matplotlib-devel Mailing List for matplotlib

On Saturday, June 4, 2011, Eric Firing <ef...@ha...> wrote:
> On 06/04/2011 06:05 AM, Benjamin Root wrote:
>> On Sat, Jun 4, 2011 at 9:09 AM, Phil Elson <phi...@ho...
>> <mailto:phi...@ho...>> wrote:
>>
>>
>>   The first line of code on the page
>>   http://matplotlib.sourceforge.net/devel/add_new_projection.html suggests
>>   that it is possible to give the projection directly to
>>   mpl.pyplot.plot but
>>   this does not work for me.
>>
>>   Should this functionality exist? I am aware that the pyplot.plot
>>   function is
>>   autogenerated by boilerplate.py, but I am a little weary of
>>   modifying that.
>>
>>   Any help gratefully received.
>>
>>
>> Hmm, this doesn't seem 100% logically correct. I think the only case
>> where this would work correctly is if the plot command would also
>> trigger the creation of a new axes object. If the axes object already
>> exists, then I seriously doubt this would work as advertised. Is that
>> the case with your code?
>>
>> Ben Root
>
> Ben,
>
> The first paragraph of that doc page needs repair. Trivially, there are
> two references to set_xscale. The problem pointed out above is more
> serious. plot() has no facility for handling a projection kwarg, and I
> don't think it would make sense to try to add one. pyplot functions
> that accept the projection kwarg are axes, subplot, and subplots.
>
> Do you want to fix the paragraph, or should I?
>
I can take a look at it tonight.
> Independently, I want to redo the FAQ discussion of draw(), show(), and
> interactive versus non-interactive mode. It is possible that this topic
> needs more attention elsewhere as well, but I want to start with the FAQ.
>
> Eric
>
Agreed, these are some fundamental functions and probably be set apart
in a "getting started" section, maybe?
Ben Root

On 06/04/2011 09:45 AM, Phil Elson wrote:
> > I think the only case where this would work correctly is if the plot
> command would also trigger the creation of a new axes object.
> Can't see how this can ever happen given the pyplot.plot code, which
> creates a standard axes without passing through any arguments.
>
> I agree that there are situations when it doesn't make sense to define
> the projection when doing a plt.plot, maybe best to fix the projections
> documentation instead of providing a keyword which is not always valid.
> Either that, or plt.plot with a projection always makes a new axes?
>
> Attempting to pass the projection currently to plot results in an exception:
>
> plt>>> plt.plot(range(10), projection='polar')
> ...
> TypeError: There is no line property "projection"
>
Yes, the behavior makes sense, the documentation doesn't; it will be fixed.
Eric

On 06/04/2011 09:08 AM, Eric Firing wrote:
> Independently, I want to redo the FAQ discussion of draw(), show(), and
> interactive versus non-interactive mode. It is possible that this topic
> needs more attention elsewhere as well, but I want to start with the FAQ.
I think I will move the "show" section entirely out of howto and put the 
revision in "usage".
Eric
>
> Eric

> I think the only case where this would work correctly is if the plot command would also trigger the creation of a new axes object. 
Can't see how this can ever happen given the pyplot.plot code, which creates a standard axes without passing through any arguments.
I agree that there are situations when it doesn't make sense to define the projection when doing a plt.plot, maybe best to fix the projections documentation instead of providing a keyword which is not always valid. Either that, or plt.plot with a projection always makes a new axes?
Attempting to pass the projection currently to plot results in an exception:
plt>>> plt.plot(range(10), projection='polar')
...
TypeError: There is no line property "projection"
 		 	 		 

On 06/04/2011 06:05 AM, Benjamin Root wrote:
> On Sat, Jun 4, 2011 at 9:09 AM, Phil Elson <phi...@ho...
> <mailto:phi...@ho...>> wrote:
>
>
> The first line of code on the page
> http://matplotlib.sourceforge.net/devel/add_new_projection.html suggests
> that it is possible to give the projection directly to
> mpl.pyplot.plot but
> this does not work for me.
>
> Should this functionality exist? I am aware that the pyplot.plot
> function is
> autogenerated by boilerplate.py, but I am a little weary of
> modifying that.
>
> Any help gratefully received.
>
>
> Hmm, this doesn't seem 100% logically correct. I think the only case
> where this would work correctly is if the plot command would also
> trigger the creation of a new axes object. If the axes object already
> exists, then I seriously doubt this would work as advertised. Is that
> the case with your code?
>
> Ben Root
Ben,
The first paragraph of that doc page needs repair. Trivially, there are 
two references to set_xscale. The problem pointed out above is more 
serious. plot() has no facility for handling a projection kwarg, and I 
don't think it would make sense to try to add one. pyplot functions 
that accept the projection kwarg are axes, subplot, and subplots.
Do you want to fix the paragraph, or should I?
Independently, I want to redo the FAQ discussion of draw(), show(), and 
interactive versus non-interactive mode. It is possible that this topic 
needs more attention elsewhere as well, but I want to start with the FAQ.
Eric

On Sat, Jun 4, 2011 at 11:09 AM, Benjamin Root <ben...@ou...> wrote:
> On Fri, Jun 3, 2011 at 6:15 PM, John Hunter <jd...@gm...> wrote:
>
>>
>>
>> On Fri, Jun 3, 2011 at 3:37 PM, Benjamin Root <ben...@ou...> wrote:
>>
>>> By the way, whoever is the one to push the docs up to sourceforge, you
>>> can now do so from the v1.0.x-maint branch. I have built it on my computer
>>> and all of my own changes appear correct. I couldn't test *everything* (my
>>> LaTeX setup isn't quite right), but everything I intended to change appears
>>> to be there.
>>>
>>> I just pushed the doc build from the maint branch up to sf. Take it for
>> a test drive. Thanks for pushing through on the doc fix patch.
>>
>> JDH
>>
>>
> No problem, glad to be of service. The pages look fine so far (although I
> want to see if I can get it to be more friendly to people with smaller
> screens such as an EeePC). There will be some additional work I will put in
> to improve the docs for the 1.1.0 release. If anybody has a wishlist or
> complaints about the current docs, just let me know.
>
> I have also tried building the docs from master after merging, and the
> debugger problem happened again using the latest from sphinx. I will look
> into it further to see if I can turn that "feature" off.
>
> Ben Root
>
Looks like it is a feature, not a bug. sphinx-build has a command-line
option "-P" which fires off pdb whenever there is an unhandled exception.
Removing that keeps the pdb from firing off, but this still doesn't solve
our problem. The multiprocessing pool does not exit out and the process
still hangs.
Right now, I am trying to figure out why the --small option to make.py is
causing an unhandled exception in master with the latest sphinx. It seems
like bad command-line parsing by sphinx.
Ben Root

On Fri, Jun 3, 2011 at 6:15 PM, John Hunter <jd...@gm...> wrote:
>
>
> On Fri, Jun 3, 2011 at 3:37 PM, Benjamin Root <ben...@ou...> wrote:
>
>> By the way, whoever is the one to push the docs up to sourceforge, you can
>> now do so from the v1.0.x-maint branch. I have built it on my computer and
>> all of my own changes appear correct. I couldn't test *everything* (my
>> LaTeX setup isn't quite right), but everything I intended to change appears
>> to be there.
>>
>> I just pushed the doc build from the maint branch up to sf. Take it for a
> test drive. Thanks for pushing through on the doc fix patch.
>
> JDH
>
>
No problem, glad to be of service. The pages look fine so far (although I
want to see if I can get it to be more friendly to people with smaller
screens such as an EeePC). There will be some additional work I will put in
to improve the docs for the 1.1.0 release. If anybody has a wishlist or
complaints about the current docs, just let me know.
I have also tried building the docs from master after merging, and the
debugger problem happened again using the latest from sphinx. I will look
into it further to see if I can turn that "feature" off.
Ben Root

On Sat, Jun 4, 2011 at 9:09 AM, Phil Elson <phi...@ho...> wrote:
>
> The first line of code on the page
> http://matplotlib.sourceforge.net/devel/add_new_projection.html suggests
> that it is possible to give the projection directly to mpl.pyplot.plot but
> this does not work for me.
>
> Should this functionality exist? I am aware that the pyplot.plot function
> is
> autogenerated by boilerplate.py, but I am a little weary of modifying that.
>
> Any help gratefully received.
>
Hmm, this doesn't seem 100% logically correct. I think the only case where
this would work correctly is if the plot command would also trigger the
creation of a new axes object. If the axes object already exists, then I
seriously doubt this would work as advertised. Is that the case with your
code?
Ben Root

The first line of code on the page
http://matplotlib.sourceforge.net/devel/add_new_projection.html suggests
that it is possible to give the projection directly to mpl.pyplot.plot but
this does not work for me.
Should this functionality exist? I am aware that the pyplot.plot function is
autogenerated by boilerplate.py, but I am a little weary of modifying that.
Any help gratefully received.
-- 
View this message in context: http://old.nabble.com/plt.plot-projection-kwarg-tp31772544p31772544.html
Sent from the matplotlib - devel mailing list archive at Nabble.com.