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
As noted in #4264, when a Model subclass defined the bounding_box property or method, if it has its own docstring that docstring ends up being lost--this is because of how the metaclass swallows the bounding_box member of the subclass and replaces it with the base Model.bounding_box property.
See for example Gaussian2D.
This actually has two negative effects:
Gaussian2D.bounding_box.__doc__ displays the docstring for Model.bounding_box, not
the docstring for Gaussian2D.bounding_box.
It is impossible to reference Gaussian2D.bounding_box in Sphinx. It results in a missing reference.
For example this line was originally a direct reference, but it didn't work. That should be fixed.
This is just a documentation issue, but is worth fixing. I think it can be fixed by slightly going back on some of my changes in #4236 and make it so that each subclass does get its own property object in its __dict__ for bounding_box, rather than just referring directly to Model.property. It's slightly wasteful, but only slightly.
The text was updated successfully, but these errors were encountered:
As noted in #4264, when a
Model
subclass defined thebounding_box
property or method, if it has its own docstring that docstring ends up being lost--this is because of how the metaclass swallows thebounding_box
member of the subclass and replaces it with the baseModel.bounding_box
property.See for example
Gaussian2D
.This actually has two negative effects:
Gaussian2D.bounding_box.__doc__
displays the docstring forModel.bounding_box
, notthe docstring for
Gaussian2D.bounding_box
.Gaussian2D.bounding_box
in Sphinx. It results in a missing reference.For example this line was originally a direct reference, but it didn't work. That should be fixed.
This is just a documentation issue, but is worth fixing. I think it can be fixed by slightly going back on some of my changes in #4236 and make it so that each subclass does get its own
property
object in its__dict__
forbounding_box
, rather than just referring directly toModel.property
. It's slightly wasteful, but only slightly.The text was updated successfully, but these errors were encountered: