> Or, is this a "statement of direction" that only GCC (syntax) compilers are (going to be) supported?

This is only the first step. Since this feature is compiler specific, we should add the support for every compiler separately. Arfrever suggested the syntax for supporting this on Microsoft compiler. But I don't know how to silence warnings on this compiler. Other compilers can support GCC syntax. We should check and switch on this.

> Which functions are bridges?

All deprecated Py_UNICODE related functions use Py_UNICODE. Most of them are bridges to new APIs. Py_UNICODE also is used in implementation of format codes 'u' and 'Z' of PyArg_Parse*. It also is used in many functions that are bridges to Windows API.

> "Py_UCS4 _PyUnicode_ToLowercase(Py_UCS4 ch)" doesn't use Py_UNICODE, what is the issue?

The problem with this function is not related to Py_UNICODE, but that correct Unicode mapping can return multiple characters.

> Do you want to deprecate functions like Py_UNICODE_ISDECIMAL(ch)?

No. It works correctly.