| 1 |
wakaba |
1.1 |
|
| 2 |
|
|
=head1 NAME
|
| 3 |
|
|
|
| 4 |
|
|
Message::MIME::EncodedWord Perl module
|
| 5 |
|
|
|
| 6 |
|
|
=head1 DESCRIPTION
|
| 7 |
|
|
|
| 8 |
|
|
Perl module for MIME C<encoded-word>.
|
| 9 |
|
|
|
| 10 |
|
|
=cut
|
| 11 |
|
|
|
| 12 |
|
|
package Message::MIME::EncodedWord;
|
| 13 |
|
|
use strict;
|
| 14 |
wakaba |
1.4 |
use vars qw(%ENCODER %DECODER %OPTION %REG $VERSION);
|
| 15 |
wakaba |
1.13 |
$VERSION=do{my @r=(q$Revision: 1.12 $=~/\d+/g);sprintf "%d."."%02d" x $#r,@r};
|
| 16 |
wakaba |
1.1 |
require Message::MIME::Charset;
|
| 17 |
|
|
|
| 18 |
|
|
$REG{WSP} = qr/[\x09\x20]/;
|
| 19 |
|
|
$REG{FWS} = qr/[\x09\x20]*/;
|
| 20 |
wakaba |
1.3 |
$REG{comment} = qr/\x28(?:\x5C[\x00-\xFF]|[\x00-\x0C\x0E-\x27\x2A-\x5B\x5D-\xFF]|(??{$REG{comment}}))*\x29/;
|
| 21 |
wakaba |
1.1 |
|
| 22 |
|
|
$REG{atext_dot} = qr/[\x21\x23-\x27\x2A\x2B\x2D-\x39\x3D\x3F\x41-\x5A\x5E-\x7E]+/;
|
| 23 |
|
|
$REG{attribute_char} = qr/[\x21\x23-\x24\x26\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+/;
|
| 24 |
wakaba |
1.3 |
$REG{M_comment} = qr/\x28((?:\x5C[\x00-\xFF]|[\x00-\x0C\x0E-\x27\x2A-\x5B\x5D-\xFF]|(??{$REG{comment}}))*)\x29/;
|
| 25 |
wakaba |
1.1 |
$REG{M_encoded_word} = qr/=\x3F($REG{attribute_char})(?:\x2A($REG{attribute_char}))?\x3F($REG{attribute_char})\x3F([\x21-\x3E\x40-\x7E]+)\x3F=/;
|
| 26 |
|
|
$REG{S_encoded_word} = qr/=\x3F$REG{atext_dot}\x3F=/;
|
| 27 |
|
|
|
| 28 |
wakaba |
1.4 |
=head1 OPTIONS
|
| 29 |
|
|
|
| 30 |
|
|
=cut
|
| 31 |
|
|
|
| 32 |
|
|
%OPTION = (
|
| 33 |
wakaba |
1.6 |
forcedecode => 0,
|
| 34 |
wakaba |
1.4 |
);
|
| 35 |
|
|
|
| 36 |
|
|
=over 4
|
| 37 |
|
|
|
| 38 |
|
|
=item $Message::MIME::EncodedWord::OPTION{forcedecode} = 1/0
|
| 39 |
|
|
|
| 40 |
|
|
When no charset decoder (See L<Message::MIME::Charset>)
|
| 41 |
|
|
for C<ISO-8859-I<n>> is defined, and this option is TRUE,
|
| 42 |
|
|
decoding C<encoded-word> functions attempt to decode
|
| 43 |
|
|
ASCII part of these charset.
|
| 44 |
|
|
|
| 45 |
wakaba |
1.7 |
RFC 2047 says that ASCII part of C<ISO-8859-I<n>> be at least
|
| 46 |
|
|
supported. This requirement is convinience for human user who
|
| 47 |
|
|
sees final rendering result. But it is not appropriate to process message.
|
| 48 |
wakaba |
1.4 |
|
| 49 |
wakaba |
1.7 |
Defalt value is C<0>, force decoding is disenabled.
|
| 50 |
wakaba |
1.4 |
|
| 51 |
|
|
=back
|
| 52 |
|
|
|
| 53 |
|
|
=head2 Note
|
| 54 |
|
|
|
| 55 |
|
|
Before you set new value for these options,
|
| 56 |
|
|
C<Message::MIME::EncodedWord> should be loaded (C<require>ed).
|
| 57 |
|
|
Other modules which use C<Message::MIME::EncodedWord>
|
| 58 |
|
|
will automatically require this module, and this module
|
| 59 |
|
|
will set initial (default) option value.
|
| 60 |
|
|
|
| 61 |
|
|
Bad example:
|
| 62 |
|
|
|
| 63 |
|
|
#! perl
|
| 64 |
|
|
$Message::MIME::EncodedWord::OPTION{forcedecode} = 0;
|
| 65 |
|
|
use Message::Field::Subject;
|
| 66 |
|
|
my $subject = Message::Field::Subject->parse ($ARGV[0]);
|
| 67 |
|
|
## At this time, M::F::Subject call M::M::EWord,
|
| 68 |
|
|
## and $OPTION{forcedecode} is set C<1>, default value.
|
| 69 |
|
|
|
| 70 |
|
|
Shold be:
|
| 71 |
|
|
|
| 72 |
|
|
#! perl
|
| 73 |
|
|
require Message::MIME::EncodedWord;
|
| 74 |
|
|
$Message::MIME::EncodedWord::OPTION{forcedecode} = 0;
|
| 75 |
|
|
use Message::Field::Subject;
|
| 76 |
|
|
my $subject = Message::Field::Subject->parse ($ARGV[0]);
|
| 77 |
|
|
## At this time, M::F::Subject call M::M::EWord,
|
| 78 |
|
|
## but perl takes no action since it has already loaded.
|
| 79 |
|
|
|
| 80 |
|
|
=cut
|
| 81 |
|
|
|
| 82 |
wakaba |
1.1 |
%DECODER = (
|
| 83 |
wakaba |
1.7 |
'*DEFAULT' => sub { $_[1] },
|
| 84 |
|
|
'7' => sub { $_[1] },
|
| 85 |
wakaba |
1.12 |
'8' => sub { $_[1] },
|
| 86 |
wakaba |
1.1 |
b => sub {require MIME::Base64; MIME::Base64::decode ($_[1])},
|
| 87 |
|
|
q => sub {my $s = $_[1]; $s =~ tr/_/\x20/;
|
| 88 |
|
|
$s=~ s/=([0-9A-Fa-f]{2})/pack("C", hex($1))/ge; $s},
|
| 89 |
|
|
);
|
| 90 |
|
|
|
| 91 |
wakaba |
1.12 |
%ENCODER = (
|
| 92 |
|
|
b => sub {require MIME::Base64; my $b = MIME::Base64::encode ($_[1]); $b =~ tr/\x00-\x20//d; ($b, {-encoded => 1})},
|
| 93 |
|
|
q => \&_encode_q_encoding,
|
| 94 |
|
|
);
|
| 95 |
|
|
|
| 96 |
wakaba |
1.13 |
=head1 $string = Message::MIME::EncodedWord::decode ($text, %option)
|
| 97 |
|
|
|
| 98 |
|
|
Decodes encoded-words in given text and return as a string (in internal code).
|
| 99 |
|
|
|
| 100 |
|
|
Options:
|
| 101 |
|
|
|
| 102 |
|
|
=over 4
|
| 103 |
|
|
|
| 104 |
|
|
=item -process_non_encoded_word => CODE / 0 (default)
|
| 105 |
|
|
|
| 106 |
|
|
Procedure for non-encoded-word string. This option can be useful
|
| 107 |
|
|
with comment decoder to unquote quoted-pair and/or to decode 8-bit string
|
| 108 |
|
|
of HTTP.
|
| 109 |
|
|
|
| 110 |
|
|
=back
|
| 111 |
|
|
|
| 112 |
|
|
=cut
|
| 113 |
|
|
|
| 114 |
|
|
sub decode ($;%) {
|
| 115 |
|
|
my @string = split /(?<=[^\x09\x20])(?=[\x09\x20])/, shift;
|
| 116 |
|
|
my %option = @_;
|
| 117 |
|
|
my @r; ## Token is encoded-word or not
|
| 118 |
|
|
for my $i (0..$#string) {
|
| 119 |
|
|
if ($string[$i] =~ m(^([\x09\x20]*)=\?([0-9A-Za-z!#\$%&'+^_`{|}~.-]{1,67})(?:\*([0-9A-Za-z!#\$&+^_`{|}~.-]{1,65}))?\?([0-9A-Za-z!#\$%&'*+^_`{|}~.-]{1,67})\?([\x21-\x3E\x40-\x7E]{1,67})\?=$)) {
|
| 120 |
wakaba |
1.3 |
my ($t, $w) = ('', $1);
|
| 121 |
|
|
($t, $r[$i]) = (_decode_eword ($2, $3, $4, $5));
|
| 122 |
wakaba |
1.13 |
$string[$i] = ($i == 0 || !$r[$i-1] ? $w : '') . $t if $r[$i];
|
| 123 |
|
|
} elsif (ref $option{-process_non_encoded_word}) {
|
| 124 |
|
|
($string[$i], $r[$i]) = &{$option{-process_non_encoded_word}} ($string[$i], \%option);
|
| 125 |
|
|
} else {
|
| 126 |
|
|
$r[$i] = 0;
|
| 127 |
wakaba |
1.1 |
}
|
| 128 |
wakaba |
1.3 |
}
|
| 129 |
wakaba |
1.13 |
join '', @string;
|
| 130 |
|
|
}
|
| 131 |
|
|
|
| 132 |
|
|
sub _decode_eword ($$$$) {
|
| 133 |
|
|
my ($charset, $lang, $encoding, $etext) = (shift, shift, lc shift, shift);
|
| 134 |
|
|
$charset = Message::MIME::Charset::name_normalize ($charset);
|
| 135 |
|
|
my ($r,%s);
|
| 136 |
|
|
if (ref $DECODER{$encoding}) { ## decode TE
|
| 137 |
|
|
$r = &{$DECODER{$encoding}} ($encoding, $etext);
|
| 138 |
|
|
($r,%s) = Message::MIME::Charset::decode ($charset, $r);
|
| 139 |
|
|
}
|
| 140 |
|
|
($r, $s{success});
|
| 141 |
wakaba |
1.1 |
}
|
| 142 |
|
|
|
| 143 |
wakaba |
1.3 |
sub decode_ccontent ($$) {
|
| 144 |
wakaba |
1.13 |
use re 'eval';
|
| 145 |
wakaba |
1.8 |
my $yourself = shift; my $s = shift;
|
| 146 |
wakaba |
1.3 |
my (@s, @r) = ();
|
| 147 |
wakaba |
1.5 |
my ($i, @t) = (-1);
|
| 148 |
|
|
$s =~ s{$REG{FWS}$REG{comment}}{$i++; $t[$i] = $&; "\x28${i}\x29"}gex;
|
| 149 |
|
|
$s =~ s{($REG{FWS}(?:\x5C[\x00-\xFF]
|
| 150 |
wakaba |
1.6 |
|[\x00-\x08\x0A-\x1F\x21-\x27\x2A-\x5B\x5D-\xFF])+)
|
| 151 |
wakaba |
1.5 |
|(\x28[0-9]+\x29)}{my ($t,$c) = ($1, $2);
|
| 152 |
|
|
if ($t) {$i++; $t[$i] = $t; "\x28${i}\x29"}
|
| 153 |
|
|
else {$c}}gex;
|
| 154 |
wakaba |
1.6 |
$s =~ s{\x28([0-9]+)\x29}{push @s, $t[$1]; ''}gex;
|
| 155 |
|
|
push @s, $s if length $s;
|
| 156 |
wakaba |
1.3 |
for my $i (0..$#s) {
|
| 157 |
|
|
if ($s[$i] =~ /^($REG{FWS})$REG{M_encoded_word}$/) {
|
| 158 |
|
|
my ($t, $w) = ('', $1);
|
| 159 |
|
|
($t, $r[$i]) = (_decode_eword ($2, $3, $4, $5));
|
| 160 |
wakaba |
1.4 |
if ($r[$i]) {
|
| 161 |
wakaba |
1.3 |
$s[$i] = $t;
|
| 162 |
wakaba |
1.4 |
if ($i == 0 || $r[$i-1] == 0) {
|
| 163 |
wakaba |
1.3 |
$s[$i] = $w.$s[$i];
|
| 164 |
|
|
}
|
| 165 |
|
|
}
|
| 166 |
|
|
} elsif ($s[$i] =~ /^($REG{FWS})$REG{M_comment}$/) {
|
| 167 |
wakaba |
1.10 |
$s[$i] = $1.'('. &decode_ccontent ($yourself, $2) .')';
|
| 168 |
wakaba |
1.3 |
} else {
|
| 169 |
|
|
$s[$i] =~ s/\x5C([\x00-\xFF])/$1/g;
|
| 170 |
|
|
my %s = &{$yourself->{option}->{hook_decode_string}} ($yourself, $s[$i],
|
| 171 |
|
|
type => 'ccontent/quoted');
|
| 172 |
|
|
$s[$i] = $s{value};
|
| 173 |
|
|
}
|
| 174 |
|
|
}
|
| 175 |
|
|
join '', @s;
|
| 176 |
|
|
}
|
| 177 |
|
|
|
| 178 |
wakaba |
1.12 |
=head1 $encoded_words = Message::MIME::EncodedWord::encode ($string, %option)
|
| 179 |
|
|
|
| 180 |
|
|
Encode given string as encoded-words if necessary.
|
| 181 |
|
|
|
| 182 |
|
|
Available options:
|
| 183 |
|
|
|
| 184 |
|
|
=over 4
|
| 185 |
|
|
|
| 186 |
|
|
=item -charset => charset (default: 'us-ascii')
|
| 187 |
|
|
|
| 188 |
|
|
Charset name (in lower cases) to be used to encode the output string.
|
| 189 |
|
|
(Currently 'us-ascii', 'iso-8859-1' and 'us-ascii' is supported.
|
| 190 |
|
|
'iso-2022-int', 'euc-jp' or other charsets are unable to co-exist with
|
| 191 |
|
|
encoded-words, in current implemention.)
|
| 192 |
|
|
|
| 193 |
|
|
=item -context => 'default' (default) / 'comment' / 'phrase' / 'quoted_string'
|
| 194 |
|
|
|
| 195 |
|
|
Context in which given string is embeded.
|
| 196 |
|
|
|
| 197 |
|
|
=item -ebcdic_safe => 0/1 (default)
|
| 198 |
|
|
|
| 199 |
|
|
Encode additional ASCII characters (shown in RFC 2047) that
|
| 200 |
|
|
are not safe in EBCDIC transports. This option is meaningful only when
|
| 201 |
|
|
"Q" encoding is used.
|
| 202 |
|
|
|
| 203 |
|
|
=item -encode_char => 1*CHAR / '' (default)
|
| 204 |
|
|
|
| 205 |
|
|
The list of ASCII characters should be encoded in encoded-word
|
| 206 |
|
|
in addition to non-ASCII characters and special ASCII characters
|
| 207 |
|
|
determined by C<-context> and C<-ebcdic_safe> options.
|
| 208 |
|
|
|
| 209 |
|
|
=item -encode_encoded_word_like => 0/1 (default)
|
| 210 |
|
|
|
| 211 |
|
|
Encode encoded-word-like tokens in given string or not.
|
| 212 |
|
|
|
| 213 |
|
|
=item -preserve_wsp => 0/1 (default)
|
| 214 |
|
|
|
| 215 |
|
|
If true, 2*WSP is encoded in encoded-word. Unless string is the content
|
| 216 |
|
|
of comment or quoted-string, this option value should be true.
|
| 217 |
|
|
|
| 218 |
|
|
=item -q_encode_char => 1*CHAR / '' (default)
|
| 219 |
|
|
|
| 220 |
|
|
The list of ASCII characters should be encoded in q encoding of encoded-word
|
| 221 |
|
|
in addition to non-ASCII characters and special ASCII characters
|
| 222 |
|
|
determined by C<-context> and C<-ebcdic_safe> options.
|
| 223 |
|
|
|
| 224 |
|
|
=item -quoted_pair => 1*CHAR / qr|(:: pattern ::)| '' (default)
|
| 225 |
|
|
|
| 226 |
|
|
A character list or a Regexp pattern for characters to be quoted as
|
| 227 |
|
|
the quoted-pairs. When '', no character is quoted.
|
| 228 |
|
|
Quoting is performed to characters NOT encoded as encoded-words.
|
| 229 |
|
|
This option should be useful if given string is to be a content
|
| 230 |
|
|
of a comment or quoted-string. Usually, "\" is also included in
|
| 231 |
|
|
the character list to represent "\" itself as "\\".
|
| 232 |
|
|
|
| 233 |
|
|
=item -source_charset => charset (default = *internal)
|
| 234 |
|
|
|
| 235 |
|
|
Charset name (in lower case) of given string.
|
| 236 |
|
|
|
| 237 |
|
|
=item -token_maxlength => 1*DIGIT / 0 (default)
|
| 238 |
|
|
|
| 239 |
|
|
Maximal length of a string (1*l<OCTET except WSP>) that can be represented
|
| 240 |
|
|
as a non-encoded-word. If 0, no length limit is implied.
|
| 241 |
|
|
Note that this option does NOT change maximal length of encoded-word.
|
| 242 |
|
|
(Maximal length of encoded-word is always 75, as defined in RFC 2047.)
|
| 243 |
|
|
|
| 244 |
|
|
=cut
|
| 245 |
|
|
|
| 246 |
|
|
sub encode ($;%) {
|
| 247 |
|
|
my $string = shift;
|
| 248 |
|
|
my %option = @_;
|
| 249 |
|
|
$option{-preserve_wsp} = 1 unless defined $option{-preserve_wsp};
|
| 250 |
|
|
$option{-source_charset} ||= '*internal';
|
| 251 |
|
|
my $re_encode = join ('|', grep {$_}
|
| 252 |
|
|
(($option{-charset} eq 'utf-8' ? '' :
|
| 253 |
|
|
$option{-charset} eq 'iso-8859-1' ? '[^\x00-\xFF]' : '[^\x00-\x7F]'),
|
| 254 |
|
|
(defined $option{-encode_char} ? qq([$option{-encode_char}]) : ''),
|
| 255 |
|
|
(!defined $option{-encode_encoded_word_like}||$option{-encode_encoded_word_like} ? '^=\?' :'')
|
| 256 |
|
|
)) || '(?!)';
|
| 257 |
|
|
if ($option{-quoted_pair} && !ref $option{-quoted_pair}) {
|
| 258 |
|
|
$option{-quoted_pair} = quotemeta $option{-quoted_pair};
|
| 259 |
|
|
$option{-quoted_pair} = qr/([$option{-quoted_pair}])/;
|
| 260 |
|
|
}
|
| 261 |
|
|
my @string = split /(?<=[^\x09\x20])(?=[\x09\x20])/, $string;
|
| 262 |
|
|
my @encoded;
|
| 263 |
|
|
for my $i (0..$#string) {
|
| 264 |
|
|
my $string_nows = $string[$i];
|
| 265 |
|
|
my $ws = ''; $ws = $1 if $string_nows =~ s/^([\x09\x20]+)//;
|
| 266 |
|
|
$encoded[$i] = -1;
|
| 267 |
|
|
if ($i == 0) { ## First component of string
|
| 268 |
|
|
if ($string_nows =~ /$re_encode/ || ($option{-preserve_wsp} && length ($ws) > 1)
|
| 269 |
|
|
|| ($option{-token_maxlength} && length ($string_nows) > $option{-token_maxlength})) {
|
| 270 |
|
|
my $estring = _encode ($string[$i], \%option);
|
| 271 |
|
|
if ($estring) {
|
| 272 |
|
|
$string[$i] = $estring;
|
| 273 |
|
|
$encoded[$i] = 1;
|
| 274 |
|
|
}
|
| 275 |
|
|
}
|
| 276 |
|
|
} elsif ($i == $#string && length ($string_nows) == 0) { ## Last component of string is 1*WSP
|
| 277 |
|
|
my $estring = _encode ($ws, \%option);
|
| 278 |
|
|
if ($estring) {
|
| 279 |
|
|
$string[$i] = ' ' . $estring;
|
| 280 |
|
|
$encoded[$i] = 1;
|
| 281 |
|
|
}
|
| 282 |
|
|
} elsif ($i == 0 || !$encoded[$i-1]) { ## Previous token is not encoded
|
| 283 |
|
|
if ($string_nows =~ /$re_encode/
|
| 284 |
|
|
|| ($option{-token_maxlength} && length ($string_nows) > $option{-token_maxlength})) {
|
| 285 |
|
|
if ($option{-preserve_wsp} && length ($ws) > 1 && $i) {
|
| 286 |
|
|
$string_nows = substr ($ws, 1) . $string_nows;
|
| 287 |
|
|
$ws = substr ($ws, 0, 1);
|
| 288 |
|
|
} elsif (!$ws) {
|
| 289 |
|
|
$ws = ' ';
|
| 290 |
|
|
}
|
| 291 |
|
|
my $estring = _encode ($string_nows, \%option);
|
| 292 |
|
|
if ($estring) {
|
| 293 |
|
|
$string[$i] = $ws . $estring;
|
| 294 |
|
|
$encoded[$i] = 1;
|
| 295 |
|
|
}
|
| 296 |
|
|
} elsif ($option{-preserve_wsp} && length ($ws) == 2) {
|
| 297 |
|
|
my $estring = _encode (substr ($ws, 1) . $string_nows, \%option);
|
| 298 |
|
|
if ($estring) {
|
| 299 |
|
|
$string[$i] = substr ($ws, 0, 1) . $estring;
|
| 300 |
|
|
$encoded[$i] = 1;
|
| 301 |
|
|
}
|
| 302 |
|
|
} elsif ($option{-preserve_wsp} && length ($ws) > 2) {
|
| 303 |
|
|
my $estring = _encode (substr ($ws, 1, length ($ws) - 2), \%option);
|
| 304 |
|
|
if ($estring) {
|
| 305 |
|
|
$string_nows =~ s/$option{-quoted_pair}/\\$1/g if $option{-quoted_pair};
|
| 306 |
|
|
$string[$i] = substr ($ws, 0, 1) . $estring . substr ($ws, -1) . $string_nows;
|
| 307 |
|
|
$encoded[$i] = 0;
|
| 308 |
|
|
}
|
| 309 |
|
|
}
|
| 310 |
|
|
} else { ## Previous token is encoded
|
| 311 |
|
|
if ($string_nows =~ /$re_encode/
|
| 312 |
|
|
|| ($option{-token_maxlength} && length ($string_nows) > $option{-token_maxlength})) {
|
| 313 |
|
|
my $estring = _encode ($string[$i], \%option);
|
| 314 |
|
|
if ($estring) {
|
| 315 |
|
|
$string[$i] = ($i!=0?' ':'') . $estring;
|
| 316 |
|
|
$encoded[$i] = 1;
|
| 317 |
|
|
}
|
| 318 |
|
|
} elsif ($option{-preserve_wsp} && length ($ws) > 1) {
|
| 319 |
|
|
my $estring = _encode (substr ($ws, 0, length ($ws) - 1), \%option);
|
| 320 |
|
|
if ($estring) {
|
| 321 |
|
|
$string_nows =~ s/$option{-quoted_pair}/\\$1/g if $option{-quoted_pair};
|
| 322 |
|
|
$string[$i] = ($i!=0?' ':'') . $estring . substr ($ws, -1) . $string_nows;
|
| 323 |
|
|
$encoded[$i] = 0;
|
| 324 |
|
|
}
|
| 325 |
|
|
}
|
| 326 |
|
|
}
|
| 327 |
|
|
if ($encoded[$i] == -1) {
|
| 328 |
|
|
$string[$i] =~ s/$option{-quoted_pair}/\\$1/g if $option{-quoted_pair};
|
| 329 |
|
|
$encoded[$i] = 0;
|
| 330 |
|
|
}
|
| 331 |
|
|
}
|
| 332 |
|
|
join '', @string;
|
| 333 |
|
|
}
|
| 334 |
|
|
|
| 335 |
|
|
## $encoded_text must be octet string (not utf8 string).
|
| 336 |
|
|
sub _encode_q_encoding ($$;\%) {
|
| 337 |
|
|
my ($encoding, $encoded_text, $option) = @_;
|
| 338 |
|
|
## -- What characters are encoded?
|
| 339 |
|
|
my $achar = {
|
| 340 |
|
|
default => q(!"#$%&'()*+,./0123456789:;<>@ABCDEFGHIJKLMNOPQRSTUVWXYZ^`abcdefghijklmnopqrstuvwxyz{|}~\\[]-),
|
| 341 |
|
|
comment => q(!"#$%&'*+,./0123456789:;<>@ABCDEFGHIJKLMNOPQRSTUVWXYZ^`abcdefghijklmnopqrstuvwxyz{|}~[]-),
|
| 342 |
|
|
phrase => q(0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz!*+/-),
|
| 343 |
|
|
quoted_string => q(!#$%&'()*+,./0123456789:;<>@ABCDEFGHIJKLMNOPQRSTUVWXYZ^`abcdefghijklmnopqrstuvwxyz{|}~[]-),
|
| 344 |
|
|
}->{$option->{-context} || 'default'};
|
| 345 |
|
|
my $echar = $option->{-q_encode_char}; $echar =~ s/([\/\\])/\\$1/g;
|
| 346 |
|
|
eval qq{\$achar =~ tr/$echar//d};
|
| 347 |
|
|
$achar =~ tr/!"#$@[\\]^`{|}~//d unless defined $option->{-ebcdic_safe} && $option->{-ebcdic_safe} == 0;
|
| 348 |
|
|
$achar = quotemeta $achar;
|
| 349 |
|
|
## -- Encode
|
| 350 |
|
|
$encoded_text =~ s/([^$achar])/sprintf '=%02X', ord $1/ge;
|
| 351 |
|
|
$encoded_text =~ s/=20/_/g;
|
| 352 |
|
|
($encoded_text, {-encoded => 1});
|
| 353 |
|
|
}
|
| 354 |
|
|
sub _encode ($$) {
|
| 355 |
|
|
my $option = $_[1];
|
| 356 |
|
|
my $charset = $option->{-source_charset};
|
| 357 |
|
|
my $encoding = Message::MIME::Charset::get_property ('cte_header_preferred', $charset);
|
| 358 |
|
|
my @estr;
|
| 359 |
|
|
for my $str (@{Message::MIME::Charset::divide_string ($charset, $_[0], -max => int ((75 - length ($charset.$encoding) - 6) * ({b=>3/4,q=>1/3}->{$encoding}||1)))}) {
|
| 360 |
|
|
my $echarset = Message::MIME::Charset::get_interchange_charset ($charset, $str, $option)->{charset} || $charset;
|
| 361 |
|
|
my ($estr, %r) = Message::MIME::Charset::encode ($echarset, $str);
|
| 362 |
|
|
do {$echarset = $charset; $estr = $str; Message::MIME::Charset::_utf8_off ($estr)} unless $r{success};
|
| 363 |
|
|
$encoding = Message::MIME::Charset::get_property ('cte_header_preferred', $echarset);
|
| 364 |
|
|
if ($encoding eq '*auto') {
|
| 365 |
|
|
$encoding = (($estr =~ tr/\x20-\x7E/\x20-\x7E/) < (length ($estr) * 0.55)) ? 'b' : 'q';
|
| 366 |
|
|
}
|
| 367 |
|
|
my ($s, $r) = &{$ENCODER{$encoding}} ($encoding, $estr, $option);
|
| 368 |
|
|
if ($r->{-encoded}) { ## Success
|
| 369 |
|
|
my $echarset = {Message::MIME::Charset::name_minimumize ($echarset, $estr, {-name_only=>1})}->{charset};
|
| 370 |
|
|
push @estr, sprintf ('=?%s?%s?%s?=', $echarset, $encoding, $s);
|
| 371 |
|
|
} else {
|
| 372 |
|
|
return undef;
|
| 373 |
|
|
}
|
| 374 |
|
|
}
|
| 375 |
|
|
join ' ', @estr;
|
| 376 |
|
|
}
|
| 377 |
|
|
|
| 378 |
wakaba |
1.3 |
|
| 379 |
wakaba |
1.1 |
=head1 LICENSE
|
| 380 |
|
|
|
| 381 |
wakaba |
1.12 |
Copyright 2002 Wakaba <[email protected]>.
|
| 382 |
wakaba |
1.1 |
|
| 383 |
|
|
This program is free software; you can redistribute it and/or modify
|
| 384 |
|
|
it under the terms of the GNU General Public License as published by
|
| 385 |
|
|
the Free Software Foundation; either version 2 of the License, or
|
| 386 |
|
|
(at your option) any later version.
|
| 387 |
|
|
|
| 388 |
|
|
This program is distributed in the hope that it will be useful,
|
| 389 |
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
| 390 |
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
| 391 |
|
|
GNU General Public License for more details.
|
| 392 |
|
|
|
| 393 |
|
|
You should have received a copy of the GNU General Public License
|
| 394 |
|
|
along with this program; see the file COPYING. If not, write to
|
| 395 |
|
|
the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
| 396 |
|
|
Boston, MA 02111-1307, USA.
|
| 397 |
|
|
|
| 398 |
|
|
=cut
|
| 399 |
|
|
|
| 400 |
wakaba |
1.13 |
1; # $Date: 2002/12/28 09:07:05 $
|
Parent Directory
| 
Revision Log