You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have looked into the documentation and believe that some problems is the result of the axes folder and its structure. I am not sure what is supposed to be automatic and what is not in sphinx but axes.SubplotBase is for example not included in the documentation even though most axes used in matplotlib are probably of that class.
.. automodule:: matplotlib.axes does not work even though Axes and SubplotBase exist in that module. .. automodule:: matplotlib.axes._subplots works as I would expect but the paths are wrong.
from matplotlib.axes import SubplotBase, Axes works but the class of actual objects are using the full path, for example matplotlib.axes._subplots.AxesSubplot
I don't know much of module structures or module aliases but I think it is kind of inconvenient when different paths are used in different places and I believe that more work than needed is necessary to write the documentation correctly. See for example #11443.
Is it possibly to get a module structure such that the obj.__class__ path is the same as the alias path, i.e. axes.Axes is always used instead of axes._axes.Axes?
The text was updated successfully, but these errors were encountered:
fredrik-1
changed the title
The axes modul structure
The axes module structure
Jun 16, 2018
One can always readjust cls.__module__ post-hoc; setting __all__ appropriately may also help with sphinx.
(A single axes.py would be sufficiently enormous that I think keeping a split implementation is more manageable.)
This is a question or feature request.
I have looked into the documentation and believe that some problems is the result of the axes folder and its structure. I am not sure what is supposed to be automatic and what is not in sphinx but
axes.SubplotBase
is for example not included in the documentation even though mostaxes
used in matplotlib are probably of that class... automodule:: matplotlib.axes
does not work even thoughAxes
andSubplotBase
exist in that module... automodule:: matplotlib.axes._subplots
works as I would expect but the paths are wrong.from matplotlib.axes import SubplotBase, Axes
works but the class of actual objects are using the full path, for examplematplotlib.axes._subplots.AxesSubplot
I don't know much of module structures or module aliases but I think it is kind of inconvenient when different paths are used in different places and I believe that more work than needed is necessary to write the documentation correctly. See for example #11443.
Is it possibly to get a module structure such that the
obj.__class__
path is the same as the alias path, i.e.axes.Axes
is always used instead ofaxes._axes.Axes
?The text was updated successfully, but these errors were encountered: