| 59177427 |
1 | <?php |
| 2ba13803 |
2 | |
| 35586184 |
3 | /** |
| 02474e43 |
4 | * mime.php |
| 5 | * |
| 6 | * Copyright (c) 1999-2005 The SquirrelMail Project Team |
| 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 | * @version $Id$ |
| 13 | * @package squirrelmail |
| 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 | /** |
| 02474e43 |
25 | * Get the MIME structure |
| 26 | * |
| 27 | * This function gets the structure of a message and stores it in the "message" class. |
| 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"); |
| 3c621ba1 |
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, |
| 02474e43 |
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, $fetch_size=0) { |
| 3d8371be |
100 | /* Do a bit of error correction. If we couldn't find the entity id, just guess |
| 02474e43 |
101 | * that it is the first one. That is usually the case anyway. |
| 102 | */ |
| 7c7b74b3 |
103 | |
| 09a4bde3 |
104 | if (!$ent_id) { |
| 08b7f7cc |
105 | $cmd = "FETCH $id BODY[]"; |
| 1035e159 |
106 | } else { |
| 08b7f7cc |
107 | $cmd = "FETCH $id BODY[$ent_id]"; |
| 09a4bde3 |
108 | } |
| 3d8371be |
109 | |
| 4d592352 |
110 | if ($fetch_size!=0) $cmd .= "<0.$fetch_size>"; |
| da2415c1 |
111 | |
| 6201339c |
112 | $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, TRUE); |
| 77b88425 |
113 | do { |
| 3d8371be |
114 | $topline = trim(array_shift($data)); |
| 115 | } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ; |
| a4a70693 |
116 | |
| 451f74a2 |
117 | $wholemessage = implode('', $data); |
| 118 | if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) { |
| 3d8371be |
119 | $ret = substr($wholemessage, 0, $regs[1]); |
| 120 | /* There is some information in the content info header that could be important |
| 02474e43 |
121 | * in order to parse html messages. Let's get them here. |
| 122 | */ |
| 0600bdf1 |
123 | // if ($ret{0} == '<') { |
| 6201339c |
124 | // $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, TRUE); |
| 0600bdf1 |
125 | // } |
| 451f74a2 |
126 | } else if (ereg('"([^"]*)"', $topline, $regs)) { |
| 127 | $ret = $regs[1]; |
| 128 | } else { |
| 129 | global $where, $what, $mailbox, $passed_id, $startMessage; |
| 3d8371be |
130 | $par = 'mailbox=' . urlencode($mailbox) . '&passed_id=' . $passed_id; |
| 451f74a2 |
131 | if (isset($where) && isset($what)) { |
| 3d8371be |
132 | $par .= '&where=' . urlencode($where) . '&what=' . urlencode($what); |
| a3daaaf3 |
133 | } else { |
| 3d8371be |
134 | $par .= '&startMessage=' . $startMessage . '&show_more=0'; |
| 451f74a2 |
135 | } |
| e5ea9327 |
136 | $par .= '&response=' . urlencode($response) . |
| 3d8371be |
137 | '&message=' . urlencode($message) . |
| 138 | '&topline=' . urlencode($topline); |
| a019eeb8 |
139 | |
| 3c621ba1 |
140 | echo '<tt><br />' . |
| 02474e43 |
141 | '<table width="80%"><tr>' . |
| 142 | '<tr><td colspan="2">' . |
| 143 | _("Body retrieval error. The reason for this is most probably that the message is malformed.") . |
| 144 | '</td></tr>' . |
| 145 | '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" . |
| 146 | '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" . |
| 147 | '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" . |
| 148 | '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" . |
| 149 | "</table><br /></tt></font><hr />"; |
| 346817d4 |
150 | |
| 6201339c |
151 | $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, TRUE); |
| 451f74a2 |
152 | array_shift($data); |
| 153 | $wholemessage = implode('', $data); |
| a019eeb8 |
154 | |
| 346817d4 |
155 | $ret = $wholemessage; |
| a3daaaf3 |
156 | } |
| 3d8371be |
157 | return $ret; |
| 451f74a2 |
158 | } |
| d4467150 |
159 | |
| 1035e159 |
160 | function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding) { |
| 1035e159 |
161 | |
| 3d8371be |
162 | /* Don't kill the connection if the browser is over a dialup |
| 02474e43 |
163 | * and it would take over 30 seconds to download it. |
| 164 | * Don't call set_time_limit in safe mode. |
| 165 | */ |
| b7206e1d |
166 | |
| 3d8371be |
167 | if (!ini_get('safe_mode')) { |
| b7206e1d |
168 | set_time_limit(0); |
| 169 | } |
| 7c7b74b3 |
170 | /* in case of base64 encoded attachments, do not buffer them. |
| 02474e43 |
171 | Instead, echo the decoded attachment directly to screen */ |
| 7c7b74b3 |
172 | if (strtolower($encoding) == 'base64') { |
| 173 | if (!$ent_id) { |
| 02474e43 |
174 | $query = "FETCH $id BODY[]"; |
| 7c7b74b3 |
175 | } else { |
| 02474e43 |
176 | $query = "FETCH $id BODY[$ent_id]"; |
| 7c7b74b3 |
177 | } |
| 6201339c |
178 | sqimap_run_command($imap_stream,$query,true,$response,$message,TRUE,'sqimap_base64_decode','php://stdout',true); |
| 1d142b8d |
179 | } else { |
| 02474e43 |
180 | $body = mime_fetch_body ($imap_stream, $id, $ent_id); |
| 181 | echo decodeBody($body, $encoding); |
| 1d142b8d |
182 | } |
| 346817d4 |
183 | |
| da2415c1 |
184 | /* |
| 02474e43 |
185 | TODO, use the same method for quoted printable. |
| 186 | However, I assume that quoted printable attachments aren't that large |
| 187 | so the performancegain / memory usage drop will be minimal. |
| 188 | If we decide to add that then we need to adapt sqimap_fread because |
| 189 | we need to split te result on \n and fread doesn't stop at \n. That |
| 190 | means we also should provide $results from sqimap_fread (by ref) to |
| 191 | te function and set $no_return to false. The $filter function for |
| 192 | quoted printable should handle unsetting of $results. |
| 7c7b74b3 |
193 | */ |
| da2415c1 |
194 | /* |
| 02474e43 |
195 | TODO 2: find out how we write to the output stream php://stdout. fwrite |
| 196 | doesn't work because 'php://stdout isn't a stream. |
| 7c7b74b3 |
197 | */ |
| 198 | |
| 5d9c6f73 |
199 | return; |
| 200 | /* |
| 451f74a2 |
201 | fputs ($imap_stream, "$sid FETCH $id BODY[$ent_id]\r\n"); |
| 202 | $cnt = 0; |
| 203 | $continue = true; |
| 1d142b8d |
204 | $read = fgets ($imap_stream,8192); |
| 205 | |
| 206 | |
| 451f74a2 |
207 | // This could be bad -- if the section has sqimap_session_id() . ' OK' |
| 208 | // or similar, it will kill the download. |
| 1d142b8d |
209 | while (!ereg("^".$sid_s." (OK|BAD|NO)(.*)$", $read, $regs)) { |
| 5d9c6f73 |
210 | if (trim($read) == ')==') { |
| 211 | $read1 = $read; |
| 212 | $read = fgets ($imap_stream,4096); |
| 213 | if (ereg("^".$sid." (OK|BAD|NO)(.*)$", $read, $regs)) { |
| 214 | return; |
| 215 | } else { |
| 216 | echo decodeBody($read1, $encoding) . |
| 02474e43 |
217 | decodeBody($read, $encoding); |
| 5d9c6f73 |
218 | } |
| 219 | } else if ($cnt) { |
| 220 | echo decodeBody($read, $encoding); |
| 221 | } |
| 222 | $read = fgets ($imap_stream,4096); |
| 223 | $cnt++; |
| 1d142b8d |
224 | // break; |
| 451f74a2 |
225 | } |
| 5d9c6f73 |
226 | */ |
| 451f74a2 |
227 | } |
| beb9e459 |
228 | |
| 451f74a2 |
229 | /* -[ END MIME DECODING ]----------------------------------------------------------- */ |
| d4467150 |
230 | |
| 3d8371be |
231 | /* This is here for debugging purposes. It will print out a list |
| 02474e43 |
232 | * of all the entity IDs that are in the $message object. |
| 233 | */ |
| 451f74a2 |
234 | function listEntities ($message) { |
| 3d8371be |
235 | if ($message) { |
| 3c621ba1 |
236 | echo "<tt>" . $message->entity_id . ' : ' . $message->type0 . '/' . $message->type1 . ' parent = '. $message->parent->entity_id. '<br />'; |
| 3d8371be |
237 | for ($i = 0; isset($message->entities[$i]); $i++) { |
| 238 | echo "$i : "; |
| 239 | $msg = listEntities($message->entities[$i]); |
| 240 | |
| 241 | if ($msg) { |
| 242 | echo "return: "; |
| 243 | return $msg; |
| 244 | } |
| 245 | } |
| a4a70693 |
246 | } |
| 451f74a2 |
247 | } |
| f0c4dc12 |
248 | |
| f792c641 |
249 | function getPriorityStr($priority) { |
| 3d8371be |
250 | $priority_level = substr($priority,0,1); |
| 251 | |
| 252 | switch($priority_level) { |
| 253 | /* Check for a higher then normal priority. */ |
| 254 | case '1': |
| 255 | case '2': |
| 256 | $priority_string = _("High"); |
| 257 | break; |
| 258 | |
| 259 | /* Check for a lower then normal priority. */ |
| 260 | case '4': |
| 261 | case '5': |
| 262 | $priority_string = _("Low"); |
| 263 | break; |
| 264 | |
| 265 | /* Check for a normal priority. */ |
| 266 | case '3': |
| 267 | default: |
| 268 | $priority_level = '3'; |
| 269 | $priority_string = _("Normal"); |
| 270 | break; |
| 271 | |
| 272 | } |
| 273 | return $priority_string; |
| f792c641 |
274 | } |
| 275 | |
| 451f74a2 |
276 | /* returns a $message object for a particular entity id */ |
| 277 | function getEntity ($message, $ent_id) { |
| a4a70693 |
278 | return $message->getEntity($ent_id); |
| 451f74a2 |
279 | } |
| 8beafbbc |
280 | |
| 3d8371be |
281 | /* translateText |
| 02474e43 |
282 | * Extracted from strings.php 23/03/2002 |
| 283 | */ |
| da4c66e8 |
284 | |
| 285 | function translateText(&$body, $wrap_at, $charset) { |
| 3d8371be |
286 | global $where, $what; /* from searching */ |
| 287 | global $color; /* color theme */ |
| da4c66e8 |
288 | |
| b68edc75 |
289 | require_once(SM_PATH . 'functions/url_parser.php'); |
| da4c66e8 |
290 | |
| 291 | $body_ary = explode("\n", $body); |
| da4c66e8 |
292 | for ($i=0; $i < count($body_ary); $i++) { |
| 293 | $line = $body_ary[$i]; |
| 294 | if (strlen($line) - 2 >= $wrap_at) { |
| c7aff938 |
295 | sqWordWrap($line, $wrap_at, $charset); |
| da4c66e8 |
296 | } |
| 297 | $line = charset_decode($charset, $line); |
| 298 | $line = str_replace("\t", ' ', $line); |
| 299 | |
| 300 | parseUrl ($line); |
| 301 | |
| 3d8371be |
302 | $quotes = 0; |
| da4c66e8 |
303 | $pos = 0; |
| 3d8371be |
304 | $j = strlen($line); |
| da4c66e8 |
305 | |
| 3d8371be |
306 | while ($pos < $j) { |
| da4c66e8 |
307 | if ($line[$pos] == ' ') { |
| 3d8371be |
308 | $pos++; |
| da4c66e8 |
309 | } else if (strpos($line, '>', $pos) === $pos) { |
| 310 | $pos += 4; |
| 3d8371be |
311 | $quotes++; |
| da4c66e8 |
312 | } else { |
| 313 | break; |
| 314 | } |
| 315 | } |
| 3d8371be |
316 | |
| 83c94382 |
317 | if ($quotes % 2) { |
| 3d8371be |
318 | if (!isset($color[13])) { |
| da4c66e8 |
319 | $color[13] = '#800000'; |
| 320 | } |
| 4c25967c |
321 | $line = '<font color="' . $color[13] . '">' . $line . '</font>'; |
| 322 | } elseif ($quotes) { |
| 323 | if (!isset($color[14])) { |
| 324 | $color[14] = '#FF0000'; |
| 325 | } |
| 326 | $line = '<font color="' . $color[14] . '">' . $line . '</font>'; |
| da4c66e8 |
327 | } |
| 3d8371be |
328 | |
| da4c66e8 |
329 | $body_ary[$i] = $line; |
| 330 | } |
| 331 | $body = '<pre>' . implode("\n", $body_ary) . '</pre>'; |
| 332 | } |
| 333 | |
| a2bfcbce |
334 | /** |
| 02474e43 |
335 | * This returns a parsed string called $body. That string can then |
| 336 | * be displayed as the actual message in the HTML. It contains |
| 337 | * everything needed, including HTML Tags, Attachments at the |
| 338 | * bottom, etc. |
| 339 | * @param clean Do not output stuff that's irrelevant for the printable version. |
| 340 | */ |
| a2bfcbce |
341 | function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) { |
| 3d8371be |
342 | /* This if statement checks for the entity to show as the |
| 02474e43 |
343 | * primary message. To add more of them, just put them in the |
| 344 | * order that is their priority. |
| 345 | */ |
| ce68b76b |
346 | global $startMessage, $languages, $squirrelmail_language, |
| 02474e43 |
347 | $show_html_default, $sort, $has_unsafe_images, $passed_ent_id; |
| 77bfbd2e |
348 | |
| 5262d9a6 |
349 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
| 350 | $view_unsafe_images = false; |
| 77bfbd2e |
351 | } |
| d03c24f4 |
352 | |
| cc34b00d |
353 | $body = ''; |
| 23bcec6f |
354 | $urlmailbox = urlencode($mailbox); |
| 451f74a2 |
355 | $body_message = getEntity($message, $ent_num); |
| 356 | if (($body_message->header->type0 == 'text') || |
| 357 | ($body_message->header->type0 == 'rfc822')) { |
| 3d8371be |
358 | $body = mime_fetch_body ($imap_stream, $id, $ent_num); |
| 451f74a2 |
359 | $body = decodeBody($body, $body_message->header->encoding); |
| e842b215 |
360 | |
| 361 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| 33a55f5a |
362 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) { |
| e842b215 |
363 | if (mb_detect_encoding($body) != 'ASCII') { |
| 33a55f5a |
364 | $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body); |
| e842b215 |
365 | } |
| 366 | } |
| 451f74a2 |
367 | $hookResults = do_hook("message_body", $body); |
| 368 | $body = $hookResults[1]; |
| 23bcec6f |
369 | |
| 3d8371be |
370 | /* If there are other types that shouldn't be formatted, add |
| 02474e43 |
371 | * them here. |
| 372 | */ |
| 3d8371be |
373 | |
| 451f74a2 |
374 | if ($body_message->header->type1 == 'html') { |
| 3d8371be |
375 | if ($show_html_default <> 1) { |
| 85015544 |
376 | $entity_conv = array(' ' => ' ', |
| 02474e43 |
377 | '<p>' => "\n", |
| 378 | '<P>' => "\n", |
| 379 | '<br>' => "\n", |
| 380 | '<BR>' => "\n", |
| 381 | '<br />' => "\n", |
| 382 | '<BR />' => "\n", |
| 383 | '>' => '>', |
| 384 | '<' => '<'); |
| 85015544 |
385 | $body = strtr($body, $entity_conv); |
| 3d8371be |
386 | $body = strip_tags($body); |
| 85015544 |
387 | $body = trim($body); |
| 388 | translateText($body, $wrap_at, |
| 02474e43 |
389 | $body_message->header->getParameter('charset')); |
| a3daaaf3 |
390 | } else { |
| 3d8371be |
391 | $body = magicHTML($body, $id, $message, $mailbox); |
| a3daaaf3 |
392 | } |
| 451f74a2 |
393 | } else { |
| 3d8371be |
394 | translateText($body, $wrap_at, |
| 02474e43 |
395 | $body_message->header->getParameter('charset')); |
| 451f74a2 |
396 | } |
| a2bfcbce |
397 | |
| 398 | // if this is the clean display (i.e. printer friendly), stop here. |
| 399 | if ( $clean ) { |
| 400 | return $body; |
| 401 | } |
| 402 | |
| 83cf04bd |
403 | $link = 'passed_id=' . $id . '&ent_id='.$ent_num. |
| 08b7f7cc |
404 | '&mailbox=' . $urlmailbox .'&sort=' . $sort . |
| da2415c1 |
405 | '&startMessage=' . $startMessage . '&show_more=0'; |
| 08b7f7cc |
406 | if (isset($passed_ent_id)) { |
| 407 | $link .= '&passed_ent_id='.$passed_ent_id; |
| 408 | } |
| 8d863f64 |
409 | $body .= '<center><small><a href="download.php?absolute_dl=true&' . |
| 02474e43 |
410 | $link . '">' . _("Download this as a file") . '</a>'; |
| 7aad7b77 |
411 | if ($view_unsafe_images) { |
| 23f617b8 |
412 | $text = _("Hide Unsafe Images"); |
| 7aad7b77 |
413 | } else { |
| 08b7f7cc |
414 | if (isset($has_unsafe_images) && $has_unsafe_images) { |
| 415 | $link .= '&view_unsafe_images=1'; |
| 416 | $text = _("View Unsafe Images"); |
| 417 | } else { |
| 418 | $text = ''; |
| 419 | } |
| 3d8371be |
420 | } |
| 83cf04bd |
421 | if($text != '') { |
| 422 | $body .= ' | <a href="read_body.php?' . $link . '">' . $text . '</a>'; |
| 423 | } |
| 3c621ba1 |
424 | $body .= '</small></center><br />' . "\n"; |
| 3d8371be |
425 | } |
| 426 | return $body; |
| 451f74a2 |
427 | } |
| b74ba498 |
428 | |
| 23bcec6f |
429 | |
| 430 | function formatAttachments($message, $exclude_id, $mailbox, $id) { |
| 80dcd298 |
431 | global $where, $what, $startMessage, $color, $passed_ent_id; |
| 451f74a2 |
432 | |
| 23bcec6f |
433 | $att_ar = $message->getAttachments($exclude_id); |
| 451f74a2 |
434 | |
| 23bcec6f |
435 | if (!count($att_ar)) return ''; |
| 2e25760a |
436 | |
| 8f5425ec |
437 | $attachments = ''; |
| 2e25760a |
438 | |
| 23bcec6f |
439 | $urlMailbox = urlencode($mailbox); |
| 440 | |
| 441 | foreach ($att_ar as $att) { |
| fdc9d9b5 |
442 | $ent = $att->entity_id; |
| 2e25760a |
443 | $header = $att->header; |
| f0c4dc12 |
444 | $type0 = strtolower($header->type0); |
| 445 | $type1 = strtolower($header->type1); |
| 2e25760a |
446 | $name = ''; |
| 21dab2dc |
447 | $links['download link']['text'] = _("Download"); |
| 6b04287c |
448 | $links['download link']['href'] = SM_PATH . |
| 449 | "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent"; |
| 2e25760a |
450 | if ($type0 =='message' && $type1 == 'rfc822') { |
| 6b04287c |
451 | $default_page = SM_PATH . 'src/read_body.php'; |
| 2e25760a |
452 | $rfc822_header = $att->rfc822_header; |
| 098ea084 |
453 | $filename = $rfc822_header->subject; |
| 6cc08d8b |
454 | if (trim( $filename ) == '') { |
| 455 | $filename = 'untitled-[' . $ent . ']' ; |
| 08b7f7cc |
456 | } |
| 2e25760a |
457 | $from_o = $rfc822_header->from; |
| 458 | if (is_object($from_o)) { |
| 04ea844e |
459 | $from_name = decodeHeader($from_o->getAddress(false)); |
| 2e25760a |
460 | } else { |
| 461 | $from_name = _("Unknown sender"); |
| f0c4dc12 |
462 | } |
| 3d8371be |
463 | $description = $from_name; |
| 23bcec6f |
464 | } else { |
| 6b04287c |
465 | $default_page = SM_PATH . 'src/download.php'; |
| 02474e43 |
466 | $filename = $att->getFilename(); |
| f810c0b2 |
467 | if ($header->description) { |
| 098ea084 |
468 | $description = decodeHeader($header->description); |
| f810c0b2 |
469 | } else { |
| 3d8371be |
470 | $description = ''; |
| 471 | } |
| 2e25760a |
472 | } |
| 473 | |
| 474 | $display_filename = $filename; |
| 3d8371be |
475 | if (isset($passed_ent_id)) { |
| 476 | $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id; |
| 477 | } else { |
| 478 | $passed_ent_id_link = ''; |
| 479 | } |
| 480 | $defaultlink = $default_page . "?startMessage=$startMessage" |
| 02474e43 |
481 | . "&passed_id=$id&mailbox=$urlMailbox" |
| 482 | . '&ent_id='.$ent.$passed_ent_id_link; |
| 2e25760a |
483 | if ($where && $what) { |
| 02474e43 |
484 | $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what); |
| 2e25760a |
485 | } |
| 21dab2dc |
486 | |
| 3d8371be |
487 | /* This executes the attachment hook with a specific MIME-type. |
| 02474e43 |
488 | * If that doesn't have results, it tries if there's a rule |
| 489 | * for a more generic type. |
| 490 | */ |
| 3d8371be |
491 | $hookresults = do_hook("attachment $type0/$type1", $links, |
| 02474e43 |
492 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
| 493 | $display_filename, $where, $what); |
| 3d8371be |
494 | if(count($hookresults[1]) <= 1) { |
| 495 | $hookresults = do_hook("attachment $type0/*", $links, |
| 02474e43 |
496 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
| 497 | $display_filename, $where, $what); |
| 2e25760a |
498 | } |
| 451f74a2 |
499 | |
| 3d8371be |
500 | $links = $hookresults[1]; |
| 501 | $defaultlink = $hookresults[6]; |
| 77b88425 |
502 | |
| 3c621ba1 |
503 | $attachments .= '<tr><td>' . |
| 504 | '<a href="'.$defaultlink.'">'.decodeHeader($display_filename).'</a> </td>' . |
| 505 | '<td><small><b>' . show_readable_size($header->size) . |
| 506 | '</b> </small></td>' . |
| 507 | '<td><small>[ '.htmlspecialchars($type0).'/'.htmlspecialchars($type1).' ] </small></td>' . |
| 508 | '<td><small>'; |
| f810c0b2 |
509 | $attachments .= '<b>' . $description . '</b>'; |
| 3c621ba1 |
510 | $attachments .= '</small></td><td><small> '; |
| b74ba498 |
511 | |
| 3d8371be |
512 | $skipspaces = 1; |
| 513 | foreach ($links as $val) { |
| 514 | if ($skipspaces) { |
| 515 | $skipspaces = 0; |
| 2e25760a |
516 | } else { |
| 517 | $attachments .= ' | '; |
| 77b88425 |
518 | } |
| ebeb8348 |
519 | $attachments .= '<a href="' . $val['href'] . '">' . (isset($val['text']) && !empty($val['text']) ? $val['text'] : '') . (isset($val['extra']) && !empty($val['extra']) ? $val['extra'] : '') . '</a>'; |
| 2e25760a |
520 | } |
| 3d8371be |
521 | unset($links); |
| 3c621ba1 |
522 | $attachments .= "</td></tr>\n"; |
| 2e25760a |
523 | } |
| 9ad17edb |
524 | $attachmentadd = do_hook_function('attachments_bottom',$attachments); |
| 525 | if ($attachmentadd != '') |
| 526 | $attachments = $attachmentadd; |
| 2e25760a |
527 | return $attachments; |
| 451f74a2 |
528 | } |
| b74ba498 |
529 | |
| 7c7b74b3 |
530 | function sqimap_base64_decode(&$string) { |
| 7c0ec1d8 |
531 | |
| b17a8968 |
532 | // Base64 encoded data goes in pairs of 4 bytes. To achieve on the |
| 7c0ec1d8 |
533 | // fly decoding (to reduce memory usage) you have to check if the |
| 534 | // data has incomplete pairs |
| 535 | |
| b17a8968 |
536 | // Remove the noise in order to check if the 4 bytes pairs are complete |
| 7c0ec1d8 |
537 | $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string); |
| 538 | |
| 42ce44f8 |
539 | $sStringRem = ''; |
| 7c0ec1d8 |
540 | $iMod = strlen($string) % 4; |
| 541 | if ($iMod) { |
| 542 | $sStringRem = substr($string,-$iMod); |
| b17a8968 |
543 | // Check if $sStringRem contains padding characters |
| 7c0ec1d8 |
544 | if (substr($sStringRem,-1) != '=') { |
| 545 | $string = substr($string,0,-$iMod); |
| 546 | } else { |
| 547 | $sStringRem = ''; |
| 548 | } |
| 549 | } |
| 7c7b74b3 |
550 | $string = base64_decode($string); |
| 7c0ec1d8 |
551 | return $sStringRem; |
| 7c7b74b3 |
552 | } |
| 553 | |
| 7c0ec1d8 |
554 | |
| 3d8371be |
555 | /* This function decodes the body depending on the encoding type. */ |
| 451f74a2 |
556 | function decodeBody($body, $encoding) { |
| 3d8371be |
557 | global $show_html_default; |
| 83be314a |
558 | |
| b583c3e8 |
559 | $body = str_replace("\r\n", "\n", $body); |
| 560 | $encoding = strtolower($encoding); |
| 3d8371be |
561 | |
| 5166f86a |
562 | $encoding_handler = do_hook_function('decode_body', $encoding); |
| 563 | |
| 564 | |
| 565 | // plugins get first shot at decoding the body |
| 566 | // |
| 567 | if (!empty($encoding_handler) && function_exists($encoding_handler)) { |
| 568 | $body = $encoding_handler('decode', $body); |
| 569 | |
| 570 | } else if ($encoding == 'quoted-printable' || |
| 3d8371be |
571 | $encoding == 'quoted_printable') { |
| b583c3e8 |
572 | $body = quoted_printable_decode($body); |
| 3d8371be |
573 | |
| b583c3e8 |
574 | while (ereg("=\n", $body)) { |
| 575 | $body = ereg_replace ("=\n", '', $body); |
| 576 | } |
| 3d8371be |
577 | |
| b583c3e8 |
578 | } else if ($encoding == 'base64') { |
| 579 | $body = base64_decode($body); |
| 580 | } |
| 3d8371be |
581 | |
| b583c3e8 |
582 | // All other encodings are returned raw. |
| 3d8371be |
583 | return $body; |
| 451f74a2 |
584 | } |
| 585 | |
| 9f7f68c3 |
586 | /** |
| 02474e43 |
587 | * Decodes headers |
| 588 | * |
| 589 | * This functions decode strings that is encoded according to |
| 590 | * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text). |
| 591 | * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002 |
| 592 | * |
| 593 | * @param string $string header string that has to be made readable |
| 594 | * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true |
| 595 | * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true |
| 596 | * @param boolean $decide decide if string can be utfencoded. defaults to false |
| 597 | * @return string decoded header string |
| 598 | */ |
| 9f7f68c3 |
599 | function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) { |
| d6f584fc |
600 | global $languages, $squirrelmail_language,$default_charset; |
| 79e07c7e |
601 | if (is_array($string)) { |
| 602 | $string = implode("\n", $string); |
| 603 | } |
| da2415c1 |
604 | |
| 10dec454 |
605 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| 33a55f5a |
606 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) { |
| 607 | $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string); |
| 08b7f7cc |
608 | // Do we need to return at this point? |
| 609 | // return $string; |
| 83be314a |
610 | } |
| 79e07c7e |
611 | $i = 0; |
| 08b7f7cc |
612 | $iLastMatch = -2; |
| db65b6b0 |
613 | $encoded = true; |
| 0a06275a |
614 | |
| 098ea084 |
615 | $aString = explode(' ',$string); |
| 08b7f7cc |
616 | $ret = ''; |
| 098ea084 |
617 | foreach ($aString as $chunk) { |
| 358a78a1 |
618 | if ($encoded && $chunk === '') { |
| 08b7f7cc |
619 | continue; |
| 358a78a1 |
620 | } elseif ($chunk === '') { |
| 08b7f7cc |
621 | $ret .= ' '; |
| 622 | continue; |
| 623 | } |
| 098ea084 |
624 | $encoded = false; |
| 08b7f7cc |
625 | /* if encoded words are not separated by a linear-space-white we still catch them */ |
| 626 | $j = $i-1; |
| 7e6ca3e8 |
627 | |
| 08b7f7cc |
628 | while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) { |
| 629 | /* if the last chunk isn't an encoded string then put back the space, otherwise don't */ |
| 630 | if ($iLastMatch !== $j) { |
| 631 | if ($htmlsave) { |
| 9f7f68c3 |
632 | $ret .= ' '; |
| 08b7f7cc |
633 | } else { |
| 634 | $ret .= ' '; |
| 635 | } |
| 636 | } |
| 637 | $iLastMatch = $i; |
| 638 | $j = $i; |
| cb718de0 |
639 | if ($htmlsave) { |
| 640 | $ret .= htmlspecialchars($res[1]); |
| 641 | } else { |
| 642 | $ret .= $res[1]; |
| 643 | } |
| 098ea084 |
644 | $encoding = ucfirst($res[3]); |
| d6f584fc |
645 | |
| 646 | /* decide about valid decoding */ |
| 647 | if ($decide && is_conversion_safe($res[2])) { |
| 02474e43 |
648 | $utfencode=true; |
| 649 | $can_be_encoded=true; |
| d6f584fc |
650 | } else { |
| 02474e43 |
651 | $can_be_encoded=false; |
| d6f584fc |
652 | } |
| 098ea084 |
653 | switch ($encoding) |
| 654 | { |
| 655 | case 'B': |
| 656 | $replace = base64_decode($res[4]); |
| fab65ca9 |
657 | if ($utfencode) { |
| 658 | if ($can_be_encoded) { |
| 659 | /* convert string to different charset, |
| 02474e43 |
660 | * if functions asks for it (usually in compose) |
| 661 | */ |
| fab65ca9 |
662 | $ret .= charset_convert($res[2],$replace,$default_charset); |
| 663 | } else { |
| 664 | // convert string to html codes in order to display it |
| 665 | $ret .= charset_decode($res[2],$replace); |
| 666 | } |
| d6f584fc |
667 | } else { |
| fab65ca9 |
668 | if ($htmlsave) { |
| 669 | $replace = htmlspecialchars($replace); |
| 670 | } |
| 671 | $ret.= $replace; |
| d6f584fc |
672 | } |
| 098ea084 |
673 | break; |
| 674 | case 'Q': |
| 098ea084 |
675 | $replace = str_replace('_', ' ', $res[4]); |
| da2415c1 |
676 | $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))', |
| 79e07c7e |
677 | $replace); |
| 098ea084 |
678 | if ($utfencode) { |
| 02474e43 |
679 | if ($can_be_encoded) { |
| d6f584fc |
680 | /* convert string to different charset, |
| 02474e43 |
681 | * if functions asks for it (usually in compose) |
| 682 | */ |
| d6f584fc |
683 | $replace = charset_convert($res[2], $replace,$default_charset); |
| 02474e43 |
684 | } else { |
| d6f584fc |
685 | // convert string to html codes in order to display it |
| 098ea084 |
686 | $replace = charset_decode($res[2], $replace); |
| 02474e43 |
687 | } |
| 098ea084 |
688 | } else { |
| 689 | if ($htmlsave) { |
| c96c32f4 |
690 | $replace = htmlspecialchars($replace); |
| 098ea084 |
691 | } |
| 692 | } |
| 08b7f7cc |
693 | $ret .= $replace; |
| 098ea084 |
694 | break; |
| 695 | default: |
| 696 | break; |
| 79e07c7e |
697 | } |
| 098ea084 |
698 | $chunk = $res[5]; |
| 699 | $encoded = true; |
| 08b7f7cc |
700 | } |
| 701 | if (!$encoded) { |
| 702 | if ($htmlsave) { |
| 9f7f68c3 |
703 | $ret .= ' '; |
| 08b7f7cc |
704 | } else { |
| 705 | $ret .= ' '; |
| da2415c1 |
706 | } |
| 08b7f7cc |
707 | } |
| dc3d13a7 |
708 | |
| 709 | if (!$encoded && $htmlsave) { |
| 710 | $ret .= htmlspecialchars($chunk); |
| 711 | } else { |
| 712 | $ret .= $chunk; |
| 713 | } |
| 098ea084 |
714 | ++$i; |
| 715 | } |
| fd81e884 |
716 | /* remove the first added space */ |
| 717 | if ($ret) { |
| 718 | if ($htmlsave) { |
| 9f7f68c3 |
719 | $ret = substr($ret,5); |
| fd81e884 |
720 | } else { |
| 721 | $ret = substr($ret,1); |
| 722 | } |
| 723 | } |
| da2415c1 |
724 | |
| 08b7f7cc |
725 | return $ret; |
| 451f74a2 |
726 | } |
| 727 | |
| 9f7f68c3 |
728 | /** |
| 02474e43 |
729 | * Encodes header as quoted-printable |
| 730 | * |
| 731 | * Encode a string according to RFC 1522 for use in headers if it |
| 732 | * contains 8-bit characters or anything that looks like it should |
| 733 | * be encoded. |
| 734 | * |
| 735 | * @param string $string header string, that has to be encoded |
| 736 | * @return string quoted-printable encoded string |
| 737 | */ |
| 451f74a2 |
738 | function encodeHeader ($string) { |
| 6fbd125b |
739 | global $default_charset, $languages, $squirrelmail_language; |
| 83be314a |
740 | |
| 10dec454 |
741 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| 33a55f5a |
742 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) { |
| 743 | return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string); |
| 83be314a |
744 | } |
| 793cc001 |
745 | |
| 451f74a2 |
746 | // Encode only if the string contains 8-bit characters or =? |
| 3d8371be |
747 | $j = strlen($string); |
| 098ea084 |
748 | $max_l = 75 - strlen($default_charset) - 7; |
| 749 | $aRet = array(); |
| 451f74a2 |
750 | $ret = ''; |
| c96c32f4 |
751 | $iEncStart = $enc_init = false; |
| 0d53f0f9 |
752 | $cur_l = $iOffset = 0; |
| 3d8371be |
753 | for($i = 0; $i < $j; ++$i) { |
| c96c32f4 |
754 | switch($string{$i}) |
| 755 | { |
| 756 | case '=': |
| 757 | case '<': |
| 758 | case '>': |
| 759 | case ',': |
| 760 | case '?': |
| 761 | case '_': |
| 762 | if ($iEncStart === false) { |
| 763 | $iEncStart = $i; |
| 764 | } |
| 765 | $cur_l+=3; |
| 766 | if ($cur_l > ($max_l-2)) { |
| 08b7f7cc |
767 | /* if there is an stringpart that doesn't need encoding, add it */ |
| c96c32f4 |
768 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| 769 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| 770 | $iOffset = $i; |
| 771 | $cur_l = 0; |
| 772 | $ret = ''; |
| 773 | $iEncStart = false; |
| 774 | } else { |
| 775 | $ret .= sprintf("=%02X",ord($string{$i})); |
| 776 | } |
| 777 | break; |
| 778 | case '(': |
| 779 | case ')': |
| 780 | if ($iEncStart !== false) { |
| 08b7f7cc |
781 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| c96c32f4 |
782 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| 783 | $iOffset = $i; |
| 784 | $cur_l = 0; |
| 785 | $ret = ''; |
| 786 | $iEncStart = false; |
| 787 | } |
| 788 | break; |
| 789 | case ' ': |
| 790 | if ($iEncStart !== false) { |
| 098ea084 |
791 | $cur_l++; |
| 792 | if ($cur_l > $max_l) { |
| 08b7f7cc |
793 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| c96c32f4 |
794 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| 795 | $iOffset = $i; |
| 796 | $cur_l = 0; |
| 797 | $ret = ''; |
| 798 | $iEncStart = false; |
| 08b7f7cc |
799 | } else { |
| c96c32f4 |
800 | $ret .= '_'; |
| 801 | } |
| 802 | } |
| 803 | break; |
| 804 | default: |
| 805 | $k = ord($string{$i}); |
| 806 | if ($k > 126) { |
| 807 | if ($iEncStart === false) { |
| 325978ac |
808 | // do not start encoding in the middle of a string, also take the rest of the word. |
| 809 | $sLeadString = substr($string,0,$i); |
| 810 | $aLeadString = explode(' ',$sLeadString); |
| da2415c1 |
811 | $sToBeEncoded = array_pop($aLeadString); |
| 325978ac |
812 | $iEncStart = $i - strlen($sToBeEncoded); |
| 813 | $ret .= $sToBeEncoded; |
| 814 | $cur_l += strlen($sToBeEncoded); |
| c96c32f4 |
815 | } |
| 816 | $cur_l += 3; |
| 08b7f7cc |
817 | /* first we add the encoded string that reached it's max size */ |
| c96c32f4 |
818 | if ($cur_l > ($max_l-2)) { |
| 08b7f7cc |
819 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| 820 | $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */ |
| c96c32f4 |
821 | $cur_l = 3; |
| 822 | $ret = ''; |
| 823 | $iOffset = $i; |
| 08b7f7cc |
824 | $iEncStart = $i; |
| c96c32f4 |
825 | } |
| 08b7f7cc |
826 | $enc_init = true; |
| c96c32f4 |
827 | $ret .= sprintf("=%02X", $k); |
| 828 | } else { |
| 829 | if ($iEncStart !== false) { |
| 098ea084 |
830 | $cur_l++; |
| 831 | if ($cur_l > $max_l) { |
| 08b7f7cc |
832 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| c96c32f4 |
833 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| 834 | $iEncStart = false; |
| 835 | $iOffset = $i; |
| 836 | $cur_l = 0; |
| 098ea084 |
837 | $ret = ''; |
| 08b7f7cc |
838 | } else { |
| c96c32f4 |
839 | $ret .= $string{$i}; |
| 840 | } |
| 3d8371be |
841 | } |
| c96c32f4 |
842 | } |
| 843 | break; |
| f7b3ba37 |
844 | } |
| 451f74a2 |
845 | } |
| 793cc001 |
846 | |
| c96c32f4 |
847 | if ($enc_init) { |
| 848 | if ($iEncStart !== false) { |
| 849 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| 850 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| 851 | } else { |
| 852 | $aRet[] = substr($string,$iOffset); |
| 853 | } |
| 854 | $string = implode('',$aRet); |
| 451f74a2 |
855 | } |
| 3d8371be |
856 | return $string; |
| 451f74a2 |
857 | } |
| b74ba498 |
858 | |
| 691a2d25 |
859 | /* This function trys to locate the entity_id of a specific mime element */ |
| 3d8371be |
860 | function find_ent_id($id, $message) { |
| a171b359 |
861 | for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) { |
| 862 | if ($message->entities[$i]->header->type0 == 'multipart') { |
| 3d8371be |
863 | $ret = find_ent_id($id, $message->entities[$i]); |
| 451f74a2 |
864 | } else { |
| 3d8371be |
865 | if (strcasecmp($message->entities[$i]->header->id, $id) == 0) { |
| d8cffbab |
866 | // if (sq_check_save_extension($message->entities[$i])) { |
| a171b359 |
867 | return $message->entities[$i]->entity_id; |
| da2415c1 |
868 | // } |
| c8f5f606 |
869 | } elseif (!empty($message->entities[$i]->header->parameters['name'])) { |
| 870 | /** |
| 871 | * This is part of a fix for Outlook Express 6.x generating |
| 872 | * cid URLs without creating content-id headers |
| 873 | * @@JA - 20050207 |
| 874 | */ |
| 875 | if (strcasecmp($message->entities[$i]->header->parameters['name'], $id) == 0) { |
| 876 | return $message->entities[$i]->entity_id; |
| 877 | } |
| 3d8371be |
878 | } |
| a3daaaf3 |
879 | } |
| 451f74a2 |
880 | } |
| 3d8371be |
881 | return $ret; |
| 451f74a2 |
882 | } |
| a3daaaf3 |
883 | |
| e5e9381a |
884 | function sq_check_save_extension($message) { |
| 885 | $filename = $message->getFilename(); |
| 886 | $ext = substr($filename, strrpos($filename,'.')+1); |
| 887 | $save_extensions = array('jpg','jpeg','gif','png','bmp'); |
| 3d8371be |
888 | return in_array($ext, $save_extensions); |
| e5e9381a |
889 | } |
| 890 | |
| 891 | |
| 691a2d25 |
892 | /** |
| 02474e43 |
893 | ** HTMLFILTER ROUTINES |
| 894 | */ |
| 451f74a2 |
895 | |
| 2dd879b8 |
896 | /** |
| 02474e43 |
897 | * This function is more or less a wrapper around stripslashes. Apparently |
| 898 | * Explorer is stupid enough to just remove the backslashes and then |
| 899 | * execute the content of the attribute as if nothing happened. |
| 900 | * Who does that? |
| 901 | * |
| 902 | * @param attvalue The value of the attribute |
| 903 | * @return attvalue The value of the attribute stripslashed. |
| 904 | */ |
| 2dd879b8 |
905 | function sq_unbackslash($attvalue){ |
| 906 | /** |
| 02474e43 |
907 | * Remove any backslashes. See if there are any first. |
| 908 | */ |
| 91220ed1 |
909 | |
| 2dd879b8 |
910 | if (strstr($attvalue, '\\') !== false){ |
| 911 | $attvalue = stripslashes($attvalue); |
| 912 | } |
| 913 | return $attvalue; |
| 914 | } |
| 915 | |
| 916 | /** |
| 02474e43 |
917 | * Kill any tabs, newlines, or carriage returns. Our friends the |
| 918 | * makers of the browser with 95% market value decided that it'd |
| 919 | * be funny to make "java[tab]script" be just as good as "javascript". |
| 920 | * |
| 921 | * @param attvalue The attribute value before extraneous spaces removed. |
| 922 | * @return attvalue The attribute value after extraneous spaces removed. |
| 923 | */ |
| 2dd879b8 |
924 | function sq_unspace($attvalue){ |
| 925 | if (strcspn($attvalue, "\t\r\n") != strlen($attvalue)){ |
| 926 | $attvalue = str_replace(Array("\t", "\r", "\n"), Array('', '', ''), |
| 927 | $attvalue); |
| 928 | } |
| 929 | return $attvalue; |
| 930 | } |
| 931 | |
| 691a2d25 |
932 | /** |
| 02474e43 |
933 | * This function returns the final tag out of the tag name, an array |
| 934 | * of attributes, and the type of the tag. This function is called by |
| 935 | * sq_sanitize internally. |
| 936 | * |
| 937 | * @param $tagname the name of the tag. |
| 938 | * @param $attary the array of attributes and their values |
| 939 | * @param $tagtype The type of the tag (see in comments). |
| 940 | * @return a string with the final tag representation. |
| 941 | */ |
| 691a2d25 |
942 | function sq_tagprint($tagname, $attary, $tagtype){ |
| b583c3e8 |
943 | $me = 'sq_tagprint'; |
| 3d8371be |
944 | |
| 691a2d25 |
945 | if ($tagtype == 2){ |
| 946 | $fulltag = '</' . $tagname . '>'; |
| 947 | } else { |
| 948 | $fulltag = '<' . $tagname; |
| 949 | if (is_array($attary) && sizeof($attary)){ |
| 950 | $atts = Array(); |
| 951 | while (list($attname, $attvalue) = each($attary)){ |
| 952 | array_push($atts, "$attname=$attvalue"); |
| 953 | } |
| 954 | $fulltag .= ' ' . join(" ", $atts); |
| 955 | } |
| 956 | if ($tagtype == 3){ |
| b583c3e8 |
957 | $fulltag .= ' /'; |
| 691a2d25 |
958 | } |
| b583c3e8 |
959 | $fulltag .= '>'; |
| 451f74a2 |
960 | } |
| 691a2d25 |
961 | return $fulltag; |
| 451f74a2 |
962 | } |
| a3daaaf3 |
963 | |
| 691a2d25 |
964 | /** |
| 02474e43 |
965 | * A small helper function to use with array_walk. Modifies a by-ref |
| 966 | * value and makes it lowercase. |
| 967 | * |
| 968 | * @param $val a value passed by-ref. |
| 969 | * @return void since it modifies a by-ref value. |
| 970 | */ |
| 691a2d25 |
971 | function sq_casenormalize(&$val){ |
| 972 | $val = strtolower($val); |
| 973 | } |
| 451f74a2 |
974 | |
| 691a2d25 |
975 | /** |
| 02474e43 |
976 | * This function skips any whitespace from the current position within |
| 977 | * a string and to the next non-whitespace value. |
| 978 | * |
| 979 | * @param $body the string |
| 980 | * @param $offset the offset within the string where we should start |
| 981 | * looking for the next non-whitespace character. |
| 982 | * @return the location within the $body where the next |
| 983 | * non-whitespace char is located. |
| 984 | */ |
| 691a2d25 |
985 | function sq_skipspace($body, $offset){ |
| b583c3e8 |
986 | $me = 'sq_skipspace'; |
| 3d8371be |
987 | preg_match('/^(\s*)/s', substr($body, $offset), $matches); |
| 691a2d25 |
988 | if (sizeof($matches{1})){ |
| 989 | $count = strlen($matches{1}); |
| 990 | $offset += $count; |
| 451f74a2 |
991 | } |
| 691a2d25 |
992 | return $offset; |
| 451f74a2 |
993 | } |
| a3daaaf3 |
994 | |
| 691a2d25 |
995 | /** |
| 02474e43 |
996 | * This function looks for the next character within a string. It's |
| 997 | * really just a glorified "strpos", except it catches if failures |
| 998 | * nicely. |
| 999 | * |
| 1000 | * @param $body The string to look for needle in. |
| 1001 | * @param $offset Start looking from this position. |
| 1002 | * @param $needle The character/string to look for. |
| 1003 | * @return location of the next occurance of the needle, or |
| 1004 | * strlen($body) if needle wasn't found. |
| 1005 | */ |
| 691a2d25 |
1006 | function sq_findnxstr($body, $offset, $needle){ |
| 3d8371be |
1007 | $me = 'sq_findnxstr'; |
| 691a2d25 |
1008 | $pos = strpos($body, $needle, $offset); |
| 1009 | if ($pos === FALSE){ |
| 1010 | $pos = strlen($body); |
| 451f74a2 |
1011 | } |
| 691a2d25 |
1012 | return $pos; |
| 451f74a2 |
1013 | } |
| a3daaaf3 |
1014 | |
| 691a2d25 |
1015 | /** |
| 02474e43 |
1016 | * This function takes a PCRE-style regexp and tries to match it |
| 1017 | * within the string. |
| 1018 | * |
| 1019 | * @param $body The string to look for needle in. |
| 1020 | * @param $offset Start looking from here. |
| 1021 | * @param $reg A PCRE-style regex to match. |
| 1022 | * @return Returns a false if no matches found, or an array |
| 1023 | * with the following members: |
| 1024 | * - integer with the location of the match within $body |
| 1025 | * - string with whatever content between offset and the match |
| 1026 | * - string with whatever it is we matched |
| 1027 | */ |
| 691a2d25 |
1028 | function sq_findnxreg($body, $offset, $reg){ |
| b583c3e8 |
1029 | $me = 'sq_findnxreg'; |
| 691a2d25 |
1030 | $matches = Array(); |
| 1031 | $retarr = Array(); |
| 7d06541f |
1032 | preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches); |
| 1033 | if (!isset($matches{0}) || !$matches{0}){ |
| 691a2d25 |
1034 | $retarr = false; |
| 1035 | } else { |
| 1036 | $retarr{0} = $offset + strlen($matches{1}); |
| 1037 | $retarr{1} = $matches{1}; |
| 1038 | $retarr{2} = $matches{2}; |
| 1039 | } |
| 1040 | return $retarr; |
| 1041 | } |
| a3daaaf3 |
1042 | |
| 691a2d25 |
1043 | /** |
| 02474e43 |
1044 | * This function looks for the next tag. |
| 1045 | * |
| 1046 | * @param $body String where to look for the next tag. |
| 1047 | * @param $offset Start looking from here. |
| 1048 | * @return false if no more tags exist in the body, or |
| 1049 | * an array with the following members: |
| 1050 | * - string with the name of the tag |
| 1051 | * - array with attributes and their values |
| 1052 | * - integer with tag type (1, 2, or 3) |
| 1053 | * - integer where the tag starts (starting "<") |
| 1054 | * - integer where the tag ends (ending ">") |
| 1055 | * first three members will be false, if the tag is invalid. |
| 1056 | */ |
| 691a2d25 |
1057 | function sq_getnxtag($body, $offset){ |
| b583c3e8 |
1058 | $me = 'sq_getnxtag'; |
| 691a2d25 |
1059 | if ($offset > strlen($body)){ |
| 1060 | return false; |
| 1061 | } |
| 1062 | $lt = sq_findnxstr($body, $offset, "<"); |
| 1063 | if ($lt == strlen($body)){ |
| 1064 | return false; |
| 1065 | } |
| 1066 | /** |
| 02474e43 |
1067 | * We are here: |
| 1068 | * blah blah <tag attribute="value"> |
| 1069 | * \---------^ |
| 1070 | */ |
| 691a2d25 |
1071 | $pos = sq_skipspace($body, $lt+1); |
| 1072 | if ($pos >= strlen($body)){ |
| 1073 | return Array(false, false, false, $lt, strlen($body)); |
| 1074 | } |
| 1075 | /** |
| 02474e43 |
1076 | * There are 3 kinds of tags: |
| 1077 | * 1. Opening tag, e.g.: |
| 1078 | * <a href="blah"> |
| 1079 | * 2. Closing tag, e.g.: |
| 1080 | * </a> |
| 1081 | * 3. XHTML-style content-less tag, e.g.: |
| 1082 | * <img src="blah" /> |
| 1083 | */ |
| 691a2d25 |
1084 | $tagtype = false; |
| 1085 | switch (substr($body, $pos, 1)){ |
| 3d8371be |
1086 | case '/': |
| 1087 | $tagtype = 2; |
| 1088 | $pos++; |
| 1089 | break; |
| 1090 | case '!': |
| 1091 | /** |
| 02474e43 |
1092 | * A comment or an SGML declaration. |
| 1093 | */ |
| 3d8371be |
1094 | if (substr($body, $pos+1, 2) == "--"){ |
| 1095 | $gt = strpos($body, "-->", $pos); |
| 1096 | if ($gt === false){ |
| 1097 | $gt = strlen($body); |
| 1098 | } else { |
| 1099 | $gt += 2; |
| 1100 | } |
| 1101 | return Array(false, false, false, $lt, $gt); |
| bb8d0799 |
1102 | } else { |
| 3d8371be |
1103 | $gt = sq_findnxstr($body, $pos, ">"); |
| 1104 | return Array(false, false, false, $lt, $gt); |
| 1105 | } |
| 1106 | break; |
| 1107 | default: |
| 1108 | /** |
| 02474e43 |
1109 | * Assume tagtype 1 for now. If it's type 3, we'll switch values |
| 1110 | * later. |
| 1111 | */ |
| 3d8371be |
1112 | $tagtype = 1; |
| 1113 | break; |
| 691a2d25 |
1114 | } |
| a3daaaf3 |
1115 | |
| 691a2d25 |
1116 | $tagname = ''; |
| 1117 | /** |
| 02474e43 |
1118 | * Look for next [\W-_], which will indicate the end of the tag name. |
| 1119 | */ |
| 691a2d25 |
1120 | $regary = sq_findnxreg($body, $pos, "[^\w\-_]"); |
| 1121 | if ($regary == false){ |
| 1122 | return Array(false, false, false, $lt, strlen($body)); |
| 1123 | } |
| 1124 | list($pos, $tagname, $match) = $regary; |
| 1125 | $tagname = strtolower($tagname); |
| 1126 | |
| 1127 | /** |
| 02474e43 |
1128 | * $match can be either of these: |
| 1129 | * '>' indicating the end of the tag entirely. |
| 1130 | * '\s' indicating the end of the tag name. |
| 1131 | * '/' indicating that this is type-3 xhtml tag. |
| 1132 | * |
| 1133 | * Whatever else we find there indicates an invalid tag. |
| 1134 | */ |
| 691a2d25 |
1135 | switch ($match){ |
| 3d8371be |
1136 | case '/': |
| 691a2d25 |
1137 | /** |
| 02474e43 |
1138 | * This is an xhtml-style tag with a closing / at the |
| 1139 | * end, like so: <img src="blah" />. Check if it's followed |
| 1140 | * by the closing bracket. If not, then this tag is invalid |
| 1141 | */ |
| 3d8371be |
1142 | if (substr($body, $pos, 2) == "/>"){ |
| 1143 | $pos++; |
| 1144 | $tagtype = 3; |
| 1145 | } else { |
| 1146 | $gt = sq_findnxstr($body, $pos, ">"); |
| 1147 | $retary = Array(false, false, false, $lt, $gt); |
| 1148 | return $retary; |
| 1149 | } |
| 1150 | case '>': |
| 1151 | return Array($tagname, false, $tagtype, $lt, $pos); |
| 1152 | break; |
| 1153 | default: |
| 1154 | /** |
| 02474e43 |
1155 | * Check if it's whitespace |
| 1156 | */ |
| 3d8371be |
1157 | if (!preg_match('/\s/', $match)){ |
| 1158 | /** |
| 02474e43 |
1159 | * This is an invalid tag! Look for the next closing ">". |
| 1160 | */ |
| 7d06541f |
1161 | $gt = sq_findnxstr($body, $lt, ">"); |
| 3d8371be |
1162 | return Array(false, false, false, $lt, $gt); |
| 1163 | } |
| 1164 | break; |
| 691a2d25 |
1165 | } |
| 3d8371be |
1166 | |
| 691a2d25 |
1167 | /** |
| 02474e43 |
1168 | * At this point we're here: |
| 1169 | * <tagname attribute='blah'> |
| 1170 | * \-------^ |
| 1171 | * |
| 1172 | * At this point we loop in order to find all attributes. |
| 1173 | */ |
| 691a2d25 |
1174 | $attname = ''; |
| 691a2d25 |
1175 | $attary = Array(); |
| 1176 | |
| 1177 | while ($pos <= strlen($body)){ |
| 1178 | $pos = sq_skipspace($body, $pos); |
| 1179 | if ($pos == strlen($body)){ |
| 1180 | /** |
| 02474e43 |
1181 | * Non-closed tag. |
| 1182 | */ |
| 691a2d25 |
1183 | return Array(false, false, false, $lt, $pos); |
| 1184 | } |
| 1185 | /** |
| 02474e43 |
1186 | * See if we arrived at a ">" or "/>", which means that we reached |
| 1187 | * the end of the tag. |
| 1188 | */ |
| 691a2d25 |
1189 | $matches = Array(); |
| 164800ad |
1190 | if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) { |
| c828931c |
1191 | /** |
| 02474e43 |
1192 | * Yep. So we did. |
| 1193 | */ |
| c828931c |
1194 | $pos += strlen($matches{1}); |
| 1195 | if ($matches{2} == "/>"){ |
| 1196 | $tagtype = 3; |
| 1197 | $pos++; |
| 1198 | } |
| 1199 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
| 1200 | } |
| a3daaaf3 |
1201 | |
| cca46357 |
1202 | /** |
| 02474e43 |
1203 | * There are several types of attributes, with optional |
| 1204 | * [:space:] between members. |
| 1205 | * Type 1: |
| 1206 | * attrname[:space:]=[:space:]'CDATA' |
| 1207 | * Type 2: |
| 1208 | * attrname[:space:]=[:space:]"CDATA" |
| 1209 | * Type 3: |
| 1210 | * attr[:space:]=[:space:]CDATA |
| 1211 | * Type 4: |
| 1212 | * attrname |
| 1213 | * |
| 1214 | * We leave types 1 and 2 the same, type 3 we check for |
| 1215 | * '"' and convert to """ if needed, then wrap in |
| 1216 | * double quotes. Type 4 we convert into: |
| 1217 | * attrname="yes". |
| 1218 | */ |
| 3f7c623f |
1219 | $regary = sq_findnxreg($body, $pos, "[^:\w\-_]"); |
| 691a2d25 |
1220 | if ($regary == false){ |
| 1221 | /** |
| 02474e43 |
1222 | * Looks like body ended before the end of tag. |
| 1223 | */ |
| 691a2d25 |
1224 | return Array(false, false, false, $lt, strlen($body)); |
| cca46357 |
1225 | } |
| 691a2d25 |
1226 | list($pos, $attname, $match) = $regary; |
| 1227 | $attname = strtolower($attname); |
| 1228 | /** |
| 02474e43 |
1229 | * We arrived at the end of attribute name. Several things possible |
| 1230 | * here: |
| 1231 | * '>' means the end of the tag and this is attribute type 4 |
| 1232 | * '/' if followed by '>' means the same thing as above |
| 1233 | * '\s' means a lot of things -- look what it's followed by. |
| 1234 | * anything else means the attribute is invalid. |
| 1235 | */ |
| 691a2d25 |
1236 | switch($match){ |
| 3d8371be |
1237 | case '/': |
| 691a2d25 |
1238 | /** |
| 02474e43 |
1239 | * This is an xhtml-style tag with a closing / at the |
| 1240 | * end, like so: <img src="blah" />. Check if it's followed |
| 1241 | * by the closing bracket. If not, then this tag is invalid |
| 1242 | */ |
| 3d8371be |
1243 | if (substr($body, $pos, 2) == "/>"){ |
| 691a2d25 |
1244 | $pos++; |
| 3d8371be |
1245 | $tagtype = 3; |
| 691a2d25 |
1246 | } else { |
| 3d8371be |
1247 | $gt = sq_findnxstr($body, $pos, ">"); |
| 1248 | $retary = Array(false, false, false, $lt, $gt); |
| 1249 | return $retary; |
| 1250 | } |
| 1251 | case '>': |
| 1252 | $attary{$attname} = '"yes"'; |
| 1253 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
| 1254 | break; |
| 1255 | default: |
| 1256 | /** |
| 02474e43 |
1257 | * Skip whitespace and see what we arrive at. |
| 1258 | */ |
| 3d8371be |
1259 | $pos = sq_skipspace($body, $pos); |
| 1260 | $char = substr($body, $pos, 1); |
| 1261 | /** |
| 02474e43 |
1262 | * Two things are valid here: |
| 1263 | * '=' means this is attribute type 1 2 or 3. |
| 1264 | * \w means this was attribute type 4. |
| 1265 | * anything else we ignore and re-loop. End of tag and |
| 1266 | * invalid stuff will be caught by our checks at the beginning |
| 1267 | * of the loop. |
| 1268 | */ |
| 3d8371be |
1269 | if ($char == "="){ |
| 1270 | $pos++; |
| 1271 | $pos = sq_skipspace($body, $pos); |
| 691a2d25 |
1272 | /** |
| 02474e43 |
1273 | * Here are 3 possibilities: |
| 1274 | * "'" attribute type 1 |
| 1275 | * '"' attribute type 2 |
| 1276 | * everything else is the content of tag type 3 |
| 1277 | */ |
| 3d8371be |
1278 | $quot = substr($body, $pos, 1); |
| 1279 | if ($quot == "'"){ |
| 1280 | $regary = sq_findnxreg($body, $pos+1, "\'"); |
| 1281 | if ($regary == false){ |
| 1282 | return Array(false, false, false, $lt, strlen($body)); |
| 1283 | } |
| 1284 | list($pos, $attval, $match) = $regary; |
| 1285 | $pos++; |
| 1286 | $attary{$attname} = "'" . $attval . "'"; |
| 1287 | } else if ($quot == '"'){ |
| 1288 | $regary = sq_findnxreg($body, $pos+1, '\"'); |
| 1289 | if ($regary == false){ |
| 1290 | return Array(false, false, false, $lt, strlen($body)); |
| 1291 | } |
| 1292 | list($pos, $attval, $match) = $regary; |
| 1293 | $pos++; |
| 1294 | $attary{$attname} = '"' . $attval . '"'; |
| 1295 | } else { |
| 1296 | /** |
| 02474e43 |
1297 | * These are hateful. Look for \s, or >. |
| 1298 | */ |
| 3d8371be |
1299 | $regary = sq_findnxreg($body, $pos, "[\s>]"); |
| 1300 | if ($regary == false){ |
| 1301 | return Array(false, false, false, $lt, strlen($body)); |
| 1302 | } |
| 1303 | list($pos, $attval, $match) = $regary; |
| 1304 | /** |
| 02474e43 |
1305 | * If it's ">" it will be caught at the top. |
| 1306 | */ |
| 3d8371be |
1307 | $attval = preg_replace("/\"/s", """, $attval); |
| 1308 | $attary{$attname} = '"' . $attval . '"'; |
| 7e235a1a |
1309 | } |
| 3d8371be |
1310 | } else if (preg_match("|[\w/>]|", $char)) { |
| 691a2d25 |
1311 | /** |
| 02474e43 |
1312 | * That was attribute type 4. |
| 1313 | */ |
| 3d8371be |
1314 | $attary{$attname} = '"yes"'; |
| 1315 | } else { |
| 1316 | /** |
| 02474e43 |
1317 | * An illegal character. Find next '>' and return. |
| 1318 | */ |
| 3d8371be |
1319 | $gt = sq_findnxstr($body, $pos, ">"); |
| 1320 | return Array(false, false, false, $lt, $gt); |
| 451f74a2 |
1321 | } |
| 3d8371be |
1322 | break; |
| 691a2d25 |
1323 | } |
| 1324 | } |
| 1325 | /** |
| 02474e43 |
1326 | * The fact that we got here indicates that the tag end was never |
| 1327 | * found. Return invalid tag indication so it gets stripped. |
| 1328 | */ |
| 691a2d25 |
1329 | return Array(false, false, false, $lt, strlen($body)); |
| 1330 | } |
| 1331 | |
| 1332 | /** |
| 02474e43 |
1333 | * This function checks attribute values for entity-encoded values |
| 1334 | * and returns them translated into 8-bit strings so we can run |
| 1335 | * checks on them. |
| 1336 | * |
| 1337 | * @param $attvalue A string to run entity check against. |
| 1338 | * @return Translated value. |
| 1339 | */ |
| 91220ed1 |
1340 | |
| 691a2d25 |
1341 | function sq_deent($attvalue){ |
| b583c3e8 |
1342 | $me = 'sq_deent'; |
| 691a2d25 |
1343 | /** |
| 02474e43 |
1344 | * See if we have to run the checks first. All entities must start |
| 1345 | * with "&". |
| 1346 | */ |
| 91220ed1 |
1347 | if (strpos($attvalue, '&') === false){ |
| 691a2d25 |
1348 | return $attvalue; |
| 1349 | } |
| 1350 | /** |
| 02474e43 |
1351 | * Check named entities first. |
| 1352 | */ |
| 691a2d25 |
1353 | $trans = get_html_translation_table(HTML_ENTITIES); |
| 1354 | /** |
| 02474e43 |
1355 | * Leave " in, as it can mess us up. |
| 1356 | */ |
| 691a2d25 |
1357 | $trans = array_flip($trans); |
| 91220ed1 |
1358 | unset($trans{'"'}); |
| 691a2d25 |
1359 | while (list($ent, $val) = each($trans)){ |
| 91220ed1 |
1360 | $attvalue = preg_replace('/' . $ent . '*/si', $val, $attvalue); |
| 691a2d25 |
1361 | } |
| 1362 | /** |
| 02474e43 |
1363 | * Now translate numbered entities from 1 to 255 if needed. |
| 1364 | */ |
| 91220ed1 |
1365 | if (strpos($attvalue, '#') !== false){ |
| 691a2d25 |
1366 | $omit = Array(34, 39); |
| 91220ed1 |
1367 | for ($asc = 256; $asc >= 0; $asc--){ |
| 691a2d25 |
1368 | if (!in_array($asc, $omit)){ |
| 1369 | $chr = chr($asc); |
| 91220ed1 |
1370 | $octrule = '/\�*' . $asc . ';*/si'; |
| 1371 | $hexrule = '/\�*' . dechex($asc) . ';*/si'; |
| 1372 | $attvalue = preg_replace($octrule, $chr, $attvalue); |
| 1373 | $attvalue = preg_replace($hexrule, $chr, $attvalue); |
| a3daaaf3 |
1374 | } |
| 691a2d25 |
1375 | } |
| 1376 | } |
| 1377 | return $attvalue; |
| 1378 | } |
| 1379 | |
| 1380 | /** |
| 02474e43 |
1381 | * This function runs various checks against the attributes. |
| 1382 | * |
| 1383 | * @param $tagname String with the name of the tag. |
| 1384 | * @param $attary Array with all tag attributes. |
| 1385 | * @param $rm_attnames See description for sq_sanitize |
| 1386 | * @param $bad_attvals See description for sq_sanitize |
| 1387 | * @param $add_attr_to_tag See description for sq_sanitize |
| 1388 | * @param $message message object |
| 1389 | * @param $id message id |
| 1390 | * @return Array with modified attributes. |
| 1391 | */ |
| da2415c1 |
1392 | function sq_fixatts($tagname, |
| 1393 | $attary, |
| 691a2d25 |
1394 | $rm_attnames, |
| 1395 | $bad_attvals, |
| 1396 | $add_attr_to_tag, |
| 1397 | $message, |
| b3af12ef |
1398 | $id, |
| 3d8371be |
1399 | $mailbox |
| 691a2d25 |
1400 | ){ |
| b583c3e8 |
1401 | $me = 'sq_fixatts'; |
| 691a2d25 |
1402 | while (list($attname, $attvalue) = each($attary)){ |
| 1403 | /** |
| 02474e43 |
1404 | * See if this attribute should be removed. |
| 1405 | */ |
| 691a2d25 |
1406 | foreach ($rm_attnames as $matchtag=>$matchattrs){ |
| 1407 | if (preg_match($matchtag, $tagname)){ |
| 1408 | foreach ($matchattrs as $matchattr){ |
| 1409 | if (preg_match($matchattr, $attname)){ |
| 1410 | unset($attary{$attname}); |
| 1411 | continue; |
| 1412 | } |
| 451f74a2 |
1413 | } |
| 451f74a2 |
1414 | } |
| 691a2d25 |
1415 | } |
| 1416 | /** |
| 02474e43 |
1417 | * Remove any backslashes, entities, and extraneous whitespace. |
| 1418 | */ |
| 2dd879b8 |
1419 | $attvalue = sq_unbackslash($attvalue); |
| 691a2d25 |
1420 | $attvalue = sq_deent($attvalue); |
| 2dd879b8 |
1421 | $attvalue = sq_unspace($attvalue); |
| 691a2d25 |
1422 | |
| af861a34 |
1423 | /** |
| 02474e43 |
1424 | * Remove \r \n \t \0 " " "\\" |
| 1425 | */ |
| 8d863f64 |
1426 | $attvalue = str_replace(Array("\r", "\n", "\t", "\0", " ", "\\"), |
| af861a34 |
1427 | Array('', '','','','',''), $attvalue); |
| 1428 | |
| 691a2d25 |
1429 | /** |
| 02474e43 |
1430 | * Now let's run checks on the attvalues. |
| 1431 | * I don't expect anyone to comprehend this. If you do, |
| 1432 | * get in touch with me so I can drive to where you live and |
| 1433 | * shake your hand personally. :) |
| 1434 | */ |
| 691a2d25 |
1435 | foreach ($bad_attvals as $matchtag=>$matchattrs){ |
| 1436 | if (preg_match($matchtag, $tagname)){ |
| 1437 | foreach ($matchattrs as $matchattr=>$valary){ |
| 1438 | if (preg_match($matchattr, $attname)){ |
| 1439 | /** |
| 02474e43 |
1440 | * There are two arrays in valary. |
| 1441 | * First is matches. |
| 1442 | * Second one is replacements |
| 1443 | */ |
| 691a2d25 |
1444 | list($valmatch, $valrepl) = $valary; |
| da2415c1 |
1445 | $newvalue = |
| 691a2d25 |
1446 | preg_replace($valmatch, $valrepl, $attvalue); |
| 1447 | if ($newvalue != $attvalue){ |
| 1448 | $attary{$attname} = $newvalue; |
| 1449 | } |
| 1450 | } |
| 1451 | } |
| 451f74a2 |
1452 | } |
| a3daaaf3 |
1453 | } |
| 834a1027 |
1454 | |
| 1455 | |
| 1456 | /** |
| 1457 | * Replace empty src tags with the blank image. src is only used |
| 1458 | * for frames, images, and image inputs. Doing a replace should |
| 1459 | * not affect them working as should be, however it will stop |
| 1460 | * IE from being kicked off when src for img tags are not set |
| 1461 | */ |
| 5a6fde9e |
1462 | if (($attname == 'src') && ($attvalue == '""')) { |
| 834a1027 |
1463 | $attary{$attname} = '"' . SM_PATH . 'images/blank.png"'; |
| 1464 | } |
| 1465 | |
| 691a2d25 |
1466 | /** |
| 02474e43 |
1467 | * Turn cid: urls into http-friendly ones. |
| 1468 | */ |
| 691a2d25 |
1469 | if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){ |
| b3af12ef |
1470 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
| 691a2d25 |
1471 | } |
| ff940ebc |
1472 | |
| 1473 | /** |
| 1474 | * "Hack" fix for Outlook using propriatary outbind:// protocol in img tags. |
| 1475 | * One day MS might actually make it match something useful, for now, falling |
| 1476 | * back to using cid2http, so we can grab the blank.png. |
| 1477 | */ |
| 1478 | if (preg_match("/^[\'\"]\s*outbind:\/\//si", $attvalue)) { |
| 1479 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
| 1480 | } |
| 1481 | |
| a3daaaf3 |
1482 | } |
| 691a2d25 |
1483 | /** |
| 02474e43 |
1484 | * See if we need to append any attributes to this tag. |
| 1485 | */ |
| 691a2d25 |
1486 | foreach ($add_attr_to_tag as $matchtag=>$addattary){ |
| 1487 | if (preg_match($matchtag, $tagname)){ |
| 1488 | $attary = array_merge($attary, $addattary); |
| 1489 | } |
| 1490 | } |
| 1491 | return $attary; |
| 451f74a2 |
1492 | } |
| a3daaaf3 |
1493 | |
| 691a2d25 |
1494 | /** |
| 02474e43 |
1495 | * This function edits the style definition to make them friendly and |
| 1496 | * usable in SquirrelMail. |
| 1497 | * |
| 1498 | * @param $message the message object |
| 1499 | * @param $id the message id |
| 1500 | * @param $content a string with whatever is between <style> and </style> |
| 1501 | * @param $mailbox the message mailbox |
| 1502 | * @return a string with edited content. |
| 1503 | */ |
| e60a299a |
1504 | function sq_fixstyle($body, $pos, $message, $id, $mailbox){ |
| 691a2d25 |
1505 | global $view_unsafe_images; |
| b583c3e8 |
1506 | $me = 'sq_fixstyle'; |
| 7d06541f |
1507 | $ret = sq_findnxreg($body, $pos, '</\s*style\s*>'); |
| 1508 | if ($ret == FALSE){ |
| 1509 | return array(FALSE, strlen($body)); |
| 1510 | } |
| 1511 | $newpos = $ret[0] + strlen($ret[2]); |
| 1512 | $content = $ret[1]; |
| 691a2d25 |
1513 | /** |
| 02474e43 |
1514 | * First look for general BODY style declaration, which would be |
| 1515 | * like so: |
| 1516 | * body {background: blah-blah} |
| 1517 | * and change it to .bodyclass so we can just assign it to a <div> |
| 1518 | */ |
| 691a2d25 |
1519 | $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content); |
| 3d8371be |
1520 | $secremoveimg = '../images/' . _("sec_remove_eng.png"); |
| 691a2d25 |
1521 | /** |
| 02474e43 |
1522 | * Fix url('blah') declarations. |
| 1523 | */ |
| 7d06541f |
1524 | $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si", |
| 691a2d25 |
1525 | "url(\\1$secremoveimg\\2)", $content); |
| 1526 | /** |
| 02474e43 |
1527 | * Fix url('https*://.*) declarations but only if $view_unsafe_images |
| 1528 | * is false. |
| 1529 | */ |
| 691a2d25 |
1530 | if (!$view_unsafe_images){ |
| 7d06541f |
1531 | $content = preg_replace("|url\s*\(\s*([\'\"])\s*https*:.*?([\'\"])\s*\)|si", |
| 691a2d25 |
1532 | "url(\\1$secremoveimg\\2)", $content); |
| 1533 | } |
| da2415c1 |
1534 | |
| 691a2d25 |
1535 | /** |
| 02474e43 |
1536 | * Fix urls that refer to cid: |
| 1537 | */ |
| da2415c1 |
1538 | while (preg_match("|url\s*\(\s*([\'\"]\s*cid:.*?[\'\"])\s*\)|si", |
| 02474e43 |
1539 | $content, $matches)){ |
| 691a2d25 |
1540 | $cidurl = $matches{1}; |
| e60a299a |
1541 | $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox); |
| 7d06541f |
1542 | $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si", |
| 691a2d25 |
1543 | "url($httpurl)", $content); |
| 1544 | } |
| a3daaaf3 |
1545 | |
| 691a2d25 |
1546 | /** |
| 02474e43 |
1547 | * Fix stupid css declarations which lead to vulnerabilities |
| 1548 | * in IE. |
| 1549 | */ |
| f83c60a2 |
1550 | $match = Array('/expression/i', |
| 02474e43 |
1551 | '/behaviou*r/i', |
| 1552 | '/binding/i', |
| 1553 | '/include-source/i'); |
| 2dd879b8 |
1554 | $replace = Array('idiocy', 'idiocy', 'idiocy', 'idiocy'); |
| bb8d0799 |
1555 | $content = preg_replace($match, $replace, $content); |
| 7d06541f |
1556 | return array($content, $newpos); |
| 691a2d25 |
1557 | } |
| a3daaaf3 |
1558 | |
| 691a2d25 |
1559 | /** |
| 02474e43 |
1560 | * This function converts cid: url's into the ones that can be viewed in |
| 1561 | * the browser. |
| 1562 | * |
| 1563 | * @param $message the message object |
| 1564 | * @param $id the message id |
| 1565 | * @param $cidurl the cid: url. |
| 1566 | * @param $mailbox the message mailbox |
| 1567 | * @return a string with a http-friendly url |
| 1568 | */ |
| b3af12ef |
1569 | function sq_cid2http($message, $id, $cidurl, $mailbox){ |
| 691a2d25 |
1570 | /** |
| 02474e43 |
1571 | * Get rid of quotes. |
| 1572 | */ |
| 691a2d25 |
1573 | $quotchar = substr($cidurl, 0, 1); |
| 2dd879b8 |
1574 | if ($quotchar == '"' || $quotchar == "'"){ |
| 1575 | $cidurl = str_replace($quotchar, "", $cidurl); |
| 1576 | } else { |
| 1577 | $quotchar = ''; |
| 1578 | } |
| 691a2d25 |
1579 | $cidurl = substr(trim($cidurl), 4); |
| e5e9381a |
1580 | $linkurl = find_ent_id($cidurl, $message); |
| 1581 | /* in case of non-save cid links $httpurl should be replaced by a sort of |
| 02474e43 |
1582 | unsave link image */ |
| e5e9381a |
1583 | $httpurl = ''; |
| c8f5f606 |
1584 | |
| 1585 | /** |
| 1586 | * This is part of a fix for Outlook Express 6.x generating |
| 1587 | * cid URLs without creating content-id headers. These images are |
| 1588 | * not part of the multipart/related html mail. The html contains |
| 1589 | * <img src="cid:{some_id}/image_filename.ext"> references to |
| 1590 | * attached images with as goal to render them inline although |
| 1591 | * the attachment disposition property is not inline. |
| 1592 | **/ |
| 1593 | |
| 1594 | if (empty($linkurl)) { |
| 1595 | if (preg_match('/{.*}\//', $cidurl)) { |
| 1596 | $cidurl = preg_replace('/{.*}\//','', $cidurl); |
| 1597 | if (!empty($cidurl)) { |
| 1598 | $linkurl = find_ent_id($cidurl, $message); |
| 1599 | } |
| 1600 | } |
| 1601 | } |
| 1602 | |
| 1603 | if (!empty($linkurl)) { |
| 6b04287c |
1604 | $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&' . |
| c8f5f606 |
1605 | "passed_id=$id&mailbox=" . urlencode($mailbox) . |
| 1606 | '&ent_id=' . $linkurl . $quotchar; |
| 1607 | } else { |
| 1608 | /** |
| 1609 | * If we couldn't generate a proper img url, drop in a blank image |
| 1610 | * instead of sending back empty, otherwise it causes unusual behaviour |
| 1611 | */ |
| 59a3dab6 |
1612 | $httpurl = $quotchar . SM_PATH . 'images/blank.png'; |
| e5e9381a |
1613 | } |
| c8f5f606 |
1614 | |
| 691a2d25 |
1615 | return $httpurl; |
| 1616 | } |
| 1617 | |
| 1618 | /** |
| 02474e43 |
1619 | * This function changes the <body> tag into a <div> tag since we |
| 1620 | * can't really have a body-within-body. |
| 1621 | * |
| 1622 | * @param $attary an array of attributes and values of <body> |
| 1623 | * @param $mailbox mailbox we're currently reading (for cid2http) |
| 1624 | * @param $message current message (for cid2http) |
| 1625 | * @param $id current message id (for cid2http) |
| 1626 | * @return a modified array of attributes to be set for <div> |
| 1627 | */ |
| 2dd879b8 |
1628 | function sq_body2div($attary, $mailbox, $message, $id){ |
| b583c3e8 |
1629 | $me = 'sq_body2div'; |
| 3d8371be |
1630 | $divattary = Array('class' => "'bodyclass'"); |
| b583c3e8 |
1631 | $text = '#000000'; |
| 80b4debd |
1632 | $has_bgc_stl = $has_txt_stl = false; |
| b583c3e8 |
1633 | $styledef = ''; |
| 691a2d25 |
1634 | if (is_array($attary) && sizeof($attary) > 0){ |
| 1635 | foreach ($attary as $attname=>$attvalue){ |
| 1636 | $quotchar = substr($attvalue, 0, 1); |
| 1637 | $attvalue = str_replace($quotchar, "", $attvalue); |
| 1638 | switch ($attname){ |
| 3d8371be |
1639 | case 'background': |
| da2415c1 |
1640 | $attvalue = sq_cid2http($message, $id, |
| 2dd879b8 |
1641 | $attvalue, $mailbox); |
| 3d8371be |
1642 | $styledef .= "background-image: url('$attvalue'); "; |
| 1643 | break; |
| 1644 | case 'bgcolor': |
| 80b4debd |
1645 | $has_bgc_stl = true; |
| 3d8371be |
1646 | $styledef .= "background-color: $attvalue; "; |
| 1647 | break; |
| 1648 | case 'text': |
| 80b4debd |
1649 | $has_txt_stl = true; |
| 3d8371be |
1650 | $styledef .= "color: $attvalue; "; |
| 1651 | break; |
| 691a2d25 |
1652 | } |
| a3daaaf3 |
1653 | } |
| 80b4debd |
1654 | // Outlook defines a white bgcolor and no text color. This can lead to |
| 1655 | // white text on a white bg with certain themes. |
| 1656 | if ($has_bgc_stl && !$has_txt_stl) { |
| 1657 | $styledef .= "color: $text; "; |
| 1658 | } |
| 691a2d25 |
1659 | if (strlen($styledef) > 0){ |
| 1660 | $divattary{"style"} = "\"$styledef\""; |
| 1661 | } |
| 1662 | } |
| 1663 | return $divattary; |
| 1664 | } |
| a3daaaf3 |
1665 | |
| 691a2d25 |
1666 | /** |
| 02474e43 |
1667 | * This is the main function and the one you should actually be calling. |
| 1668 | * There are several variables you should be aware of an which need |
| 1669 | * special description. |
| 1670 | * |
| 1671 | * Since the description is quite lengthy, see it here: |
| 1672 | * http://linux.duke.edu/projects/mini/htmlfilter/ |
| 1673 | * |
| 1674 | * @param $body the string with HTML you wish to filter |
| 1675 | * @param $tag_list see description above |
| 1676 | * @param $rm_tags_with_content see description above |
| 1677 | * @param $self_closing_tags see description above |
| 1678 | * @param $force_tag_closing see description above |
| 1679 | * @param $rm_attnames see description above |
| 1680 | * @param $bad_attvals see description above |
| 1681 | * @param $add_attr_to_tag see description above |
| 1682 | * @param $message message object |
| 1683 | * @param $id message id |
| 1684 | * @return sanitized html safe to show on your pages. |
| 1685 | */ |
| da2415c1 |
1686 | function sq_sanitize($body, |
| 02474e43 |
1687 | $tag_list, |
| 1688 | $rm_tags_with_content, |
| 1689 | $self_closing_tags, |
| 1690 | $force_tag_closing, |
| 1691 | $rm_attnames, |
| 1692 | $bad_attvals, |
| 1693 | $add_attr_to_tag, |
| 1694 | $message, |
| 1695 | $id, |
| 1696 | $mailbox |
| 1697 | ){ |
| b583c3e8 |
1698 | $me = 'sq_sanitize'; |
| 7d06541f |
1699 | $rm_tags = array_shift($tag_list); |
| 691a2d25 |
1700 | /** |
| 02474e43 |
1701 | * Normalize rm_tags and rm_tags_with_content. |
| 1702 | */ |
| 7d06541f |
1703 | @array_walk($tag_list, 'sq_casenormalize'); |
| 691a2d25 |
1704 | @array_walk($rm_tags_with_content, 'sq_casenormalize'); |
| 1705 | @array_walk($self_closing_tags, 'sq_casenormalize'); |
| 1706 | /** |
| 02474e43 |
1707 | * See if tag_list is of tags to remove or tags to allow. |
| 1708 | * false means remove these tags |
| 1709 | * true means allow these tags |
| 1710 | */ |
| 691a2d25 |
1711 | $curpos = 0; |
| 1712 | $open_tags = Array(); |
| 2dd879b8 |
1713 | $trusted = "\n<!-- begin sanitized html -->\n"; |
| 691a2d25 |
1714 | $skip_content = false; |
| bb8d0799 |
1715 | /** |
| 02474e43 |
1716 | * Take care of netscape's stupid javascript entities like |
| 1717 | * &{alert('boo')}; |
| 1718 | */ |
| bb8d0799 |
1719 | $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body); |
| 691a2d25 |
1720 | |
| 7d06541f |
1721 | while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){ |
| 691a2d25 |
1722 | list($tagname, $attary, $tagtype, $lt, $gt) = $curtag; |
| 1723 | $free_content = substr($body, $curpos, $lt-$curpos); |
| 1724 | /** |
| 02474e43 |
1725 | * Take care of <style> |
| 1726 | */ |
| 7d06541f |
1727 | if ($tagname == "style" && $tagtype == 1){ |
| da2415c1 |
1728 | list($free_content, $curpos) = |
| e60a299a |
1729 | sq_fixstyle($body, $gt+1, $message, $id, $mailbox); |
| 7d06541f |
1730 | if ($free_content != FALSE){ |
| 1731 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
| 1732 | $trusted .= $free_content; |
| 1733 | $trusted .= sq_tagprint($tagname, false, 2); |
| 1734 | } |
| 1735 | continue; |
| 691a2d25 |
1736 | } |
| 1737 | if ($skip_content == false){ |
| 1738 | $trusted .= $free_content; |
| 691a2d25 |
1739 | } |
| 1740 | if ($tagname != FALSE){ |
| 1741 | if ($tagtype == 2){ |
| 1742 | if ($skip_content == $tagname){ |
| 1743 | /** |
| 02474e43 |
1744 | * Got to the end of tag we needed to remove. |
| 1745 | */ |
| 691a2d25 |
1746 | $tagname = false; |
| 1747 | $skip_content = false; |
| 1748 | } else { |
| 1749 | if ($skip_content == false){ |
| c828931c |
1750 | if ($tagname == "body"){ |
| 1751 | $tagname = "div"; |
| 2dd879b8 |
1752 | } |
| da2415c1 |
1753 | if (isset($open_tags{$tagname}) && |
| 2dd879b8 |
1754 | $open_tags{$tagname} > 0){ |
| 1755 | $open_tags{$tagname}--; |
| 691a2d25 |
1756 | } else { |
| 2dd879b8 |
1757 | $tagname = false; |
| 691a2d25 |
1758 | } |
| 691a2d25 |
1759 | } |
| 1760 | } |
| 1761 | } else { |
| 1762 | /** |
| 02474e43 |
1763 | * $rm_tags_with_content |
| 1764 | */ |
| 691a2d25 |
1765 | if ($skip_content == false){ |
| 1766 | /** |
| 02474e43 |
1767 | * See if this is a self-closing type and change |
| 1768 | * tagtype appropriately. |
| 1769 | */ |
| 691a2d25 |
1770 | if ($tagtype == 1 |
| 1771 | && in_array($tagname, $self_closing_tags)){ |
| 2dd879b8 |
1772 | $tagtype = 3; |
| 691a2d25 |
1773 | } |
| 1774 | /** |
| 02474e43 |
1775 | * See if we should skip this tag and any content |
| 1776 | * inside it. |
| 1777 | */ |
| 691a2d25 |
1778 | if ($tagtype == 1 && |
| 1779 | in_array($tagname, $rm_tags_with_content)){ |
| 1780 | $skip_content = $tagname; |
| 1781 | } else { |
| da2415c1 |
1782 | if (($rm_tags == false |
| 02474e43 |
1783 | && in_array($tagname, $tag_list)) || |
| 691a2d25 |
1784 | ($rm_tags == true && |
| 02474e43 |
1785 | !in_array($tagname, $tag_list))){ |
| 691a2d25 |
1786 | $tagname = false; |
| 1787 | } else { |
| 2dd879b8 |
1788 | /** |
| 02474e43 |
1789 | * Convert body into div. |
| 1790 | */ |
| 2dd879b8 |
1791 | if ($tagname == "body"){ |
| 1792 | $tagname = "div"; |
| da2415c1 |
1793 | $attary = sq_body2div($attary, $mailbox, |
| 02474e43 |
1794 | $message, $id); |
| 2dd879b8 |
1795 | } |
| 691a2d25 |
1796 | if ($tagtype == 1){ |
| 1797 | if (isset($open_tags{$tagname})){ |
| 1798 | $open_tags{$tagname}++; |
| 1799 | } else { |
| 1800 | $open_tags{$tagname}=1; |
| 1801 | } |
| 1802 | } |
| 1803 | /** |
| 02474e43 |
1804 | * This is where we run other checks. |
| 1805 | */ |
| 691a2d25 |
1806 | if (is_array($attary) && sizeof($attary) > 0){ |
| 1807 | $attary = sq_fixatts($tagname, |
| 02474e43 |
1808 | $attary, |
| 1809 | $rm_attnames, |
| 1810 | $bad_attvals, |
| 1811 | $add_attr_to_tag, |
| 1812 | $message, |
| 1813 | $id, |
| 1814 | $mailbox |
| 1815 | ); |
| 691a2d25 |
1816 | } |
| 1817 | } |
| 1818 | } |
| 691a2d25 |
1819 | } |
| 1820 | } |
| 1821 | if ($tagname != false && $skip_content == false){ |
| 1822 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
| 1823 | } |
| 691a2d25 |
1824 | } |
| 1825 | $curpos = $gt+1; |
| a3daaaf3 |
1826 | } |
| 691a2d25 |
1827 | $trusted .= substr($body, $curpos, strlen($body)-$curpos); |
| 1828 | if ($force_tag_closing == true){ |
| 1829 | foreach ($open_tags as $tagname=>$opentimes){ |
| 1830 | while ($opentimes > 0){ |
| 1831 | $trusted .= '</' . $tagname . '>'; |
| 1832 | $opentimes--; |
| 1833 | } |
| 1834 | } |
| 1835 | $trusted .= "\n"; |
| 1836 | } |
| 1837 | $trusted .= "<!-- end sanitized html -->\n"; |
| 1838 | return $trusted; |
| 1839 | } |
| 451f74a2 |
1840 | |
| 691a2d25 |
1841 | /** |
| 02474e43 |
1842 | * This is a wrapper function to call html sanitizing routines. |
| 1843 | * |
| 1844 | * @param $body the body of the message |
| 1845 | * @param $id the id of the message |
| 1814c39b |
1846 | * @param $message |
| 1847 | * @param $mailbox |
| 1848 | * @param boolean $take_mailto_links When TRUE, converts mailto: links |
| 1849 | * into internal SM compose links |
| 1850 | * (optional; default = TRUE) |
| 02474e43 |
1851 | * @return a string with html safe to display in the browser. |
| 1852 | */ |
| 1814c39b |
1853 | function magicHTML($body, $id, $message, $mailbox = 'INBOX', $take_mailto_links = true) { |
| 376667f7 |
1854 | |
| 1855 | require_once(SM_PATH . 'functions/url_parser.php'); // for $MailTo_PReg_Match |
| 1856 | |
| 691a2d25 |
1857 | global $attachment_common_show_images, $view_unsafe_images, |
| 02474e43 |
1858 | $has_unsafe_images; |
| 376667f7 |
1859 | |
| 691a2d25 |
1860 | /** |
| 02474e43 |
1861 | * Don't display attached images in HTML mode. |
| 1862 | */ |
| 691a2d25 |
1863 | $attachment_common_show_images = false; |
| 1864 | $tag_list = Array( |
| 02474e43 |
1865 | false, |
| 1866 | "object", |
| 1867 | "meta", |
| 1868 | "html", |
| 1869 | "head", |
| 1870 | "base", |
| 1871 | "link", |
| 1872 | "frame", |
| 1873 | "iframe", |
| 1874 | "plaintext", |
| 1875 | "marquee" |
| 1876 | ); |
| 691a2d25 |
1877 | |
| 1878 | $rm_tags_with_content = Array( |
| 02474e43 |
1879 | "script", |
| 1880 | "applet", |
| 1881 | "embed", |
| 1882 | "title", |
| 1883 | "frameset", |
| 1884 | "xml" |
| 1885 | ); |
| 691a2d25 |
1886 | |
| 1887 | $self_closing_tags = Array( |
| 1888 | "img", |
| 1889 | "br", |
| 1890 | "hr", |
| ff940ebc |
1891 | "input", |
| 1892 | "outbind" |
| 691a2d25 |
1893 | ); |
| 1894 | |
| 2dd879b8 |
1895 | $force_tag_closing = true; |
| 691a2d25 |
1896 | |
| 1897 | $rm_attnames = Array( |
| 02474e43 |
1898 | "/.*/" => |
| 1899 | Array( |
| 1900 | "/target/i", |
| 1901 | "/^on.*/i", |
| 1902 | "/^dynsrc/i", |
| 1903 | "/^data.*/i", |
| 1904 | "/^lowsrc.*/i" |
| 1905 | ) |
| 1906 | ); |
| 691a2d25 |
1907 | |
| 1908 | $secremoveimg = "../images/" . _("sec_remove_eng.png"); |
| 1909 | $bad_attvals = Array( |
| 1910 | "/.*/" => |
| 1911 | Array( |
| 0a6ec9b5 |
1912 | "/^src|background/i" => |
| 691a2d25 |
1913 | Array( |
| 02474e43 |
1914 | Array( |
| f83c60a2 |
1915 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
| 1916 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
| 1917 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
| 691a2d25 |
1918 | ), |
| 02474e43 |
1919 | Array( |
| 691a2d25 |
1920 | "\\1$secremoveimg\\2", |
| bb8d0799 |
1921 | "\\1$secremoveimg\\2", |
| 3d8371be |
1922 | "\\1$secremoveimg\\2", |
| 1923 | "\\1$secremoveimg\\2" |
| 691a2d25 |
1924 | ) |
| 1925 | ), |
| 0a6ec9b5 |
1926 | "/^href|action/i" => |
| 1927 | Array( |
| 02474e43 |
1928 | Array( |
| f83c60a2 |
1929 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
| 1930 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
| 1931 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
| 0a6ec9b5 |
1932 | ), |
| 02474e43 |
1933 | Array( |
| 2dd879b8 |
1934 | "\\1#\\1", |
| 1935 | "\\1#\\1", |
| 1936 | "\\1#\\1", |
| 1937 | "\\1#\\1" |
| 0a6ec9b5 |
1938 | ) |
| 1939 | ), |
| f83c60a2 |
1940 | "/^style/i" => |
| 691a2d25 |
1941 | Array( |
| 02474e43 |
1942 | Array( |
| f83c60a2 |
1943 | "/expression/i", |
| 1944 | "/binding/i", |
| 1945 | "/behaviou*r/i", |
| 2dd879b8 |
1946 | "/include-source/i", |
| 4eacb6ed |
1947 | "/position\s*:\s*absolute/i", |
| 7d06541f |
1948 | "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si", |
| 1949 | "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si", |
| af861a34 |
1950 | "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si", |
| 1951 | "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si" |
| 02474e43 |
1952 | ), |
| 1953 | Array( |
| bb8d0799 |
1954 | "idiocy", |
| 3d8371be |
1955 | "idiocy", |
| 1956 | "idiocy", |
| 2dd879b8 |
1957 | "idiocy", |
| 4eacb6ed |
1958 | "", |
| 2dd879b8 |
1959 | "url(\\1#\\1)", |
| 1960 | "url(\\1#\\1)", |
| 1961 | "url(\\1#\\1)", |
| af861a34 |
1962 | "url(\\1#\\1)", |
| 1963 | "\\1:url(\\2#\\3)" |
| 02474e43 |
1964 | ) |
| 1965 | ) |
| 691a2d25 |
1966 | ) |
| 1967 | ); |
| 5262d9a6 |
1968 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
| 2dd879b8 |
1969 | $view_unsafe_images = false; |
| 45071bd6 |
1970 | } |
| 691a2d25 |
1971 | if (!$view_unsafe_images){ |
| 1972 | /** |
| 02474e43 |
1973 | * Remove any references to http/https if view_unsafe_images set |
| 1974 | * to false. |
| 1975 | */ |
| 1976 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0], |
| f83c60a2 |
1977 | '/^([\'\"])\s*https*:.*([\'\"])/si'); |
| 02474e43 |
1978 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1], |
| 2dd879b8 |
1979 | "\\1$secremoveimg\\1"); |
| 02474e43 |
1980 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[0], |
| 43d678b6 |
1981 | '/url\(([\'\"])\s*https*:.*([\'\"])\)/si'); |
| 02474e43 |
1982 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[1], |
| 2dd879b8 |
1983 | "url(\\1$secremoveimg\\1)"); |
| 691a2d25 |
1984 | } |
| 451f74a2 |
1985 | |
| 691a2d25 |
1986 | $add_attr_to_tag = Array( |
| da2415c1 |
1987 | "/^a$/i" => |
| c25f2fbb |
1988 | Array('target'=>'"_blank"', |
| 02474e43 |
1989 | 'title'=>'"'._("This external link will open in a new window").'"' |
| 2dd879b8 |
1990 | ) |
| 1991 | ); |
| da2415c1 |
1992 | $trusted = sq_sanitize($body, |
| 02474e43 |
1993 | $tag_list, |
| 1994 | $rm_tags_with_content, |
| 1995 | $self_closing_tags, |
| 1996 | $force_tag_closing, |
| 1997 | $rm_attnames, |
| 1998 | $bad_attvals, |
| 1999 | $add_attr_to_tag, |
| 2000 | $message, |
| 2001 | $id, |
| 2002 | $mailbox |
| 2003 | ); |
| f83c60a2 |
2004 | if (preg_match("|$secremoveimg|i", $trusted)){ |
| 691a2d25 |
2005 | $has_unsafe_images = true; |
| da2415c1 |
2006 | } |
| ebeb8348 |
2007 | |
| 2008 | |
| 376667f7 |
2009 | // we want to parse mailto's in HTML output, change to SM compose links |
| 2010 | // this is a modified version of code from url_parser.php... but Marc is |
| 2011 | // right: we need a better filtering implementation; adding this randomly |
| 2012 | // here is not a great solution |
| 2013 | // |
| 1814c39b |
2014 | if ($take_mailto_links) { |
| 2015 | // parseUrl($trusted); // this even parses URLs inside of tags... too aggressive |
| 2016 | global $MailTo_PReg_Match; |
| 2017 | $MailTo_PReg_Match = '/mailto:' . substr($MailTo_PReg_Match, 1); |
| 2018 | if ((preg_match_all($MailTo_PReg_Match, $trusted, $regs)) && ($regs[0][0] != '')) { |
| 2019 | foreach ($regs[0] as $i => $mailto_before) { |
| 2020 | $mailto_params = $regs[10][$i]; |
| 2021 | |
| 2022 | // get rid of any tailing quote since we have to add send_to to the end |
| 2023 | // |
| 2024 | if (substr($mailto_before, strlen($mailto_before) - 1) == '"') |
| 2025 | $mailto_before = substr($mailto_before, 0, strlen($mailto_before) - 1); |
| 2026 | if (substr($mailto_params, strlen($mailto_params) - 1) == '"') |
| 2027 | $mailto_params = substr($mailto_params, 0, strlen($mailto_params) - 1); |
| 2028 | |
| 2029 | if ($regs[1][$i]) { //if there is an email addr before '?', we need to merge it with the params |
| 2030 | $to = 'to=' . $regs[1][$i]; |
| 2031 | if (strpos($mailto_params, 'to=') > -1) //already a 'to=' |
| 2032 | $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params); |
| 2033 | else { |
| 2034 | if ($mailto_params) //already some params, append to them |
| 2035 | $mailto_params .= '&' . $to; |
| 2036 | else |
| 2037 | $mailto_params .= '?' . $to; |
| 2038 | } |
| 376667f7 |
2039 | } |
| 1814c39b |
2040 | |
| 2041 | $url_str = preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params); |
| 2042 | |
| 2043 | // we'll already have target=_blank, no need to allow comp_in_new |
| 2044 | // here (which would be a lot more work anyway) |
| 2045 | // |
| 2046 | global $compose_new_win; |
| 2047 | $temp_comp_in_new = $compose_new_win; |
| 2048 | $compose_new_win = 0; |
| 2049 | $comp_uri = makeComposeLink('src/compose.php' . $url_str, $mailto_before); |
| 2050 | $compose_new_win = $temp_comp_in_new; |
| 2051 | |
| 2052 | // remove <a href=" and anything after the next quote (we only |
| 2053 | // need the uri, not the link HTML) in compose uri |
| 2054 | // |
| 2055 | $comp_uri = substr($comp_uri, 9); |
| 2056 | $comp_uri = substr($comp_uri, 0, strpos($comp_uri, '"', 1)); |
| 2057 | $trusted = str_replace($mailto_before, $comp_uri, $trusted); |
| 376667f7 |
2058 | } |
| 376667f7 |
2059 | } |
| 2060 | } |
| ebeb8348 |
2061 | |
| 691a2d25 |
2062 | return $trusted; |
| 451f74a2 |
2063 | } |
| a4a70693 |
2064 | |
| da2415c1 |
2065 | /** |
| 02474e43 |
2066 | * function SendDownloadHeaders - send file to the browser |
| 2067 | * |
| 2068 | * Original Source: SM core src/download.php |
| 2069 | * moved here to make it available to other code, and separate |
| 2070 | * front end from back end functionality. |
| 2071 | * |
| 2072 | * @param string $type0 first half of mime type |
| 2073 | * @param string $type1 second half of mime type |
| 2074 | * @param string $filename filename to tell the browser for downloaded file |
| 2075 | * @param boolean $force whether to force the download dialog to pop |
| 2076 | * @param optional integer $filesize send the Content-Header and length to the browser |
| 2077 | * @return void |
| 2078 | */ |
| 2079 | function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) { |
| 2080 | global $languages, $squirrelmail_language; |
| 2081 | $isIE = $isIE6 = 0; |
| 2082 | |
| 2083 | sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER); |
| 2084 | |
| 2085 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false && |
| 2086 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
| 2087 | $isIE = 1; |
| 2088 | } |
| 2089 | |
| 2090 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE 6') !== false && |
| 2091 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
| 2092 | $isIE6 = 1; |
| 2093 | } |
| 2094 | |
| 2095 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| 2096 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) { |
| 2097 | $filename = |
| 2098 | call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT); |
| 2099 | } else { |
| 2100 | $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename)); |
| 2101 | } |
| 2102 | |
| 2103 | // A Pox on Microsoft and it's Internet Explorer! |
| 2104 | // |
| 2105 | // IE has lots of bugs with file downloads. |
| 2106 | // It also has problems with SSL. Both of these cause problems |
| 2107 | // for us in this function. |
| 2108 | // |
| 2109 | // See this article on Cache Control headers and SSL |
| 2110 | // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308 |
| 2111 | // |
| 2112 | // The best thing you can do for IE is to upgrade to the latest |
| 2113 | // version |
| 2114 | //set all the Cache Control Headers for IE |
| 2115 | if ($isIE) { |
| 2116 | $filename=rawurlencode($filename); |
| 2117 | header ("Pragma: public"); |
| 2118 | header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); # HTTP/1.1 |
| 2119 | header ("Cache-Control: post-check=0, pre-check=0", false); |
| 2120 | header ("Cache-control: private"); |
| 2121 | |
| 2122 | //set the inline header for IE, we'll add the attachment header later if we need it |
| 2123 | header ("Content-Disposition: inline; filename=$filename"); |
| 2124 | } |
| 2125 | |
| 2126 | if (!$force) { |
| 2127 | // Try to show in browser window |
| 2128 | header ("Content-Disposition: inline; filename=\"$filename\""); |
| 2129 | header ("Content-Type: $type0/$type1; name=\"$filename\""); |
| 2130 | } else { |
| 2131 | // Try to pop up the "save as" box |
| 2132 | |
| 2133 | // IE makes this hard. It pops up 2 save boxes, or none. |
| 2134 | // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP |
| 2135 | // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519 |
| 2136 | // But, according to Microsoft, it is "RFC compliant but doesn't |
| 2137 | // take into account some deviations that allowed within the |
| 2138 | // specification." Doesn't that mean RFC non-compliant? |
| 2139 | // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP |
| 2140 | |
| 2141 | // all browsers need the application/octet-stream header for this |
| 2142 | header ("Content-Type: application/octet-stream; name=\"$filename\""); |
| 2143 | |
| 2144 | // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp |
| 2145 | // Do not have quotes around filename, but that applied to |
| 2146 | // "attachment"... does it apply to inline too? |
| 2147 | header ("Content-Disposition: attachment; filename=\"$filename\""); |
| 2148 | |
| 2149 | if ($isIE && !$isIE6) { |
| 2150 | // This combination seems to work mostly. IE 5.5 SP 1 has |
| 2151 | // known issues (see the Microsoft Knowledge Base) |
| 2152 | |
| 2153 | // This works for most types, but doesn't work with Word files |
| 2154 | header ("Content-Type: application/download; name=\"$filename\""); |
| 2155 | |
| 2156 | // These are spares, just in case. :-) |
| 2157 | //header("Content-Type: $type0/$type1; name=\"$filename\""); |
| 2158 | //header("Content-Type: application/x-msdownload; name=\"$filename\""); |
| 2159 | //header("Content-Type: application/octet-stream; name=\"$filename\""); |
| 2160 | } else { |
| 2161 | // another application/octet-stream forces download for Netscape |
| 2162 | header ("Content-Type: application/octet-stream; name=\"$filename\""); |
| 2163 | } |
| 2164 | } |
| 2165 | |
| 2166 | //send the content-length header if the calling function provides it |
| 2167 | if ($filesize > 0) { |
| 2168 | header("Content-Length: $filesize"); |
| 2169 | } |
| 07c49f57 |
2170 | |
| 8d863f64 |
2171 | } // end fn SendDownloadHeaders |
| da2415c1 |
2172 | |
| c8f5f606 |
2173 | ?> |
projects / squirrelmail.git / blame