But what about winelib apps? Isn't it likely that there'll be a few
assumptions about signed chars made in those, and shouldn't we generate
makefiles for them that just make all chars signed, to ease porting?
And if we do that, why not do it inside wine, too?
Josh.
Alexandre Julliard wrote:
> "Eric Pouech" <[EMAIL PROTECTED]> writes:
>
> > just a quick thought
> > wouldn't it be better to change the default sign-ness of char on PPC ?
> > All code has been written with char being signed in mind. there may be
> > lots of other spots where the assumption is made and will have to be
> > delt with. also supporting both signed and unsigned char may be a heavy
> > process.
>
> IMO it would be better to define CHAR as 'signed char', which should
> address most of the issues WRT the windows API. And where our code
> uses 'char' to mean signed char this is a bug that should be fixed.
>
> --
> Alexandre Julliard
> [EMAIL PROTECTED]
