Re: [v3] Variadic templates, third revision (3/3)

[Paolo Carlini <pcarlini at suse dot de>]

Doug Gregor wrote:
> The compiler only complains about the use of variadic templates
> outside system headers, so we're free to use variadic templates in
> libstdc++; no #ifdef necessary.
Yeah!
> However, variadic templates will change the ABI of the TR1 facilities
> that use them. For example, instead of mangling "tuple<int, float>" as
> "tuple<int, float, _NullClass, _NullClass, _NullClass, _NullClass,
> _NullClass, _NullClass, _NullClass, _NullClass>" (which happens now),
> it will mangle as "tuple<int, float>." If this is a problem we'll
> either have to do what Benjamin suggests or (*shudder*) select between
> the variadic and non-variadic implementations based on
> __GXX_EXPERIMENTAL_CPP0X__.
In my opinion, and consistently with my personal "philosophical" point 
of view on TR1 features, already well known to the libstdc++ list, I 
don't think such complexities are necessary: only a note in the relevant 
(4.3, I hope ;) release notes, mentioning the ABI breakage caused, as a 
minor negative fall out, by the new up to date variadic templates-based 
implementation, will be more than enough.

Paolo.