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 | } |
83cf04bd |
396 | $link = 'passed_id=' . $id . '&ent_id='.$ent_num. |
08b7f7cc |
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 | } |
83cf04bd |
402 | $body .= '<center><small><a href="download.php?absolute_dl=true&' . |
403 | $link . '">' . _("Download this as a file") . '</a>'; |
7aad7b77 |
404 | if ($view_unsafe_images) { |
23f617b8 |
405 | $text = _("Hide Unsafe Images"); |
7aad7b77 |
406 | } else { |
08b7f7cc |
407 | if (isset($has_unsafe_images) && $has_unsafe_images) { |
408 | $link .= '&view_unsafe_images=1'; |
409 | $text = _("View Unsafe Images"); |
410 | } else { |
411 | $text = ''; |
412 | } |
3d8371be |
413 | } |
83cf04bd |
414 | if($text != '') { |
415 | $body .= ' | <a href="read_body.php?' . $link . '">' . $text . '</a>'; |
416 | } |
417 | $body .= '</small></center><br>' . "\n"; |
3d8371be |
418 | } |
419 | return $body; |
451f74a2 |
420 | } |
b74ba498 |
421 | |
23bcec6f |
422 | |
423 | function formatAttachments($message, $exclude_id, $mailbox, $id) { |
b583c3e8 |
424 | global $where, $what, $startMessage, $color; |
77b88425 |
425 | static $ShownHTML = 0; |
451f74a2 |
426 | |
23bcec6f |
427 | $att_ar = $message->getAttachments($exclude_id); |
451f74a2 |
428 | |
23bcec6f |
429 | if (!count($att_ar)) return ''; |
2e25760a |
430 | |
8f5425ec |
431 | $attachments = ''; |
2e25760a |
432 | |
23bcec6f |
433 | $urlMailbox = urlencode($mailbox); |
434 | |
435 | foreach ($att_ar as $att) { |
fdc9d9b5 |
436 | $ent = $att->entity_id; |
2e25760a |
437 | $header = $att->header; |
f0c4dc12 |
438 | $type0 = strtolower($header->type0); |
439 | $type1 = strtolower($header->type1); |
2e25760a |
440 | $name = ''; |
3d8371be |
441 | $links['download link']['text'] = _("download"); |
6b04287c |
442 | $links['download link']['href'] = SM_PATH . |
443 | "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent"; |
2e25760a |
444 | $ImageURL = ''; |
445 | if ($type0 =='message' && $type1 == 'rfc822') { |
6b04287c |
446 | $default_page = SM_PATH . 'src/read_body.php'; |
2e25760a |
447 | $rfc822_header = $att->rfc822_header; |
098ea084 |
448 | $filename = $rfc822_header->subject; |
6cc08d8b |
449 | if (trim( $filename ) == '') { |
450 | $filename = 'untitled-[' . $ent . ']' ; |
08b7f7cc |
451 | } |
2e25760a |
452 | $from_o = $rfc822_header->from; |
453 | if (is_object($from_o)) { |
04ea844e |
454 | $from_name = decodeHeader($from_o->getAddress(false)); |
2e25760a |
455 | } else { |
456 | $from_name = _("Unknown sender"); |
f0c4dc12 |
457 | } |
3d8371be |
458 | $description = $from_name; |
23bcec6f |
459 | } else { |
6b04287c |
460 | $default_page = SM_PATH . 'src/download.php'; |
3d8371be |
461 | if (is_object($header->disposition)) { |
098ea084 |
462 | $filename = $header->disposition->getProperty('filename'); |
3d8371be |
463 | if (trim($filename) == '') { |
464 | $name = decodeHeader($header->disposition->getProperty('name')); |
465 | if (trim($name) == '') { |
098ea084 |
466 | $name = $header->getParameter('name'); |
1035e159 |
467 | if(trim($name) == '') { |
468 | if (trim( $header->id ) == '') { |
469 | $filename = 'untitled-[' . $ent . ']' ; |
470 | } else { |
471 | $filename = 'cid: ' . $header->id; |
08b7f7cc |
472 | } |
3d8371be |
473 | } else { |
08b7f7cc |
474 | $filename = $name; |
3d8371be |
475 | } |
476 | } else { |
477 | $filename = $name; |
478 | } |
479 | } |
480 | } else { |
08b7f7cc |
481 | $filename = $header->getParameter('name'); |
482 | if (!trim($filename)) { |
483 | if (trim( $header->id ) == '') { |
484 | $filename = 'untitled-[' . $ent . ']' ; |
485 | } else { |
486 | $filename = 'cid: ' . $header->id; |
487 | } |
488 | } |
489 | } |
f810c0b2 |
490 | if ($header->description) { |
098ea084 |
491 | $description = decodeHeader($header->description); |
f810c0b2 |
492 | } else { |
3d8371be |
493 | $description = ''; |
494 | } |
2e25760a |
495 | } |
496 | |
497 | $display_filename = $filename; |
3d8371be |
498 | if (isset($passed_ent_id)) { |
499 | $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id; |
500 | } else { |
501 | $passed_ent_id_link = ''; |
502 | } |
503 | $defaultlink = $default_page . "?startMessage=$startMessage" |
2e25760a |
504 | . "&passed_id=$id&mailbox=$urlMailbox" |
6b04287c |
505 | . '&ent_id='.$ent.$passed_ent_id_link; |
2e25760a |
506 | if ($where && $what) { |
09f4707e |
507 | $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what); |
2e25760a |
508 | } |
3d8371be |
509 | /* This executes the attachment hook with a specific MIME-type. |
510 | * If that doesn't have results, it tries if there's a rule |
511 | * for a more generic type. |
512 | */ |
513 | $hookresults = do_hook("attachment $type0/$type1", $links, |
514 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
515 | $display_filename, $where, $what); |
516 | if(count($hookresults[1]) <= 1) { |
517 | $hookresults = do_hook("attachment $type0/*", $links, |
518 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
519 | $display_filename, $where, $what); |
2e25760a |
520 | } |
451f74a2 |
521 | |
3d8371be |
522 | $links = $hookresults[1]; |
523 | $defaultlink = $hookresults[6]; |
77b88425 |
524 | |
2e25760a |
525 | $attachments .= '<TR><TD>' . |
098ea084 |
526 | '<A HREF="'.$defaultlink.'">'.decodeHeader($display_filename).'</A> </TD>' . |
23bcec6f |
527 | '<TD><SMALL><b>' . show_readable_size($header->size) . |
77b88425 |
528 | '</b> </small></TD>' . |
529 | "<TD><SMALL>[ $type0/$type1 ] </SMALL></TD>" . |
530 | '<TD><SMALL>'; |
f810c0b2 |
531 | $attachments .= '<b>' . $description . '</b>'; |
2e25760a |
532 | $attachments .= '</SMALL></TD><TD><SMALL> '; |
b74ba498 |
533 | |
3d8371be |
534 | $skipspaces = 1; |
535 | foreach ($links as $val) { |
536 | if ($skipspaces) { |
537 | $skipspaces = 0; |
2e25760a |
538 | } else { |
539 | $attachments .= ' | '; |
77b88425 |
540 | } |
3d8371be |
541 | $attachments .= '<a href="' . $val['href'] . '">' . $val['text'] . '</a>'; |
2e25760a |
542 | } |
3d8371be |
543 | unset($links); |
2e25760a |
544 | $attachments .= "</TD></TR>\n"; |
545 | } |
9ad17edb |
546 | $attachmentadd = do_hook_function('attachments_bottom',$attachments); |
547 | if ($attachmentadd != '') |
548 | $attachments = $attachmentadd; |
2e25760a |
549 | return $attachments; |
451f74a2 |
550 | } |
b74ba498 |
551 | |
7c7b74b3 |
552 | function sqimap_base64_decode(&$string) { |
325978ac |
553 | $string = str_replace("\r\n", "\n", $string); |
7c7b74b3 |
554 | $string = base64_decode($string); |
555 | } |
556 | |
3d8371be |
557 | /* This function decodes the body depending on the encoding type. */ |
451f74a2 |
558 | function decodeBody($body, $encoding) { |
3d8371be |
559 | global $show_html_default; |
83be314a |
560 | |
b583c3e8 |
561 | $body = str_replace("\r\n", "\n", $body); |
562 | $encoding = strtolower($encoding); |
3d8371be |
563 | |
5166f86a |
564 | $encoding_handler = do_hook_function('decode_body', $encoding); |
565 | |
566 | |
567 | // plugins get first shot at decoding the body |
568 | // |
569 | if (!empty($encoding_handler) && function_exists($encoding_handler)) { |
570 | $body = $encoding_handler('decode', $body); |
571 | |
572 | } else if ($encoding == 'quoted-printable' || |
3d8371be |
573 | $encoding == 'quoted_printable') { |
b583c3e8 |
574 | $body = quoted_printable_decode($body); |
3d8371be |
575 | |
b583c3e8 |
576 | while (ereg("=\n", $body)) { |
577 | $body = ereg_replace ("=\n", '', $body); |
578 | } |
3d8371be |
579 | |
b583c3e8 |
580 | } else if ($encoding == 'base64') { |
581 | $body = base64_decode($body); |
582 | } |
3d8371be |
583 | |
b583c3e8 |
584 | // All other encodings are returned raw. |
3d8371be |
585 | return $body; |
451f74a2 |
586 | } |
587 | |
588 | /* |
589 | * This functions decode strings that is encoded according to |
590 | * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text). |
79e07c7e |
591 | * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002 |
451f74a2 |
592 | */ |
098ea084 |
593 | function decodeHeader ($string, $utfencode=true,$htmlsave=true) { |
d7448126 |
594 | global $languages, $squirrelmail_language; |
79e07c7e |
595 | if (is_array($string)) { |
596 | $string = implode("\n", $string); |
597 | } |
da2415c1 |
598 | |
10dec454 |
599 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
600 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
6fbd125b |
601 | $string = $languages[$squirrelmail_language]['XTRA_CODE']('decodeheader', $string); |
08b7f7cc |
602 | // Do we need to return at this point? |
603 | // return $string; |
83be314a |
604 | } |
79e07c7e |
605 | $i = 0; |
08b7f7cc |
606 | $iLastMatch = -2; |
db65b6b0 |
607 | $encoded = true; |
0a06275a |
608 | |
098ea084 |
609 | $aString = explode(' ',$string); |
08b7f7cc |
610 | $ret = ''; |
098ea084 |
611 | foreach ($aString as $chunk) { |
358a78a1 |
612 | if ($encoded && $chunk === '') { |
08b7f7cc |
613 | continue; |
358a78a1 |
614 | } elseif ($chunk === '') { |
08b7f7cc |
615 | $ret .= ' '; |
616 | continue; |
617 | } |
098ea084 |
618 | $encoded = false; |
08b7f7cc |
619 | /* if encoded words are not separated by a linear-space-white we still catch them */ |
620 | $j = $i-1; |
7e6ca3e8 |
621 | |
08b7f7cc |
622 | while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) { |
623 | /* if the last chunk isn't an encoded string then put back the space, otherwise don't */ |
624 | if ($iLastMatch !== $j) { |
625 | if ($htmlsave) { |
626 | $ret .= ' '; |
627 | } else { |
628 | $ret .= ' '; |
629 | } |
630 | } |
631 | $iLastMatch = $i; |
632 | $j = $i; |
633 | $ret .= $res[1]; |
098ea084 |
634 | $encoding = ucfirst($res[3]); |
635 | switch ($encoding) |
636 | { |
637 | case 'B': |
638 | $replace = base64_decode($res[4]); |
08b7f7cc |
639 | $ret .= charset_decode($res[2],$replace); |
098ea084 |
640 | break; |
641 | case 'Q': |
098ea084 |
642 | $replace = str_replace('_', ' ', $res[4]); |
da2415c1 |
643 | $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))', |
79e07c7e |
644 | $replace); |
098ea084 |
645 | /* Only encode into entities by default. Some places |
646 | * don't need the encoding, like the compose form. |
647 | */ |
648 | if ($utfencode) { |
649 | $replace = charset_decode($res[2], $replace); |
650 | } else { |
651 | if ($htmlsave) { |
c96c32f4 |
652 | $replace = htmlspecialchars($replace); |
098ea084 |
653 | } |
654 | } |
08b7f7cc |
655 | $ret .= $replace; |
098ea084 |
656 | break; |
657 | default: |
658 | break; |
79e07c7e |
659 | } |
098ea084 |
660 | $chunk = $res[5]; |
661 | $encoded = true; |
08b7f7cc |
662 | } |
7e6ca3e8 |
663 | |
664 | if (!$encoded && $htmlsave) { |
665 | $ret .= htmlspecialchars($chunk); |
666 | } else { |
667 | $ret .= $chunk; |
668 | } |
669 | |
08b7f7cc |
670 | if (!$encoded) { |
671 | if ($htmlsave) { |
672 | $ret .= ' '; |
673 | } else { |
674 | $ret .= ' '; |
da2415c1 |
675 | } |
08b7f7cc |
676 | } |
098ea084 |
677 | ++$i; |
678 | } |
da2415c1 |
679 | |
08b7f7cc |
680 | return $ret; |
451f74a2 |
681 | } |
682 | |
683 | /* |
684 | * Encode a string according to RFC 1522 for use in headers if it |
685 | * contains 8-bit characters or anything that looks like it should |
686 | * be encoded. |
687 | */ |
688 | function encodeHeader ($string) { |
6fbd125b |
689 | global $default_charset, $languages, $squirrelmail_language; |
83be314a |
690 | |
10dec454 |
691 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
692 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
6fbd125b |
693 | return $languages[$squirrelmail_language]['XTRA_CODE']('encodeheader', $string); |
83be314a |
694 | } |
c96c32f4 |
695 | if (strtolower($default_charset) == 'iso-8859-1') { |
696 | $string = str_replace("\240",' ',$string); |
697 | } |
793cc001 |
698 | |
451f74a2 |
699 | // Encode only if the string contains 8-bit characters or =? |
3d8371be |
700 | $j = strlen($string); |
098ea084 |
701 | $max_l = 75 - strlen($default_charset) - 7; |
702 | $aRet = array(); |
451f74a2 |
703 | $ret = ''; |
c96c32f4 |
704 | $iEncStart = $enc_init = false; |
0d53f0f9 |
705 | $cur_l = $iOffset = 0; |
3d8371be |
706 | for($i = 0; $i < $j; ++$i) { |
c96c32f4 |
707 | switch($string{$i}) |
708 | { |
709 | case '=': |
710 | case '<': |
711 | case '>': |
712 | case ',': |
713 | case '?': |
714 | case '_': |
715 | if ($iEncStart === false) { |
716 | $iEncStart = $i; |
717 | } |
718 | $cur_l+=3; |
719 | if ($cur_l > ($max_l-2)) { |
08b7f7cc |
720 | /* if there is an stringpart that doesn't need encoding, add it */ |
c96c32f4 |
721 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
722 | $aRet[] = "=?$default_charset?Q?$ret?="; |
723 | $iOffset = $i; |
724 | $cur_l = 0; |
725 | $ret = ''; |
726 | $iEncStart = false; |
727 | } else { |
728 | $ret .= sprintf("=%02X",ord($string{$i})); |
729 | } |
730 | break; |
731 | case '(': |
732 | case ')': |
733 | if ($iEncStart !== false) { |
08b7f7cc |
734 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
735 | $aRet[] = "=?$default_charset?Q?$ret?="; |
736 | $iOffset = $i; |
737 | $cur_l = 0; |
738 | $ret = ''; |
739 | $iEncStart = false; |
740 | } |
741 | break; |
742 | case ' ': |
743 | if ($iEncStart !== false) { |
098ea084 |
744 | $cur_l++; |
745 | if ($cur_l > $max_l) { |
08b7f7cc |
746 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
747 | $aRet[] = "=?$default_charset?Q?$ret?="; |
748 | $iOffset = $i; |
749 | $cur_l = 0; |
750 | $ret = ''; |
751 | $iEncStart = false; |
08b7f7cc |
752 | } else { |
c96c32f4 |
753 | $ret .= '_'; |
754 | } |
755 | } |
756 | break; |
757 | default: |
758 | $k = ord($string{$i}); |
759 | if ($k > 126) { |
760 | if ($iEncStart === false) { |
325978ac |
761 | // do not start encoding in the middle of a string, also take the rest of the word. |
762 | $sLeadString = substr($string,0,$i); |
763 | $aLeadString = explode(' ',$sLeadString); |
da2415c1 |
764 | $sToBeEncoded = array_pop($aLeadString); |
325978ac |
765 | $iEncStart = $i - strlen($sToBeEncoded); |
766 | $ret .= $sToBeEncoded; |
767 | $cur_l += strlen($sToBeEncoded); |
c96c32f4 |
768 | } |
769 | $cur_l += 3; |
08b7f7cc |
770 | /* first we add the encoded string that reached it's max size */ |
c96c32f4 |
771 | if ($cur_l > ($max_l-2)) { |
08b7f7cc |
772 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
773 | $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */ |
c96c32f4 |
774 | $cur_l = 3; |
775 | $ret = ''; |
776 | $iOffset = $i; |
08b7f7cc |
777 | $iEncStart = $i; |
c96c32f4 |
778 | } |
08b7f7cc |
779 | $enc_init = true; |
c96c32f4 |
780 | $ret .= sprintf("=%02X", $k); |
781 | } else { |
782 | if ($iEncStart !== false) { |
098ea084 |
783 | $cur_l++; |
784 | if ($cur_l > $max_l) { |
08b7f7cc |
785 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
786 | $aRet[] = "=?$default_charset?Q?$ret?="; |
787 | $iEncStart = false; |
788 | $iOffset = $i; |
789 | $cur_l = 0; |
098ea084 |
790 | $ret = ''; |
08b7f7cc |
791 | } else { |
c96c32f4 |
792 | $ret .= $string{$i}; |
793 | } |
3d8371be |
794 | } |
c96c32f4 |
795 | } |
796 | break; |
f7b3ba37 |
797 | } |
451f74a2 |
798 | } |
793cc001 |
799 | |
c96c32f4 |
800 | if ($enc_init) { |
801 | if ($iEncStart !== false) { |
802 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
803 | $aRet[] = "=?$default_charset?Q?$ret?="; |
804 | } else { |
805 | $aRet[] = substr($string,$iOffset); |
806 | } |
807 | $string = implode('',$aRet); |
451f74a2 |
808 | } |
3d8371be |
809 | return $string; |
451f74a2 |
810 | } |
b74ba498 |
811 | |
691a2d25 |
812 | /* This function trys to locate the entity_id of a specific mime element */ |
3d8371be |
813 | function find_ent_id($id, $message) { |
a171b359 |
814 | for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) { |
815 | if ($message->entities[$i]->header->type0 == 'multipart') { |
3d8371be |
816 | $ret = find_ent_id($id, $message->entities[$i]); |
451f74a2 |
817 | } else { |
3d8371be |
818 | if (strcasecmp($message->entities[$i]->header->id, $id) == 0) { |
d8cffbab |
819 | // if (sq_check_save_extension($message->entities[$i])) { |
a171b359 |
820 | return $message->entities[$i]->entity_id; |
da2415c1 |
821 | // } |
3d8371be |
822 | } |
a3daaaf3 |
823 | } |
451f74a2 |
824 | } |
3d8371be |
825 | return $ret; |
451f74a2 |
826 | } |
a3daaaf3 |
827 | |
e5e9381a |
828 | function sq_check_save_extension($message) { |
829 | $filename = $message->getFilename(); |
830 | $ext = substr($filename, strrpos($filename,'.')+1); |
831 | $save_extensions = array('jpg','jpeg','gif','png','bmp'); |
3d8371be |
832 | return in_array($ext, $save_extensions); |
e5e9381a |
833 | } |
834 | |
835 | |
691a2d25 |
836 | /** |
837 | ** HTMLFILTER ROUTINES |
838 | */ |
451f74a2 |
839 | |
2dd879b8 |
840 | /** |
841 | * This function is more or less a wrapper around stripslashes. Apparently |
842 | * Explorer is stupid enough to just remove the backslashes and then |
843 | * execute the content of the attribute as if nothing happened. |
844 | * Who does that? |
845 | * |
846 | * @param attvalue The value of the attribute |
847 | * @return attvalue The value of the attribute stripslashed. |
848 | */ |
849 | function sq_unbackslash($attvalue){ |
850 | /** |
851 | * Remove any backslashes. See if there are any first. |
852 | */ |
91220ed1 |
853 | |
2dd879b8 |
854 | if (strstr($attvalue, '\\') !== false){ |
855 | $attvalue = stripslashes($attvalue); |
856 | } |
857 | return $attvalue; |
858 | } |
859 | |
860 | /** |
861 | * Kill any tabs, newlines, or carriage returns. Our friends the |
862 | * makers of the browser with 95% market value decided that it'd |
863 | * be funny to make "java[tab]script" be just as good as "javascript". |
da2415c1 |
864 | * |
2dd879b8 |
865 | * @param attvalue The attribute value before extraneous spaces removed. |
866 | * @return attvalue The attribute value after extraneous spaces removed. |
867 | */ |
868 | function sq_unspace($attvalue){ |
869 | if (strcspn($attvalue, "\t\r\n") != strlen($attvalue)){ |
870 | $attvalue = str_replace(Array("\t", "\r", "\n"), Array('', '', ''), |
871 | $attvalue); |
872 | } |
873 | return $attvalue; |
874 | } |
875 | |
691a2d25 |
876 | /** |
877 | * This function returns the final tag out of the tag name, an array |
da2415c1 |
878 | * of attributes, and the type of the tag. This function is called by |
691a2d25 |
879 | * sq_sanitize internally. |
880 | * |
881 | * @param $tagname the name of the tag. |
882 | * @param $attary the array of attributes and their values |
883 | * @param $tagtype The type of the tag (see in comments). |
884 | * @return a string with the final tag representation. |
885 | */ |
886 | function sq_tagprint($tagname, $attary, $tagtype){ |
b583c3e8 |
887 | $me = 'sq_tagprint'; |
3d8371be |
888 | |
691a2d25 |
889 | if ($tagtype == 2){ |
890 | $fulltag = '</' . $tagname . '>'; |
891 | } else { |
892 | $fulltag = '<' . $tagname; |
893 | if (is_array($attary) && sizeof($attary)){ |
894 | $atts = Array(); |
895 | while (list($attname, $attvalue) = each($attary)){ |
896 | array_push($atts, "$attname=$attvalue"); |
897 | } |
898 | $fulltag .= ' ' . join(" ", $atts); |
899 | } |
900 | if ($tagtype == 3){ |
b583c3e8 |
901 | $fulltag .= ' /'; |
691a2d25 |
902 | } |
b583c3e8 |
903 | $fulltag .= '>'; |
451f74a2 |
904 | } |
691a2d25 |
905 | return $fulltag; |
451f74a2 |
906 | } |
a3daaaf3 |
907 | |
691a2d25 |
908 | /** |
909 | * A small helper function to use with array_walk. Modifies a by-ref |
910 | * value and makes it lowercase. |
911 | * |
912 | * @param $val a value passed by-ref. |
913 | * @return void since it modifies a by-ref value. |
914 | */ |
915 | function sq_casenormalize(&$val){ |
916 | $val = strtolower($val); |
917 | } |
451f74a2 |
918 | |
691a2d25 |
919 | /** |
920 | * This function skips any whitespace from the current position within |
921 | * a string and to the next non-whitespace value. |
da2415c1 |
922 | * |
691a2d25 |
923 | * @param $body the string |
924 | * @param $offset the offset within the string where we should start |
925 | * looking for the next non-whitespace character. |
926 | * @return the location within the $body where the next |
927 | * non-whitespace char is located. |
928 | */ |
929 | function sq_skipspace($body, $offset){ |
b583c3e8 |
930 | $me = 'sq_skipspace'; |
3d8371be |
931 | preg_match('/^(\s*)/s', substr($body, $offset), $matches); |
691a2d25 |
932 | if (sizeof($matches{1})){ |
933 | $count = strlen($matches{1}); |
934 | $offset += $count; |
451f74a2 |
935 | } |
691a2d25 |
936 | return $offset; |
451f74a2 |
937 | } |
a3daaaf3 |
938 | |
691a2d25 |
939 | /** |
940 | * This function looks for the next character within a string. It's |
941 | * really just a glorified "strpos", except it catches if failures |
942 | * nicely. |
943 | * |
944 | * @param $body The string to look for needle in. |
945 | * @param $offset Start looking from this position. |
946 | * @param $needle The character/string to look for. |
947 | * @return location of the next occurance of the needle, or |
948 | * strlen($body) if needle wasn't found. |
949 | */ |
950 | function sq_findnxstr($body, $offset, $needle){ |
3d8371be |
951 | $me = 'sq_findnxstr'; |
691a2d25 |
952 | $pos = strpos($body, $needle, $offset); |
953 | if ($pos === FALSE){ |
954 | $pos = strlen($body); |
451f74a2 |
955 | } |
691a2d25 |
956 | return $pos; |
451f74a2 |
957 | } |
a3daaaf3 |
958 | |
691a2d25 |
959 | /** |
960 | * This function takes a PCRE-style regexp and tries to match it |
961 | * within the string. |
962 | * |
963 | * @param $body The string to look for needle in. |
964 | * @param $offset Start looking from here. |
965 | * @param $reg A PCRE-style regex to match. |
966 | * @return Returns a false if no matches found, or an array |
967 | * with the following members: |
968 | * - integer with the location of the match within $body |
969 | * - string with whatever content between offset and the match |
970 | * - string with whatever it is we matched |
971 | */ |
972 | function sq_findnxreg($body, $offset, $reg){ |
b583c3e8 |
973 | $me = 'sq_findnxreg'; |
691a2d25 |
974 | $matches = Array(); |
975 | $retarr = Array(); |
7d06541f |
976 | preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches); |
977 | if (!isset($matches{0}) || !$matches{0}){ |
691a2d25 |
978 | $retarr = false; |
979 | } else { |
980 | $retarr{0} = $offset + strlen($matches{1}); |
981 | $retarr{1} = $matches{1}; |
982 | $retarr{2} = $matches{2}; |
983 | } |
984 | return $retarr; |
985 | } |
a3daaaf3 |
986 | |
691a2d25 |
987 | /** |
988 | * This function looks for the next tag. |
989 | * |
990 | * @param $body String where to look for the next tag. |
991 | * @param $offset Start looking from here. |
992 | * @return false if no more tags exist in the body, or |
993 | * an array with the following members: |
994 | * - string with the name of the tag |
995 | * - array with attributes and their values |
996 | * - integer with tag type (1, 2, or 3) |
997 | * - integer where the tag starts (starting "<") |
998 | * - integer where the tag ends (ending ">") |
999 | * first three members will be false, if the tag is invalid. |
1000 | */ |
1001 | function sq_getnxtag($body, $offset){ |
b583c3e8 |
1002 | $me = 'sq_getnxtag'; |
691a2d25 |
1003 | if ($offset > strlen($body)){ |
1004 | return false; |
1005 | } |
1006 | $lt = sq_findnxstr($body, $offset, "<"); |
1007 | if ($lt == strlen($body)){ |
1008 | return false; |
1009 | } |
1010 | /** |
1011 | * We are here: |
1012 | * blah blah <tag attribute="value"> |
1013 | * \---------^ |
1014 | */ |
1015 | $pos = sq_skipspace($body, $lt+1); |
1016 | if ($pos >= strlen($body)){ |
1017 | return Array(false, false, false, $lt, strlen($body)); |
1018 | } |
1019 | /** |
1020 | * There are 3 kinds of tags: |
1021 | * 1. Opening tag, e.g.: |
1022 | * <a href="blah"> |
1023 | * 2. Closing tag, e.g.: |
1024 | * </a> |
1025 | * 3. XHTML-style content-less tag, e.g.: |
1026 | * <img src="blah"/> |
1027 | */ |
1028 | $tagtype = false; |
1029 | switch (substr($body, $pos, 1)){ |
3d8371be |
1030 | case '/': |
1031 | $tagtype = 2; |
1032 | $pos++; |
1033 | break; |
1034 | case '!': |
1035 | /** |
1036 | * A comment or an SGML declaration. |
1037 | */ |
1038 | if (substr($body, $pos+1, 2) == "--"){ |
1039 | $gt = strpos($body, "-->", $pos); |
1040 | if ($gt === false){ |
1041 | $gt = strlen($body); |
1042 | } else { |
1043 | $gt += 2; |
1044 | } |
1045 | return Array(false, false, false, $lt, $gt); |
bb8d0799 |
1046 | } else { |
3d8371be |
1047 | $gt = sq_findnxstr($body, $pos, ">"); |
1048 | return Array(false, false, false, $lt, $gt); |
1049 | } |
1050 | break; |
1051 | default: |
1052 | /** |
1053 | * Assume tagtype 1 for now. If it's type 3, we'll switch values |
1054 | * later. |
1055 | */ |
1056 | $tagtype = 1; |
1057 | break; |
691a2d25 |
1058 | } |
a3daaaf3 |
1059 | |
691a2d25 |
1060 | $tag_start = $pos; |
1061 | $tagname = ''; |
1062 | /** |
1063 | * Look for next [\W-_], which will indicate the end of the tag name. |
1064 | */ |
1065 | $regary = sq_findnxreg($body, $pos, "[^\w\-_]"); |
1066 | if ($regary == false){ |
1067 | return Array(false, false, false, $lt, strlen($body)); |
1068 | } |
1069 | list($pos, $tagname, $match) = $regary; |
1070 | $tagname = strtolower($tagname); |
1071 | |
1072 | /** |
1073 | * $match can be either of these: |
1074 | * '>' indicating the end of the tag entirely. |
1075 | * '\s' indicating the end of the tag name. |
1076 | * '/' indicating that this is type-3 xhtml tag. |
da2415c1 |
1077 | * |
691a2d25 |
1078 | * Whatever else we find there indicates an invalid tag. |
1079 | */ |
1080 | switch ($match){ |
3d8371be |
1081 | case '/': |
691a2d25 |
1082 | /** |
3d8371be |
1083 | * This is an xhtml-style tag with a closing / at the |
1084 | * end, like so: <img src="blah"/>. Check if it's followed |
1085 | * by the closing bracket. If not, then this tag is invalid |
691a2d25 |
1086 | */ |
3d8371be |
1087 | if (substr($body, $pos, 2) == "/>"){ |
1088 | $pos++; |
1089 | $tagtype = 3; |
1090 | } else { |
1091 | $gt = sq_findnxstr($body, $pos, ">"); |
1092 | $retary = Array(false, false, false, $lt, $gt); |
1093 | return $retary; |
1094 | } |
1095 | case '>': |
1096 | return Array($tagname, false, $tagtype, $lt, $pos); |
1097 | break; |
1098 | default: |
1099 | /** |
1100 | * Check if it's whitespace |
1101 | */ |
1102 | if (!preg_match('/\s/', $match)){ |
1103 | /** |
1104 | * This is an invalid tag! Look for the next closing ">". |
1105 | */ |
7d06541f |
1106 | $gt = sq_findnxstr($body, $lt, ">"); |
3d8371be |
1107 | return Array(false, false, false, $lt, $gt); |
1108 | } |
1109 | break; |
691a2d25 |
1110 | } |
3d8371be |
1111 | |
691a2d25 |
1112 | /** |
1113 | * At this point we're here: |
1114 | * <tagname attribute='blah'> |
1115 | * \-------^ |
1116 | * |
1117 | * At this point we loop in order to find all attributes. |
1118 | */ |
1119 | $attname = ''; |
1120 | $atttype = false; |
1121 | $attary = Array(); |
1122 | |
1123 | while ($pos <= strlen($body)){ |
1124 | $pos = sq_skipspace($body, $pos); |
1125 | if ($pos == strlen($body)){ |
1126 | /** |
1127 | * Non-closed tag. |
1128 | */ |
1129 | return Array(false, false, false, $lt, $pos); |
1130 | } |
1131 | /** |
1132 | * See if we arrived at a ">" or "/>", which means that we reached |
1133 | * the end of the tag. |
1134 | */ |
1135 | $matches = Array(); |
164800ad |
1136 | if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) { |
c828931c |
1137 | /** |
1138 | * Yep. So we did. |
1139 | */ |
1140 | $pos += strlen($matches{1}); |
1141 | if ($matches{2} == "/>"){ |
1142 | $tagtype = 3; |
1143 | $pos++; |
1144 | } |
1145 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
1146 | } |
a3daaaf3 |
1147 | |
cca46357 |
1148 | /** |
691a2d25 |
1149 | * There are several types of attributes, with optional |
1150 | * [:space:] between members. |
1151 | * Type 1: |
1152 | * attrname[:space:]=[:space:]'CDATA' |
1153 | * Type 2: |
1154 | * attrname[:space:]=[:space:]"CDATA" |
1155 | * Type 3: |
1156 | * attr[:space:]=[:space:]CDATA |
1157 | * Type 4: |
1158 | * attrname |
cca46357 |
1159 | * |
691a2d25 |
1160 | * We leave types 1 and 2 the same, type 3 we check for |
1161 | * '"' and convert to """ if needed, then wrap in |
1162 | * double quotes. Type 4 we convert into: |
1163 | * attrname="yes". |
cca46357 |
1164 | */ |
3f7c623f |
1165 | $regary = sq_findnxreg($body, $pos, "[^:\w\-_]"); |
691a2d25 |
1166 | if ($regary == false){ |
1167 | /** |
1168 | * Looks like body ended before the end of tag. |
1169 | */ |
1170 | return Array(false, false, false, $lt, strlen($body)); |
cca46357 |
1171 | } |
691a2d25 |
1172 | list($pos, $attname, $match) = $regary; |
1173 | $attname = strtolower($attname); |
1174 | /** |
1175 | * We arrived at the end of attribute name. Several things possible |
1176 | * here: |
1177 | * '>' means the end of the tag and this is attribute type 4 |
1178 | * '/' if followed by '>' means the same thing as above |
1179 | * '\s' means a lot of things -- look what it's followed by. |
1180 | * anything else means the attribute is invalid. |
1181 | */ |
1182 | switch($match){ |
3d8371be |
1183 | case '/': |
691a2d25 |
1184 | /** |
3d8371be |
1185 | * This is an xhtml-style tag with a closing / at the |
1186 | * end, like so: <img src="blah"/>. Check if it's followed |
1187 | * by the closing bracket. If not, then this tag is invalid |
691a2d25 |
1188 | */ |
3d8371be |
1189 | if (substr($body, $pos, 2) == "/>"){ |
691a2d25 |
1190 | $pos++; |
3d8371be |
1191 | $tagtype = 3; |
691a2d25 |
1192 | } else { |
3d8371be |
1193 | $gt = sq_findnxstr($body, $pos, ">"); |
1194 | $retary = Array(false, false, false, $lt, $gt); |
1195 | return $retary; |
1196 | } |
1197 | case '>': |
1198 | $attary{$attname} = '"yes"'; |
1199 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
1200 | break; |
1201 | default: |
1202 | /** |
1203 | * Skip whitespace and see what we arrive at. |
1204 | */ |
1205 | $pos = sq_skipspace($body, $pos); |
1206 | $char = substr($body, $pos, 1); |
1207 | /** |
1208 | * Two things are valid here: |
1209 | * '=' means this is attribute type 1 2 or 3. |
1210 | * \w means this was attribute type 4. |
1211 | * anything else we ignore and re-loop. End of tag and |
1212 | * invalid stuff will be caught by our checks at the beginning |
1213 | * of the loop. |
1214 | */ |
1215 | if ($char == "="){ |
1216 | $pos++; |
1217 | $pos = sq_skipspace($body, $pos); |
691a2d25 |
1218 | /** |
3d8371be |
1219 | * Here are 3 possibilities: |
1220 | * "'" attribute type 1 |
1221 | * '"' attribute type 2 |
1222 | * everything else is the content of tag type 3 |
691a2d25 |
1223 | */ |
3d8371be |
1224 | $quot = substr($body, $pos, 1); |
1225 | if ($quot == "'"){ |
1226 | $regary = sq_findnxreg($body, $pos+1, "\'"); |
1227 | if ($regary == false){ |
1228 | return Array(false, false, false, $lt, strlen($body)); |
1229 | } |
1230 | list($pos, $attval, $match) = $regary; |
1231 | $pos++; |
1232 | $attary{$attname} = "'" . $attval . "'"; |
1233 | } else if ($quot == '"'){ |
1234 | $regary = sq_findnxreg($body, $pos+1, '\"'); |
1235 | if ($regary == false){ |
1236 | return Array(false, false, false, $lt, strlen($body)); |
1237 | } |
1238 | list($pos, $attval, $match) = $regary; |
1239 | $pos++; |
1240 | $attary{$attname} = '"' . $attval . '"'; |
1241 | } else { |
1242 | /** |
1243 | * These are hateful. Look for \s, or >. |
1244 | */ |
1245 | $regary = sq_findnxreg($body, $pos, "[\s>]"); |
1246 | if ($regary == false){ |
1247 | return Array(false, false, false, $lt, strlen($body)); |
1248 | } |
1249 | list($pos, $attval, $match) = $regary; |
1250 | /** |
1251 | * If it's ">" it will be caught at the top. |
1252 | */ |
1253 | $attval = preg_replace("/\"/s", """, $attval); |
1254 | $attary{$attname} = '"' . $attval . '"'; |
7e235a1a |
1255 | } |
3d8371be |
1256 | } else if (preg_match("|[\w/>]|", $char)) { |
691a2d25 |
1257 | /** |
3d8371be |
1258 | * That was attribute type 4. |
691a2d25 |
1259 | */ |
3d8371be |
1260 | $attary{$attname} = '"yes"'; |
1261 | } else { |
1262 | /** |
1263 | * An illegal character. Find next '>' and return. |
1264 | */ |
1265 | $gt = sq_findnxstr($body, $pos, ">"); |
1266 | return Array(false, false, false, $lt, $gt); |
451f74a2 |
1267 | } |
3d8371be |
1268 | break; |
691a2d25 |
1269 | } |
1270 | } |
1271 | /** |
1272 | * The fact that we got here indicates that the tag end was never |
1273 | * found. Return invalid tag indication so it gets stripped. |
1274 | */ |
1275 | return Array(false, false, false, $lt, strlen($body)); |
1276 | } |
1277 | |
1278 | /** |
1279 | * This function checks attribute values for entity-encoded values |
1280 | * and returns them translated into 8-bit strings so we can run |
1281 | * checks on them. |
1282 | * |
1283 | * @param $attvalue A string to run entity check against. |
1284 | * @return Translated value. |
1285 | */ |
91220ed1 |
1286 | |
691a2d25 |
1287 | function sq_deent($attvalue){ |
b583c3e8 |
1288 | $me = 'sq_deent'; |
691a2d25 |
1289 | /** |
1290 | * See if we have to run the checks first. All entities must start |
1291 | * with "&". |
1292 | */ |
91220ed1 |
1293 | if (strpos($attvalue, '&') === false){ |
691a2d25 |
1294 | return $attvalue; |
1295 | } |
1296 | /** |
1297 | * Check named entities first. |
1298 | */ |
1299 | $trans = get_html_translation_table(HTML_ENTITIES); |
1300 | /** |
1301 | * Leave " in, as it can mess us up. |
1302 | */ |
1303 | $trans = array_flip($trans); |
91220ed1 |
1304 | unset($trans{'"'}); |
691a2d25 |
1305 | while (list($ent, $val) = each($trans)){ |
91220ed1 |
1306 | $attvalue = preg_replace('/' . $ent . '*/si', $val, $attvalue); |
691a2d25 |
1307 | } |
1308 | /** |
1309 | * Now translate numbered entities from 1 to 255 if needed. |
1310 | */ |
91220ed1 |
1311 | if (strpos($attvalue, '#') !== false){ |
691a2d25 |
1312 | $omit = Array(34, 39); |
91220ed1 |
1313 | for ($asc = 256; $asc >= 0; $asc--){ |
691a2d25 |
1314 | if (!in_array($asc, $omit)){ |
1315 | $chr = chr($asc); |
91220ed1 |
1316 | $octrule = '/\�*' . $asc . ';*/si'; |
1317 | $hexrule = '/\�*' . dechex($asc) . ';*/si'; |
1318 | $attvalue = preg_replace($octrule, $chr, $attvalue); |
1319 | $attvalue = preg_replace($hexrule, $chr, $attvalue); |
a3daaaf3 |
1320 | } |
691a2d25 |
1321 | } |
1322 | } |
1323 | return $attvalue; |
1324 | } |
1325 | |
1326 | /** |
1327 | * This function runs various checks against the attributes. |
1328 | * |
1329 | * @param $tagname String with the name of the tag. |
1330 | * @param $attary Array with all tag attributes. |
1331 | * @param $rm_attnames See description for sq_sanitize |
1332 | * @param $bad_attvals See description for sq_sanitize |
1333 | * @param $add_attr_to_tag See description for sq_sanitize |
1334 | * @param $message message object |
1335 | * @param $id message id |
1336 | * @return Array with modified attributes. |
1337 | */ |
da2415c1 |
1338 | function sq_fixatts($tagname, |
1339 | $attary, |
691a2d25 |
1340 | $rm_attnames, |
1341 | $bad_attvals, |
1342 | $add_attr_to_tag, |
1343 | $message, |
b3af12ef |
1344 | $id, |
3d8371be |
1345 | $mailbox |
691a2d25 |
1346 | ){ |
b583c3e8 |
1347 | $me = 'sq_fixatts'; |
691a2d25 |
1348 | while (list($attname, $attvalue) = each($attary)){ |
1349 | /** |
1350 | * See if this attribute should be removed. |
1351 | */ |
1352 | foreach ($rm_attnames as $matchtag=>$matchattrs){ |
1353 | if (preg_match($matchtag, $tagname)){ |
1354 | foreach ($matchattrs as $matchattr){ |
1355 | if (preg_match($matchattr, $attname)){ |
1356 | unset($attary{$attname}); |
1357 | continue; |
1358 | } |
451f74a2 |
1359 | } |
451f74a2 |
1360 | } |
691a2d25 |
1361 | } |
1362 | /** |
2dd879b8 |
1363 | * Remove any backslashes, entities, and extraneous whitespace. |
691a2d25 |
1364 | */ |
2dd879b8 |
1365 | $attvalue = sq_unbackslash($attvalue); |
691a2d25 |
1366 | $attvalue = sq_deent($attvalue); |
2dd879b8 |
1367 | $attvalue = sq_unspace($attvalue); |
691a2d25 |
1368 | |
1369 | /** |
1370 | * Now let's run checks on the attvalues. |
1371 | * I don't expect anyone to comprehend this. If you do, |
1372 | * get in touch with me so I can drive to where you live and |
1373 | * shake your hand personally. :) |
1374 | */ |
1375 | foreach ($bad_attvals as $matchtag=>$matchattrs){ |
1376 | if (preg_match($matchtag, $tagname)){ |
1377 | foreach ($matchattrs as $matchattr=>$valary){ |
1378 | if (preg_match($matchattr, $attname)){ |
1379 | /** |
1380 | * There are two arrays in valary. |
1381 | * First is matches. |
1382 | * Second one is replacements |
1383 | */ |
1384 | list($valmatch, $valrepl) = $valary; |
da2415c1 |
1385 | $newvalue = |
691a2d25 |
1386 | preg_replace($valmatch, $valrepl, $attvalue); |
1387 | if ($newvalue != $attvalue){ |
1388 | $attary{$attname} = $newvalue; |
1389 | } |
1390 | } |
1391 | } |
451f74a2 |
1392 | } |
a3daaaf3 |
1393 | } |
691a2d25 |
1394 | /** |
1395 | * Turn cid: urls into http-friendly ones. |
1396 | */ |
1397 | if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){ |
b3af12ef |
1398 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
691a2d25 |
1399 | } |
a3daaaf3 |
1400 | } |
691a2d25 |
1401 | /** |
1402 | * See if we need to append any attributes to this tag. |
1403 | */ |
1404 | foreach ($add_attr_to_tag as $matchtag=>$addattary){ |
1405 | if (preg_match($matchtag, $tagname)){ |
1406 | $attary = array_merge($attary, $addattary); |
1407 | } |
1408 | } |
1409 | return $attary; |
451f74a2 |
1410 | } |
a3daaaf3 |
1411 | |
691a2d25 |
1412 | /** |
1413 | * This function edits the style definition to make them friendly and |
1414 | * usable in squirrelmail. |
da2415c1 |
1415 | * |
691a2d25 |
1416 | * @param $message the message object |
1417 | * @param $id the message id |
1418 | * @param $content a string with whatever is between <style> and </style> |
e60a299a |
1419 | * @param $mailbox the message mailbox |
691a2d25 |
1420 | * @return a string with edited content. |
1421 | */ |
e60a299a |
1422 | function sq_fixstyle($body, $pos, $message, $id, $mailbox){ |
691a2d25 |
1423 | global $view_unsafe_images; |
b583c3e8 |
1424 | $me = 'sq_fixstyle'; |
7d06541f |
1425 | $ret = sq_findnxreg($body, $pos, '</\s*style\s*>'); |
1426 | if ($ret == FALSE){ |
1427 | return array(FALSE, strlen($body)); |
1428 | } |
1429 | $newpos = $ret[0] + strlen($ret[2]); |
1430 | $content = $ret[1]; |
691a2d25 |
1431 | /** |
1432 | * First look for general BODY style declaration, which would be |
1433 | * like so: |
1434 | * body {background: blah-blah} |
1435 | * and change it to .bodyclass so we can just assign it to a <div> |
1436 | */ |
1437 | $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content); |
3d8371be |
1438 | $secremoveimg = '../images/' . _("sec_remove_eng.png"); |
691a2d25 |
1439 | /** |
1440 | * Fix url('blah') declarations. |
1441 | */ |
7d06541f |
1442 | $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si", |
691a2d25 |
1443 | "url(\\1$secremoveimg\\2)", $content); |
1444 | /** |
1445 | * Fix url('https*://.*) declarations but only if $view_unsafe_images |
1446 | * is false. |
1447 | */ |
1448 | if (!$view_unsafe_images){ |
7d06541f |
1449 | $content = preg_replace("|url\s*\(\s*([\'\"])\s*https*:.*?([\'\"])\s*\)|si", |
691a2d25 |
1450 | "url(\\1$secremoveimg\\2)", $content); |
1451 | } |
da2415c1 |
1452 | |
691a2d25 |
1453 | /** |
1454 | * Fix urls that refer to cid: |
1455 | */ |
da2415c1 |
1456 | while (preg_match("|url\s*\(\s*([\'\"]\s*cid:.*?[\'\"])\s*\)|si", |
7d06541f |
1457 | $content, $matches)){ |
691a2d25 |
1458 | $cidurl = $matches{1}; |
e60a299a |
1459 | $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox); |
7d06541f |
1460 | $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si", |
691a2d25 |
1461 | "url($httpurl)", $content); |
1462 | } |
a3daaaf3 |
1463 | |
691a2d25 |
1464 | /** |
bb8d0799 |
1465 | * Fix stupid css declarations which lead to vulnerabilities |
691a2d25 |
1466 | * in IE. |
1467 | */ |
f83c60a2 |
1468 | $match = Array('/expression/i', |
1469 | '/behaviou*r/i', |
2dd879b8 |
1470 | '/binding/i', |
1471 | '/include-source/i'); |
1472 | $replace = Array('idiocy', 'idiocy', 'idiocy', 'idiocy'); |
bb8d0799 |
1473 | $content = preg_replace($match, $replace, $content); |
7d06541f |
1474 | return array($content, $newpos); |
691a2d25 |
1475 | } |
a3daaaf3 |
1476 | |
691a2d25 |
1477 | /** |
1478 | * This function converts cid: url's into the ones that can be viewed in |
1479 | * the browser. |
1480 | * |
1481 | * @param $message the message object |
1482 | * @param $id the message id |
1483 | * @param $cidurl the cid: url. |
e60a299a |
1484 | * @param $mailbox the message mailbox |
691a2d25 |
1485 | * @return a string with a http-friendly url |
1486 | */ |
b3af12ef |
1487 | function sq_cid2http($message, $id, $cidurl, $mailbox){ |
691a2d25 |
1488 | /** |
1489 | * Get rid of quotes. |
1490 | */ |
1491 | $quotchar = substr($cidurl, 0, 1); |
2dd879b8 |
1492 | if ($quotchar == '"' || $quotchar == "'"){ |
1493 | $cidurl = str_replace($quotchar, "", $cidurl); |
1494 | } else { |
1495 | $quotchar = ''; |
1496 | } |
691a2d25 |
1497 | $cidurl = substr(trim($cidurl), 4); |
e5e9381a |
1498 | $linkurl = find_ent_id($cidurl, $message); |
1499 | /* in case of non-save cid links $httpurl should be replaced by a sort of |
1500 | unsave link image */ |
1501 | $httpurl = ''; |
1502 | if ($linkurl) { |
6b04287c |
1503 | $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&' . |
e5e9381a |
1504 | "passed_id=$id&mailbox=" . urlencode($mailbox) . |
3d8371be |
1505 | '&ent_id=' . $linkurl . $quotchar; |
e5e9381a |
1506 | } |
691a2d25 |
1507 | return $httpurl; |
1508 | } |
1509 | |
1510 | /** |
1511 | * This function changes the <body> tag into a <div> tag since we |
1512 | * can't really have a body-within-body. |
1513 | * |
2dd879b8 |
1514 | * @param $attary an array of attributes and values of <body> |
1515 | * @param $mailbox mailbox we're currently reading (for cid2http) |
1516 | * @param $message current message (for cid2http) |
1517 | * @param $id current message id (for cid2http) |
1518 | * @return a modified array of attributes to be set for <div> |
691a2d25 |
1519 | */ |
2dd879b8 |
1520 | function sq_body2div($attary, $mailbox, $message, $id){ |
b583c3e8 |
1521 | $me = 'sq_body2div'; |
3d8371be |
1522 | $divattary = Array('class' => "'bodyclass'"); |
b583c3e8 |
1523 | $bgcolor = '#ffffff'; |
1524 | $text = '#000000'; |
80b4debd |
1525 | $has_bgc_stl = $has_txt_stl = false; |
b583c3e8 |
1526 | $styledef = ''; |
691a2d25 |
1527 | if (is_array($attary) && sizeof($attary) > 0){ |
1528 | foreach ($attary as $attname=>$attvalue){ |
1529 | $quotchar = substr($attvalue, 0, 1); |
1530 | $attvalue = str_replace($quotchar, "", $attvalue); |
1531 | switch ($attname){ |
3d8371be |
1532 | case 'background': |
da2415c1 |
1533 | $attvalue = sq_cid2http($message, $id, |
2dd879b8 |
1534 | $attvalue, $mailbox); |
3d8371be |
1535 | $styledef .= "background-image: url('$attvalue'); "; |
1536 | break; |
1537 | case 'bgcolor': |
80b4debd |
1538 | $has_bgc_stl = true; |
3d8371be |
1539 | $styledef .= "background-color: $attvalue; "; |
1540 | break; |
1541 | case 'text': |
80b4debd |
1542 | $has_txt_stl = true; |
3d8371be |
1543 | $styledef .= "color: $attvalue; "; |
1544 | break; |
691a2d25 |
1545 | } |
a3daaaf3 |
1546 | } |
80b4debd |
1547 | // Outlook defines a white bgcolor and no text color. This can lead to |
1548 | // white text on a white bg with certain themes. |
1549 | if ($has_bgc_stl && !$has_txt_stl) { |
1550 | $styledef .= "color: $text; "; |
1551 | } |
691a2d25 |
1552 | if (strlen($styledef) > 0){ |
1553 | $divattary{"style"} = "\"$styledef\""; |
1554 | } |
1555 | } |
1556 | return $divattary; |
1557 | } |
a3daaaf3 |
1558 | |
691a2d25 |
1559 | /** |
1560 | * This is the main function and the one you should actually be calling. |
1561 | * There are several variables you should be aware of an which need |
1562 | * special description. |
1563 | * |
1564 | * Since the description is quite lengthy, see it here: |
1565 | * http://www.mricon.com/html/phpfilter.html |
1566 | * |
1567 | * @param $body the string with HTML you wish to filter |
1568 | * @param $tag_list see description above |
1569 | * @param $rm_tags_with_content see description above |
1570 | * @param $self_closing_tags see description above |
1571 | * @param $force_tag_closing see description above |
1572 | * @param $rm_attnames see description above |
1573 | * @param $bad_attvals see description above |
1574 | * @param $add_attr_to_tag see description above |
1575 | * @param $message message object |
1576 | * @param $id message id |
1577 | * @return sanitized html safe to show on your pages. |
1578 | */ |
da2415c1 |
1579 | function sq_sanitize($body, |
1580 | $tag_list, |
691a2d25 |
1581 | $rm_tags_with_content, |
1582 | $self_closing_tags, |
1583 | $force_tag_closing, |
1584 | $rm_attnames, |
1585 | $bad_attvals, |
1586 | $add_attr_to_tag, |
1587 | $message, |
b3af12ef |
1588 | $id, |
3d8371be |
1589 | $mailbox |
691a2d25 |
1590 | ){ |
b583c3e8 |
1591 | $me = 'sq_sanitize'; |
7d06541f |
1592 | $rm_tags = array_shift($tag_list); |
691a2d25 |
1593 | /** |
1594 | * Normalize rm_tags and rm_tags_with_content. |
1595 | */ |
7d06541f |
1596 | @array_walk($tag_list, 'sq_casenormalize'); |
691a2d25 |
1597 | @array_walk($rm_tags_with_content, 'sq_casenormalize'); |
1598 | @array_walk($self_closing_tags, 'sq_casenormalize'); |
1599 | /** |
1600 | * See if tag_list is of tags to remove or tags to allow. |
1601 | * false means remove these tags |
1602 | * true means allow these tags |
1603 | */ |
691a2d25 |
1604 | $curpos = 0; |
1605 | $open_tags = Array(); |
2dd879b8 |
1606 | $trusted = "\n<!-- begin sanitized html -->\n"; |
691a2d25 |
1607 | $skip_content = false; |
bb8d0799 |
1608 | /** |
1609 | * Take care of netscape's stupid javascript entities like |
1610 | * &{alert('boo')}; |
1611 | */ |
1612 | $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body); |
691a2d25 |
1613 | |
7d06541f |
1614 | while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){ |
691a2d25 |
1615 | list($tagname, $attary, $tagtype, $lt, $gt) = $curtag; |
1616 | $free_content = substr($body, $curpos, $lt-$curpos); |
1617 | /** |
1618 | * Take care of <style> |
1619 | */ |
7d06541f |
1620 | if ($tagname == "style" && $tagtype == 1){ |
da2415c1 |
1621 | list($free_content, $curpos) = |
e60a299a |
1622 | sq_fixstyle($body, $gt+1, $message, $id, $mailbox); |
7d06541f |
1623 | if ($free_content != FALSE){ |
1624 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
1625 | $trusted .= $free_content; |
1626 | $trusted .= sq_tagprint($tagname, false, 2); |
1627 | } |
1628 | continue; |
691a2d25 |
1629 | } |
1630 | if ($skip_content == false){ |
1631 | $trusted .= $free_content; |
691a2d25 |
1632 | } |
1633 | if ($tagname != FALSE){ |
1634 | if ($tagtype == 2){ |
1635 | if ($skip_content == $tagname){ |
1636 | /** |
1637 | * Got to the end of tag we needed to remove. |
1638 | */ |
1639 | $tagname = false; |
1640 | $skip_content = false; |
1641 | } else { |
1642 | if ($skip_content == false){ |
c828931c |
1643 | if ($tagname == "body"){ |
1644 | $tagname = "div"; |
2dd879b8 |
1645 | } |
da2415c1 |
1646 | if (isset($open_tags{$tagname}) && |
2dd879b8 |
1647 | $open_tags{$tagname} > 0){ |
1648 | $open_tags{$tagname}--; |
691a2d25 |
1649 | } else { |
2dd879b8 |
1650 | $tagname = false; |
691a2d25 |
1651 | } |
691a2d25 |
1652 | } |
1653 | } |
1654 | } else { |
1655 | /** |
1656 | * $rm_tags_with_content |
1657 | */ |
1658 | if ($skip_content == false){ |
1659 | /** |
1660 | * See if this is a self-closing type and change |
1661 | * tagtype appropriately. |
1662 | */ |
1663 | if ($tagtype == 1 |
1664 | && in_array($tagname, $self_closing_tags)){ |
2dd879b8 |
1665 | $tagtype = 3; |
691a2d25 |
1666 | } |
1667 | /** |
1668 | * See if we should skip this tag and any content |
1669 | * inside it. |
1670 | */ |
1671 | if ($tagtype == 1 && |
1672 | in_array($tagname, $rm_tags_with_content)){ |
1673 | $skip_content = $tagname; |
1674 | } else { |
da2415c1 |
1675 | if (($rm_tags == false |
691a2d25 |
1676 | && in_array($tagname, $tag_list)) || |
1677 | ($rm_tags == true && |
1678 | !in_array($tagname, $tag_list))){ |
1679 | $tagname = false; |
1680 | } else { |
2dd879b8 |
1681 | /** |
1682 | * Convert body into div. |
1683 | */ |
1684 | if ($tagname == "body"){ |
1685 | $tagname = "div"; |
da2415c1 |
1686 | $attary = sq_body2div($attary, $mailbox, |
2dd879b8 |
1687 | $message, $id); |
1688 | } |
691a2d25 |
1689 | if ($tagtype == 1){ |
1690 | if (isset($open_tags{$tagname})){ |
1691 | $open_tags{$tagname}++; |
1692 | } else { |
1693 | $open_tags{$tagname}=1; |
1694 | } |
1695 | } |
1696 | /** |
1697 | * This is where we run other checks. |
1698 | */ |
1699 | if (is_array($attary) && sizeof($attary) > 0){ |
1700 | $attary = sq_fixatts($tagname, |
1701 | $attary, |
1702 | $rm_attnames, |
1703 | $bad_attvals, |
1704 | $add_attr_to_tag, |
1705 | $message, |
b3af12ef |
1706 | $id, |
3d8371be |
1707 | $mailbox |
691a2d25 |
1708 | ); |
1709 | } |
1710 | } |
1711 | } |
691a2d25 |
1712 | } |
1713 | } |
1714 | if ($tagname != false && $skip_content == false){ |
1715 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
1716 | } |
691a2d25 |
1717 | } |
1718 | $curpos = $gt+1; |
a3daaaf3 |
1719 | } |
691a2d25 |
1720 | $trusted .= substr($body, $curpos, strlen($body)-$curpos); |
1721 | if ($force_tag_closing == true){ |
1722 | foreach ($open_tags as $tagname=>$opentimes){ |
1723 | while ($opentimes > 0){ |
1724 | $trusted .= '</' . $tagname . '>'; |
1725 | $opentimes--; |
1726 | } |
1727 | } |
1728 | $trusted .= "\n"; |
1729 | } |
1730 | $trusted .= "<!-- end sanitized html -->\n"; |
1731 | return $trusted; |
1732 | } |
451f74a2 |
1733 | |
691a2d25 |
1734 | /** |
1735 | * This is a wrapper function to call html sanitizing routines. |
1736 | * |
1737 | * @param $body the body of the message |
1738 | * @param $id the id of the message |
1739 | * @return a string with html safe to display in the browser. |
1740 | */ |
7aad7b77 |
1741 | function magicHTML($body, $id, $message, $mailbox = 'INBOX') { |
691a2d25 |
1742 | global $attachment_common_show_images, $view_unsafe_images, |
3d8371be |
1743 | $has_unsafe_images; |
691a2d25 |
1744 | /** |
1745 | * Don't display attached images in HTML mode. |
1746 | */ |
1747 | $attachment_common_show_images = false; |
1748 | $tag_list = Array( |
1749 | false, |
1750 | "object", |
1751 | "meta", |
1752 | "html", |
1753 | "head", |
cc34b00d |
1754 | "base", |
e18cb31b |
1755 | "link", |
45071bd6 |
1756 | "frame", |
2dd879b8 |
1757 | "iframe", |
1758 | "plaintext", |
1759 | "marquee" |
691a2d25 |
1760 | ); |
1761 | |
1762 | $rm_tags_with_content = Array( |
1763 | "script", |
1764 | "applet", |
1765 | "embed", |
2dd879b8 |
1766 | "title", |
1767 | "frameset", |
1768 | "xml" |
691a2d25 |
1769 | ); |
1770 | |
1771 | $self_closing_tags = Array( |
1772 | "img", |
1773 | "br", |
1774 | "hr", |
1775 | "input" |
1776 | ); |
1777 | |
2dd879b8 |
1778 | $force_tag_closing = true; |
691a2d25 |
1779 | |
1780 | $rm_attnames = Array( |
1781 | "/.*/" => |
1782 | Array( |
3a50c8d2 |
1783 | "/target/i", |
1784 | "/^on.*/i", |
1785 | "/^dynsrc/i", |
1786 | "/^data.*/i", |
1787 | "/^lowsrc.*/i" |
691a2d25 |
1788 | ) |
1789 | ); |
1790 | |
1791 | $secremoveimg = "../images/" . _("sec_remove_eng.png"); |
1792 | $bad_attvals = Array( |
1793 | "/.*/" => |
1794 | Array( |
0a6ec9b5 |
1795 | "/^src|background/i" => |
691a2d25 |
1796 | Array( |
1797 | Array( |
f83c60a2 |
1798 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
1799 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
1800 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
691a2d25 |
1801 | ), |
1802 | Array( |
1803 | "\\1$secremoveimg\\2", |
bb8d0799 |
1804 | "\\1$secremoveimg\\2", |
3d8371be |
1805 | "\\1$secremoveimg\\2", |
1806 | "\\1$secremoveimg\\2" |
691a2d25 |
1807 | ) |
1808 | ), |
0a6ec9b5 |
1809 | "/^href|action/i" => |
1810 | Array( |
1811 | Array( |
f83c60a2 |
1812 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
1813 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
1814 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
0a6ec9b5 |
1815 | ), |
1816 | Array( |
2dd879b8 |
1817 | "\\1#\\1", |
1818 | "\\1#\\1", |
1819 | "\\1#\\1", |
1820 | "\\1#\\1" |
0a6ec9b5 |
1821 | ) |
1822 | ), |
f83c60a2 |
1823 | "/^style/i" => |
691a2d25 |
1824 | Array( |
1825 | Array( |
f83c60a2 |
1826 | "/expression/i", |
1827 | "/binding/i", |
1828 | "/behaviou*r/i", |
2dd879b8 |
1829 | "/include-source/i", |
7d06541f |
1830 | "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si", |
1831 | "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si", |
1832 | "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si" |
691a2d25 |
1833 | ), |
1834 | Array( |
bb8d0799 |
1835 | "idiocy", |
3d8371be |
1836 | "idiocy", |
1837 | "idiocy", |
2dd879b8 |
1838 | "idiocy", |
1839 | "url(\\1#\\1)", |
1840 | "url(\\1#\\1)", |
1841 | "url(\\1#\\1)", |
1842 | "url(\\1#\\1)" |
691a2d25 |
1843 | ) |
1844 | ) |
1845 | ) |
1846 | ); |
5262d9a6 |
1847 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
2dd879b8 |
1848 | $view_unsafe_images = false; |
45071bd6 |
1849 | } |
691a2d25 |
1850 | if (!$view_unsafe_images){ |
1851 | /** |
1852 | * Remove any references to http/https if view_unsafe_images set |
1853 | * to false. |
1854 | */ |
0a6ec9b5 |
1855 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0], |
f83c60a2 |
1856 | '/^([\'\"])\s*https*:.*([\'\"])/si'); |
0a6ec9b5 |
1857 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1], |
2dd879b8 |
1858 | "\\1$secremoveimg\\1"); |
3a50c8d2 |
1859 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[0], |
43d678b6 |
1860 | '/url\(([\'\"])\s*https*:.*([\'\"])\)/si'); |
3a50c8d2 |
1861 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[1], |
2dd879b8 |
1862 | "url(\\1$secremoveimg\\1)"); |
691a2d25 |
1863 | } |
451f74a2 |
1864 | |
691a2d25 |
1865 | $add_attr_to_tag = Array( |
da2415c1 |
1866 | "/^a$/i" => |
2dd879b8 |
1867 | Array('target'=>'"_new"', |
1868 | 'title'=>'"'._("This external link will open in a new window").'"' |
1869 | ) |
1870 | ); |
da2415c1 |
1871 | $trusted = sq_sanitize($body, |
1872 | $tag_list, |
691a2d25 |
1873 | $rm_tags_with_content, |
1874 | $self_closing_tags, |
1875 | $force_tag_closing, |
1876 | $rm_attnames, |
1877 | $bad_attvals, |
1878 | $add_attr_to_tag, |
1879 | $message, |
b3af12ef |
1880 | $id, |
3d8371be |
1881 | $mailbox |
691a2d25 |
1882 | ); |
f83c60a2 |
1883 | if (preg_match("|$secremoveimg|i", $trusted)){ |
691a2d25 |
1884 | $has_unsafe_images = true; |
da2415c1 |
1885 | } |
691a2d25 |
1886 | return $trusted; |
451f74a2 |
1887 | } |
a4a70693 |
1888 | |
da2415c1 |
1889 | /** |
1890 | * function SendDownloadHeaders - send file to the browser |
1891 | * |
1892 | * Original Source: SM core src/download.php |
1893 | * moved here to make it available to other code, and separate |
1894 | * front end from back end functionality. |
1895 | * |
1896 | * @param string $type0 first half of mime type |
1897 | * @param string $type1 second half of mime type |
1898 | * @param string $filename filename to tell the browser for downloaded file |
1899 | * @param boolean $force whether to force the download dialog to pop |
07c49f57 |
1900 | * @param integer $filesize optional, send the Content-Header and length to the browser |
da2415c1 |
1901 | * @return void |
1902 | */ |
07c49f57 |
1903 | function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) { |
da2415c1 |
1904 | global $languages, $squirrelmail_language; |
1905 | $isIE = $isIE6 = 0; |
1906 | |
1907 | sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER); |
1908 | |
1909 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false && |
1910 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
1911 | $isIE = 1; |
1912 | } |
1913 | |
1914 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE 6') !== false && |
1915 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
1916 | $isIE6 = 1; |
1917 | } |
1918 | |
1919 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
1920 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) { |
1921 | $filename = |
1922 | $languages[$squirrelmail_language]['XTRA_CODE']('downloadfilename', $filename, $HTTP_USER_AGENT); |
1923 | } else { |
1924 | $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename)); |
1925 | } |
1926 | |
1927 | // A Pox on Microsoft and it's Office! |
1928 | if (!$force) { |
1929 | // Try to show in browser window |
1930 | header("Content-Disposition: inline; filename=\"$filename\""); |
1931 | header("Content-Type: $type0/$type1; name=\"$filename\""); |
1932 | } else { |
1933 | // Try to pop up the "save as" box |
1934 | // IE makes this hard. It pops up 2 save boxes, or none. |
1935 | // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP |
1936 | // But, accordint to Microsoft, it is "RFC compliant but doesn't |
1937 | // take into account some deviations that allowed within the |
1938 | // specification." Doesn't that mean RFC non-compliant? |
1939 | // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP |
1940 | // |
1941 | // The best thing you can do for IE is to upgrade to the latest |
1942 | // version |
1943 | if ($isIE && !$isIE6) { |
1944 | // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp |
1945 | // Do not have quotes around filename, but that applied to |
1946 | // "attachment"... does it apply to inline too? |
1947 | // |
1948 | // This combination seems to work mostly. IE 5.5 SP 1 has |
1949 | // known issues (see the Microsoft Knowledge Base) |
1950 | header("Content-Disposition: inline; filename=$filename"); |
1951 | // This works for most types, but doesn't work with Word files |
1952 | header("Content-Type: application/download; name=\"$filename\""); |
1953 | |
1954 | // These are spares, just in case. :-) |
1955 | //header("Content-Type: $type0/$type1; name=\"$filename\""); |
1956 | //header("Content-Type: application/x-msdownload; name=\"$filename\""); |
1957 | //header("Content-Type: application/octet-stream; name=\"$filename\""); |
1958 | } else { |
1959 | header("Content-Disposition: attachment; filename=\"$filename\""); |
1960 | // application/octet-stream forces download for Netscape |
1961 | header("Content-Type: application/octet-stream; name=\"$filename\""); |
1962 | } |
1963 | } |
07c49f57 |
1964 | |
1965 | //send the content-length header if the calling function provides it |
1966 | if ($filesize > 0) { |
1967 | header("Content-Length: $filesize"); |
1968 | } |
1969 | |
1970 | } // end fn SendDownlaodHeaders |
da2415c1 |
1971 | |
d6c32258 |
1972 | ?> |