Commit | Line | Data |
af1f55d9 |
1 | package Encode::MIME::Header; |
2 | use strict; |
3 | # use warnings; |
4 | our $VERSION = do { my @r = (q$Revision: 1.1 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r }; |
5 | |
6 | use Encode qw(find_encoding encode_utf8); |
7 | use MIME::Base64; |
8 | use Carp; |
9 | |
10 | my %seed = |
11 | ( |
12 | decode_b => '1', # decodes 'B' encoding ? |
13 | decode_q => '1', # decodes 'Q' encoding ? |
14 | encode => 'B', # encode with 'B' or 'Q' ? |
15 | bpl => 75, # bytes per line |
16 | ); |
17 | |
18 | $Encode::Encoding{'MIME-Header'} = |
19 | bless { |
20 | %seed, |
21 | Name => 'MIME-Header', |
22 | } => __PACKAGE__; |
23 | |
24 | $Encode::Encoding{'MIME-B'} = |
25 | bless { |
26 | %seed, |
27 | decode_q => 0, |
28 | Name => 'MIME-B', |
29 | } => __PACKAGE__; |
30 | |
31 | $Encode::Encoding{'MIME-Q'} = |
32 | bless { |
33 | %seed, |
34 | decode_q => 1, |
35 | encode => 'Q', |
36 | Name => 'MIME-Q', |
37 | } => __PACKAGE__; |
38 | |
39 | sub name { shift->{'Name'} } |
40 | sub new_sequence { $_[0] } |
41 | sub needs_lines { 1 } |
42 | sub perlio_ok{ 0 }; |
43 | |
44 | sub decode($$;$){ |
45 | use utf8; |
46 | my ($obj, $str, $chk) = @_; |
47 | # zap spaces between encoded words |
48 | $str =~ s/\?=\s+=\?/\?==\?/gos; |
49 | # multi-line header to single line |
50 | $str =~ s/(:?\r|\n|\r\n)[ \t]//gos; |
51 | $str =~ |
52 | s{ |
53 | =\? # begin encoded word |
54 | ([0-9A-Za-z\-]+) # charset (encoding) |
55 | \?([QqBb])\? # delimiter |
56 | (.*?) # Base64-encodede contents |
57 | \?= # end encoded word |
58 | }{ |
59 | if (uc($2) eq 'B'){ |
60 | $obj->{decode_b} or croak qq(MIME "B" unsupported); |
61 | decode_b($1, $3); |
62 | }elsif(uc($2) eq 'Q'){ |
63 | $obj->{decode_q} or croak qq(MIME "Q" unsupported); |
64 | decode_q($1, $3); |
65 | }else{ |
66 | croak qq(MIME "$2" encoding is nonexistent!); |
67 | } |
68 | }egox; |
69 | $_[1] = '' if $chk; |
70 | return $str; |
71 | } |
72 | |
73 | sub decode_b{ |
74 | my $enc = shift; |
75 | my $d = find_encoding($enc) or croak(Unknown encoding "$enc"); |
76 | my $db64 = decode_base64(shift); |
77 | return $d->decode($db64, Encode::FB_PERLQQ); |
78 | } |
79 | |
80 | sub decode_q{ |
81 | my ($enc, $q) = @_; |
82 | my $d = find_encoding($enc) or croak(Unknown encoding "$enc"); |
83 | $q =~ s/_/ /go; |
84 | $q =~ s/=([0-9A-Fa-f]{2})/pack("C", hex($1))/ego; |
85 | return $d->decode($q, Encode::FB_PERLQQ); |
86 | } |
87 | |
88 | my $especials = |
89 | join('|' => |
90 | map {quotemeta(chr($_))} |
91 | unpack("C*", qq{()<>@,;:\"\'/[]?.=})); |
92 | |
93 | my $re_especials = qr/$especials/o; |
94 | |
95 | sub encode($$;$){ |
96 | my ($obj, $str, $chk) = @_; |
97 | my @line = (); |
98 | for my $line (split /\r|\n|\r\n/o, $str){ |
99 | my (@word, @subline); |
100 | for my $word (split /($re_especials)/o, $line){ |
101 | if ($word =~ /[^\x00-\x7f]/o){ |
102 | push @word, $obj->_encode($word); |
103 | }else{ |
104 | push @word, $word; |
105 | } |
106 | } |
107 | my $subline = ''; |
108 | for my $word (@word){ |
109 | use bytes (); |
110 | if (bytes::length($subline) + bytes::length($word) > $obj->{bpl}){ |
111 | push @subline, $subline; |
112 | $subline = ''; |
113 | } |
114 | $subline .= $word; |
115 | } |
116 | $subline and push @subline, $subline; |
117 | push @line, join("\n " => @subline); |
118 | } |
119 | $_[1] = '' if $chk; |
120 | return join("\n", @line); |
121 | } |
122 | |
123 | use constant HEAD => '=?UTF-8?'; |
124 | use constant TAIL => '?='; |
125 | use constant SINGLE => { B => \&_encode_b, Q => \&_encode_q, }; |
126 | |
127 | sub _encode{ |
128 | my ($o, $str) = @_; |
129 | my $enc = $o->{encode}; |
130 | my $llen = ($o->{bpl} - length(HEAD) - 2 - length(TAIL)); |
131 | $llen *= $enc eq 'B' ? 3/4 : 1/3; |
132 | my @result = (); |
133 | my $chunk = ''; |
134 | while(my $chr = substr($str, 0, 1, '')){ |
135 | use bytes (); |
136 | if (bytes::length($chunk) + bytes::length($chr) > $llen){ |
137 | push @result, SINGLE->{$enc}($chunk); |
138 | $chunk = ''; |
139 | } |
140 | $chunk .= $chr; |
141 | } |
142 | $chunk and push @result, SINGLE->{$enc}($chunk); |
143 | return @result; |
144 | } |
145 | |
146 | sub _encode_b{ |
147 | HEAD . 'B?' . encode_base64(encode_utf8(shift), '') . TAIL; |
148 | } |
149 | |
150 | sub _encode_q{ |
151 | my $chunk = shift; |
152 | $chunk =~ s{ |
153 | ([^0-9A-Za-z]) |
154 | }{ |
155 | join("" => map {sprintf "=%02X", $_} unpack("C*", $1)) |
156 | }egox; |
157 | return HEAD . 'Q?' . $chunk . TAIL; |
158 | } |
159 | |
160 | 1; |
161 | __END__ |
162 | |
163 | =head1 NAME |
164 | |
165 | Encode::MIME::Header -- MIME 'B' and 'Q' header encoding |
166 | |
167 | =head1 SYNOPSIS |
168 | |
169 | use Encode qw/encode decode/; |
170 | $utf8 = decode('MIME-Header', $header); |
171 | $header = encode('MIME-Header', $utf8); |
172 | |
173 | =head1 ABSTRACT |
174 | |
175 | This module implements RFC 2047 Mime Header Encoding. There are 3 |
176 | variant encoding names; C<MIME-Header>, C<MIME-B> and C<MIME-Q>. The |
177 | difference is described below |
178 | |
179 | decode() encode() |
180 | ---------------------------------------------- |
181 | MIME-Header Both B and Q =?UTF-8?B?....?= |
182 | MIME-B B only; Q croaks =?UTF-8?B?....?= |
183 | MIME-Q Q only; B croaks =?UTF-8?Q?....?= |
184 | |
185 | =head1 DESCRIPTION |
186 | |
187 | When you decode(=?I<encoding>?I<X>?I<ENCODED WORD>?=), I<ENCODED WORD> |
188 | is extracted and decoded for I<X> encoding (B for Base64, Q for |
189 | Quoted-Printable). Then the decoded chunk is fed to |
190 | decode(I<encoding>). So long as I<encoding> is supported by Encode, |
191 | any source encoding is fine. |
192 | |
193 | When you encode, it just encodes UTF-8 string with I<X> encoding then |
194 | quoted with =?UTF-8?I<X>?....?= . The parts that RFC 2047 forbids to |
195 | encode are left as is and long lines are folded within 76 bytes per |
196 | line. |
197 | |
198 | =head1 BUGS |
199 | |
200 | It would be nice to support non-UTF8 encoding, such as =?ISO-2022-JP? |
201 | and =?ISO-8859-1?= but that makes the implementation too complicated. |
202 | These days major mail agents all support =?UTF-8? so I think it is |
203 | just good enough. |
204 | |
205 | =head1 SEE ALSO |
206 | |
207 | L<Encode> |
208 | |
209 | RFC 2047, L<http://www.faqs.org/rfcs/rfc2047.html> and many other |
210 | locations. |
211 | |
212 | =cut |