(original) (raw)

On Sat, 3 Mar 2012 12:59:13 -0800
Thomas Wouters <thomas@python.org> wrote:
>
> Why even have separate tp\_as\_sequence and tp\_as\_mapping anymore? That
> particular distinction never existed for Python types, so why should it
> exist for C types at all? I forget if there was ever a real point to it,
> but all it seems to do now is create confusion, what with many sequence
> types implementing both, and PyMapping\_Check() and PySequence\_Check() doing
> seemingly random things to come up with somewhat sensible answers.

Ironically, most of the confusion stems from sequence types
implementing the mapping protocol for extended slicing.

> Do note
> that the dict type actually implements tp\_as\_sequence (in order to support
> containtment tests) and that PySequence\_Check() has to explicitly return 0
> for dicts -- which means that it will give the "wrong" answer for another
> type that behaves exactly like dicts.

It seems to be a leftover:

int
PySequence\_Check(PyObject \*s)
{
if (PyDict\_Check(s))
return 0;
return s != NULL && s->ob\_type->tp\_as\_sequence &&
s->ob\_type->tp\_as\_sequence->sq\_item != NULL;
}

Dict objects have a NULL sq\_item so even removing the explicit check
would still return the right answer.

> Getting rid of the misleading distinction seems like a much better idea
> than trying to re-conflate some of the issues.

This proposal sounds rather backwards, given that we now have separate
Mapping and Sequence ABCs.