1 # This module is part of debbugs, and is released
2 # under the terms of the GPL version 2, or any later
3 # version at your option.
4 # See the file README and COPYING for more information.
6 # [Other people have contributed to this file; their copyrights should
8 # Copyright 2006 by Don Armstrong <don@donarmstrong.com>.
11 package Debbugs::MIME;
15 Debbugs::MIME -- Mime handling routines for debbugs
19 use Debbugs::MIME qw(parse decode_rfc1522);
33 use base qw(Exporter);
34 use vars qw($VERSION @EXPORT_OK);
39 @EXPORT_OK = qw(parse decode_rfc1522 encode_rfc1522 convert_to_utf8 create_mime_message getmailbody);
45 use POSIX qw(strftime);
46 use List::MoreUtils qw(apply);
49 use MIME::WordDecoder qw();
50 use Encode qw(decode encode encode_utf8 decode_utf8 is_utf8);
58 my $type = $entity->effective_type;
59 if ($type eq 'text/plain' or
60 ($type =~ m#text/# and $type ne 'text/html') or
61 $type eq 'application/pgp') {
62 return $entity->bodyhandle;
63 } elsif ($type eq 'multipart/alternative') {
64 # RFC 2046 says we should use the last part we recognize.
65 for my $part (reverse $entity->parts) {
66 my $ret = getmailbody($part);
70 # For other multipart types, we just pretend they're
71 # multipart/mixed and run through in order.
72 for my $part ($entity->parts) {
73 my $ret = getmailbody($part);
82 # header and decoded body respectively
83 my (@headerlines, @bodylines);
85 my $parser = MIME::Parser->new();
86 mkdir "mime.tmp.$$", 0777;
87 $parser->output_under("mime.tmp.$$");
88 my $entity = eval { $parser->parse_data($_[0]) };
90 if ($entity and $entity->head->tags) {
91 @headerlines = @{$entity->head->header};
94 my $entity_body = getmailbody($entity);
95 @bodylines = $entity_body ? $entity_body->as_lines() : ();
98 # Legacy pre-MIME code, kept around in case MIME::Parser fails.
99 my @msg = split /\n/, $_[0];
102 for ($i = 0; $i <= $#msg; ++$i) {
105 while ($msg[$i + 1] =~ /^\s/) {
107 $_ .= "\n" . $msg[$i];
109 push @headerlines, $_;
112 @bodylines = @msg[$i .. $#msg];
115 rmtree "mime.tmp.$$", 0, 1;
117 # Remove blank lines.
118 shift @bodylines while @bodylines and $bodylines[0] !~ /\S/;
120 # Strip off RFC2440-style PGP clearsigning.
121 if (@bodylines and $bodylines[0] =~ /^-----BEGIN PGP SIGNED/) {
122 shift @bodylines while @bodylines and length $bodylines[0];
123 shift @bodylines while @bodylines and $bodylines[0] !~ /\S/;
124 for my $findsig (0 .. $#bodylines) {
125 if ($bodylines[$findsig] =~ /^-----BEGIN PGP SIGNATURE/) {
126 $#bodylines = $findsig - 1;
130 map { s/^- // } @bodylines;
133 return { header => [@headerlines], body => [@bodylines]};
136 =head2 create_mime_message
138 create_mime_message([To=>'don@debian.org'],$body,[$attach1, $attach2],$include_date);
140 Creates a MIME encoded message with headers given by the first
141 argument, and a message given by the second.
143 Optional attachments can be specified in the third arrayref argument.
145 Whether to include the date in the header is the final argument; it
146 defaults to true, setting the Date header if one is not already
149 Headers are passed directly to MIME::Entity::build, the message is the
152 Each of the elements of the attachment arrayref is attached as an
153 rfc822 message if it is a scalar or an arrayref; otherwise if it is a
154 hashref, the contents are passed as an argument to
159 sub create_mime_message{
160 my ($headers,$body,$attachments,$include_date) = @_;
161 $attachments = [] if not defined $attachments;
162 $include_date = 1 if not defined $include_date;
164 die "The first argument to create_mime_message must be an arrayref" unless ref($headers) eq 'ARRAY';
165 die "The third argument to create_mime_message must be an arrayref" unless ref($attachments) eq 'ARRAY';
168 my %headers = apply {lc($_)} @{$headers};
169 if (not exists $headers{date}) {
172 strftime("%a, %d %b %Y %H:%M:%S +0000",gmtime)
178 # MIME::Entity is stupid, and doesn't rfc1522 encode its headers, so we do it for it.
179 my $msg = MIME::Entity->build('Content-Type' => 'text/plain; charset=utf-8',
180 'Encoding' => 'quoted-printable',
181 (map{encode_rfc1522($_)} @{$headers}),
185 # Attach the attachments
186 for my $attachment (@{$attachments}) {
187 if (ref($attachment) eq 'HASH') {
188 $msg->attach(%{$attachment});
191 # This is *craptacular*, but because various MTAs
192 # (sendmail and exim4, at least) appear to eat From
193 # lines in message/rfc822 attachments, we need eat
194 # the entire From line ourselves so the MTA doesn't
195 # leave \n detrius around.
196 if (ref($attachment) eq 'ARRAY' and $attachment->[1] =~ /^From /) {
197 # make a copy so that we don't screw up anything
198 # that is expecting this arrayref to stay constant
199 $attachment = [@{$attachment}];
200 # remove the from line
201 splice @$attachment, 1, 1;
203 elsif (not ref($attachment)) {
204 # It's a scalar; remove the from line
205 $attachment =~ s/^(Received:[^\n]+\n)(From [^\n]+\n)/$1/s;
207 $msg->attach(Type => 'message/rfc822',
213 return $msg->as_string;
219 sub convert_to_utf8 {
220 my ($data, $charset) = @_;
221 # raw data just gets returned (that's the charset WordDecorder
222 # uses when it doesn't know what to do)
223 return $data if $charset eq 'raw' or is_utf8($data,1);
226 # this encode/decode madness is to make sure that the data
227 # really is valid utf8 and that the is_utf8 flag is off.
228 $result = encode("utf8",decode($charset,$data))
231 warn "Unable to decode charset; '$charset' and '$data': $@";
238 =head2 decode_rfc1522
240 decode_rfc1522('=?iso-8859-1?Q?D=F6n_Armstr=F3ng?= <don@donarmstrong.com>')
242 Turn RFC-1522 names into the UTF-8 equivalent.
247 # Set up the default RFC1522 decoder, which turns all charsets that
248 # are supported into the appropriate UTF-8 charset.
249 MIME::WordDecoder->default(new MIME::WordDecoder(
250 ['*' => \&convert_to_utf8,
257 # this is craptacular, but leading space is hacked off by unmime.
259 my $leading_space = '';
260 $leading_space = $1 if $string =~ s/^(\s+)//;
261 # unmime calls the default MIME::WordDecoder handler set up at
262 # initialization time.
263 return $leading_space . MIME::WordDecoder::unmime($string);
266 =head2 encode_rfc1522
268 encode_rfc1522('Dön Armströng <don@donarmstrong.com>')
270 Encodes headers according to the RFC1522 standard by calling
271 MIME::Words::encode_mimeword on distinct words as appropriate.
275 # We cannot use MIME::Words::encode_mimewords because that function
276 # does not handle spaces properly at all.
281 # handle being passed undef properly
282 return undef if not defined $rawstr;
283 # We process words in reverse so we can preserve spacing between
284 # encoded words. This regex splits on word|nonword boundaries and
285 # nonword|nonword boundaries.
286 my @words = reverse split /(?:(?<=[\s\n])|(?=[\s\n]))/m, $rawstr;
288 my $previous_word_encoded = 0;
290 for my $word (@words) {
291 if ($word !~ m#[\x00-\x1F\x7F-\xFF]#o and $word ne ' ') {
292 $string = $word.$string;
293 $previous_word_encoded=0;
295 elsif ($word =~ /^[\s\n]$/) {
296 $string = $word.$string;
297 $previous_word_encoded = 0 if $word eq "\n";
300 my $encoded = MIME::Words::encode_mimeword($word, 'q', 'UTF-8');
301 # RFC 1522 mandates that segments be at most 76 characters
302 # long. If that's the case, we split the word up into 10
303 # character pieces and encode it. We must use the Encode
304 # magic here to avoid breaking on bit boundaries here.
305 if (length $encoded > 75) {
306 # Turn utf8 into the internal perl representation
307 # so . is a character, not a byte.
308 my $tempstr = decode_utf8($word,Encode::FB_DEFAULT);
310 # Strip it into 10 character long segments, and encode
312 # XXX It's possible that these segments are > 76 characters
313 while ($tempstr =~ s/(.{1,10})$//) {
314 # turn the character back into the utf8 representation.
315 my $tempword = encode_utf8($1);
316 # It may actually be better to eventually use
317 # the base64 encoding here, but I'm not sure
318 # if that's as widely supported as quoted
320 unshift @encoded, MIME::Words::encode_mimeword($tempword,'q','UTF-8');
322 $encoded = join(" ",@encoded);
323 # If the previous word was encoded, we must
324 # include a trailing _ that gets encoded as a
326 $encoded =~ s/\?\=$/_\?\=/ if $previous_word_encoded;
327 $string = $encoded.$string;
330 # If the previous word was encoded, we must
331 # include a trailing _ that gets encoded as a
333 $encoded =~ s/\?\=$/_\?\=/ if $previous_word_encoded;
334 $string = $encoded.$string;
336 $previous_word_encoded = 1;