CJKV Information Processing, 2nd Edition

742

|

Appendix C: Perl Code Examples

sub han2zen {

my ($hkana) = @_;

if ($hkana =~ /^$euc\xB3$euc\xDE/o) { # Special "u + dakuten" case

if ($euc) {

$hkana =~ s/$euc\xB3$euc\xDE/\xA5\xF4/go;

} else {

$hkana =~ s/\xB3\xDE/\x83\x94/g;

}

} elsif ($hkana =~ /^${euc}[\xB6-\xC4\xCA-\xCE]${euc}[\xDE\xDF]/o) {

$prefix = $kana_one; # First byte for katakana

if ($hkana =~ /^${euc}[\xCA-\xCE]$euc\xDF/o) {

$suffix = 2; # Increment value for handakuten

} else {

$suffix = 1; # Increment value for dakuten

}

$hkana =~ s/$euc([\xB6-\xC4\xCA-\xCE])${euc}[\xDE\xDF]/

pack("n",unpack("n","$prefix$char_hash{$1}") +

$suffix)/egox;

} else {

if ($hkana =~ /^${euc}[\xA0-\xA5\xB0\xDE\xDF]/o) {

$prefix = $symbol_one; # First byte for symbol

} else {

$prefix = $kana_one; # First byte for katakana

}

$hkana =~ s/$euc([\xA0-\xDF])/$prefix$char_hash{$1}/go;

}

return $hkana;

}

Korean Code Conversion

Although this section does not include any complete Perl programs, the most dicult al-

gorithms for handling Korean encodings are included as workable subroutines that can be

used in Perl programs. e main focus of this section is Johab encoding, which requires

an algorithm for handling conversion to and from ISO-2022-KR or EUC-KR encodings,

which is strikingly similar to that used for handling conversion to and from Shi-JIS en-

coding. Because handling Johab encoding also requires mapping tables (for handling the

2,350 ...