59177427 |
1 | <?php |
2ba13803 |
2 | |
35586184 |
3 | /** |
4 | * mime.php |
5 | * |
76911253 |
6 | * Copyright (c) 1999-2003 The SquirrelMail Project Team |
35586184 |
7 | * Licensed under the GNU GPL. For full terms see the file COPYING. |
8 | * |
9 | * This contains the functions necessary to detect and decode MIME |
10 | * messages. |
11 | * |
12 | * $Id$ |
d6c32258 |
13 | * @package squirrelmail |
35586184 |
14 | */ |
b74ba498 |
15 | |
d6c32258 |
16 | /** The typical includes... */ |
b68edc75 |
17 | require_once(SM_PATH . 'functions/imap.php'); |
18 | require_once(SM_PATH . 'functions/attachment_common.php'); |
8beafbbc |
19 | |
7c7b74b3 |
20 | /* -------------------------------------------------------------------------- */ |
21 | /* MIME DECODING */ |
22 | /* -------------------------------------------------------------------------- */ |
b74ba498 |
23 | |
d6c32258 |
24 | /** |
25 | * Get the MIME structure |
07c49f57 |
26 | * |
d6c32258 |
27 | * This function gets the structure of a message and stores it in the "message" class. |
451f74a2 |
28 | * It will return this object for use with all relevant header information and |
29 | * fully parsed into the standard "message" object format. |
30 | */ |
a4a70693 |
31 | function mime_structure ($bodystructure, $flags=array()) { |
c9d78ab4 |
32 | |
3d8371be |
33 | /* Isolate the body structure and remove beginning and end parenthesis. */ |
a4a70693 |
34 | $read = trim(substr ($bodystructure, strpos(strtolower($bodystructure), 'bodystructure') + 13)); |
451f74a2 |
35 | $read = trim(substr ($read, 0, -1)); |
22efa9fb |
36 | $i = 0; |
37 | $msg = Message::parseStructure($read,$i); |
9de42168 |
38 | if (!is_object($msg)) { |
bd9c880b |
39 | include_once(SM_PATH . 'functions/display_messages.php'); |
3d8371be |
40 | global $color, $mailbox; |
c96c32f4 |
41 | /* removed urldecode because $_GET is auto urldecoded ??? */ |
a48eba8f |
42 | displayPageHeader( $color, $mailbox ); |
5e8de8b6 |
43 | $errormessage = _("SquirrelMail could not decode the bodystructure of the message"); |
3d8371be |
44 | $errormessage .= '<BR>'._("the provided bodystructure by your imap-server").':<BR><BR>'; |
472e7acb |
45 | $errormessage .= '<pre>' . htmlspecialchars($read) . '</pre>'; |
9de42168 |
46 | plain_error_message( $errormessage, $color ); |
3d8371be |
47 | echo '</body></html>'; |
9de42168 |
48 | exit; |
49 | } |
a4a70693 |
50 | if (count($flags)) { |
7a9e9c89 |
51 | foreach ($flags as $flag) { |
52 | $char = strtoupper($flag{1}); |
53 | switch ($char) { |
3d8371be |
54 | case 'S': |
55 | if (strtolower($flag) == '\\seen') { |
56 | $msg->is_seen = true; |
57 | } |
58 | break; |
59 | case 'A': |
60 | if (strtolower($flag) == '\\answered') { |
61 | $msg->is_answered = true; |
62 | } |
63 | break; |
64 | case 'D': |
65 | if (strtolower($flag) == '\\deleted') { |
66 | $msg->is_deleted = true; |
67 | } |
68 | break; |
69 | case 'F': |
70 | if (strtolower($flag) == '\\flagged') { |
71 | $msg->is_flagged = true; |
72 | } |
73 | break; |
74 | case 'M': |
75 | if (strtolower($flag) == '$mdnsent') { |
76 | $msg->is_mdnsent = true; |
77 | } |
78 | break; |
79 | default: |
80 | break; |
7a9e9c89 |
81 | } |
82 | } |
451f74a2 |
83 | } |
7a9e9c89 |
84 | // listEntities($msg); |
3d8371be |
85 | return $msg; |
451f74a2 |
86 | } |
b74ba498 |
87 | |
22efa9fb |
88 | |
89 | |
3d8371be |
90 | /* This starts the parsing of a particular structure. It is called recursively, |
451f74a2 |
91 | * so it can be passed different structures. It returns an object of type |
92 | * $message. |
93 | * First, it checks to see if it is a multipart message. If it is, then it |
94 | * handles that as it sees is necessary. If it is just a regular entity, |
95 | * then it parses it and adds the necessary header information (by calling out |
96 | * to mime_get_elements() |
97 | */ |
451f74a2 |
98 | |
4d592352 |
99 | function mime_fetch_body($imap_stream, $id, $ent_id=1, $fetch_size=0) { |
da2415c1 |
100 | global $uid_support; |
3d8371be |
101 | /* Do a bit of error correction. If we couldn't find the entity id, just guess |
09a4bde3 |
102 | * that it is the first one. That is usually the case anyway. |
103 | */ |
7c7b74b3 |
104 | |
09a4bde3 |
105 | if (!$ent_id) { |
08b7f7cc |
106 | $cmd = "FETCH $id BODY[]"; |
1035e159 |
107 | } else { |
08b7f7cc |
108 | $cmd = "FETCH $id BODY[$ent_id]"; |
09a4bde3 |
109 | } |
3d8371be |
110 | |
4d592352 |
111 | if ($fetch_size!=0) $cmd .= "<0.$fetch_size>"; |
da2415c1 |
112 | |
a4a70693 |
113 | $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, $uid_support); |
77b88425 |
114 | do { |
3d8371be |
115 | $topline = trim(array_shift($data)); |
116 | } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ; |
a4a70693 |
117 | |
451f74a2 |
118 | $wholemessage = implode('', $data); |
119 | if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) { |
3d8371be |
120 | $ret = substr($wholemessage, 0, $regs[1]); |
121 | /* There is some information in the content info header that could be important |
122 | * in order to parse html messages. Let's get them here. |
123 | */ |
0600bdf1 |
124 | // if ($ret{0} == '<') { |
125 | // $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, $uid_support); |
126 | // } |
451f74a2 |
127 | } else if (ereg('"([^"]*)"', $topline, $regs)) { |
128 | $ret = $regs[1]; |
129 | } else { |
130 | global $where, $what, $mailbox, $passed_id, $startMessage; |
3d8371be |
131 | $par = 'mailbox=' . urlencode($mailbox) . '&passed_id=' . $passed_id; |
451f74a2 |
132 | if (isset($where) && isset($what)) { |
3d8371be |
133 | $par .= '&where=' . urlencode($where) . '&what=' . urlencode($what); |
a3daaaf3 |
134 | } else { |
3d8371be |
135 | $par .= '&startMessage=' . $startMessage . '&show_more=0'; |
451f74a2 |
136 | } |
e5ea9327 |
137 | $par .= '&response=' . urlencode($response) . |
3d8371be |
138 | '&message=' . urlencode($message) . |
139 | '&topline=' . urlencode($topline); |
a019eeb8 |
140 | |
346817d4 |
141 | echo '<tt><br>' . |
142 | '<table width="80%"><tr>' . |
143 | '<tr><td colspan=2>' . |
0e5b61b4 |
144 | _("Body retrieval error. The reason for this is most probably that the message is malformed.") . |
346817d4 |
145 | '</td></tr>' . |
0e5b61b4 |
146 | '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" . |
147 | '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" . |
148 | '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" . |
149 | '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" . |
346817d4 |
150 | "</table><BR></tt></font><hr>"; |
151 | |
a4a70693 |
152 | $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, $uid_support); |
451f74a2 |
153 | array_shift($data); |
154 | $wholemessage = implode('', $data); |
a019eeb8 |
155 | |
346817d4 |
156 | $ret = $wholemessage; |
a3daaaf3 |
157 | } |
3d8371be |
158 | return $ret; |
451f74a2 |
159 | } |
d4467150 |
160 | |
1035e159 |
161 | function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding) { |
a4a70693 |
162 | global $uid_support; |
1035e159 |
163 | |
3d8371be |
164 | /* Don't kill the connection if the browser is over a dialup |
165 | * and it would take over 30 seconds to download it. |
7e6ca3e8 |
166 | * Dont call set_time_limit in safe mode. |
3d8371be |
167 | */ |
b7206e1d |
168 | |
3d8371be |
169 | if (!ini_get('safe_mode')) { |
b7206e1d |
170 | set_time_limit(0); |
171 | } |
7c7b74b3 |
172 | /* in case of base64 encoded attachments, do not buffer them. |
173 | Instead, echo the decoded attachment directly to screen */ |
174 | if (strtolower($encoding) == 'base64') { |
175 | if (!$ent_id) { |
176 | $query = "FETCH $id BODY[]"; |
177 | } else { |
178 | $query = "FETCH $id BODY[$ent_id]"; |
179 | } |
180 | sqimap_run_command($imap_stream,$query,true,$response,$message,$uid_support,'sqimap_base64_decode','php://stdout',true); |
1d142b8d |
181 | } else { |
7c7b74b3 |
182 | $body = mime_fetch_body ($imap_stream, $id, $ent_id); |
183 | echo decodeBody($body, $encoding); |
1d142b8d |
184 | } |
346817d4 |
185 | |
da2415c1 |
186 | /* |
7c7b74b3 |
187 | TODO, use the same method for quoted printable. |
188 | However, I assume that quoted printable attachments aren't that large |
189 | so the performancegain / memory usage drop will be minimal. |
190 | If we decide to add that then we need to adapt sqimap_fread because |
da2415c1 |
191 | we need to split te result on \n and fread doesn't stop at \n. That |
7c7b74b3 |
192 | means we also should provide $results from sqimap_fread (by ref) to |
193 | te function and set $no_return to false. The $filter function for |
da2415c1 |
194 | quoted printable should handle unsetting of $results. |
7c7b74b3 |
195 | */ |
da2415c1 |
196 | /* |
7c7b74b3 |
197 | TODO 2: find out how we write to the output stream php://stdout. fwrite |
198 | doesn't work because 'php://stdout isn't a stream. |
199 | */ |
200 | |
5d9c6f73 |
201 | return; |
202 | /* |
451f74a2 |
203 | fputs ($imap_stream, "$sid FETCH $id BODY[$ent_id]\r\n"); |
204 | $cnt = 0; |
205 | $continue = true; |
1d142b8d |
206 | $read = fgets ($imap_stream,8192); |
207 | |
208 | |
451f74a2 |
209 | // This could be bad -- if the section has sqimap_session_id() . ' OK' |
210 | // or similar, it will kill the download. |
1d142b8d |
211 | while (!ereg("^".$sid_s." (OK|BAD|NO)(.*)$", $read, $regs)) { |
5d9c6f73 |
212 | if (trim($read) == ')==') { |
213 | $read1 = $read; |
214 | $read = fgets ($imap_stream,4096); |
215 | if (ereg("^".$sid." (OK|BAD|NO)(.*)$", $read, $regs)) { |
216 | return; |
217 | } else { |
218 | echo decodeBody($read1, $encoding) . |
219 | decodeBody($read, $encoding); |
220 | } |
221 | } else if ($cnt) { |
222 | echo decodeBody($read, $encoding); |
223 | } |
224 | $read = fgets ($imap_stream,4096); |
225 | $cnt++; |
1d142b8d |
226 | // break; |
451f74a2 |
227 | } |
5d9c6f73 |
228 | */ |
451f74a2 |
229 | } |
beb9e459 |
230 | |
451f74a2 |
231 | /* -[ END MIME DECODING ]----------------------------------------------------------- */ |
d4467150 |
232 | |
3d8371be |
233 | /* This is here for debugging purposes. It will print out a list |
234 | * of all the entity IDs that are in the $message object. |
235 | */ |
451f74a2 |
236 | function listEntities ($message) { |
3d8371be |
237 | if ($message) { |
238 | echo "<tt>" . $message->entity_id . ' : ' . $message->type0 . '/' . $message->type1 . ' parent = '. $message->parent->entity_id. '<br>'; |
239 | for ($i = 0; isset($message->entities[$i]); $i++) { |
240 | echo "$i : "; |
241 | $msg = listEntities($message->entities[$i]); |
242 | |
243 | if ($msg) { |
244 | echo "return: "; |
245 | return $msg; |
246 | } |
247 | } |
a4a70693 |
248 | } |
451f74a2 |
249 | } |
f0c4dc12 |
250 | |
f792c641 |
251 | function getPriorityStr($priority) { |
3d8371be |
252 | $priority_level = substr($priority,0,1); |
253 | |
254 | switch($priority_level) { |
255 | /* Check for a higher then normal priority. */ |
256 | case '1': |
257 | case '2': |
258 | $priority_string = _("High"); |
259 | break; |
260 | |
261 | /* Check for a lower then normal priority. */ |
262 | case '4': |
263 | case '5': |
264 | $priority_string = _("Low"); |
265 | break; |
266 | |
267 | /* Check for a normal priority. */ |
268 | case '3': |
269 | default: |
270 | $priority_level = '3'; |
271 | $priority_string = _("Normal"); |
272 | break; |
273 | |
274 | } |
275 | return $priority_string; |
f792c641 |
276 | } |
277 | |
451f74a2 |
278 | /* returns a $message object for a particular entity id */ |
279 | function getEntity ($message, $ent_id) { |
a4a70693 |
280 | return $message->getEntity($ent_id); |
451f74a2 |
281 | } |
8beafbbc |
282 | |
3d8371be |
283 | /* translateText |
da4c66e8 |
284 | * Extracted from strings.php 23/03/2002 |
285 | */ |
286 | |
287 | function translateText(&$body, $wrap_at, $charset) { |
3d8371be |
288 | global $where, $what; /* from searching */ |
289 | global $color; /* color theme */ |
da4c66e8 |
290 | |
b68edc75 |
291 | require_once(SM_PATH . 'functions/url_parser.php'); |
da4c66e8 |
292 | |
293 | $body_ary = explode("\n", $body); |
da4c66e8 |
294 | for ($i=0; $i < count($body_ary); $i++) { |
295 | $line = $body_ary[$i]; |
296 | if (strlen($line) - 2 >= $wrap_at) { |
297 | sqWordWrap($line, $wrap_at); |
298 | } |
299 | $line = charset_decode($charset, $line); |
300 | $line = str_replace("\t", ' ', $line); |
301 | |
302 | parseUrl ($line); |
303 | |
3d8371be |
304 | $quotes = 0; |
da4c66e8 |
305 | $pos = 0; |
3d8371be |
306 | $j = strlen($line); |
da4c66e8 |
307 | |
3d8371be |
308 | while ($pos < $j) { |
da4c66e8 |
309 | if ($line[$pos] == ' ') { |
3d8371be |
310 | $pos++; |
da4c66e8 |
311 | } else if (strpos($line, '>', $pos) === $pos) { |
312 | $pos += 4; |
3d8371be |
313 | $quotes++; |
da4c66e8 |
314 | } else { |
315 | break; |
316 | } |
317 | } |
3d8371be |
318 | |
83c94382 |
319 | if ($quotes % 2) { |
3d8371be |
320 | if (!isset($color[13])) { |
da4c66e8 |
321 | $color[13] = '#800000'; |
322 | } |
4c25967c |
323 | $line = '<font color="' . $color[13] . '">' . $line . '</font>'; |
324 | } elseif ($quotes) { |
325 | if (!isset($color[14])) { |
326 | $color[14] = '#FF0000'; |
327 | } |
328 | $line = '<font color="' . $color[14] . '">' . $line . '</font>'; |
da4c66e8 |
329 | } |
3d8371be |
330 | |
da4c66e8 |
331 | $body_ary[$i] = $line; |
332 | } |
333 | $body = '<pre>' . implode("\n", $body_ary) . '</pre>'; |
334 | } |
335 | |
451f74a2 |
336 | /* This returns a parsed string called $body. That string can then |
3d8371be |
337 | * be displayed as the actual message in the HTML. It contains |
338 | * everything needed, including HTML Tags, Attachments at the |
339 | * bottom, etc. |
340 | */ |
b3af12ef |
341 | function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX') { |
3d8371be |
342 | /* This if statement checks for the entity to show as the |
343 | * primary message. To add more of them, just put them in the |
344 | * order that is their priority. |
345 | */ |
008989da |
346 | global $startMessage, $username, $key, $imapServerAddress, $imapPort, |
23f617b8 |
347 | $show_html_default, $sort, $has_unsafe_images, $passed_ent_id; |
08b7f7cc |
348 | global $languages, $squirrelmail_language; |
77bfbd2e |
349 | |
5262d9a6 |
350 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
351 | $view_unsafe_images = false; |
77bfbd2e |
352 | } |
d03c24f4 |
353 | |
cc34b00d |
354 | $body = ''; |
23bcec6f |
355 | $urlmailbox = urlencode($mailbox); |
451f74a2 |
356 | $body_message = getEntity($message, $ent_num); |
357 | if (($body_message->header->type0 == 'text') || |
358 | ($body_message->header->type0 == 'rfc822')) { |
3d8371be |
359 | $body = mime_fetch_body ($imap_stream, $id, $ent_num); |
451f74a2 |
360 | $body = decodeBody($body, $body_message->header->encoding); |
e842b215 |
361 | |
362 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
363 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
364 | if (mb_detect_encoding($body) != 'ASCII') { |
365 | $body = $languages[$squirrelmail_language]['XTRA_CODE']('decode', $body); |
366 | } |
367 | } |
451f74a2 |
368 | $hookResults = do_hook("message_body", $body); |
369 | $body = $hookResults[1]; |
23bcec6f |
370 | |
3d8371be |
371 | /* If there are other types that shouldn't be formatted, add |
372 | * them here. |
373 | */ |
374 | |
451f74a2 |
375 | if ($body_message->header->type1 == 'html') { |
3d8371be |
376 | if ($show_html_default <> 1) { |
85015544 |
377 | $entity_conv = array(' ' => ' ', |
3d8371be |
378 | '<p>' => "\n", |
379 | '<br>' => "\n", |
380 | '<P>' => "\n", |
381 | '<BR>' => "\n", |
85015544 |
382 | '>' => '>', |
383 | '<' => '<'); |
85015544 |
384 | $body = strtr($body, $entity_conv); |
3d8371be |
385 | $body = strip_tags($body); |
85015544 |
386 | $body = trim($body); |
387 | translateText($body, $wrap_at, |
807e884a |
388 | $body_message->header->getParameter('charset')); |
a3daaaf3 |
389 | } else { |
3d8371be |
390 | $body = magicHTML($body, $id, $message, $mailbox); |
a3daaaf3 |
391 | } |
451f74a2 |
392 | } else { |
3d8371be |
393 | translateText($body, $wrap_at, |
394 | $body_message->header->getParameter('charset')); |
451f74a2 |
395 | } |
08b7f7cc |
396 | $link = 'read_body.php?passed_id=' . $id . '&ent_id='.$ent_num. |
397 | '&mailbox=' . $urlmailbox .'&sort=' . $sort . |
da2415c1 |
398 | '&startMessage=' . $startMessage . '&show_more=0'; |
08b7f7cc |
399 | if (isset($passed_ent_id)) { |
400 | $link .= '&passed_ent_id='.$passed_ent_id; |
401 | } |
7aad7b77 |
402 | if ($view_unsafe_images) { |
23f617b8 |
403 | $text = _("Hide Unsafe Images"); |
7aad7b77 |
404 | } else { |
08b7f7cc |
405 | if (isset($has_unsafe_images) && $has_unsafe_images) { |
406 | $link .= '&view_unsafe_images=1'; |
407 | $text = _("View Unsafe Images"); |
408 | } else { |
409 | $text = ''; |
410 | } |
3d8371be |
411 | } |
23f617b8 |
412 | $body .= '<center><small><a href="'.$link.'">'.$text. |
08b7f7cc |
413 | '</a></small></center><br>' . "\n"; |
3d8371be |
414 | } |
415 | return $body; |
451f74a2 |
416 | } |
b74ba498 |
417 | |
23bcec6f |
418 | |
419 | function formatAttachments($message, $exclude_id, $mailbox, $id) { |
b583c3e8 |
420 | global $where, $what, $startMessage, $color; |
77b88425 |
421 | static $ShownHTML = 0; |
451f74a2 |
422 | |
23bcec6f |
423 | $att_ar = $message->getAttachments($exclude_id); |
451f74a2 |
424 | |
23bcec6f |
425 | if (!count($att_ar)) return ''; |
2e25760a |
426 | |
8f5425ec |
427 | $attachments = ''; |
2e25760a |
428 | |
23bcec6f |
429 | $urlMailbox = urlencode($mailbox); |
430 | |
431 | foreach ($att_ar as $att) { |
fdc9d9b5 |
432 | $ent = $att->entity_id; |
2e25760a |
433 | $header = $att->header; |
f0c4dc12 |
434 | $type0 = strtolower($header->type0); |
435 | $type1 = strtolower($header->type1); |
2e25760a |
436 | $name = ''; |
3d8371be |
437 | $links['download link']['text'] = _("download"); |
6b04287c |
438 | $links['download link']['href'] = SM_PATH . |
439 | "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent"; |
2e25760a |
440 | $ImageURL = ''; |
441 | if ($type0 =='message' && $type1 == 'rfc822') { |
6b04287c |
442 | $default_page = SM_PATH . 'src/read_body.php'; |
2e25760a |
443 | $rfc822_header = $att->rfc822_header; |
098ea084 |
444 | $filename = $rfc822_header->subject; |
6cc08d8b |
445 | if (trim( $filename ) == '') { |
446 | $filename = 'untitled-[' . $ent . ']' ; |
08b7f7cc |
447 | } |
2e25760a |
448 | $from_o = $rfc822_header->from; |
449 | if (is_object($from_o)) { |
04ea844e |
450 | $from_name = decodeHeader($from_o->getAddress(false)); |
2e25760a |
451 | } else { |
452 | $from_name = _("Unknown sender"); |
f0c4dc12 |
453 | } |
3d8371be |
454 | $description = $from_name; |
23bcec6f |
455 | } else { |
6b04287c |
456 | $default_page = SM_PATH . 'src/download.php'; |
3d8371be |
457 | if (is_object($header->disposition)) { |
098ea084 |
458 | $filename = $header->disposition->getProperty('filename'); |
3d8371be |
459 | if (trim($filename) == '') { |
460 | $name = decodeHeader($header->disposition->getProperty('name')); |
461 | if (trim($name) == '') { |
098ea084 |
462 | $name = $header->getParameter('name'); |
1035e159 |
463 | if(trim($name) == '') { |
464 | if (trim( $header->id ) == '') { |
465 | $filename = 'untitled-[' . $ent . ']' ; |
466 | } else { |
467 | $filename = 'cid: ' . $header->id; |
08b7f7cc |
468 | } |
3d8371be |
469 | } else { |
08b7f7cc |
470 | $filename = $name; |
3d8371be |
471 | } |
472 | } else { |
473 | $filename = $name; |
474 | } |
475 | } |
476 | } else { |
08b7f7cc |
477 | $filename = $header->getParameter('name'); |
478 | if (!trim($filename)) { |
479 | if (trim( $header->id ) == '') { |
480 | $filename = 'untitled-[' . $ent . ']' ; |
481 | } else { |
482 | $filename = 'cid: ' . $header->id; |
483 | } |
484 | } |
485 | } |
f810c0b2 |
486 | if ($header->description) { |
098ea084 |
487 | $description = decodeHeader($header->description); |
f810c0b2 |
488 | } else { |
3d8371be |
489 | $description = ''; |
490 | } |
2e25760a |
491 | } |
492 | |
493 | $display_filename = $filename; |
3d8371be |
494 | if (isset($passed_ent_id)) { |
495 | $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id; |
496 | } else { |
497 | $passed_ent_id_link = ''; |
498 | } |
499 | $defaultlink = $default_page . "?startMessage=$startMessage" |
2e25760a |
500 | . "&passed_id=$id&mailbox=$urlMailbox" |
6b04287c |
501 | . '&ent_id='.$ent.$passed_ent_id_link; |
2e25760a |
502 | if ($where && $what) { |
09f4707e |
503 | $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what); |
2e25760a |
504 | } |
3d8371be |
505 | /* This executes the attachment hook with a specific MIME-type. |
506 | * If that doesn't have results, it tries if there's a rule |
507 | * for a more generic type. |
508 | */ |
509 | $hookresults = do_hook("attachment $type0/$type1", $links, |
510 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
511 | $display_filename, $where, $what); |
512 | if(count($hookresults[1]) <= 1) { |
513 | $hookresults = do_hook("attachment $type0/*", $links, |
514 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
515 | $display_filename, $where, $what); |
2e25760a |
516 | } |
451f74a2 |
517 | |
3d8371be |
518 | $links = $hookresults[1]; |
519 | $defaultlink = $hookresults[6]; |
77b88425 |
520 | |
2e25760a |
521 | $attachments .= '<TR><TD>' . |
098ea084 |
522 | '<A HREF="'.$defaultlink.'">'.decodeHeader($display_filename).'</A> </TD>' . |
23bcec6f |
523 | '<TD><SMALL><b>' . show_readable_size($header->size) . |
77b88425 |
524 | '</b> </small></TD>' . |
525 | "<TD><SMALL>[ $type0/$type1 ] </SMALL></TD>" . |
526 | '<TD><SMALL>'; |
f810c0b2 |
527 | $attachments .= '<b>' . $description . '</b>'; |
2e25760a |
528 | $attachments .= '</SMALL></TD><TD><SMALL> '; |
b74ba498 |
529 | |
3d8371be |
530 | $skipspaces = 1; |
531 | foreach ($links as $val) { |
532 | if ($skipspaces) { |
533 | $skipspaces = 0; |
2e25760a |
534 | } else { |
535 | $attachments .= ' | '; |
77b88425 |
536 | } |
3d8371be |
537 | $attachments .= '<a href="' . $val['href'] . '">' . $val['text'] . '</a>'; |
2e25760a |
538 | } |
3d8371be |
539 | unset($links); |
2e25760a |
540 | $attachments .= "</TD></TR>\n"; |
541 | } |
9ad17edb |
542 | $attachmentadd = do_hook_function('attachments_bottom',$attachments); |
543 | if ($attachmentadd != '') |
544 | $attachments = $attachmentadd; |
2e25760a |
545 | return $attachments; |
451f74a2 |
546 | } |
b74ba498 |
547 | |
7c7b74b3 |
548 | function sqimap_base64_decode(&$string) { |
325978ac |
549 | $string = str_replace("\r\n", "\n", $string); |
7c7b74b3 |
550 | $string = base64_decode($string); |
551 | } |
552 | |
3d8371be |
553 | /* This function decodes the body depending on the encoding type. */ |
451f74a2 |
554 | function decodeBody($body, $encoding) { |
3d8371be |
555 | global $show_html_default; |
83be314a |
556 | |
b583c3e8 |
557 | $body = str_replace("\r\n", "\n", $body); |
558 | $encoding = strtolower($encoding); |
3d8371be |
559 | |
5166f86a |
560 | $encoding_handler = do_hook_function('decode_body', $encoding); |
561 | |
562 | |
563 | // plugins get first shot at decoding the body |
564 | // |
565 | if (!empty($encoding_handler) && function_exists($encoding_handler)) { |
566 | $body = $encoding_handler('decode', $body); |
567 | |
568 | } else if ($encoding == 'quoted-printable' || |
3d8371be |
569 | $encoding == 'quoted_printable') { |
b583c3e8 |
570 | $body = quoted_printable_decode($body); |
3d8371be |
571 | |
b583c3e8 |
572 | while (ereg("=\n", $body)) { |
573 | $body = ereg_replace ("=\n", '', $body); |
574 | } |
3d8371be |
575 | |
b583c3e8 |
576 | } else if ($encoding == 'base64') { |
577 | $body = base64_decode($body); |
578 | } |
3d8371be |
579 | |
b583c3e8 |
580 | // All other encodings are returned raw. |
3d8371be |
581 | return $body; |
451f74a2 |
582 | } |
583 | |
584 | /* |
585 | * This functions decode strings that is encoded according to |
586 | * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text). |
79e07c7e |
587 | * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002 |
451f74a2 |
588 | */ |
098ea084 |
589 | function decodeHeader ($string, $utfencode=true,$htmlsave=true) { |
d7448126 |
590 | global $languages, $squirrelmail_language; |
79e07c7e |
591 | if (is_array($string)) { |
592 | $string = implode("\n", $string); |
593 | } |
da2415c1 |
594 | |
10dec454 |
595 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
596 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
6fbd125b |
597 | $string = $languages[$squirrelmail_language]['XTRA_CODE']('decodeheader', $string); |
08b7f7cc |
598 | // Do we need to return at this point? |
599 | // return $string; |
83be314a |
600 | } |
79e07c7e |
601 | $i = 0; |
08b7f7cc |
602 | $iLastMatch = -2; |
db65b6b0 |
603 | $encoded = true; |
0a06275a |
604 | |
098ea084 |
605 | $aString = explode(' ',$string); |
08b7f7cc |
606 | $ret = ''; |
098ea084 |
607 | foreach ($aString as $chunk) { |
358a78a1 |
608 | if ($encoded && $chunk === '') { |
08b7f7cc |
609 | continue; |
358a78a1 |
610 | } elseif ($chunk === '') { |
08b7f7cc |
611 | $ret .= ' '; |
612 | continue; |
613 | } |
098ea084 |
614 | $encoded = false; |
08b7f7cc |
615 | /* if encoded words are not separated by a linear-space-white we still catch them */ |
616 | $j = $i-1; |
7e6ca3e8 |
617 | |
08b7f7cc |
618 | while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) { |
619 | /* if the last chunk isn't an encoded string then put back the space, otherwise don't */ |
620 | if ($iLastMatch !== $j) { |
621 | if ($htmlsave) { |
622 | $ret .= ' '; |
623 | } else { |
624 | $ret .= ' '; |
625 | } |
626 | } |
627 | $iLastMatch = $i; |
628 | $j = $i; |
629 | $ret .= $res[1]; |
098ea084 |
630 | $encoding = ucfirst($res[3]); |
631 | switch ($encoding) |
632 | { |
633 | case 'B': |
634 | $replace = base64_decode($res[4]); |
08b7f7cc |
635 | $ret .= charset_decode($res[2],$replace); |
098ea084 |
636 | break; |
637 | case 'Q': |
098ea084 |
638 | $replace = str_replace('_', ' ', $res[4]); |
da2415c1 |
639 | $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))', |
79e07c7e |
640 | $replace); |
098ea084 |
641 | /* Only encode into entities by default. Some places |
642 | * don't need the encoding, like the compose form. |
643 | */ |
644 | if ($utfencode) { |
645 | $replace = charset_decode($res[2], $replace); |
646 | } else { |
647 | if ($htmlsave) { |
c96c32f4 |
648 | $replace = htmlspecialchars($replace); |
098ea084 |
649 | } |
650 | } |
08b7f7cc |
651 | $ret .= $replace; |
098ea084 |
652 | break; |
653 | default: |
654 | break; |
79e07c7e |
655 | } |
098ea084 |
656 | $chunk = $res[5]; |
657 | $encoded = true; |
08b7f7cc |
658 | } |
7e6ca3e8 |
659 | |
660 | if (!$encoded && $htmlsave) { |
661 | $ret .= htmlspecialchars($chunk); |
662 | } else { |
663 | $ret .= $chunk; |
664 | } |
665 | |
08b7f7cc |
666 | if (!$encoded) { |
667 | if ($htmlsave) { |
668 | $ret .= ' '; |
669 | } else { |
670 | $ret .= ' '; |
da2415c1 |
671 | } |
08b7f7cc |
672 | } |
098ea084 |
673 | ++$i; |
674 | } |
da2415c1 |
675 | |
08b7f7cc |
676 | return $ret; |
451f74a2 |
677 | } |
678 | |
679 | /* |
680 | * Encode a string according to RFC 1522 for use in headers if it |
681 | * contains 8-bit characters or anything that looks like it should |
682 | * be encoded. |
683 | */ |
684 | function encodeHeader ($string) { |
6fbd125b |
685 | global $default_charset, $languages, $squirrelmail_language; |
83be314a |
686 | |
10dec454 |
687 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
688 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
6fbd125b |
689 | return $languages[$squirrelmail_language]['XTRA_CODE']('encodeheader', $string); |
83be314a |
690 | } |
c96c32f4 |
691 | if (strtolower($default_charset) == 'iso-8859-1') { |
692 | $string = str_replace("\240",' ',$string); |
693 | } |
793cc001 |
694 | |
451f74a2 |
695 | // Encode only if the string contains 8-bit characters or =? |
3d8371be |
696 | $j = strlen($string); |
098ea084 |
697 | $max_l = 75 - strlen($default_charset) - 7; |
698 | $aRet = array(); |
451f74a2 |
699 | $ret = ''; |
c96c32f4 |
700 | $iEncStart = $enc_init = false; |
0d53f0f9 |
701 | $cur_l = $iOffset = 0; |
3d8371be |
702 | for($i = 0; $i < $j; ++$i) { |
c96c32f4 |
703 | switch($string{$i}) |
704 | { |
705 | case '=': |
706 | case '<': |
707 | case '>': |
708 | case ',': |
709 | case '?': |
710 | case '_': |
711 | if ($iEncStart === false) { |
712 | $iEncStart = $i; |
713 | } |
714 | $cur_l+=3; |
715 | if ($cur_l > ($max_l-2)) { |
08b7f7cc |
716 | /* if there is an stringpart that doesn't need encoding, add it */ |
c96c32f4 |
717 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
718 | $aRet[] = "=?$default_charset?Q?$ret?="; |
719 | $iOffset = $i; |
720 | $cur_l = 0; |
721 | $ret = ''; |
722 | $iEncStart = false; |
723 | } else { |
724 | $ret .= sprintf("=%02X",ord($string{$i})); |
725 | } |
726 | break; |
727 | case '(': |
728 | case ')': |
729 | if ($iEncStart !== false) { |
08b7f7cc |
730 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
731 | $aRet[] = "=?$default_charset?Q?$ret?="; |
732 | $iOffset = $i; |
733 | $cur_l = 0; |
734 | $ret = ''; |
735 | $iEncStart = false; |
736 | } |
737 | break; |
738 | case ' ': |
739 | if ($iEncStart !== false) { |
098ea084 |
740 | $cur_l++; |
741 | if ($cur_l > $max_l) { |
08b7f7cc |
742 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
743 | $aRet[] = "=?$default_charset?Q?$ret?="; |
744 | $iOffset = $i; |
745 | $cur_l = 0; |
746 | $ret = ''; |
747 | $iEncStart = false; |
08b7f7cc |
748 | } else { |
c96c32f4 |
749 | $ret .= '_'; |
750 | } |
751 | } |
752 | break; |
753 | default: |
754 | $k = ord($string{$i}); |
755 | if ($k > 126) { |
756 | if ($iEncStart === false) { |
325978ac |
757 | // do not start encoding in the middle of a string, also take the rest of the word. |
758 | $sLeadString = substr($string,0,$i); |
759 | $aLeadString = explode(' ',$sLeadString); |
da2415c1 |
760 | $sToBeEncoded = array_pop($aLeadString); |
325978ac |
761 | $iEncStart = $i - strlen($sToBeEncoded); |
762 | $ret .= $sToBeEncoded; |
763 | $cur_l += strlen($sToBeEncoded); |
c96c32f4 |
764 | } |
765 | $cur_l += 3; |
08b7f7cc |
766 | /* first we add the encoded string that reached it's max size */ |
c96c32f4 |
767 | if ($cur_l > ($max_l-2)) { |
08b7f7cc |
768 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
769 | $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */ |
c96c32f4 |
770 | $cur_l = 3; |
771 | $ret = ''; |
772 | $iOffset = $i; |
08b7f7cc |
773 | $iEncStart = $i; |
c96c32f4 |
774 | } |
08b7f7cc |
775 | $enc_init = true; |
c96c32f4 |
776 | $ret .= sprintf("=%02X", $k); |
777 | } else { |
778 | if ($iEncStart !== false) { |
098ea084 |
779 | $cur_l++; |
780 | if ($cur_l > $max_l) { |
08b7f7cc |
781 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
782 | $aRet[] = "=?$default_charset?Q?$ret?="; |
783 | $iEncStart = false; |
784 | $iOffset = $i; |
785 | $cur_l = 0; |
098ea084 |
786 | $ret = ''; |
08b7f7cc |
787 | } else { |
c96c32f4 |
788 | $ret .= $string{$i}; |
789 | } |
3d8371be |
790 | } |
c96c32f4 |
791 | } |
792 | break; |
f7b3ba37 |
793 | } |
451f74a2 |
794 | } |
793cc001 |
795 | |
c96c32f4 |
796 | if ($enc_init) { |
797 | if ($iEncStart !== false) { |
798 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
799 | $aRet[] = "=?$default_charset?Q?$ret?="; |
800 | } else { |
801 | $aRet[] = substr($string,$iOffset); |
802 | } |
803 | $string = implode('',$aRet); |
451f74a2 |
804 | } |
3d8371be |
805 | return $string; |
451f74a2 |
806 | } |
b74ba498 |
807 | |
691a2d25 |
808 | /* This function trys to locate the entity_id of a specific mime element */ |
3d8371be |
809 | function find_ent_id($id, $message) { |
a171b359 |
810 | for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) { |
811 | if ($message->entities[$i]->header->type0 == 'multipart') { |
3d8371be |
812 | $ret = find_ent_id($id, $message->entities[$i]); |
451f74a2 |
813 | } else { |
3d8371be |
814 | if (strcasecmp($message->entities[$i]->header->id, $id) == 0) { |
d8cffbab |
815 | // if (sq_check_save_extension($message->entities[$i])) { |
a171b359 |
816 | return $message->entities[$i]->entity_id; |
da2415c1 |
817 | // } |
3d8371be |
818 | } |
a3daaaf3 |
819 | } |
451f74a2 |
820 | } |
3d8371be |
821 | return $ret; |
451f74a2 |
822 | } |
a3daaaf3 |
823 | |
e5e9381a |
824 | function sq_check_save_extension($message) { |
825 | $filename = $message->getFilename(); |
826 | $ext = substr($filename, strrpos($filename,'.')+1); |
827 | $save_extensions = array('jpg','jpeg','gif','png','bmp'); |
3d8371be |
828 | return in_array($ext, $save_extensions); |
e5e9381a |
829 | } |
830 | |
831 | |
691a2d25 |
832 | /** |
833 | ** HTMLFILTER ROUTINES |
834 | */ |
451f74a2 |
835 | |
2dd879b8 |
836 | /** |
837 | * This function is more or less a wrapper around stripslashes. Apparently |
838 | * Explorer is stupid enough to just remove the backslashes and then |
839 | * execute the content of the attribute as if nothing happened. |
840 | * Who does that? |
841 | * |
842 | * @param attvalue The value of the attribute |
843 | * @return attvalue The value of the attribute stripslashed. |
844 | */ |
845 | function sq_unbackslash($attvalue){ |
846 | /** |
847 | * Remove any backslashes. See if there are any first. |
848 | */ |
849 | if (strstr($attvalue, '\\') !== false){ |
850 | $attvalue = stripslashes($attvalue); |
851 | } |
852 | return $attvalue; |
853 | } |
854 | |
855 | /** |
856 | * Kill any tabs, newlines, or carriage returns. Our friends the |
857 | * makers of the browser with 95% market value decided that it'd |
858 | * be funny to make "java[tab]script" be just as good as "javascript". |
da2415c1 |
859 | * |
2dd879b8 |
860 | * @param attvalue The attribute value before extraneous spaces removed. |
861 | * @return attvalue The attribute value after extraneous spaces removed. |
862 | */ |
863 | function sq_unspace($attvalue){ |
864 | if (strcspn($attvalue, "\t\r\n") != strlen($attvalue)){ |
865 | $attvalue = str_replace(Array("\t", "\r", "\n"), Array('', '', ''), |
866 | $attvalue); |
867 | } |
868 | return $attvalue; |
869 | } |
870 | |
691a2d25 |
871 | /** |
872 | * This function returns the final tag out of the tag name, an array |
da2415c1 |
873 | * of attributes, and the type of the tag. This function is called by |
691a2d25 |
874 | * sq_sanitize internally. |
875 | * |
876 | * @param $tagname the name of the tag. |
877 | * @param $attary the array of attributes and their values |
878 | * @param $tagtype The type of the tag (see in comments). |
879 | * @return a string with the final tag representation. |
880 | */ |
881 | function sq_tagprint($tagname, $attary, $tagtype){ |
b583c3e8 |
882 | $me = 'sq_tagprint'; |
3d8371be |
883 | |
691a2d25 |
884 | if ($tagtype == 2){ |
885 | $fulltag = '</' . $tagname . '>'; |
886 | } else { |
887 | $fulltag = '<' . $tagname; |
888 | if (is_array($attary) && sizeof($attary)){ |
889 | $atts = Array(); |
890 | while (list($attname, $attvalue) = each($attary)){ |
891 | array_push($atts, "$attname=$attvalue"); |
892 | } |
893 | $fulltag .= ' ' . join(" ", $atts); |
894 | } |
895 | if ($tagtype == 3){ |
b583c3e8 |
896 | $fulltag .= ' /'; |
691a2d25 |
897 | } |
b583c3e8 |
898 | $fulltag .= '>'; |
451f74a2 |
899 | } |
691a2d25 |
900 | return $fulltag; |
451f74a2 |
901 | } |
a3daaaf3 |
902 | |
691a2d25 |
903 | /** |
904 | * A small helper function to use with array_walk. Modifies a by-ref |
905 | * value and makes it lowercase. |
906 | * |
907 | * @param $val a value passed by-ref. |
908 | * @return void since it modifies a by-ref value. |
909 | */ |
910 | function sq_casenormalize(&$val){ |
911 | $val = strtolower($val); |
912 | } |
451f74a2 |
913 | |
691a2d25 |
914 | /** |
915 | * This function skips any whitespace from the current position within |
916 | * a string and to the next non-whitespace value. |
da2415c1 |
917 | * |
691a2d25 |
918 | * @param $body the string |
919 | * @param $offset the offset within the string where we should start |
920 | * looking for the next non-whitespace character. |
921 | * @return the location within the $body where the next |
922 | * non-whitespace char is located. |
923 | */ |
924 | function sq_skipspace($body, $offset){ |
b583c3e8 |
925 | $me = 'sq_skipspace'; |
3d8371be |
926 | preg_match('/^(\s*)/s', substr($body, $offset), $matches); |
691a2d25 |
927 | if (sizeof($matches{1})){ |
928 | $count = strlen($matches{1}); |
929 | $offset += $count; |
451f74a2 |
930 | } |
691a2d25 |
931 | return $offset; |
451f74a2 |
932 | } |
a3daaaf3 |
933 | |
691a2d25 |
934 | /** |
935 | * This function looks for the next character within a string. It's |
936 | * really just a glorified "strpos", except it catches if failures |
937 | * nicely. |
938 | * |
939 | * @param $body The string to look for needle in. |
940 | * @param $offset Start looking from this position. |
941 | * @param $needle The character/string to look for. |
942 | * @return location of the next occurance of the needle, or |
943 | * strlen($body) if needle wasn't found. |
944 | */ |
945 | function sq_findnxstr($body, $offset, $needle){ |
3d8371be |
946 | $me = 'sq_findnxstr'; |
691a2d25 |
947 | $pos = strpos($body, $needle, $offset); |
948 | if ($pos === FALSE){ |
949 | $pos = strlen($body); |
451f74a2 |
950 | } |
691a2d25 |
951 | return $pos; |
451f74a2 |
952 | } |
a3daaaf3 |
953 | |
691a2d25 |
954 | /** |
955 | * This function takes a PCRE-style regexp and tries to match it |
956 | * within the string. |
957 | * |
958 | * @param $body The string to look for needle in. |
959 | * @param $offset Start looking from here. |
960 | * @param $reg A PCRE-style regex to match. |
961 | * @return Returns a false if no matches found, or an array |
962 | * with the following members: |
963 | * - integer with the location of the match within $body |
964 | * - string with whatever content between offset and the match |
965 | * - string with whatever it is we matched |
966 | */ |
967 | function sq_findnxreg($body, $offset, $reg){ |
b583c3e8 |
968 | $me = 'sq_findnxreg'; |
691a2d25 |
969 | $matches = Array(); |
970 | $retarr = Array(); |
7d06541f |
971 | preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches); |
972 | if (!isset($matches{0}) || !$matches{0}){ |
691a2d25 |
973 | $retarr = false; |
974 | } else { |
975 | $retarr{0} = $offset + strlen($matches{1}); |
976 | $retarr{1} = $matches{1}; |
977 | $retarr{2} = $matches{2}; |
978 | } |
979 | return $retarr; |
980 | } |
a3daaaf3 |
981 | |
691a2d25 |
982 | /** |
983 | * This function looks for the next tag. |
984 | * |
985 | * @param $body String where to look for the next tag. |
986 | * @param $offset Start looking from here. |
987 | * @return false if no more tags exist in the body, or |
988 | * an array with the following members: |
989 | * - string with the name of the tag |
990 | * - array with attributes and their values |
991 | * - integer with tag type (1, 2, or 3) |
992 | * - integer where the tag starts (starting "<") |
993 | * - integer where the tag ends (ending ">") |
994 | * first three members will be false, if the tag is invalid. |
995 | */ |
996 | function sq_getnxtag($body, $offset){ |
b583c3e8 |
997 | $me = 'sq_getnxtag'; |
691a2d25 |
998 | if ($offset > strlen($body)){ |
999 | return false; |
1000 | } |
1001 | $lt = sq_findnxstr($body, $offset, "<"); |
1002 | if ($lt == strlen($body)){ |
1003 | return false; |
1004 | } |
1005 | /** |
1006 | * We are here: |
1007 | * blah blah <tag attribute="value"> |
1008 | * \---------^ |
1009 | */ |
1010 | $pos = sq_skipspace($body, $lt+1); |
1011 | if ($pos >= strlen($body)){ |
1012 | return Array(false, false, false, $lt, strlen($body)); |
1013 | } |
1014 | /** |
1015 | * There are 3 kinds of tags: |
1016 | * 1. Opening tag, e.g.: |
1017 | * <a href="blah"> |
1018 | * 2. Closing tag, e.g.: |
1019 | * </a> |
1020 | * 3. XHTML-style content-less tag, e.g.: |
1021 | * <img src="blah"/> |
1022 | */ |
1023 | $tagtype = false; |
1024 | switch (substr($body, $pos, 1)){ |
3d8371be |
1025 | case '/': |
1026 | $tagtype = 2; |
1027 | $pos++; |
1028 | break; |
1029 | case '!': |
1030 | /** |
1031 | * A comment or an SGML declaration. |
1032 | */ |
1033 | if (substr($body, $pos+1, 2) == "--"){ |
1034 | $gt = strpos($body, "-->", $pos); |
1035 | if ($gt === false){ |
1036 | $gt = strlen($body); |
1037 | } else { |
1038 | $gt += 2; |
1039 | } |
1040 | return Array(false, false, false, $lt, $gt); |
bb8d0799 |
1041 | } else { |
3d8371be |
1042 | $gt = sq_findnxstr($body, $pos, ">"); |
1043 | return Array(false, false, false, $lt, $gt); |
1044 | } |
1045 | break; |
1046 | default: |
1047 | /** |
1048 | * Assume tagtype 1 for now. If it's type 3, we'll switch values |
1049 | * later. |
1050 | */ |
1051 | $tagtype = 1; |
1052 | break; |
691a2d25 |
1053 | } |
a3daaaf3 |
1054 | |
691a2d25 |
1055 | $tag_start = $pos; |
1056 | $tagname = ''; |
1057 | /** |
1058 | * Look for next [\W-_], which will indicate the end of the tag name. |
1059 | */ |
1060 | $regary = sq_findnxreg($body, $pos, "[^\w\-_]"); |
1061 | if ($regary == false){ |
1062 | return Array(false, false, false, $lt, strlen($body)); |
1063 | } |
1064 | list($pos, $tagname, $match) = $regary; |
1065 | $tagname = strtolower($tagname); |
1066 | |
1067 | /** |
1068 | * $match can be either of these: |
1069 | * '>' indicating the end of the tag entirely. |
1070 | * '\s' indicating the end of the tag name. |
1071 | * '/' indicating that this is type-3 xhtml tag. |
da2415c1 |
1072 | * |
691a2d25 |
1073 | * Whatever else we find there indicates an invalid tag. |
1074 | */ |
1075 | switch ($match){ |
3d8371be |
1076 | case '/': |
691a2d25 |
1077 | /** |
3d8371be |
1078 | * This is an xhtml-style tag with a closing / at the |
1079 | * end, like so: <img src="blah"/>. Check if it's followed |
1080 | * by the closing bracket. If not, then this tag is invalid |
691a2d25 |
1081 | */ |
3d8371be |
1082 | if (substr($body, $pos, 2) == "/>"){ |
1083 | $pos++; |
1084 | $tagtype = 3; |
1085 | } else { |
1086 | $gt = sq_findnxstr($body, $pos, ">"); |
1087 | $retary = Array(false, false, false, $lt, $gt); |
1088 | return $retary; |
1089 | } |
1090 | case '>': |
1091 | return Array($tagname, false, $tagtype, $lt, $pos); |
1092 | break; |
1093 | default: |
1094 | /** |
1095 | * Check if it's whitespace |
1096 | */ |
1097 | if (!preg_match('/\s/', $match)){ |
1098 | /** |
1099 | * This is an invalid tag! Look for the next closing ">". |
1100 | */ |
7d06541f |
1101 | $gt = sq_findnxstr($body, $lt, ">"); |
3d8371be |
1102 | return Array(false, false, false, $lt, $gt); |
1103 | } |
1104 | break; |
691a2d25 |
1105 | } |
3d8371be |
1106 | |
691a2d25 |
1107 | /** |
1108 | * At this point we're here: |
1109 | * <tagname attribute='blah'> |
1110 | * \-------^ |
1111 | * |
1112 | * At this point we loop in order to find all attributes. |
1113 | */ |
1114 | $attname = ''; |
1115 | $atttype = false; |
1116 | $attary = Array(); |
1117 | |
1118 | while ($pos <= strlen($body)){ |
1119 | $pos = sq_skipspace($body, $pos); |
1120 | if ($pos == strlen($body)){ |
1121 | /** |
1122 | * Non-closed tag. |
1123 | */ |
1124 | return Array(false, false, false, $lt, $pos); |
1125 | } |
1126 | /** |
1127 | * See if we arrived at a ">" or "/>", which means that we reached |
1128 | * the end of the tag. |
1129 | */ |
1130 | $matches = Array(); |
164800ad |
1131 | if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) { |
c828931c |
1132 | /** |
1133 | * Yep. So we did. |
1134 | */ |
1135 | $pos += strlen($matches{1}); |
1136 | if ($matches{2} == "/>"){ |
1137 | $tagtype = 3; |
1138 | $pos++; |
1139 | } |
1140 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
1141 | } |
a3daaaf3 |
1142 | |
cca46357 |
1143 | /** |
691a2d25 |
1144 | * There are several types of attributes, with optional |
1145 | * [:space:] between members. |
1146 | * Type 1: |
1147 | * attrname[:space:]=[:space:]'CDATA' |
1148 | * Type 2: |
1149 | * attrname[:space:]=[:space:]"CDATA" |
1150 | * Type 3: |
1151 | * attr[:space:]=[:space:]CDATA |
1152 | * Type 4: |
1153 | * attrname |
cca46357 |
1154 | * |
691a2d25 |
1155 | * We leave types 1 and 2 the same, type 3 we check for |
1156 | * '"' and convert to """ if needed, then wrap in |
1157 | * double quotes. Type 4 we convert into: |
1158 | * attrname="yes". |
cca46357 |
1159 | */ |
3f7c623f |
1160 | $regary = sq_findnxreg($body, $pos, "[^:\w\-_]"); |
691a2d25 |
1161 | if ($regary == false){ |
1162 | /** |
1163 | * Looks like body ended before the end of tag. |
1164 | */ |
1165 | return Array(false, false, false, $lt, strlen($body)); |
cca46357 |
1166 | } |
691a2d25 |
1167 | list($pos, $attname, $match) = $regary; |
1168 | $attname = strtolower($attname); |
1169 | /** |
1170 | * We arrived at the end of attribute name. Several things possible |
1171 | * here: |
1172 | * '>' means the end of the tag and this is attribute type 4 |
1173 | * '/' if followed by '>' means the same thing as above |
1174 | * '\s' means a lot of things -- look what it's followed by. |
1175 | * anything else means the attribute is invalid. |
1176 | */ |
1177 | switch($match){ |
3d8371be |
1178 | case '/': |
691a2d25 |
1179 | /** |
3d8371be |
1180 | * This is an xhtml-style tag with a closing / at the |
1181 | * end, like so: <img src="blah"/>. Check if it's followed |
1182 | * by the closing bracket. If not, then this tag is invalid |
691a2d25 |
1183 | */ |
3d8371be |
1184 | if (substr($body, $pos, 2) == "/>"){ |
691a2d25 |
1185 | $pos++; |
3d8371be |
1186 | $tagtype = 3; |
691a2d25 |
1187 | } else { |
3d8371be |
1188 | $gt = sq_findnxstr($body, $pos, ">"); |
1189 | $retary = Array(false, false, false, $lt, $gt); |
1190 | return $retary; |
1191 | } |
1192 | case '>': |
1193 | $attary{$attname} = '"yes"'; |
1194 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
1195 | break; |
1196 | default: |
1197 | /** |
1198 | * Skip whitespace and see what we arrive at. |
1199 | */ |
1200 | $pos = sq_skipspace($body, $pos); |
1201 | $char = substr($body, $pos, 1); |
1202 | /** |
1203 | * Two things are valid here: |
1204 | * '=' means this is attribute type 1 2 or 3. |
1205 | * \w means this was attribute type 4. |
1206 | * anything else we ignore and re-loop. End of tag and |
1207 | * invalid stuff will be caught by our checks at the beginning |
1208 | * of the loop. |
1209 | */ |
1210 | if ($char == "="){ |
1211 | $pos++; |
1212 | $pos = sq_skipspace($body, $pos); |
691a2d25 |
1213 | /** |
3d8371be |
1214 | * Here are 3 possibilities: |
1215 | * "'" attribute type 1 |
1216 | * '"' attribute type 2 |
1217 | * everything else is the content of tag type 3 |
691a2d25 |
1218 | */ |
3d8371be |
1219 | $quot = substr($body, $pos, 1); |
1220 | if ($quot == "'"){ |
1221 | $regary = sq_findnxreg($body, $pos+1, "\'"); |
1222 | if ($regary == false){ |
1223 | return Array(false, false, false, $lt, strlen($body)); |
1224 | } |
1225 | list($pos, $attval, $match) = $regary; |
1226 | $pos++; |
1227 | $attary{$attname} = "'" . $attval . "'"; |
1228 | } else if ($quot == '"'){ |
1229 | $regary = sq_findnxreg($body, $pos+1, '\"'); |
1230 | if ($regary == false){ |
1231 | return Array(false, false, false, $lt, strlen($body)); |
1232 | } |
1233 | list($pos, $attval, $match) = $regary; |
1234 | $pos++; |
1235 | $attary{$attname} = '"' . $attval . '"'; |
1236 | } else { |
1237 | /** |
1238 | * These are hateful. Look for \s, or >. |
1239 | */ |
1240 | $regary = sq_findnxreg($body, $pos, "[\s>]"); |
1241 | if ($regary == false){ |
1242 | return Array(false, false, false, $lt, strlen($body)); |
1243 | } |
1244 | list($pos, $attval, $match) = $regary; |
1245 | /** |
1246 | * If it's ">" it will be caught at the top. |
1247 | */ |
1248 | $attval = preg_replace("/\"/s", """, $attval); |
1249 | $attary{$attname} = '"' . $attval . '"'; |
7e235a1a |
1250 | } |
3d8371be |
1251 | } else if (preg_match("|[\w/>]|", $char)) { |
691a2d25 |
1252 | /** |
3d8371be |
1253 | * That was attribute type 4. |
691a2d25 |
1254 | */ |
3d8371be |
1255 | $attary{$attname} = '"yes"'; |
1256 | } else { |
1257 | /** |
1258 | * An illegal character. Find next '>' and return. |
1259 | */ |
1260 | $gt = sq_findnxstr($body, $pos, ">"); |
1261 | return Array(false, false, false, $lt, $gt); |
451f74a2 |
1262 | } |
3d8371be |
1263 | break; |
691a2d25 |
1264 | } |
1265 | } |
1266 | /** |
1267 | * The fact that we got here indicates that the tag end was never |
1268 | * found. Return invalid tag indication so it gets stripped. |
1269 | */ |
1270 | return Array(false, false, false, $lt, strlen($body)); |
1271 | } |
1272 | |
1273 | /** |
1274 | * This function checks attribute values for entity-encoded values |
1275 | * and returns them translated into 8-bit strings so we can run |
1276 | * checks on them. |
1277 | * |
1278 | * @param $attvalue A string to run entity check against. |
1279 | * @return Translated value. |
1280 | */ |
1281 | function sq_deent($attvalue){ |
b583c3e8 |
1282 | $me = 'sq_deent'; |
691a2d25 |
1283 | /** |
1284 | * See if we have to run the checks first. All entities must start |
1285 | * with "&". |
1286 | */ |
1287 | if (strpos($attvalue, "&") === false){ |
1288 | return $attvalue; |
1289 | } |
1290 | /** |
1291 | * Check named entities first. |
1292 | */ |
1293 | $trans = get_html_translation_table(HTML_ENTITIES); |
1294 | /** |
1295 | * Leave " in, as it can mess us up. |
1296 | */ |
1297 | $trans = array_flip($trans); |
1298 | unset($trans{"""}); |
1299 | while (list($ent, $val) = each($trans)){ |
1300 | $attvalue = preg_replace("/$ent*(\W)/si", "$val\\1", $attvalue); |
1301 | } |
1302 | /** |
1303 | * Now translate numbered entities from 1 to 255 if needed. |
1304 | */ |
1305 | if (strpos($attvalue, "#") !== false){ |
1306 | $omit = Array(34, 39); |
1307 | for ($asc=1; $asc<256; $asc++){ |
1308 | if (!in_array($asc, $omit)){ |
1309 | $chr = chr($asc); |
da2415c1 |
1310 | $attvalue = preg_replace("/\�*$asc;*(\D)/si", "$chr\\1", |
691a2d25 |
1311 | $attvalue); |
1312 | $attvalue = preg_replace("/\�*".dechex($asc).";*(\W)/si", |
1313 | "$chr\\1", $attvalue); |
a3daaaf3 |
1314 | } |
691a2d25 |
1315 | } |
1316 | } |
1317 | return $attvalue; |
1318 | } |
1319 | |
1320 | /** |
1321 | * This function runs various checks against the attributes. |
1322 | * |
1323 | * @param $tagname String with the name of the tag. |
1324 | * @param $attary Array with all tag attributes. |
1325 | * @param $rm_attnames See description for sq_sanitize |
1326 | * @param $bad_attvals See description for sq_sanitize |
1327 | * @param $add_attr_to_tag See description for sq_sanitize |
1328 | * @param $message message object |
1329 | * @param $id message id |
1330 | * @return Array with modified attributes. |
1331 | */ |
da2415c1 |
1332 | function sq_fixatts($tagname, |
1333 | $attary, |
691a2d25 |
1334 | $rm_attnames, |
1335 | $bad_attvals, |
1336 | $add_attr_to_tag, |
1337 | $message, |
b3af12ef |
1338 | $id, |
3d8371be |
1339 | $mailbox |
691a2d25 |
1340 | ){ |
b583c3e8 |
1341 | $me = 'sq_fixatts'; |
691a2d25 |
1342 | while (list($attname, $attvalue) = each($attary)){ |
1343 | /** |
1344 | * See if this attribute should be removed. |
1345 | */ |
1346 | foreach ($rm_attnames as $matchtag=>$matchattrs){ |
1347 | if (preg_match($matchtag, $tagname)){ |
1348 | foreach ($matchattrs as $matchattr){ |
1349 | if (preg_match($matchattr, $attname)){ |
1350 | unset($attary{$attname}); |
1351 | continue; |
1352 | } |
451f74a2 |
1353 | } |
451f74a2 |
1354 | } |
691a2d25 |
1355 | } |
1356 | /** |
2dd879b8 |
1357 | * Remove any backslashes, entities, and extraneous whitespace. |
691a2d25 |
1358 | */ |
2dd879b8 |
1359 | $attvalue = sq_unbackslash($attvalue); |
691a2d25 |
1360 | $attvalue = sq_deent($attvalue); |
2dd879b8 |
1361 | $attvalue = sq_unspace($attvalue); |
691a2d25 |
1362 | |
1363 | /** |
1364 | * Now let's run checks on the attvalues. |
1365 | * I don't expect anyone to comprehend this. If you do, |
1366 | * get in touch with me so I can drive to where you live and |
1367 | * shake your hand personally. :) |
1368 | */ |
1369 | foreach ($bad_attvals as $matchtag=>$matchattrs){ |
1370 | if (preg_match($matchtag, $tagname)){ |
1371 | foreach ($matchattrs as $matchattr=>$valary){ |
1372 | if (preg_match($matchattr, $attname)){ |
1373 | /** |
1374 | * There are two arrays in valary. |
1375 | * First is matches. |
1376 | * Second one is replacements |
1377 | */ |
1378 | list($valmatch, $valrepl) = $valary; |
da2415c1 |
1379 | $newvalue = |
691a2d25 |
1380 | preg_replace($valmatch, $valrepl, $attvalue); |
1381 | if ($newvalue != $attvalue){ |
1382 | $attary{$attname} = $newvalue; |
1383 | } |
1384 | } |
1385 | } |
451f74a2 |
1386 | } |
a3daaaf3 |
1387 | } |
691a2d25 |
1388 | /** |
1389 | * Turn cid: urls into http-friendly ones. |
1390 | */ |
1391 | if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){ |
b3af12ef |
1392 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
691a2d25 |
1393 | } |
a3daaaf3 |
1394 | } |
691a2d25 |
1395 | /** |
1396 | * See if we need to append any attributes to this tag. |
1397 | */ |
1398 | foreach ($add_attr_to_tag as $matchtag=>$addattary){ |
1399 | if (preg_match($matchtag, $tagname)){ |
1400 | $attary = array_merge($attary, $addattary); |
1401 | } |
1402 | } |
1403 | return $attary; |
451f74a2 |
1404 | } |
a3daaaf3 |
1405 | |
691a2d25 |
1406 | /** |
1407 | * This function edits the style definition to make them friendly and |
1408 | * usable in squirrelmail. |
da2415c1 |
1409 | * |
691a2d25 |
1410 | * @param $message the message object |
1411 | * @param $id the message id |
1412 | * @param $content a string with whatever is between <style> and </style> |
e60a299a |
1413 | * @param $mailbox the message mailbox |
691a2d25 |
1414 | * @return a string with edited content. |
1415 | */ |
e60a299a |
1416 | function sq_fixstyle($body, $pos, $message, $id, $mailbox){ |
691a2d25 |
1417 | global $view_unsafe_images; |
b583c3e8 |
1418 | $me = 'sq_fixstyle'; |
7d06541f |
1419 | $ret = sq_findnxreg($body, $pos, '</\s*style\s*>'); |
1420 | if ($ret == FALSE){ |
1421 | return array(FALSE, strlen($body)); |
1422 | } |
1423 | $newpos = $ret[0] + strlen($ret[2]); |
1424 | $content = $ret[1]; |
691a2d25 |
1425 | /** |
1426 | * First look for general BODY style declaration, which would be |
1427 | * like so: |
1428 | * body {background: blah-blah} |
1429 | * and change it to .bodyclass so we can just assign it to a <div> |
1430 | */ |
1431 | $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content); |
3d8371be |
1432 | $secremoveimg = '../images/' . _("sec_remove_eng.png"); |
691a2d25 |
1433 | /** |
1434 | * Fix url('blah') declarations. |
1435 | */ |
7d06541f |
1436 | $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si", |
691a2d25 |
1437 | "url(\\1$secremoveimg\\2)", $content); |
1438 | /** |
1439 | * Fix url('https*://.*) declarations but only if $view_unsafe_images |
1440 | * is false. |
1441 | */ |
1442 | if (!$view_unsafe_images){ |
7d06541f |
1443 | $content = preg_replace("|url\s*\(\s*([\'\"])\s*https*:.*?([\'\"])\s*\)|si", |
691a2d25 |
1444 | "url(\\1$secremoveimg\\2)", $content); |
1445 | } |
da2415c1 |
1446 | |
691a2d25 |
1447 | /** |
1448 | * Fix urls that refer to cid: |
1449 | */ |
da2415c1 |
1450 | while (preg_match("|url\s*\(\s*([\'\"]\s*cid:.*?[\'\"])\s*\)|si", |
7d06541f |
1451 | $content, $matches)){ |
691a2d25 |
1452 | $cidurl = $matches{1}; |
e60a299a |
1453 | $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox); |
7d06541f |
1454 | $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si", |
691a2d25 |
1455 | "url($httpurl)", $content); |
1456 | } |
a3daaaf3 |
1457 | |
691a2d25 |
1458 | /** |
bb8d0799 |
1459 | * Fix stupid css declarations which lead to vulnerabilities |
691a2d25 |
1460 | * in IE. |
1461 | */ |
f83c60a2 |
1462 | $match = Array('/expression/i', |
1463 | '/behaviou*r/i', |
2dd879b8 |
1464 | '/binding/i', |
1465 | '/include-source/i'); |
1466 | $replace = Array('idiocy', 'idiocy', 'idiocy', 'idiocy'); |
bb8d0799 |
1467 | $content = preg_replace($match, $replace, $content); |
7d06541f |
1468 | return array($content, $newpos); |
691a2d25 |
1469 | } |
a3daaaf3 |
1470 | |
691a2d25 |
1471 | /** |
1472 | * This function converts cid: url's into the ones that can be viewed in |
1473 | * the browser. |
1474 | * |
1475 | * @param $message the message object |
1476 | * @param $id the message id |
1477 | * @param $cidurl the cid: url. |
e60a299a |
1478 | * @param $mailbox the message mailbox |
691a2d25 |
1479 | * @return a string with a http-friendly url |
1480 | */ |
b3af12ef |
1481 | function sq_cid2http($message, $id, $cidurl, $mailbox){ |
691a2d25 |
1482 | /** |
1483 | * Get rid of quotes. |
1484 | */ |
1485 | $quotchar = substr($cidurl, 0, 1); |
2dd879b8 |
1486 | if ($quotchar == '"' || $quotchar == "'"){ |
1487 | $cidurl = str_replace($quotchar, "", $cidurl); |
1488 | } else { |
1489 | $quotchar = ''; |
1490 | } |
691a2d25 |
1491 | $cidurl = substr(trim($cidurl), 4); |
e5e9381a |
1492 | $linkurl = find_ent_id($cidurl, $message); |
1493 | /* in case of non-save cid links $httpurl should be replaced by a sort of |
1494 | unsave link image */ |
1495 | $httpurl = ''; |
1496 | if ($linkurl) { |
6b04287c |
1497 | $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&' . |
e5e9381a |
1498 | "passed_id=$id&mailbox=" . urlencode($mailbox) . |
3d8371be |
1499 | '&ent_id=' . $linkurl . $quotchar; |
e5e9381a |
1500 | } |
691a2d25 |
1501 | return $httpurl; |
1502 | } |
1503 | |
1504 | /** |
1505 | * This function changes the <body> tag into a <div> tag since we |
1506 | * can't really have a body-within-body. |
1507 | * |
2dd879b8 |
1508 | * @param $attary an array of attributes and values of <body> |
1509 | * @param $mailbox mailbox we're currently reading (for cid2http) |
1510 | * @param $message current message (for cid2http) |
1511 | * @param $id current message id (for cid2http) |
1512 | * @return a modified array of attributes to be set for <div> |
691a2d25 |
1513 | */ |
2dd879b8 |
1514 | function sq_body2div($attary, $mailbox, $message, $id){ |
b583c3e8 |
1515 | $me = 'sq_body2div'; |
3d8371be |
1516 | $divattary = Array('class' => "'bodyclass'"); |
b583c3e8 |
1517 | $bgcolor = '#ffffff'; |
1518 | $text = '#000000'; |
80b4debd |
1519 | $has_bgc_stl = $has_txt_stl = false; |
b583c3e8 |
1520 | $styledef = ''; |
691a2d25 |
1521 | if (is_array($attary) && sizeof($attary) > 0){ |
1522 | foreach ($attary as $attname=>$attvalue){ |
1523 | $quotchar = substr($attvalue, 0, 1); |
1524 | $attvalue = str_replace($quotchar, "", $attvalue); |
1525 | switch ($attname){ |
3d8371be |
1526 | case 'background': |
da2415c1 |
1527 | $attvalue = sq_cid2http($message, $id, |
2dd879b8 |
1528 | $attvalue, $mailbox); |
3d8371be |
1529 | $styledef .= "background-image: url('$attvalue'); "; |
1530 | break; |
1531 | case 'bgcolor': |
80b4debd |
1532 | $has_bgc_stl = true; |
3d8371be |
1533 | $styledef .= "background-color: $attvalue; "; |
1534 | break; |
1535 | case 'text': |
80b4debd |
1536 | $has_txt_stl = true; |
3d8371be |
1537 | $styledef .= "color: $attvalue; "; |
1538 | break; |
691a2d25 |
1539 | } |
a3daaaf3 |
1540 | } |
80b4debd |
1541 | // Outlook defines a white bgcolor and no text color. This can lead to |
1542 | // white text on a white bg with certain themes. |
1543 | if ($has_bgc_stl && !$has_txt_stl) { |
1544 | $styledef .= "color: $text; "; |
1545 | } |
691a2d25 |
1546 | if (strlen($styledef) > 0){ |
1547 | $divattary{"style"} = "\"$styledef\""; |
1548 | } |
1549 | } |
1550 | return $divattary; |
1551 | } |
a3daaaf3 |
1552 | |
691a2d25 |
1553 | /** |
1554 | * This is the main function and the one you should actually be calling. |
1555 | * There are several variables you should be aware of an which need |
1556 | * special description. |
1557 | * |
1558 | * Since the description is quite lengthy, see it here: |
1559 | * http://www.mricon.com/html/phpfilter.html |
1560 | * |
1561 | * @param $body the string with HTML you wish to filter |
1562 | * @param $tag_list see description above |
1563 | * @param $rm_tags_with_content see description above |
1564 | * @param $self_closing_tags see description above |
1565 | * @param $force_tag_closing see description above |
1566 | * @param $rm_attnames see description above |
1567 | * @param $bad_attvals see description above |
1568 | * @param $add_attr_to_tag see description above |
1569 | * @param $message message object |
1570 | * @param $id message id |
1571 | * @return sanitized html safe to show on your pages. |
1572 | */ |
da2415c1 |
1573 | function sq_sanitize($body, |
1574 | $tag_list, |
691a2d25 |
1575 | $rm_tags_with_content, |
1576 | $self_closing_tags, |
1577 | $force_tag_closing, |
1578 | $rm_attnames, |
1579 | $bad_attvals, |
1580 | $add_attr_to_tag, |
1581 | $message, |
b3af12ef |
1582 | $id, |
3d8371be |
1583 | $mailbox |
691a2d25 |
1584 | ){ |
b583c3e8 |
1585 | $me = 'sq_sanitize'; |
7d06541f |
1586 | $rm_tags = array_shift($tag_list); |
691a2d25 |
1587 | /** |
1588 | * Normalize rm_tags and rm_tags_with_content. |
1589 | */ |
7d06541f |
1590 | @array_walk($tag_list, 'sq_casenormalize'); |
691a2d25 |
1591 | @array_walk($rm_tags_with_content, 'sq_casenormalize'); |
1592 | @array_walk($self_closing_tags, 'sq_casenormalize'); |
1593 | /** |
1594 | * See if tag_list is of tags to remove or tags to allow. |
1595 | * false means remove these tags |
1596 | * true means allow these tags |
1597 | */ |
691a2d25 |
1598 | $curpos = 0; |
1599 | $open_tags = Array(); |
2dd879b8 |
1600 | $trusted = "\n<!-- begin sanitized html -->\n"; |
691a2d25 |
1601 | $skip_content = false; |
bb8d0799 |
1602 | /** |
1603 | * Take care of netscape's stupid javascript entities like |
1604 | * &{alert('boo')}; |
1605 | */ |
1606 | $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body); |
691a2d25 |
1607 | |
7d06541f |
1608 | while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){ |
691a2d25 |
1609 | list($tagname, $attary, $tagtype, $lt, $gt) = $curtag; |
1610 | $free_content = substr($body, $curpos, $lt-$curpos); |
1611 | /** |
1612 | * Take care of <style> |
1613 | */ |
7d06541f |
1614 | if ($tagname == "style" && $tagtype == 1){ |
da2415c1 |
1615 | list($free_content, $curpos) = |
e60a299a |
1616 | sq_fixstyle($body, $gt+1, $message, $id, $mailbox); |
7d06541f |
1617 | if ($free_content != FALSE){ |
1618 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
1619 | $trusted .= $free_content; |
1620 | $trusted .= sq_tagprint($tagname, false, 2); |
1621 | } |
1622 | continue; |
691a2d25 |
1623 | } |
1624 | if ($skip_content == false){ |
1625 | $trusted .= $free_content; |
691a2d25 |
1626 | } |
1627 | if ($tagname != FALSE){ |
1628 | if ($tagtype == 2){ |
1629 | if ($skip_content == $tagname){ |
1630 | /** |
1631 | * Got to the end of tag we needed to remove. |
1632 | */ |
1633 | $tagname = false; |
1634 | $skip_content = false; |
1635 | } else { |
1636 | if ($skip_content == false){ |
c828931c |
1637 | if ($tagname == "body"){ |
1638 | $tagname = "div"; |
2dd879b8 |
1639 | } |
da2415c1 |
1640 | if (isset($open_tags{$tagname}) && |
2dd879b8 |
1641 | $open_tags{$tagname} > 0){ |
1642 | $open_tags{$tagname}--; |
691a2d25 |
1643 | } else { |
2dd879b8 |
1644 | $tagname = false; |
691a2d25 |
1645 | } |
691a2d25 |
1646 | } |
1647 | } |
1648 | } else { |
1649 | /** |
1650 | * $rm_tags_with_content |
1651 | */ |
1652 | if ($skip_content == false){ |
1653 | /** |
1654 | * See if this is a self-closing type and change |
1655 | * tagtype appropriately. |
1656 | */ |
1657 | if ($tagtype == 1 |
1658 | && in_array($tagname, $self_closing_tags)){ |
2dd879b8 |
1659 | $tagtype = 3; |
691a2d25 |
1660 | } |
1661 | /** |
1662 | * See if we should skip this tag and any content |
1663 | * inside it. |
1664 | */ |
1665 | if ($tagtype == 1 && |
1666 | in_array($tagname, $rm_tags_with_content)){ |
1667 | $skip_content = $tagname; |
1668 | } else { |
da2415c1 |
1669 | if (($rm_tags == false |
691a2d25 |
1670 | && in_array($tagname, $tag_list)) || |
1671 | ($rm_tags == true && |
1672 | !in_array($tagname, $tag_list))){ |
1673 | $tagname = false; |
1674 | } else { |
2dd879b8 |
1675 | /** |
1676 | * Convert body into div. |
1677 | */ |
1678 | if ($tagname == "body"){ |
1679 | $tagname = "div"; |
da2415c1 |
1680 | $attary = sq_body2div($attary, $mailbox, |
2dd879b8 |
1681 | $message, $id); |
1682 | } |
691a2d25 |
1683 | if ($tagtype == 1){ |
1684 | if (isset($open_tags{$tagname})){ |
1685 | $open_tags{$tagname}++; |
1686 | } else { |
1687 | $open_tags{$tagname}=1; |
1688 | } |
1689 | } |
1690 | /** |
1691 | * This is where we run other checks. |
1692 | */ |
1693 | if (is_array($attary) && sizeof($attary) > 0){ |
1694 | $attary = sq_fixatts($tagname, |
1695 | $attary, |
1696 | $rm_attnames, |
1697 | $bad_attvals, |
1698 | $add_attr_to_tag, |
1699 | $message, |
b3af12ef |
1700 | $id, |
3d8371be |
1701 | $mailbox |
691a2d25 |
1702 | ); |
1703 | } |
1704 | } |
1705 | } |
691a2d25 |
1706 | } |
1707 | } |
1708 | if ($tagname != false && $skip_content == false){ |
1709 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
1710 | } |
691a2d25 |
1711 | } |
1712 | $curpos = $gt+1; |
a3daaaf3 |
1713 | } |
691a2d25 |
1714 | $trusted .= substr($body, $curpos, strlen($body)-$curpos); |
1715 | if ($force_tag_closing == true){ |
1716 | foreach ($open_tags as $tagname=>$opentimes){ |
1717 | while ($opentimes > 0){ |
1718 | $trusted .= '</' . $tagname . '>'; |
1719 | $opentimes--; |
1720 | } |
1721 | } |
1722 | $trusted .= "\n"; |
1723 | } |
1724 | $trusted .= "<!-- end sanitized html -->\n"; |
1725 | return $trusted; |
1726 | } |
451f74a2 |
1727 | |
691a2d25 |
1728 | /** |
1729 | * This is a wrapper function to call html sanitizing routines. |
1730 | * |
1731 | * @param $body the body of the message |
1732 | * @param $id the id of the message |
1733 | * @return a string with html safe to display in the browser. |
1734 | */ |
7aad7b77 |
1735 | function magicHTML($body, $id, $message, $mailbox = 'INBOX') { |
691a2d25 |
1736 | global $attachment_common_show_images, $view_unsafe_images, |
3d8371be |
1737 | $has_unsafe_images; |
691a2d25 |
1738 | /** |
1739 | * Don't display attached images in HTML mode. |
1740 | */ |
1741 | $attachment_common_show_images = false; |
1742 | $tag_list = Array( |
1743 | false, |
1744 | "object", |
1745 | "meta", |
1746 | "html", |
1747 | "head", |
cc34b00d |
1748 | "base", |
e18cb31b |
1749 | "link", |
45071bd6 |
1750 | "frame", |
2dd879b8 |
1751 | "iframe", |
1752 | "plaintext", |
1753 | "marquee" |
691a2d25 |
1754 | ); |
1755 | |
1756 | $rm_tags_with_content = Array( |
1757 | "script", |
1758 | "applet", |
1759 | "embed", |
2dd879b8 |
1760 | "title", |
1761 | "frameset", |
1762 | "xml" |
691a2d25 |
1763 | ); |
1764 | |
1765 | $self_closing_tags = Array( |
1766 | "img", |
1767 | "br", |
1768 | "hr", |
1769 | "input" |
1770 | ); |
1771 | |
2dd879b8 |
1772 | $force_tag_closing = true; |
691a2d25 |
1773 | |
1774 | $rm_attnames = Array( |
1775 | "/.*/" => |
1776 | Array( |
3a50c8d2 |
1777 | "/target/i", |
1778 | "/^on.*/i", |
1779 | "/^dynsrc/i", |
1780 | "/^data.*/i", |
1781 | "/^lowsrc.*/i" |
691a2d25 |
1782 | ) |
1783 | ); |
1784 | |
1785 | $secremoveimg = "../images/" . _("sec_remove_eng.png"); |
1786 | $bad_attvals = Array( |
1787 | "/.*/" => |
1788 | Array( |
0a6ec9b5 |
1789 | "/^src|background/i" => |
691a2d25 |
1790 | Array( |
1791 | Array( |
f83c60a2 |
1792 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
1793 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
1794 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
691a2d25 |
1795 | ), |
1796 | Array( |
1797 | "\\1$secremoveimg\\2", |
bb8d0799 |
1798 | "\\1$secremoveimg\\2", |
3d8371be |
1799 | "\\1$secremoveimg\\2", |
1800 | "\\1$secremoveimg\\2" |
691a2d25 |
1801 | ) |
1802 | ), |
0a6ec9b5 |
1803 | "/^href|action/i" => |
1804 | Array( |
1805 | Array( |
f83c60a2 |
1806 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
1807 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
1808 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
0a6ec9b5 |
1809 | ), |
1810 | Array( |
2dd879b8 |
1811 | "\\1#\\1", |
1812 | "\\1#\\1", |
1813 | "\\1#\\1", |
1814 | "\\1#\\1" |
0a6ec9b5 |
1815 | ) |
1816 | ), |
f83c60a2 |
1817 | "/^style/i" => |
691a2d25 |
1818 | Array( |
1819 | Array( |
f83c60a2 |
1820 | "/expression/i", |
1821 | "/binding/i", |
1822 | "/behaviou*r/i", |
2dd879b8 |
1823 | "/include-source/i", |
7d06541f |
1824 | "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si", |
1825 | "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si", |
1826 | "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si" |
691a2d25 |
1827 | ), |
1828 | Array( |
bb8d0799 |
1829 | "idiocy", |
3d8371be |
1830 | "idiocy", |
1831 | "idiocy", |
2dd879b8 |
1832 | "idiocy", |
1833 | "url(\\1#\\1)", |
1834 | "url(\\1#\\1)", |
1835 | "url(\\1#\\1)", |
1836 | "url(\\1#\\1)" |
691a2d25 |
1837 | ) |
1838 | ) |
1839 | ) |
1840 | ); |
5262d9a6 |
1841 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
2dd879b8 |
1842 | $view_unsafe_images = false; |
45071bd6 |
1843 | } |
691a2d25 |
1844 | if (!$view_unsafe_images){ |
1845 | /** |
1846 | * Remove any references to http/https if view_unsafe_images set |
1847 | * to false. |
1848 | */ |
0a6ec9b5 |
1849 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0], |
f83c60a2 |
1850 | '/^([\'\"])\s*https*:.*([\'\"])/si'); |
0a6ec9b5 |
1851 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1], |
2dd879b8 |
1852 | "\\1$secremoveimg\\1"); |
3a50c8d2 |
1853 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[0], |
43d678b6 |
1854 | '/url\(([\'\"])\s*https*:.*([\'\"])\)/si'); |
3a50c8d2 |
1855 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[1], |
2dd879b8 |
1856 | "url(\\1$secremoveimg\\1)"); |
691a2d25 |
1857 | } |
451f74a2 |
1858 | |
691a2d25 |
1859 | $add_attr_to_tag = Array( |
da2415c1 |
1860 | "/^a$/i" => |
2dd879b8 |
1861 | Array('target'=>'"_new"', |
1862 | 'title'=>'"'._("This external link will open in a new window").'"' |
1863 | ) |
1864 | ); |
da2415c1 |
1865 | $trusted = sq_sanitize($body, |
1866 | $tag_list, |
691a2d25 |
1867 | $rm_tags_with_content, |
1868 | $self_closing_tags, |
1869 | $force_tag_closing, |
1870 | $rm_attnames, |
1871 | $bad_attvals, |
1872 | $add_attr_to_tag, |
1873 | $message, |
b3af12ef |
1874 | $id, |
3d8371be |
1875 | $mailbox |
691a2d25 |
1876 | ); |
f83c60a2 |
1877 | if (preg_match("|$secremoveimg|i", $trusted)){ |
691a2d25 |
1878 | $has_unsafe_images = true; |
da2415c1 |
1879 | } |
691a2d25 |
1880 | return $trusted; |
451f74a2 |
1881 | } |
a4a70693 |
1882 | |
da2415c1 |
1883 | /** |
1884 | * function SendDownloadHeaders - send file to the browser |
1885 | * |
1886 | * Original Source: SM core src/download.php |
1887 | * moved here to make it available to other code, and separate |
1888 | * front end from back end functionality. |
1889 | * |
1890 | * @param string $type0 first half of mime type |
1891 | * @param string $type1 second half of mime type |
1892 | * @param string $filename filename to tell the browser for downloaded file |
1893 | * @param boolean $force whether to force the download dialog to pop |
07c49f57 |
1894 | * @param integer $filesize optional, send the Content-Header and length to the browser |
da2415c1 |
1895 | * @return void |
1896 | */ |
07c49f57 |
1897 | function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) { |
da2415c1 |
1898 | global $languages, $squirrelmail_language; |
1899 | $isIE = $isIE6 = 0; |
1900 | |
1901 | sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER); |
1902 | |
1903 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false && |
1904 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
1905 | $isIE = 1; |
1906 | } |
1907 | |
1908 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE 6') !== false && |
1909 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
1910 | $isIE6 = 1; |
1911 | } |
1912 | |
1913 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
1914 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
1915 | $filename = |
1916 | $languages[$squirrelmail_language]['XTRA_CODE']('downloadfilename', $filename, $HTTP_USER_AGENT); |
1917 | } else { |
1918 | $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename)); |
1919 | } |
1920 | |
1921 | // A Pox on Microsoft and it's Office! |
1922 | if (!$force) { |
1923 | // Try to show in browser window |
1924 | header("Content-Disposition: inline; filename=\"$filename\""); |
1925 | header("Content-Type: $type0/$type1; name=\"$filename\""); |
1926 | } else { |
1927 | // Try to pop up the "save as" box |
1928 | // IE makes this hard. It pops up 2 save boxes, or none. |
1929 | // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP |
1930 | // But, accordint to Microsoft, it is "RFC compliant but doesn't |
1931 | // take into account some deviations that allowed within the |
1932 | // specification." Doesn't that mean RFC non-compliant? |
1933 | // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP |
1934 | // |
1935 | // The best thing you can do for IE is to upgrade to the latest |
1936 | // version |
1937 | if ($isIE && !$isIE6) { |
1938 | // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp |
1939 | // Do not have quotes around filename, but that applied to |
1940 | // "attachment"... does it apply to inline too? |
1941 | // |
1942 | // This combination seems to work mostly. IE 5.5 SP 1 has |
1943 | // known issues (see the Microsoft Knowledge Base) |
1944 | header("Content-Disposition: inline; filename=$filename"); |
1945 | // This works for most types, but doesn't work with Word files |
1946 | header("Content-Type: application/download; name=\"$filename\""); |
1947 | |
1948 | // These are spares, just in case. :-) |
1949 | //header("Content-Type: $type0/$type1; name=\"$filename\""); |
1950 | //header("Content-Type: application/x-msdownload; name=\"$filename\""); |
1951 | //header("Content-Type: application/octet-stream; name=\"$filename\""); |
1952 | } else { |
1953 | header("Content-Disposition: attachment; filename=\"$filename\""); |
1954 | // application/octet-stream forces download for Netscape |
1955 | header("Content-Type: application/octet-stream; name=\"$filename\""); |
1956 | } |
1957 | } |
07c49f57 |
1958 | |
1959 | //send the content-length header if the calling function provides it |
1960 | if ($filesize > 0) { |
1961 | header("Content-Length: $filesize"); |
1962 | } |
1963 | |
1964 | } // end fn SendDownlaodHeaders |
da2415c1 |
1965 | |
d6c32258 |
1966 | ?> |