[capi-sig] Re: associate lifetime of PyMethodDef[] and PyTypeObject

On 2019-06-12 03:42, Joe Jevnik via capi-sig wrote:

One idea I have is to change PyCFunctionObject to hold a PyMethodDef *value* instead of a pointer.

We might as well go further and rethink how we store the data provided by PyMethodDef. For example, we could store the function name as Python str object (instead of a C string). I did that in my PEP 580 implementation (which was rejected) and it simplified some code.