_M_fill initialization in basic_ios::init

[Paolo Carlini]

Richard Burkert wrote:

> So I
> guess the problem is in initializing the fill character in
> this way, since there is no guarantee that the locale has
> the ctype facet.

Well, in the meanwhile I studied myself a little bit of locales (as I already
told you, I'm not an expert) and, I must say, I cannot find in the standard
support for this hypothesis of yours.

For instance, Langer/Kreft (0-201-18395-1), which I usually trust for iostreams
and locale issues (and is much more easier to read than the standard ;-)
explicitly says at the beginning of chapter 5 that ctype is a *standard* facet,
not a user-defined one, so it is "automatically contained in every locale".

So, please, come up with a self-contained seg-faulting testcase: it will help
focusing the real problem here.

Cheers,
Paolo.