On Thu, Apr 27, 2006 at 05:16:10PM -0700, Joe Buck wrote:
> On Thu, Apr 27, 2006 at 07:58:29PM -0400, Zack Weinberg wrote:
> [ Unicode, UTF-{8,16}, BOMs, etc ]
> > It would also be good to take advantage of the fact that 95+% of C
> > source files start with "/*", "//", "#i", or "#d" to distinguish
> > ASCII from EBCDIC. (This is in fact necessary in order to have any
> > hope of detecting and processing an editor's code page marker in an
> > EBCDIC source file.)
>
> Is this really an important problem? EBCDIC is still around (barely),
> but code page markers in EBCDIC source files? Perhaps I'm missing
> something.
I don't know how useful it would be to process -*- type marks in EBCDIC
source files, and I certainly wouldn't make any EBCDIC-related features
requirements for inclusion of a patch that met the original poster's
needs. However, I do think basic ASCII-vs-EBCDIC sniffing would be
useful in a mixed environment, and I know folks at IBM cared enough
about it to make EBCDIC input files work (in an appropriate locale or
with -finput-charset), a few years back.
zw
