Re: [v3] container requirements cleanups

[Benjamin Kosnik <bkoz at redhat dot com>]

> For the record, that was certainly on purpose: if you go through the
> archive you will find a number of statements of mine to the effect
> that since we would switch to a completely different implementation
> in C++0x and the current one is also so intertwined with the rest of
> the *.so, we would rather not experiment with the C++0x features, eg,
> I decided not to add move constructor and assignment. Lately Jason
> "broke" that with initializer_list things, I pointed out the
> reasoning, but was too late... Really, if at all possible, let's not
> overload the refcounted implementation.

Thanks for the clarification.

Sorry, I did not realize this. How do you propose going about
using __versa_string as std::string (and presumably moving the existing
std::string code to __gnu_cxx::basic_string) on trunk? I am in support
of this change, but consider this a separate issue.

The status quo is untenable: C++0x public mf definitions that are extern
template suppressed and unexported. Unusable. I think it makes sense to
audit and export the new bits. Do you agree, agree with string
special-cased, or other?

-benjamin