This is the mail archive of the
mailing list for the GCC project.
Re: gcc compile-time (multibyte issue)
- From: dewar at gnat dot com (Robert Dewar)
- To: gcc at gcc dot gnu dot org, starner at okstate dot edu
- Date: Sun, 19 May 2002 18:37:21 -0400 (EDT)
- Subject: Re: gcc compile-time (multibyte issue)
<<It would be nice to be error tolerant in comments, though. I think most
use of non-ASCII text in source code is in comments, and the compiler
doesn't care either way.
You are not allowed to be tolerant in an Ada compiler.
<<All text files are encoded in some character set. All code that GCC
would be expected to compile is in an ASCII superset, where any ASCII
file is a valid file with the same meaning in that character set. Given
that GCC should be able to use the locale character set without problem,
assuming that multibyte character set reading is fast enough.
This is only true if you consider ASCII to be the lower half (00-7F)
Ada programs are typically written in Latin-1, and that is the default
character set for most programs. Latin-1 and (for example) shift-JIS
are not compatible, you have to know which of the two you are scanning.