Re: [gtkmm] Does Gtkmm, Glibmm work on systems where sizeof(wchar_t) < sizeof(gunichar)?

[Anastasios Bassoukos <abas aix meng auth gr>]

On Wed, 2004-03-24 at 16:31, Murray Cumming wrote:
> On Wed, 2004-03-24 at 10:08, Anastasios Bassoukos wrote:
> > On Tue, 2004-03-23 at 23:56, Paul Elliott wrote:
> > 
> > > My original question is: "Does gtkmm, glibmm run on any systems
> > > where sizeof(wchar_t) == 2?" (I want my code to use another 
> > > library  (boost::regex) that needs these characters to be in wchar_t.)
> > 
> > Yes. If you can accept the fact that your program will be broken for
> > uncommon locales, you can simply assign gunichars to wchar_t. 
> 
> I think there is almost certainly some encoding conversion to do here,
> rather than just forcing characters into different bytes. There's more
> to a encoding than how bit its bytes are, or whether they are multibyte.

Hmm... IIRC, UCS-2 and UCS-4 should overlap for the lower 64k
characters; That's why I mentioned that it will be broken: it will work
for most of the languages he will work with, but it will break silently.

> If it was me, I would want to know what encoding the library uses, and
> whether that encoding can support the locales that I want.

I've just skimmed the documentation for boost::regex, and while it
accepts the character type as a template parameter, there seem to be
some other localization issues that complicate the issue.

What does regexxer use ?

Tassos
-- 
Beware of he who would deny you access to information, for in his
heart he dreams himself your master." 
	-- Commissioner Pravin Lal, Sid Meier's Alpha Centauri