This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: cpplib: locale-sensitive lexing


On Thu, Oct 25, 2001 at 09:34:45AM -0600, Tom Tromey wrote:
> 
> No, it doesn't.  The Java compiler uses Unicode internally (some
> places UTF-8, some places UCS-2).  My understanding is that the
> multibyte functions can't be used if you need a specific wide or
> multibyte encoding.

That's also my understanding.

> Note that the Java and C/C++ cases aren't entirely analogous.  In Java
> the execution character set is fixed.

I think it'd be reasonable for the C/C++ compiler to use Unicode
internally as well, and convert back to the user's desired execution
charset on output.

For interoperability, we want to say that identifiers are always
UTF-8, no exceptions (re-encoded if the assembler wants a more
restricted set).

zw


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]