preprocessor bug: @ + MACRO does not work in gcc-3.0

[Neil Booth]

Zack Weinberg wrote:-

> > 1) Is it a good idea to introduce yet another token type, when CPP_OTHER
> >    would do (from what I can see)?
> 
> It makes life easier for the front end, which doesn't have access to
> the cpp_token at the point we have to detect '@'.  And it's
> consistent; all other significant punctuators get their own token types.

Good point.

Neil.