| 1 |
wakaba |
1.1 |
=head1 NAME
|
| 2 |
|
|
|
| 3 |
|
|
Encode::Unicode::Escape --- Encode/decode of Unicode escape (x-u-escaped)
|
| 4 |
|
|
of Java
|
| 5 |
|
|
|
| 6 |
|
|
=head1 ENCODINGS
|
| 7 |
|
|
|
| 8 |
|
|
=over 4
|
| 9 |
|
|
|
| 10 |
|
|
=cut
|
| 11 |
|
|
|
| 12 |
|
|
package Encode::Unicode::Escape;
|
| 13 |
|
|
use strict;
|
| 14 |
|
|
use vars qw($VERSION);
|
| 15 |
|
|
$VERSION=do{my @r=(q$Revision: 1.2 $=~/\d+/g);sprintf "%d."."%02d" x $#r,@r};
|
| 16 |
|
|
|
| 17 |
|
|
use base qw(Encode::Encoding);
|
| 18 |
|
|
__PACKAGE__->Define (qw/unicode-escape x-u-escaped/);
|
| 19 |
|
|
|
| 20 |
|
|
=item unicode-escape
|
| 21 |
|
|
|
| 22 |
|
|
Unicode escape encoding used with Java, Mozilla, etc. (Alias: x-u-escaped)
|
| 23 |
|
|
|
| 24 |
|
|
=cut
|
| 25 |
|
|
|
| 26 |
|
|
sub encode ($$;$) {
|
| 27 |
|
|
use integer;
|
| 28 |
|
|
my ($obj, $str, $chk) = @_;
|
| 29 |
|
|
$_[1] = '' if $chk;
|
| 30 |
|
|
$str =~ s{([^\x00-\x7F])|\\\\|\\(u+[0-9A-Fa-f]{4})}{
|
| 31 |
|
|
my ($C,$u) = ($1,$2);
|
| 32 |
|
|
if ($C) { ## U+0080-U+FFFF. See L</BUGS>.
|
| 33 |
|
|
my $c = ord $C;
|
| 34 |
|
|
if ($c < 0x10000) {
|
| 35 |
|
|
sprintf '\u%04X', $c;
|
| 36 |
|
|
} else {
|
| 37 |
|
|
sprintf '\U%08X', $c;
|
| 38 |
|
|
}
|
| 39 |
|
|
} elsif ($u) { ## \u+....
|
| 40 |
|
|
'\u'.$u;
|
| 41 |
|
|
} else {
|
| 42 |
|
|
'\\\\';
|
| 43 |
|
|
}
|
| 44 |
|
|
}ge;
|
| 45 |
|
|
Encode::_utf8_off ($str);
|
| 46 |
|
|
$str;
|
| 47 |
|
|
}
|
| 48 |
|
|
# chr ((ord ($u) - 0x10000) / 0x400 + 0xD800)
|
| 49 |
|
|
# chr ((ord ($u) - 0x10000) % 0x400 + 0xDC00)
|
| 50 |
|
|
|
| 51 |
|
|
sub decode ($$;$) {
|
| 52 |
|
|
no warnings;
|
| 53 |
|
|
my ($obj, $str, $chk) = @_;
|
| 54 |
|
|
$_[1] = '' if $chk;
|
| 55 |
|
|
$str =~ s{\\\\|\\(u+)([0-9A-Fa-f]{4})}{
|
| 56 |
|
|
my ($u,$h) = ($1, $2);
|
| 57 |
|
|
if ($u eq 'u') {
|
| 58 |
|
|
chr (hex $h);
|
| 59 |
|
|
} elsif ($u) {
|
| 60 |
|
|
'\\'.substr ($u, 1).$h;
|
| 61 |
|
|
} else {
|
| 62 |
|
|
'\\\\';
|
| 63 |
|
|
}
|
| 64 |
|
|
}ge;
|
| 65 |
|
|
#Encode::_utf8_on ($str);
|
| 66 |
|
|
return $str;
|
| 67 |
|
|
}
|
| 68 |
|
|
# chr (0x10000+(ord($u1)-0xD800)*0x400+(ord($u2)-0xDC00))
|
| 69 |
|
|
|
| 70 |
|
|
=back
|
| 71 |
|
|
|
| 72 |
|
|
=head1 BUGS
|
| 73 |
|
|
|
| 74 |
|
|
I don't know how U+10000 or higher characters should be treated,
|
| 75 |
|
|
so this implemention may be incorect...
|
| 76 |
|
|
|
| 77 |
|
|
=head1 SEE ALSO
|
| 78 |
|
|
|
| 79 |
|
|
"Java Language Specification" -- 3.3 Unicode Escapes.
|
| 80 |
|
|
<http://java.sun.com/docs/books/jls/second_edition/html/lexical.doc.html#100850>,
|
| 81 |
|
|
<http://java.sun.com/docs/books/jls/first_edition/html/3.doc.html#100850>.
|
| 82 |
|
|
Japanese translation is available as TR X 0005:2002. See
|
| 83 |
|
|
<http://www.y-adagio.com/public/standards/tr_javalang2/lexical.doc.html>,
|
| 84 |
|
|
<http://www.y-adagio.com/public/standards/tr_javalang/3.doc.htm>.
|
| 85 |
|
|
|
| 86 |
|
|
=head1 LICENSE
|
| 87 |
|
|
|
| 88 |
|
|
Copyright 2002 Nanashi-san <nanashi-san@nanashi.invalid>
|
| 89 |
|
|
|
| 90 |
|
|
This program is free software; you can redistribute it and/or modify
|
| 91 |
|
|
it under the terms of the GNU General Public License as published by
|
| 92 |
|
|
the Free Software Foundation; either version 2 of the License, or
|
| 93 |
|
|
(at your option) any later version.
|
| 94 |
|
|
|
| 95 |
|
|
This program is distributed in the hope that it will be useful,
|
| 96 |
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
| 97 |
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
| 98 |
|
|
GNU General Public License for more details.
|
| 99 |
|
|
|
| 100 |
|
|
You should have received a copy of the GNU General Public License
|
| 101 |
|
|
along with this program; see the file COPYING. If not, write to
|
| 102 |
|
|
the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
| 103 |
|
|
Boston, MA 02111-1307, USA.
|
| 104 |
|
|
|
| 105 |
|
|
=cut
|
| 106 |
|
|
|
| 107 |
|
|
1; # $Date: 2002/09/15 04:15:51 $
|
Parent Directory
| 
Revision Log