__ import division
>>>> import matplotlib
>>>> matplotlib.use('PDF')
>>>> import matplotlib.pyplot as plt
>>>> import matplotlib.font_manager
>>>> import matplotlib.backends.backend_pdf
>>>> import DataDef as DB
>>>> Inf = DB.Inf
>>>> NaN = DB.NaN
>>>> inf = DB.inf
>>>> nan = DB.nan
>>>> HandlePDF =
>>>> matplotlib.backends.backend_pdf.PdfPages('Temp\SelectedAssembledPlots.pdf')
>>>> HandleFigure = plt.figure()
>>>> HandleAxes = HandleFigure.add_subplot(111)
>>>> DefaultLegendFontSize =
>>>> matplotlib.font_manager.FontProperties(size=matplotlib.rcParams['legend.fontsize']).get_size_in_points()
>>>> # New plot sequence
>>>> HandleAxes.clear()
>>>> HandleAxes.clear()
>>>> HandleAxes.set_title('Age - Avg All')
>>>> HandleAxes.set_xlabel('Time')
>>>> HandleAxes.plot( [0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3]
>>>> ,[None, 30.0, None, 30.0, None, 31.0, None, 31.0, 32.0, 32.0, 32.0, 32.0,
>>>> 33.0, 33.0, 33.0, 33.0], 'ko-' , label = 'File 1')
>>>> HandleAxes.plot( [] ,[], 'k:' , label = 'File 2')
>>>> HandleSmallerFont = matplotlib.font_manager.FontProperties(size=
>>>> DefaultLegendFontSize*1.0 )
>>>> LineHandles, LabelHandles = HandleAxes.get_legend_handles_labels()
>>>> TheLegend = HandleAxes.legend(LineHandles, LabelHandles, 0, prop =
>>>> HandleSmallerFont, handlelength = 2.0/1.0)
>>>> TheLegend.get_frame().set_alpha(0.5)
>>>> HandlePDF.savefig(HandleFigure)
>>>> HandleAxes.clear()
>>>> HandleAxes.set_title("""Create Plots Warning: No parameter entry :
>>>> ('Alive', 'Avg All', '')""")
>>>> HandlePDF.savefig(HandleFigure)
>>>> HandleAxes.clear()
>>>> HandleAxes.set_title(' - Rec Count')
>>>> HandleAxes.set_xlabel('Time')
>>>> HandleAxes.plot( [0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3]
>>>> ,[None, 900, None, 1000, None, 900, None, 900, 63, 835, 772, 835, 61, 772,
>>>> 711, 772], 'ko-' , label = 'File 1')
>>>> HandleAxes.plot( [] ,[], 'k:' , label = 'File 2')
>>>> HandleSmallerFont = matplotlib.font_manager.FontProperties(size=
>>>> DefaultLegendFontSize*1.0 )
>>>> LineHandles, LabelHandles = HandleAxes.get_legend_handles_labels()
>>>> TheLegend = HandleAxes.legend(LineHandles, LabelHandles, 0, prop =
>>>> HandleSmallerFont, handlelength = 2.0/1.0)
>>>> TheLegend.get_frame().set_alpha(0.5)
>>>> HandlePDF.savefig(HandleFigure)
>>>> # New plot sequence
>>>> HandleAxes.clear()
>>>> HandleAxes.set_title("""Create Plots Warning: No parameter entry :
>>>> ('Alive', 'Avg All', '')""")
>>>> HandlePDF.savefig(HandleFigure)
>>>> HandleAxes.plot( [None, 30.0, None, 30.0, None, 31.0, None, 31.0, 32.0,
>>>> 32.0, 32.0, 32.0, 33.0, 33.0, 33.0, 33.0] ,[None, 900, None, 1000, None,
>>>> 900, None, 900, 63, 835, 772, 835, 61, 772, 711, 772], 'ko-' , label =
>>>> 'File 1 - nested record count')
>>>> HandleAxes.plot( [] ,[], 'k:' , label = 'File 2 - nested record count')
>>>> HandleSmallerFont = matplotlib.font_manager.FontProperties(size=
>>>> DefaultLegendFontSize*0.714285714286 )
>>>> LineHandles, LabelHandles = HandleAxes.get_legend_handles_labels()
>>>> TheLegend = HandleAxes.legend(LineHandles, LabelHandles, 0, prop =
>>>> HandleSmallerFont, handlelength = 2.0/0.714285714286)
>>>> TheLegend.get_frame().set_alpha(0.5)
>>>> HandlePDF.savefig(HandleFigure)
>>>> HandleAxes.clear()
>>>> HandleAxes.set_title("""Create Plots Warning: No parameter entry :
>>>> ('AnError', 'Avg All', '')""")
>>>> HandlePDF.savefig(HandleFigure)
>>>> HandleAxes.clear()
>>>> HandleAxes.set_title("""Create Plots Warning: No parameter entry :
>>>> ('Alive', 'AlsoAnError', '')""")
>>>> HandlePDF.savefig(HandleFigure)
>>>> HandlePDF.close()
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Matplotlib-users mailing list
>>>> Matplotlib-users at python.org
>>>> https://mail.python.org/mailman/listinfo/matplotlib-users
>>>>
>>>>
>>>
>>
>> _______________________________________________
>> Matplotlib-users mailing list
>> Matplotlib-users at python.org
>> https://mail.python.org/mailman/listinfo/matplotlib-users
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-users/attachments/20160324/f8649a34/attachment-0001.html>