4 ** This contains the functions necessary to detect and decode MIME
10 if (defined('mime_php'))
12 define('mime_php', true);
14 require_once('../functions/imap.php');
16 /** Setting up the objects that have the structure for the message **/
19 /** msg_header contains generic variables for values that **/
20 /** could be in a header. **/
22 var $type0 = '', $type1 = '', $boundary = '', $charset = '';
23 var $encoding = '', $size = 0, $to = array(), $from = '', $date = '';
24 var $cc = array(), $bcc = array(), $reply_to = '', $subject = '';
25 var $id = 0, $mailbox = '', $description = '', $filename = '';
26 var $entity_id = 0, $message_id = 0, $name = '';
30 /** message is the object that contains messages. It is a recursive
31 object in that through the $entities variable, it can contain
32 more objects of type message. See documentation in mime.txt for
33 a better description of how this works.
36 var $entities = array();
38 function addEntity ($msg) {
39 $this->entities
[] = $msg;
43 /* --------------------------------------------------------------------------------- */
45 /* --------------------------------------------------------------------------------- */
47 // This function gets the structure of a message and stores it in the "message" class.
48 // It will return this object for use with all relevant header information and
49 // fully parsed into the standard "message" object format.
50 function mime_structure ($imap_stream, $header) {
52 sqimap_messages_flag ($imap_stream, $header->id
, $header->id
, 'Seen');
53 $ssid = sqimap_session_id();
54 $lsid = strlen( $ssid );
56 fputs ($imap_stream, "$ssid FETCH $id BODYSTRUCTURE\r\n");
58 // This should use sqimap_read_data instead of reading it itself
60 $read = fgets ($imap_stream, 10000);
62 while( substr($read, 0, $lsid) <> $ssid &&
63 !feof( $imap_stream ) ) {
64 $bodystructure .= $read;
65 $read = fgets ($imap_stream, 10000);
67 $read = $bodystructure;
69 // isolate the body structure and remove beginning and end parenthesis
70 $read = trim(substr ($read, strpos(strtolower($read), 'bodystructure') +
13));
71 $read = trim(substr ($read, 0, -1));
72 $end = mime_match_parenthesis(0, $read);
73 while ($end == strlen($read)-1) {
74 $read = trim(substr ($read, 0, -1));
75 $read = trim(substr ($read, 1));
76 $end = mime_match_parenthesis(0, $read);
79 $msg = mime_parse_structure ($read, 0);
80 $msg->header
= $header;
84 // this starts the parsing of a particular structure. It is called recursively,
85 // so it can be passed different structures. It returns an object of type
87 // First, it checks to see if it is a multipart message. If it is, then it
88 // handles that as it sees is necessary. If it is just a regular entity,
89 // then it parses it and adds the necessary header information (by calling out
90 // to mime_get_elements()
91 function mime_parse_structure ($structure, $ent_id) {
94 if ($structure{0} == '(') {
95 $ent_id = mime_new_element_level($ent_id);
99 $end = mime_match_parenthesis ($start, $structure);
101 $element = substr($structure, $start+
1, ($end - $start)-1);
102 $ent_id = mime_increment_id ($ent_id);
103 $newmsg = mime_parse_structure ($element, $ent_id);
104 $msg->addEntity ($newmsg);
105 } while ($structure{$end+
1} == '(');
107 // parse the elements
108 $msg = mime_get_element ($structure, $msg, $ent_id);
113 // Increments the element ID. An element id can look like any of
114 // the following: 1, 1.2, 4.3.2.4.1, etc. This function increments
115 // the last number of the element id, changing 1.2 to 1.3.
116 function mime_increment_id ($id) {
118 if (strpos($id, ".")) {
119 $first = substr($id, 0, strrpos($id, "."));
120 $last = substr($id, strrpos($id, ".")+
1);
122 $new = $first . "." .$last;
130 // See comment for mime_increment_id().
131 // This adds another level on to the entity_id changing 1.3 to 1.3.0
132 // NOTE: 1.3.0 is not a valid element ID. It MUST be incremented
133 // before it can be used. I left it this way so as not to have
134 // to make a special case if it is the first entity_id. It
135 // always increments it, and that works fine.
136 function mime_new_element_level ($id) {
147 function mime_get_element (&$structure, $msg, $ent_id) {
150 $msg->header
= new msg_header();
151 $msg->header
->entity_id
= $ent_id;
152 $properties = array();
154 while (strlen($structure) > 0) {
155 $structure = trim($structure);
156 $char = $structure{0};
158 if (strtolower(substr($structure, 0, 3)) == 'nil') {
160 $structure = substr($structure, 3);
161 } else if ($char == '"') {
162 // loop through until we find the matching quote, and return that as a string
165 while ( ($char = $structure{$pos} ) <> '"' && $pos < strlen($structure)) {
169 $structure = substr($structure, strlen($text) +
2);
170 } else if ($char == '(') {
172 $end = mime_match_parenthesis (0, $structure);
173 $sub = substr($structure, 1, $end-1);
174 $properties = mime_get_props($properties, $sub);
175 $structure = substr($structure, strlen($sub) +
2);
177 // loop through until we find a space or an end parenthesis
179 $char = $structure{$pos};
181 while ($char != ' ' && $char != ')' && $pos < strlen($structure)) {
184 $char = $structure{$pos};
186 $structure = substr($structure, strlen($text));
189 // This is where all the text parts get put into the header
192 $msg->header
->type0
= strtolower($text);
195 $msg->header
->type1
= strtolower($text);
198 // Invisimail enclose images with <>
199 $msg->header
->id
= str_replace( '<', '', str_replace( '>', '', $text ) );
202 $msg->header
->description
= $text;
205 $msg->header
->encoding
= strtolower($text);
208 $msg->header
->size
= $text;
211 if ($msg->header
->type0
== 'text' && $elem_num == 8) {
212 // This is a plain text message, so lets get the number of lines
214 $msg->header
->num_lines
= $text;
216 } else if ($msg->header
->type0
== 'message' && $msg->header
->type1
== 'rfc822' && $elem_num == 8) {
217 // This is an encapsulated message, so lets start all over again and
218 // parse this message adding it on to the existing one.
219 $structure = trim($structure);
220 if ( $structure{0} == '(' ) {
221 $e = mime_match_parenthesis (0, $structure);
222 $structure = substr($structure, 0, $e);
223 $structure = substr($structure, 1);
224 $m = mime_parse_structure($structure, $msg->header
->entity_id
);
226 // the following conditional is there to correct a bug that wasn't
227 // incrementing the entity IDs correctly because of the special case
228 // that message/rfc822 is. This fixes it fine.
229 if (substr($structure, 1, 1) != '(')
230 $m->header
->entity_id
= mime_increment_id(mime_new_element_level($ent_id));
232 // Now we'll go through and reformat the results.
234 for ($i=0; $i < count($m->entities
); $i++
) {
235 $msg->addEntity($m->entities
[$i]);
248 // loop through the additional properties and put those in the various headers
249 if ($msg->header
->type0
!= 'message') {
250 for ($i=0; $i < count($properties); $i++
) {
251 $msg->header
->{$properties[$i]['name']} = $properties[$i]['value'];
258 // I did most of the MIME stuff yesterday (June 20, 2000), but I couldn't
259 // figure out how to do this part, so I decided to go to bed. I woke up
260 // in the morning and had a flash of insight. I went to the white-board
261 // and scribbled it out, then spent a bit programming it, and this is the
262 // result. Nothing complicated, but I think my brain was fried yesterday.
263 // Funny how that happens some times.
265 // This gets properties in a nested parenthesisized list. For example,
266 // this would get passed something like: ("attachment" ("filename" "luke.tar.gz"))
267 // This returns an array called $props with all paired up properties.
268 // It ignores the "attachment" for now, maybe that should change later
269 // down the road. In this case, what is returned is:
270 // $props[0]["name"] = "filename";
271 // $props[0]["value"] = "luke.tar.gz";
272 function mime_get_props ($props, $structure) {
274 while (strlen($structure) > 0) {
275 $structure = trim($structure);
276 $char = $structure{0};
281 while ( ( $char = $structure{$pos} ) != '"' &&
282 $pos < strlen($structure)) {
286 $structure = trim(substr($structure, strlen($tmp) +
2));
287 $char = $structure{0};
292 while ( ( $char = $structure{$pos} ) != '"' &&
293 $pos < strlen($structure) ) {
297 $structure = trim(substr($structure, strlen($tmp) +
2));
300 $props[$k]['name'] = strtolower($tmp);
301 $props[$k]['value'] = $value;
302 } else if ($char == '(') {
303 $end = mime_match_parenthesis (0, $structure);
304 $sub = substr($structure, 1, $end-1);
307 $props = mime_get_props($props, $sub);
308 $structure = substr($structure, strlen($sub) +
2);
311 } else if ($char == '(') {
312 $end = mime_match_parenthesis (0, $structure);
313 $sub = substr($structure, 1, $end-1);
314 $props = mime_get_props($props, $sub);
315 $structure = substr($structure, strlen($sub) +
2);
323 // Matches parenthesis. It will return the position of the matching
324 // parenthesis in $structure. For instance, if $structure was:
325 // ("text" "plain" ("val1name", "1") nil ... )
327 // then this would return 42 to match up those two.
328 function mime_match_parenthesis ($pos, $structure) {
330 $j = strlen( $structure );
332 // ignore all extra characters
333 // If inside of a string, skip string -- Boundary IDs and other
334 // things can have ) in them.
335 if( $structure{$pos} != '(' )
340 if ($structure{$pos} == ')') {
342 } elseif ($structure{$pos} == '"') {
344 while( $structure{$pos} != '"' &&
346 if (substr($structure, $pos, 2) == '\\"')
348 elseif (substr($structure, $pos, 2) == '\\\\')
352 } elseif ( $structure{$pos} == '(' ) {
353 $pos = mime_match_parenthesis ($pos, $structure);
356 echo "Error decoding mime structure. Report this as a bug!<br>\n";
360 function mime_fetch_body ($imap_stream, $id, $ent_id) {
361 // do a bit of error correction. If we couldn't find the entity id, just guess
362 // that it is the first one. That is usually the case anyway.
363 if (!$ent_id) $ent_id = 1;
365 fputs ($imap_stream, sqimap_session_id() . " FETCH $id BODY[$ent_id]\r\n");
366 $data = sqimap_read_data ($imap_stream, sqimap_session_id(), true, $response, $message);
367 $topline = array_shift($data);
368 while (! ereg('\\* [0-9]+ FETCH ', $topline) && $data)
369 $topline = array_shift($data);
370 $wholemessage = implode('', $data);
372 if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) {
373 return substr($wholemessage, 0, $regs[1]);
375 else if (ereg('"([^"]*)"', $topline, $regs)) {
379 $str = "Body retrieval error. Please report this bug!\n" .
380 "Response: $response\n" .
381 "Message: $message\n" .
382 "FETCH line: $topline" .
383 "---------------\n$wholemessage";
384 foreach ($data as $d) {
385 $str .= htmlspecialchars($d) . "\n";
390 function mime_print_body_lines ($imap_stream, $id, $ent_id, $encoding) {
391 // do a bit of error correction. If we couldn't find the entity id, just guess
392 // that it is the first one. That is usually the case anyway.
393 if (!$ent_id) $ent_id = 1;
394 $sid = sqimap_session_id();
395 // Don't kill the connection if the browser is over a dialup
396 // and it would take over 30 seconds to download it.
399 fputs ($imap_stream, "$sid FETCH $id BODY[$ent_id]\r\n");
402 $read = fgets ($imap_stream,4096);
403 // This could be bad -- if the section has sqimap_session_id() . ' OK'
404 // or similar, it will kill the download.
405 while (!ereg("^".$sid." (OK|BAD|NO)(.*)$", $read, $regs)) {
406 if (trim($read) == ')==') {
408 $read = fgets ($imap_stream,4096);
409 if (ereg("^".$sid." (OK|BAD|NO)(.*)$", $read, $regs)) {
412 echo decodeBody($read1, $encoding) .
413 decodeBody($read, $encoding);
416 echo decodeBody($read, $encoding);
418 $read = fgets ($imap_stream,4096);
423 /* -[ END MIME DECODING ]----------------------------------------------------------- */
427 /** This is the first function called. It decides if this is a multipart
428 message or if it should be handled as a single entity
430 function decodeMime ($imap_stream, &$header) {
431 global $username, $key, $imapServerAddress, $imapPort;
432 return mime_structure ($imap_stream, $header);
435 // This is here for debugging purposese. It will print out a list
436 // of all the entity IDs that are in the $message object.
438 function listEntities ($message) {
440 if ($message->header->entity_id)
441 echo "<tt>" . $message->header->entity_id . ' : ' . $message->header->type0 . '/' . $message->header->type1 . '<br>';
442 for ($i = 0; $message->entities[$i]; $i++) {
443 $msg = listEntities($message->entities[$i], $ent_id);
451 // returns a $message object for a particular entity id
452 function getEntity ($message, $ent_id) {
454 if ($message->header
->entity_id
== $ent_id && strlen($ent_id) == strlen($message->header
->entity_id
)) {
457 for ($i = 0; isset($message->entities
[$i]); $i++
) {
458 $msg = getEntity ($message->entities
[$i], $ent_id);
466 // figures out what entity to display and returns the $message object
468 function findDisplayEntity ($message, $textOnly = 1)
470 global $show_html_default;
475 if ($message->header
->type0
== 'multipart' &&
476 $message->header
->type1
== 'alternative' &&
477 $show_html_default && ! $textOnly) {
478 $entity = findDisplayEntityHTML($message);
483 // Show text/plain or text/html -- the first one we find.
484 if ( $message->header
->type0
== 'text' &&
485 ( $message->header
->type1
== 'plain' ||
486 $message->header
->type1
== 'html' ) &&
487 isset($message->header
->entity_id
) )
488 return $message->header
->entity_id
;
490 for ($i=0; isset($message->entities
[$i]); $i++
) {
491 $entity = findDisplayEntity($message->entities
[$i], $textOnly);
499 // Shows the HTML version
500 function findDisplayEntityHTML ($message) {
501 if ($message->header
->type0
== 'text' &&
502 $message->header
->type1
== 'html' &&
503 isset($message->header
->entity_id
))
504 return $message->header
->entity_id
;
505 for ($i = 0; isset($message->entities
[$i]); $i ++
) {
506 $entity = findDisplayEntityHTML($message->entities
[$i]);
513 /** This returns a parsed string called $body. That string can then
514 be displayed as the actual message in the HTML. It contains
515 everything needed, including HTML Tags, Attachments at the
518 function formatBody($imap_stream, $message, $color, $wrap_at) {
519 // this if statement checks for the entity to show as the
520 // primary message. To add more of them, just put them in the
521 // order that is their priority.
522 global $startMessage, $username, $key, $imapServerAddress, $imapPort,
525 $id = $message->header
->id
;
526 $urlmailbox = urlencode($message->header
->mailbox
);
528 // Get the right entity and redefine message to be this entity
529 // Pass the 0 to mean that we want the 'best' viewable one
530 $ent_num = findDisplayEntity ($message, 0);
531 $body_message = getEntity($message, $ent_num);
532 if (($body_message->header
->type0
== 'text') ||
533 ($body_message->header
->type0
== 'rfc822')) {
535 $body = mime_fetch_body ($imap_stream, $id, $ent_num);
536 $body = decodeBody($body, $body_message->header
->encoding
);
537 $hookResults = do_hook("message_body", $body);
538 $body = $hookResults[1];
540 // If there are other types that shouldn't be formatted, add
542 if ($body_message->header
->type1
!= "html" ||
! $show_html_default) {
543 translateText($body, $wrap_at, $body_message->header
->charset
);
546 $body .= "<SMALL><CENTER><A HREF=\"../src/download.php?absolute_dl=true&passed_id=$id&passed_ent_id=$ent_num&mailbox=$urlmailbox&showHeaders=1\">". _("Download this as a file") ."</A></CENTER><BR></SMALL>";
548 /** Display the ATTACHMENTS: message if there's more than one part **/
549 $body .= "</TD></TR></TABLE>";
550 if (isset($message->entities
[0])) {
551 $body .= formatAttachments ($message, $ent_num, $message->header
->mailbox
, $id);
553 $body .= "</TD></TR></TABLE>";
555 $body = formatAttachments ($message, -1, $message->header
->mailbox
, $id);
560 // A recursive function that returns a list of attachments with links
561 // to where to download these attachments
562 function formatAttachments ($message, $ent_id, $mailbox, $id) {
563 global $where, $what;
564 global $startMessage, $color;
565 static $ShownHTML = 0;
568 if ($ShownHTML == 0) {
571 $body .= "<TABLE WIDTH=100% CELLSPACING=0 CELLPADDING=2 BORDER=0 BGCOLOR=\"$color[0]\"><TR>\n" .
572 "<TH ALIGN=\"left\" BGCOLOR=\"$color[9]\"><B>\n" .
573 _("Attachments") . ':' .
574 "</B></TH></TR><TR><TD>\n" .
575 "<TABLE CELLSPACING=0 CELLPADDING=1 BORDER=0>\n" .
576 formatAttachments ($message, $ent_id, $mailbox, $id) .
577 "</TABLE></TD></TR></TABLE>";
583 if (!$message->entities
) {
584 $type0 = strtolower($message->header
->type0
);
585 $type1 = strtolower($message->header
->type1
);
586 $name = decodeHeader($message->header
->name
);
588 if ($message->header
->entity_id
!= $ent_id) {
589 $filename = decodeHeader($message->header
->filename
);
590 if (trim($filename) == '') {
591 if (trim($name) == '') {
592 if( trim( $message->header
->id
) == '' )
593 $display_filename = 'untitled-[' . $message->header
->entity_id
. ']' ;
595 $display_filename = 'cid: ' . $message->header
->id
;
596 // $display_filename = 'untitled-[' . $message->header->entity_id . ']' ;
598 $display_filename = $name;
602 $display_filename = $filename;
605 $urlMailbox = urlencode($mailbox);
606 $ent = urlencode($message->header
->entity_id
);
609 "../src/download.php?startMessage=$startMessage&passed_id=$id&mailbox=$urlMailbox&passed_ent_id=$ent";
611 $DefaultLink .= '&where=' . urlencode($where) . '&what=' . urlencode($what);
612 $Links['download link']['text'] = _("download");
613 $Links['download link']['href'] =
614 "../src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&passed_ent_id=$ent";
617 $HookResults = do_hook("attachment $type0/$type1", $Links,
618 $startMessage, $id, $urlMailbox, $ent, $DefaultLink,
619 $display_filename, $where, $what);
621 $Links = $HookResults[1];
622 $DefaultLink = $HookResults[6];
624 $body .= '<TR><TD> </TD><TD>' .
625 "<A HREF=\"$DefaultLink\">$display_filename</A> </TD>" .
626 '<TD><SMALL><b>' . show_readable_size($message->header
->size
) .
627 '</b> </small></TD>' .
628 "<TD><SMALL>[ $type0/$type1 ] </SMALL></TD>" .
630 if ($message->header
->description
)
631 $body .= '<b>' . htmlspecialchars($message->header
->description
) . '</b>';
632 $body .= '</SMALL></TD><TD><SMALL> ';
636 foreach ($Links as $Val) {
640 $body .= ' | ';
642 $body .= '<a href="' . $Val['href'] . '">' . $Val['text'] . '</a>';
647 $body .= "</SMALL></TD></TR>\n";
650 for ($i = 0; $i < count($message->entities
); $i++
) {
651 $body .= formatAttachments ($message->entities
[$i], $ent_id, $mailbox, $id);
659 /** this function decodes the body depending on the encoding type. **/
660 function decodeBody($body, $encoding) {
661 $body = str_replace("\r\n", "\n", $body);
662 $encoding = strtolower($encoding);
664 global $show_html_default;
666 if ($encoding == 'quoted-printable') {
667 $body = quoted_printable_decode($body);
671 Following code has been comented as I see no reason for it.
672 If there is any please tell me a mingo@rotedic.com
674 while (ereg("=\n", $body))
675 $body = ereg_replace ("=\n", "", $body);
677 } else if ($encoding == 'base64') {
678 $body = base64_decode($body);
681 // All other encodings are returned raw.
686 // This functions decode strings that is encoded according to
687 // RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text).
688 function decodeHeader ($string) {
689 if (eregi('=\\?([^?]+)\\?(q|b)\\?([^?]+)\\?=',
691 if (ucfirst($res[2]) == "B") {
692 $replace = base64_decode($res[3]);
694 $replace = ereg_replace("_", " ", $res[3]);
695 // Convert lowercase Quoted Printable to uppercase for
696 // quoted_printable_decode to understand it.
697 while (ereg("(=(([0-9][abcdef])|([abcdef][0-9])|([abcdef][abcdef])))", $replace, $res)) {
698 $replace = str_replace($res[1], strtoupper($res[1]), $replace);
700 $replace = quoted_printable_decode($replace);
703 $replace = charset_decode ($res[1], $replace);
705 // Remove the name of the character set.
706 $string = eregi_replace ('=\\?([^?]+)\\?(q|b)\\?([^?]+)\\?=',
709 // In case there should be more encoding in the string: recurse
710 return (decodeHeader($string));
715 // Encode a string according to RFC 1522 for use in headers if it
716 // contains 8-bit characters or anything that looks like it should
718 function encodeHeader ($string) {
719 global $default_charset;
721 // Encode only if the string contains 8-bit characters or =?
722 $j = strlen( $string );
723 $l = FALSE; // Must be encoded ?
725 for( $i=0; $i < $j; ++
$i) {
726 switch( $string{$i} ) {
741 $k = ord( $string{$i} );
743 $ret .= sprintf("=%02X", $k);
751 $string = "=?$default_charset?Q?$ret?=";