I prefer the tables as-is. The _ne_ method does exist; it has just been moved farther along the _mro_. From a user's point of view, the method still flows automatically from the abstract methods.

There are several methods which "exist" because they are inherited from the object class.

>>> sorted(set(object.__dict__) - set(collections.abc.Mapping.__dict__))
['__class__', '__delattr__', '__dir__', '__format__', '__ge__', '__getattribute__', '__gt__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__']

But there are no __lt__, __reduce__ or __format__ in the table.

I now think that it would be better to keep __ne__ in the table, but change the code (bpo-41416).

I now think that it would be better to keep __ne__ in the table, but change the code (bpo-41416).

@serhiy-storchaka So shall we close this PR (and bpo-41400) in favour of doing bpo-41416 instead?

When you're done making the requested changes, leave the comment: I have made the requested changes; please review again.