On Fri, 27 Apr 2001, Jarkko Hietaniemi wrote:
> What's the thinking now on the 'utf8' vs 'utfebcdic' issue? I'm
> slowly starting to lean towards 'utfebdic'. If it ain't UTF-8, we
> shouldn't call it 'utf8'.
I think that Nick did not want to see such distinction as to have to
start utf8 aware programs or modules with
BEGIN {
if (ord("A") == 193) {
use utfebcdic;
}
else {
use utf8;
}
}
Hence there may be yet another possibility: how about a rename to
something like:
mv utf8.pm wide_characters.pm
or some similar name such as F<wide_chars.pm>, F<uni_chars.pm>, or ...
somesuch. Then the applications could say:
use wide_chars;
We might even have symlinks, or copies for the symlinkless, such as:
utf8.pm -> wide_chars.pm
utfebcdic.pm -> wide_chars.pm
So that any script that went through the convolution of the BEGIN
block above, or a simple:
use utf8;
could work "as expected" - provided of course that you are advised not to
expect UTF-8 on EBCDIC platforms. Reasonable? In violation of the
Camel-III? Comments?
Peter Prymmer
