Re: Undocumented internal function

[Jordi Gutiérrez Hermoso]

On 1 September 2010 14:08, John W. Eaton <address@hidden> wrote:
> What we are saying is that if we could, we would make these functions
> inaccessible to ordinary uses, similar to the way C++ classes have
> private methods.

Private members and functions are usually documented if the public
ones are. Documenting them isn't an invitation to try to subvert the
access specifiers of C++.

If the current API demands a docstring, why not give it? Or would you
prefer to create a PRIVDEFUN macro that makes the function uncallable
from the interpreter and doesn't let readline complete its name? Even
so, this isn't an excuse to not document it.