package Encode::JP::JIS7;
use strict;
-our $VERSION = do { my @r = (q$Revision: 1.2 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
+our $VERSION = do { my @r = (q$Revision: 2.0 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
+
+use Encode qw(:fallbacks);
-require Encode;
for my $name ('7bit-jis', 'iso-2022-jp', 'iso-2022-jp-1'){
my $h2z = ($name eq '7bit-jis') ? 0 : 1;
my $jis0212 = ($name eq 'iso-2022-jp') ? 0 : 1;
} => __PACKAGE__;
}
-sub name { shift->{'Name'} }
-
-sub new_sequence { $_[0] }
+use base qw(Encode::Encoding);
+# we override this to 1 so PerlIO works
sub needs_lines { 1 }
use Encode::CJKConstants qw(:all);
-our $DEBUG = 0;
-
#
# decode is identical for all 2022 variants
#
-sub decode
+sub decode($$;$)
{
- my ($obj,$str,$chk) = @_;
- my $residue = jis_euc(\$str);
- # This is for PerlIO
+ my ($obj, $str, $chk) = @_;
+ my $residue = '';
+ if ($chk){
+ $str =~ s/([^\x00-\x7f].*)$//so and $residue = $1;
+ }
+ $residue .= jis_euc(\$str);
$_[1] = $residue if $chk;
- # use perlqq fallback for euc-jp -> utf8
- return Encode::decode('euc-jp', $str, 0);
+ return Encode::decode('euc-jp', $str, FB_PERLQQ);
}
#
# encode is different
#
-sub encode
+sub encode($$;$)
{
require Encode::JP::H2Z;
my ($obj, $utf8, $chk) = @_;
# empty the input string in the stack so perlio is ok
$_[1] = '' if $chk;
my ($h2z, $jis0212) = @$obj{qw(h2z jis0212)};
- my $octet = Encode::encode('euc-jp', $utf8, 0) ;
+ my $octet = Encode::encode('euc-jp', $utf8, FB_PERLQQ) ;
$h2z and &Encode::JP::H2Z::h2z(\$octet);
euc_jis(\$octet, $jis0212);
return $octet;
}
+#
+# cat_decode
+#
+my $re_scan_jis_g = qr{
+ \G ( ($RE{JIS_0212}) | $RE{JIS_0208} |
+ ($RE{ISO_ASC}) | ($RE{JIS_KANA}) | )
+ ([^\e]*)
+}x;
+sub cat_decode { # ($obj, $dst, $src, $pos, $trm, $chk)
+ my ($obj, undef, undef, $pos, $trm) = @_; # currently ignores $chk
+ my ($rdst, $rsrc, $rpos) = \@_[1,2,3];
+ local ${^ENCODING};
+ use bytes;
+ my $opos = pos($$rsrc);
+ pos($$rsrc) = $pos;
+ while ($$rsrc =~ /$re_scan_jis_g/gc) {
+ my ($esc, $esc_0212, $esc_asc, $esc_kana, $chunk) =
+ ($1, $2, $3, $4, $5);
+
+ unless ($chunk) { $esc or last; next; }
+
+ if ($esc && !$esc_asc) {
+ $chunk =~ tr/\x21-\x7e/\xa1-\xfe/;
+ if ($esc_kana) {
+ $chunk =~ s/([\xa1-\xdf])/\x8e$1/og;
+ } elsif ($esc_0212) {
+ $chunk =~ s/([\xa1-\xfe][\xa1-\xfe])/\x8f$1/og;
+ }
+ $chunk = Encode::decode('euc-jp', $chunk, 0);
+ }
+ elsif ((my $npos = index($chunk, $trm)) >= 0) {
+ $$rdst .= substr($chunk, 0, $npos + length($trm));
+ $$rpos += length($esc) + $npos + length($trm);
+ pos($$rsrc) = $opos;
+ return 1;
+ }
+ $$rdst .= $chunk;
+ $$rpos = pos($$rsrc);
+ }
+ $$rpos = pos($$rsrc);
+ pos($$rsrc) = $opos;
+ return '';
+}
# JIS<->EUC
+my $re_scan_jis = qr{
+ (?:($RE{JIS_0212})|$RE{JIS_0208}|($RE{ISO_ASC})|($RE{JIS_KANA}))([^\e]*)
+}x;
sub jis_euc {
+ local ${^ENCODING};
my $r_str = shift;
- $$r_str =~ s(
- ($RE{JIS_0212}|$RE{JIS_0208}|$RE{ISO_ASC}|$RE{JIS_KANA})
- ([^\e]*)
- )
+ $$r_str =~ s($re_scan_jis)
{
- my ($esc, $chunk) = ($1, $2);
- if ($esc !~ /$RE{ISO_ASC}/o) {
+ my ($esc_0212, $esc_asc, $esc_kana, $chunk) =
+ ($1, $2, $3, $4);
+ if (!$esc_asc) {
$chunk =~ tr/\x21-\x7e/\xa1-\xfe/;
- if ($esc =~ /$RE{JIS_KANA}/o) {
+ if ($esc_kana) {
$chunk =~ s/([\xa1-\xdf])/\x8e$1/og;
}
- elsif ($esc =~ /$RE{JIS_0212}/o) {
+ elsif ($esc_0212) {
$chunk =~ s/([\xa1-\xfe][\xa1-\xfe])/\x8f$1/og;
}
}
}
sub euc_jis{
+ no warnings qw(uninitialized);
my $r_str = shift;
my $jis0212 = shift;
$$r_str =~ s{