59177427 |
1 | <?php |
2ba13803 |
2 | |
35586184 |
3 | /** |
8bd0068d |
4 | * mime.php |
5 | * |
8bd0068d |
6 | * This contains the functions necessary to detect and decode MIME |
7 | * messages. |
8 | * |
47ccfad4 |
9 | * @copyright © 1999-2006 The SquirrelMail Project Team |
4b4abf93 |
10 | * @license http://opensource.org/licenses/gpl-license.php GNU Public License |
8bd0068d |
11 | * @version $Id$ |
12 | * @package squirrelmail |
13 | */ |
b74ba498 |
14 | |
202bcbcc |
15 | /** |
16 | * dependency information |
17 | functions dependency |
18 | mime_structure |
19 | class/mime/Message.class.php |
20 | Message::parseStructure |
21 | functions/page_header.php |
22 | displayPageHeader |
23 | functions/display_messages.php |
24 | plain_error_message |
25 | mime_fetch_body |
26 | functions/imap_general.php |
27 | sqimap_run_command |
28 | mime_print_body_lines |
29 | |
30 | |
31 | |
32 | functions/imap.php |
33 | functions/attachment_common.php |
34 | functions/display_messages.php |
35 | |
36 | magicHtml => url_parser |
37 | translateText => url_parser |
38 | |
39 | */ |
40 | |
8beafbbc |
41 | |
7c7b74b3 |
42 | /* -------------------------------------------------------------------------- */ |
43 | /* MIME DECODING */ |
44 | /* -------------------------------------------------------------------------- */ |
b74ba498 |
45 | |
d6c32258 |
46 | /** |
8bd0068d |
47 | * Get the MIME structure |
48 | * |
49 | * This function gets the structure of a message and stores it in the "message" class. |
50 | * It will return this object for use with all relevant header information and |
51 | * fully parsed into the standard "message" object format. |
52 | */ |
a4a70693 |
53 | function mime_structure ($bodystructure, $flags=array()) { |
c9d78ab4 |
54 | |
3d8371be |
55 | /* Isolate the body structure and remove beginning and end parenthesis. */ |
a4a70693 |
56 | $read = trim(substr ($bodystructure, strpos(strtolower($bodystructure), 'bodystructure') + 13)); |
451f74a2 |
57 | $read = trim(substr ($read, 0, -1)); |
22efa9fb |
58 | $i = 0; |
59 | $msg = Message::parseStructure($read,$i); |
9de42168 |
60 | if (!is_object($msg)) { |
3d8371be |
61 | global $color, $mailbox; |
c96c32f4 |
62 | /* removed urldecode because $_GET is auto urldecoded ??? */ |
a48eba8f |
63 | displayPageHeader( $color, $mailbox ); |
5e8de8b6 |
64 | $errormessage = _("SquirrelMail could not decode the bodystructure of the message"); |
8bd0068d |
65 | $errormessage .= '<br />'._("The bodystructure provided by your IMAP server:").'<br /><br />'; |
472e7acb |
66 | $errormessage .= '<pre>' . htmlspecialchars($read) . '</pre>'; |
9de42168 |
67 | plain_error_message( $errormessage, $color ); |
3d8371be |
68 | echo '</body></html>'; |
9de42168 |
69 | exit; |
70 | } |
a4a70693 |
71 | if (count($flags)) { |
7a9e9c89 |
72 | foreach ($flags as $flag) { |
73 | $char = strtoupper($flag{1}); |
74 | switch ($char) { |
3d8371be |
75 | case 'S': |
76 | if (strtolower($flag) == '\\seen') { |
77 | $msg->is_seen = true; |
78 | } |
79 | break; |
80 | case 'A': |
81 | if (strtolower($flag) == '\\answered') { |
82 | $msg->is_answered = true; |
83 | } |
84 | break; |
85 | case 'D': |
86 | if (strtolower($flag) == '\\deleted') { |
87 | $msg->is_deleted = true; |
88 | } |
89 | break; |
90 | case 'F': |
91 | if (strtolower($flag) == '\\flagged') { |
92 | $msg->is_flagged = true; |
93 | } |
94 | break; |
95 | case 'M': |
96 | if (strtolower($flag) == '$mdnsent') { |
97 | $msg->is_mdnsent = true; |
98 | } |
99 | break; |
100 | default: |
101 | break; |
7a9e9c89 |
102 | } |
103 | } |
451f74a2 |
104 | } |
7a9e9c89 |
105 | // listEntities($msg); |
3d8371be |
106 | return $msg; |
451f74a2 |
107 | } |
b74ba498 |
108 | |
22efa9fb |
109 | |
110 | |
3d8371be |
111 | /* This starts the parsing of a particular structure. It is called recursively, |
8bd0068d |
112 | * so it can be passed different structures. It returns an object of type |
113 | * $message. |
114 | * First, it checks to see if it is a multipart message. If it is, then it |
115 | * handles that as it sees is necessary. If it is just a regular entity, |
116 | * then it parses it and adds the necessary header information (by calling out |
117 | * to mime_get_elements() |
118 | */ |
451f74a2 |
119 | |
4d592352 |
120 | function mime_fetch_body($imap_stream, $id, $ent_id=1, $fetch_size=0) { |
3d8371be |
121 | /* Do a bit of error correction. If we couldn't find the entity id, just guess |
8bd0068d |
122 | * that it is the first one. That is usually the case anyway. |
123 | */ |
7c7b74b3 |
124 | |
09a4bde3 |
125 | if (!$ent_id) { |
08b7f7cc |
126 | $cmd = "FETCH $id BODY[]"; |
1035e159 |
127 | } else { |
08b7f7cc |
128 | $cmd = "FETCH $id BODY[$ent_id]"; |
09a4bde3 |
129 | } |
3d8371be |
130 | |
4d592352 |
131 | if ($fetch_size!=0) $cmd .= "<0.$fetch_size>"; |
da2415c1 |
132 | |
6201339c |
133 | $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, TRUE); |
77b88425 |
134 | do { |
3d8371be |
135 | $topline = trim(array_shift($data)); |
136 | } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ; |
a4a70693 |
137 | |
451f74a2 |
138 | $wholemessage = implode('', $data); |
139 | if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) { |
3d8371be |
140 | $ret = substr($wholemessage, 0, $regs[1]); |
141 | /* There is some information in the content info header that could be important |
8bd0068d |
142 | * in order to parse html messages. Let's get them here. |
143 | */ |
0600bdf1 |
144 | // if ($ret{0} == '<') { |
6201339c |
145 | // $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, TRUE); |
0600bdf1 |
146 | // } |
451f74a2 |
147 | } else if (ereg('"([^"]*)"', $topline, $regs)) { |
148 | $ret = $regs[1]; |
149 | } else { |
150 | global $where, $what, $mailbox, $passed_id, $startMessage; |
3d8371be |
151 | $par = 'mailbox=' . urlencode($mailbox) . '&passed_id=' . $passed_id; |
451f74a2 |
152 | if (isset($where) && isset($what)) { |
3d8371be |
153 | $par .= '&where=' . urlencode($where) . '&what=' . urlencode($what); |
a3daaaf3 |
154 | } else { |
3d8371be |
155 | $par .= '&startMessage=' . $startMessage . '&show_more=0'; |
451f74a2 |
156 | } |
e5ea9327 |
157 | $par .= '&response=' . urlencode($response) . |
8bd0068d |
158 | '&message=' . urlencode($message) . |
159 | '&topline=' . urlencode($topline); |
a019eeb8 |
160 | |
8bd0068d |
161 | echo '<tt><br />' . |
02474e43 |
162 | '<table width="80%"><tr>' . |
163 | '<tr><td colspan="2">' . |
164 | _("Body retrieval error. The reason for this is most probably that the message is malformed.") . |
165 | '</td></tr>' . |
166 | '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" . |
167 | '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" . |
168 | '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" . |
169 | '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" . |
170 | "</table><br /></tt></font><hr />"; |
346817d4 |
171 | |
6201339c |
172 | $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, TRUE); |
451f74a2 |
173 | array_shift($data); |
174 | $wholemessage = implode('', $data); |
a019eeb8 |
175 | |
346817d4 |
176 | $ret = $wholemessage; |
a3daaaf3 |
177 | } |
3d8371be |
178 | return $ret; |
451f74a2 |
179 | } |
d4467150 |
180 | |
6cc670de |
181 | function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding, $rStream='php://stdout') { |
1035e159 |
182 | |
3d8371be |
183 | /* Don't kill the connection if the browser is over a dialup |
8bd0068d |
184 | * and it would take over 30 seconds to download it. |
185 | * Don't call set_time_limit in safe mode. |
186 | */ |
b7206e1d |
187 | |
3d8371be |
188 | if (!ini_get('safe_mode')) { |
b7206e1d |
189 | set_time_limit(0); |
190 | } |
7c7b74b3 |
191 | /* in case of base64 encoded attachments, do not buffer them. |
8bd0068d |
192 | Instead, echo the decoded attachment directly to screen */ |
7c7b74b3 |
193 | if (strtolower($encoding) == 'base64') { |
194 | if (!$ent_id) { |
7591f143 |
195 | $query = "FETCH $id BODY[]"; |
7c7b74b3 |
196 | } else { |
7591f143 |
197 | $query = "FETCH $id BODY[$ent_id]"; |
7c7b74b3 |
198 | } |
7591f143 |
199 | sqimap_run_command($imap_stream,$query,true,$response,$message,TRUE,'sqimap_base64_decode',$rStream,true); |
1d142b8d |
200 | } else { |
7591f143 |
201 | $body = mime_fetch_body ($imap_stream, $id, $ent_id); |
202 | if (is_resource($rStream)) { |
203 | fputs($rStream,decodeBody($body,$encoding)); |
204 | } else { |
205 | echo decodeBody($body, $encoding); |
206 | } |
1d142b8d |
207 | } |
346817d4 |
208 | |
da2415c1 |
209 | /* |
8bd0068d |
210 | TODO, use the same method for quoted printable. |
211 | However, I assume that quoted printable attachments aren't that large |
212 | so the performancegain / memory usage drop will be minimal. |
213 | If we decide to add that then we need to adapt sqimap_fread because |
214 | we need to split te result on \n and fread doesn't stop at \n. That |
215 | means we also should provide $results from sqimap_fread (by ref) to |
216 | te function and set $no_return to false. The $filter function for |
217 | quoted printable should handle unsetting of $results. |
218 | */ |
da2415c1 |
219 | /* |
8bd0068d |
220 | TODO 2: find out how we write to the output stream php://stdout. fwrite |
221 | doesn't work because 'php://stdout isn't a stream. |
222 | */ |
7c7b74b3 |
223 | |
5d9c6f73 |
224 | return; |
451f74a2 |
225 | } |
beb9e459 |
226 | |
451f74a2 |
227 | /* -[ END MIME DECODING ]----------------------------------------------------------- */ |
d4467150 |
228 | |
3d8371be |
229 | /* This is here for debugging purposes. It will print out a list |
8bd0068d |
230 | * of all the entity IDs that are in the $message object. |
231 | */ |
451f74a2 |
232 | function listEntities ($message) { |
3d8371be |
233 | if ($message) { |
3c621ba1 |
234 | echo "<tt>" . $message->entity_id . ' : ' . $message->type0 . '/' . $message->type1 . ' parent = '. $message->parent->entity_id. '<br />'; |
3d8371be |
235 | for ($i = 0; isset($message->entities[$i]); $i++) { |
236 | echo "$i : "; |
237 | $msg = listEntities($message->entities[$i]); |
238 | |
239 | if ($msg) { |
240 | echo "return: "; |
241 | return $msg; |
242 | } |
243 | } |
a4a70693 |
244 | } |
451f74a2 |
245 | } |
f0c4dc12 |
246 | |
f792c641 |
247 | function getPriorityStr($priority) { |
3d8371be |
248 | $priority_level = substr($priority,0,1); |
249 | |
250 | switch($priority_level) { |
251 | /* Check for a higher then normal priority. */ |
252 | case '1': |
253 | case '2': |
254 | $priority_string = _("High"); |
255 | break; |
256 | |
257 | /* Check for a lower then normal priority. */ |
258 | case '4': |
259 | case '5': |
260 | $priority_string = _("Low"); |
261 | break; |
262 | |
263 | /* Check for a normal priority. */ |
264 | case '3': |
265 | default: |
266 | $priority_level = '3'; |
267 | $priority_string = _("Normal"); |
268 | break; |
269 | |
270 | } |
271 | return $priority_string; |
f792c641 |
272 | } |
273 | |
451f74a2 |
274 | /* returns a $message object for a particular entity id */ |
275 | function getEntity ($message, $ent_id) { |
a4a70693 |
276 | return $message->getEntity($ent_id); |
451f74a2 |
277 | } |
8beafbbc |
278 | |
3d8371be |
279 | /* translateText |
8bd0068d |
280 | * Extracted from strings.php 23/03/2002 |
281 | */ |
da4c66e8 |
282 | |
283 | function translateText(&$body, $wrap_at, $charset) { |
3d8371be |
284 | global $where, $what; /* from searching */ |
285 | global $color; /* color theme */ |
da4c66e8 |
286 | |
202bcbcc |
287 | // require_once(SM_PATH . 'functions/url_parser.php'); |
da4c66e8 |
288 | |
289 | $body_ary = explode("\n", $body); |
da4c66e8 |
290 | for ($i=0; $i < count($body_ary); $i++) { |
291 | $line = $body_ary[$i]; |
292 | if (strlen($line) - 2 >= $wrap_at) { |
c7aff938 |
293 | sqWordWrap($line, $wrap_at, $charset); |
da4c66e8 |
294 | } |
295 | $line = charset_decode($charset, $line); |
296 | $line = str_replace("\t", ' ', $line); |
297 | |
298 | parseUrl ($line); |
299 | |
3d8371be |
300 | $quotes = 0; |
da4c66e8 |
301 | $pos = 0; |
3d8371be |
302 | $j = strlen($line); |
da4c66e8 |
303 | |
3d8371be |
304 | while ($pos < $j) { |
da4c66e8 |
305 | if ($line[$pos] == ' ') { |
3d8371be |
306 | $pos++; |
da4c66e8 |
307 | } else if (strpos($line, '>', $pos) === $pos) { |
308 | $pos += 4; |
3d8371be |
309 | $quotes++; |
da4c66e8 |
310 | } else { |
311 | break; |
312 | } |
313 | } |
3d8371be |
314 | |
83c94382 |
315 | if ($quotes % 2) { |
3d8371be |
316 | if (!isset($color[13])) { |
da4c66e8 |
317 | $color[13] = '#800000'; |
318 | } |
4c25967c |
319 | $line = '<font color="' . $color[13] . '">' . $line . '</font>'; |
320 | } elseif ($quotes) { |
321 | if (!isset($color[14])) { |
322 | $color[14] = '#FF0000'; |
323 | } |
324 | $line = '<font color="' . $color[14] . '">' . $line . '</font>'; |
da4c66e8 |
325 | } |
3d8371be |
326 | |
da4c66e8 |
327 | $body_ary[$i] = $line; |
328 | } |
329 | $body = '<pre>' . implode("\n", $body_ary) . '</pre>'; |
330 | } |
331 | |
a2bfcbce |
332 | /** |
da1b55ad |
333 | * This returns a parsed string called $body. That string can then |
334 | * be displayed as the actual message in the HTML. It contains |
335 | * everything needed, including HTML Tags, Attachments at the |
336 | * bottom, etc. |
f8a1ed5a |
337 | * |
da1b55ad |
338 | * Since 1.2.0 function uses message_body hook. |
339 | * Till 1.3.0 function included output of formatAttachments(). |
340 | * |
341 | * @param resource $imap_stream imap connection resource |
342 | * @param object $message squirrelmail message object |
343 | * @param array $color squirrelmail color theme array |
344 | * @param integer $wrap_at number of characters per line |
345 | * @param string $ent_num (since 1.3.0) message part id |
346 | * @param integer $id (since 1.3.0) message id |
347 | * @param string $mailbox (since 1.3.0) imap folder name |
41f701c1 |
348 | * @param boolean $clean (since 1.5.1) Do not output stuff that's irrelevant for the printable version. |
da1b55ad |
349 | * @return string html formated message text |
350 | */ |
a2bfcbce |
351 | function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) { |
3d8371be |
352 | /* This if statement checks for the entity to show as the |
8bd0068d |
353 | * primary message. To add more of them, just put them in the |
354 | * order that is their priority. |
355 | */ |
ce68b76b |
356 | global $startMessage, $languages, $squirrelmail_language, |
40a34e57 |
357 | $show_html_default, $sort, $has_unsafe_images, $passed_ent_id, |
358 | $use_iframe, $iframe_height, $download_and_unsafe_link; |
2c25d36a |
359 | |
360 | // workaround for not updated config.php |
361 | if (! isset($use_iframe)) $use_iframe = false; |
77bfbd2e |
362 | |
5262d9a6 |
363 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
364 | $view_unsafe_images = false; |
77bfbd2e |
365 | } |
d03c24f4 |
366 | |
cc34b00d |
367 | $body = ''; |
23bcec6f |
368 | $urlmailbox = urlencode($mailbox); |
451f74a2 |
369 | $body_message = getEntity($message, $ent_num); |
370 | if (($body_message->header->type0 == 'text') || |
8bd0068d |
371 | ($body_message->header->type0 == 'rfc822')) { |
3d8371be |
372 | $body = mime_fetch_body ($imap_stream, $id, $ent_num); |
451f74a2 |
373 | $body = decodeBody($body, $body_message->header->encoding); |
e842b215 |
374 | |
375 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
8bd0068d |
376 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) { |
e842b215 |
377 | if (mb_detect_encoding($body) != 'ASCII') { |
33a55f5a |
378 | $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body); |
e842b215 |
379 | } |
380 | } |
451f74a2 |
381 | $hookResults = do_hook("message_body", $body); |
382 | $body = $hookResults[1]; |
23bcec6f |
383 | |
3d8371be |
384 | /* If there are other types that shouldn't be formatted, add |
8bd0068d |
385 | * them here. |
386 | */ |
3d8371be |
387 | |
451f74a2 |
388 | if ($body_message->header->type1 == 'html') { |
3d8371be |
389 | if ($show_html_default <> 1) { |
85015544 |
390 | $entity_conv = array(' ' => ' ', |
8bd0068d |
391 | '<p>' => "\n", |
392 | '<P>' => "\n", |
393 | '<br>' => "\n", |
394 | '<BR>' => "\n", |
395 | '<br />' => "\n", |
396 | '<BR />' => "\n", |
397 | '>' => '>', |
398 | '<' => '<'); |
85015544 |
399 | $body = strtr($body, $entity_conv); |
3d8371be |
400 | $body = strip_tags($body); |
85015544 |
401 | $body = trim($body); |
402 | translateText($body, $wrap_at, |
8bd0068d |
403 | $body_message->header->getParameter('charset')); |
2c25d36a |
404 | } elseif ($use_iframe && ! $clean) { |
405 | // $clean is used to remove iframe in printable view. |
406 | |
84410f31 |
407 | /** |
408 | * If we don't add html message between iframe tags, |
409 | * we must detect unsafe images and modify $has_unsafe_images. |
f8a1ed5a |
410 | */ |
758a7889 |
411 | $html_body = magicHTML($body, $id, $message, $mailbox); |
b6c52e61 |
412 | // Convert character set in order to display html mails in different character set |
413 | $html_body = charset_decode($body_message->header->getParameter('charset'),$html_body,false,true); |
84410f31 |
414 | |
2c25d36a |
415 | // creating iframe url |
416 | $iframeurl=sqm_baseuri().'src/view_html.php?' |
f8a1ed5a |
417 | . 'mailbox=' . $urlmailbox |
2c25d36a |
418 | . '&passed_id=' . $id |
419 | . '&ent_id=' . $ent_num |
420 | . '&view_unsafe_images=' . (int) $view_unsafe_images; |
421 | |
422 | // adding warning message |
423 | $body = html_tag('div',_("Viewing HTML formatted email"),'center'); |
424 | |
425 | /** |
f8a1ed5a |
426 | * height can't be set to 100%, because it does not work as expected when |
427 | * iframe is inside the table. Browsers do not create full height objects |
428 | * even when iframe is not nested. Maybe there is some way to get full size |
2c25d36a |
429 | * with CSS. Tested in firefox 1.02 and opera 7.53 |
430 | * |
431 | * width="100%" does not work as expected, when table width is not set (automatic) |
432 | * |
433 | * tokul: I think <iframe> are safer sandbox than <object>. Objects might |
434 | * need special handling for IE and IE6SP2. |
435 | */ |
436 | $body.= "<div><iframe name=\"message_frame\" width=\"100%\" height=\"$iframe_height\" src=\"$iframeurl\"" |
5d374baf |
437 | .' frameborder="1" marginwidth="0" marginheight="0" scrolling="auto">' . "\n"; |
2c25d36a |
438 | |
439 | // Message for browsers without iframe support |
f8a1ed5a |
440 | //$body.= _("Your browser does not support inline frames. |
84410f31 |
441 | // You can view HTML formated message by following below link."); |
f8a1ed5a |
442 | //$body.= "<br /><a href=\"$iframeurl\">"._("View HTML Message")."</a>"; |
84410f31 |
443 | |
444 | // if browser can't render iframe, it renders html message. |
445 | $body.= $html_body; |
2c25d36a |
446 | |
447 | // close iframe |
448 | $body.="</iframe></div>\n"; |
a3daaaf3 |
449 | } else { |
2c25d36a |
450 | // old way of html rendering |
f8a1ed5a |
451 | $body = magicHTML($body, $id, $message, $mailbox); |
b6c52e61 |
452 | /** |
758a7889 |
453 | * convert character set. charset_decode does not remove html special chars |
b6c52e61 |
454 | * applied by magicHTML functions and does not sanitize them second time if |
758a7889 |
455 | * fourth argument is true. |
456 | */ |
b6c52e61 |
457 | $body = charset_decode($body_message->header->getParameter('charset'),$body,false,true); |
a3daaaf3 |
458 | } |
451f74a2 |
459 | } else { |
3d8371be |
460 | translateText($body, $wrap_at, |
8bd0068d |
461 | $body_message->header->getParameter('charset')); |
451f74a2 |
462 | } |
a2bfcbce |
463 | |
464 | // if this is the clean display (i.e. printer friendly), stop here. |
465 | if ( $clean ) { |
466 | return $body; |
467 | } |
468 | |
40a34e57 |
469 | $download_and_unsafe_link = ''; |
470 | |
83cf04bd |
471 | $link = 'passed_id=' . $id . '&ent_id='.$ent_num. |
8bd0068d |
472 | '&mailbox=' . $urlmailbox .'&sort=' . $sort . |
473 | '&startMessage=' . $startMessage . '&show_more=0'; |
08b7f7cc |
474 | if (isset($passed_ent_id)) { |
475 | $link .= '&passed_ent_id='.$passed_ent_id; |
476 | } |
40a34e57 |
477 | $download_and_unsafe_link .= ' | <a href="download.php?absolute_dl=true&' . |
8bd0068d |
478 | $link . '">' . _("Download this as a file") . '</a>'; |
7aad7b77 |
479 | if ($view_unsafe_images) { |
23f617b8 |
480 | $text = _("Hide Unsafe Images"); |
7aad7b77 |
481 | } else { |
08b7f7cc |
482 | if (isset($has_unsafe_images) && $has_unsafe_images) { |
483 | $link .= '&view_unsafe_images=1'; |
484 | $text = _("View Unsafe Images"); |
485 | } else { |
486 | $text = ''; |
487 | } |
3d8371be |
488 | } |
83cf04bd |
489 | if($text != '') { |
40a34e57 |
490 | $download_and_unsafe_link .= ' | <a href="read_body.php?' . $link . '">' . $text . '</a>'; |
83cf04bd |
491 | } |
3d8371be |
492 | } |
493 | return $body; |
451f74a2 |
494 | } |
b74ba498 |
495 | |
da1b55ad |
496 | /** |
497 | * Displays attachment links and information |
da1b55ad |
498 | * |
499 | * Since 1.3.0 function is not included in formatBody() call. |
500 | * |
501 | * Since 1.0.2 uses attachment $type0/$type1 hook. |
502 | * Since 1.2.5 uses attachment $type0/* hook. |
503 | * Since 1.5.0 uses attachments_bottom hook. |
f8a1ed5a |
504 | * |
da1b55ad |
505 | * @param object $message SquirrelMail message object |
506 | * @param array $exclude_id message parts that are not attachments. |
507 | * @param string $mailbox mailbox name |
508 | * @param integer $id message id |
509 | * @return string html formated attachment information. |
510 | */ |
23bcec6f |
511 | function formatAttachments($message, $exclude_id, $mailbox, $id) { |
202bcbcc |
512 | global $where, $what, $startMessage, $color, $passed_ent_id, $base_uri; |
451f74a2 |
513 | |
23bcec6f |
514 | $att_ar = $message->getAttachments($exclude_id); |
451f74a2 |
515 | |
23bcec6f |
516 | if (!count($att_ar)) return ''; |
2e25760a |
517 | |
8f5425ec |
518 | $attachments = ''; |
2e25760a |
519 | |
23bcec6f |
520 | $urlMailbox = urlencode($mailbox); |
521 | |
522 | foreach ($att_ar as $att) { |
fdc9d9b5 |
523 | $ent = $att->entity_id; |
2e25760a |
524 | $header = $att->header; |
f0c4dc12 |
525 | $type0 = strtolower($header->type0); |
526 | $type1 = strtolower($header->type1); |
2e25760a |
527 | $name = ''; |
21dab2dc |
528 | $links['download link']['text'] = _("Download"); |
202bcbcc |
529 | $links['download link']['href'] = $base_uri . |
8bd0068d |
530 | "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent"; |
2e25760a |
531 | if ($type0 =='message' && $type1 == 'rfc822') { |
202bcbcc |
532 | $default_page = $base_uri . 'src/read_body.php'; |
2e25760a |
533 | $rfc822_header = $att->rfc822_header; |
098ea084 |
534 | $filename = $rfc822_header->subject; |
6cc08d8b |
535 | if (trim( $filename ) == '') { |
536 | $filename = 'untitled-[' . $ent . ']' ; |
08b7f7cc |
537 | } |
2e25760a |
538 | $from_o = $rfc822_header->from; |
539 | if (is_object($from_o)) { |
04ea844e |
540 | $from_name = decodeHeader($from_o->getAddress(false)); |
7e697748 |
541 | } elseif (is_array($from_o) && count($from_o) && is_object($from_o[0])) { |
542 | // something weird happens when a digest message is opened and you return to the digest |
543 | // now the from object is part of an array. Probably the parseHeader call overwrites the info |
544 | // retrieved from the bodystructure in a different way. We need to fix this later. |
545 | // possible starting point, do not fetch header we already have and inspect how |
546 | // the rfc822_header object behaves. |
547 | $from_name = decodeHeader($from_o[0]->getAddress(false)); |
2e25760a |
548 | } else { |
549 | $from_name = _("Unknown sender"); |
f0c4dc12 |
550 | } |
3d8371be |
551 | $description = $from_name; |
23bcec6f |
552 | } else { |
202bcbcc |
553 | $default_page = $base_uri . 'src/download.php'; |
02474e43 |
554 | $filename = $att->getFilename(); |
f810c0b2 |
555 | if ($header->description) { |
098ea084 |
556 | $description = decodeHeader($header->description); |
f810c0b2 |
557 | } else { |
3d8371be |
558 | $description = ''; |
559 | } |
2e25760a |
560 | } |
561 | |
562 | $display_filename = $filename; |
3d8371be |
563 | if (isset($passed_ent_id)) { |
564 | $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id; |
565 | } else { |
566 | $passed_ent_id_link = ''; |
567 | } |
568 | $defaultlink = $default_page . "?startMessage=$startMessage" |
8bd0068d |
569 | . "&passed_id=$id&mailbox=$urlMailbox" |
570 | . '&ent_id='.$ent.$passed_ent_id_link; |
2e25760a |
571 | if ($where && $what) { |
8bd0068d |
572 | $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what); |
2e25760a |
573 | } |
21dab2dc |
574 | |
3d8371be |
575 | /* This executes the attachment hook with a specific MIME-type. |
8bd0068d |
576 | * If that doesn't have results, it tries if there's a rule |
577 | * for a more generic type. |
578 | */ |
3d8371be |
579 | $hookresults = do_hook("attachment $type0/$type1", $links, |
8bd0068d |
580 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
581 | $display_filename, $where, $what); |
3d8371be |
582 | if(count($hookresults[1]) <= 1) { |
583 | $hookresults = do_hook("attachment $type0/*", $links, |
8bd0068d |
584 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
585 | $display_filename, $where, $what); |
2e25760a |
586 | } |
451f74a2 |
587 | |
3d8371be |
588 | $links = $hookresults[1]; |
589 | $defaultlink = $hookresults[6]; |
77b88425 |
590 | |
3c621ba1 |
591 | $attachments .= '<tr><td>' . |
8bd0068d |
592 | '<a href="'.$defaultlink.'">'.decodeHeader($display_filename).'</a> </td>' . |
593 | '<td><small><b>' . show_readable_size($header->size) . |
594 | '</b> </small></td>' . |
595 | '<td><small>[ '.htmlspecialchars($type0).'/'.htmlspecialchars($type1).' ] </small></td>' . |
596 | '<td><small>'; |
f810c0b2 |
597 | $attachments .= '<b>' . $description . '</b>'; |
3c621ba1 |
598 | $attachments .= '</small></td><td><small> '; |
b74ba498 |
599 | |
3d8371be |
600 | $skipspaces = 1; |
601 | foreach ($links as $val) { |
602 | if ($skipspaces) { |
603 | $skipspaces = 0; |
2e25760a |
604 | } else { |
605 | $attachments .= ' | '; |
77b88425 |
606 | } |
f8a1ed5a |
607 | $attachments .= '<a href="' . $val['href'] . '">' |
608 | . (isset($val['text']) && !empty($val['text']) ? $val['text'] : '') |
609 | . (isset($val['extra']) && !empty($val['extra']) ? $val['extra'] : '') |
2c25d36a |
610 | . '</a>'; |
2e25760a |
611 | } |
3d8371be |
612 | unset($links); |
3c621ba1 |
613 | $attachments .= "</td></tr>\n"; |
2e25760a |
614 | } |
9ad17edb |
615 | $attachmentadd = do_hook_function('attachments_bottom',$attachments); |
616 | if ($attachmentadd != '') |
617 | $attachments = $attachmentadd; |
2e25760a |
618 | return $attachments; |
451f74a2 |
619 | } |
b74ba498 |
620 | |
7c7b74b3 |
621 | function sqimap_base64_decode(&$string) { |
7c0ec1d8 |
622 | |
b17a8968 |
623 | // Base64 encoded data goes in pairs of 4 bytes. To achieve on the |
7c0ec1d8 |
624 | // fly decoding (to reduce memory usage) you have to check if the |
625 | // data has incomplete pairs |
626 | |
b17a8968 |
627 | // Remove the noise in order to check if the 4 bytes pairs are complete |
7c0ec1d8 |
628 | $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string); |
629 | |
42ce44f8 |
630 | $sStringRem = ''; |
7c0ec1d8 |
631 | $iMod = strlen($string) % 4; |
632 | if ($iMod) { |
633 | $sStringRem = substr($string,-$iMod); |
b17a8968 |
634 | // Check if $sStringRem contains padding characters |
7c0ec1d8 |
635 | if (substr($sStringRem,-1) != '=') { |
636 | $string = substr($string,0,-$iMod); |
637 | } else { |
638 | $sStringRem = ''; |
639 | } |
640 | } |
7c7b74b3 |
641 | $string = base64_decode($string); |
7c0ec1d8 |
642 | return $sStringRem; |
7c7b74b3 |
643 | } |
644 | |
fdf7cef1 |
645 | /** |
646 | * Decodes encoded message body |
647 | * |
648 | * This function decodes the body depending on the encoding type. |
649 | * Currently quoted-printable and base64 encodings are supported. |
650 | * decode_body hook was added to this function in 1.4.2/1.5.0 |
651 | * @param string $body encoded message body |
652 | * @param string $encoding used encoding |
653 | * @return string decoded string |
654 | * @since 1.0 |
655 | */ |
451f74a2 |
656 | function decodeBody($body, $encoding) { |
83be314a |
657 | |
b583c3e8 |
658 | $body = str_replace("\r\n", "\n", $body); |
659 | $encoding = strtolower($encoding); |
3d8371be |
660 | |
5166f86a |
661 | $encoding_handler = do_hook_function('decode_body', $encoding); |
662 | |
663 | |
664 | // plugins get first shot at decoding the body |
665 | // |
666 | if (!empty($encoding_handler) && function_exists($encoding_handler)) { |
667 | $body = $encoding_handler('decode', $body); |
668 | |
fdf7cef1 |
669 | } elseif ($encoding == 'quoted-printable' || |
8bd0068d |
670 | $encoding == 'quoted_printable') { |
fdf7cef1 |
671 | /** |
672 | * quoted_printable_decode() function is broken in older |
673 | * php versions. Text with \r\n decoding was fixed only |
758a7889 |
674 | * in php 4.3.0. Minimal code requirement 4.0.4 + |
fdf7cef1 |
675 | * str_replace("\r\n", "\n", $body); call. |
676 | */ |
b583c3e8 |
677 | $body = quoted_printable_decode($body); |
fdf7cef1 |
678 | } elseif ($encoding == 'base64') { |
b583c3e8 |
679 | $body = base64_decode($body); |
680 | } |
3d8371be |
681 | |
b583c3e8 |
682 | // All other encodings are returned raw. |
3d8371be |
683 | return $body; |
451f74a2 |
684 | } |
685 | |
9f7f68c3 |
686 | /** |
8bd0068d |
687 | * Decodes headers |
688 | * |
689 | * This functions decode strings that is encoded according to |
690 | * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text). |
691 | * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002 |
692 | * |
693 | * @param string $string header string that has to be made readable |
694 | * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true |
695 | * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true |
696 | * @param boolean $decide decide if string can be utfencoded. defaults to false |
697 | * @return string decoded header string |
698 | */ |
9f7f68c3 |
699 | function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) { |
d6f584fc |
700 | global $languages, $squirrelmail_language,$default_charset; |
79e07c7e |
701 | if (is_array($string)) { |
702 | $string = implode("\n", $string); |
703 | } |
da2415c1 |
704 | |
10dec454 |
705 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
8bd0068d |
706 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) { |
33a55f5a |
707 | $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string); |
08b7f7cc |
708 | // Do we need to return at this point? |
709 | // return $string; |
83be314a |
710 | } |
79e07c7e |
711 | $i = 0; |
08b7f7cc |
712 | $iLastMatch = -2; |
db65b6b0 |
713 | $encoded = true; |
0a06275a |
714 | |
098ea084 |
715 | $aString = explode(' ',$string); |
08b7f7cc |
716 | $ret = ''; |
098ea084 |
717 | foreach ($aString as $chunk) { |
358a78a1 |
718 | if ($encoded && $chunk === '') { |
08b7f7cc |
719 | continue; |
358a78a1 |
720 | } elseif ($chunk === '') { |
08b7f7cc |
721 | $ret .= ' '; |
722 | continue; |
723 | } |
098ea084 |
724 | $encoded = false; |
08b7f7cc |
725 | /* if encoded words are not separated by a linear-space-white we still catch them */ |
726 | $j = $i-1; |
7e6ca3e8 |
727 | |
08b7f7cc |
728 | while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) { |
729 | /* if the last chunk isn't an encoded string then put back the space, otherwise don't */ |
730 | if ($iLastMatch !== $j) { |
731 | if ($htmlsave) { |
9f7f68c3 |
732 | $ret .= ' '; |
08b7f7cc |
733 | } else { |
734 | $ret .= ' '; |
735 | } |
736 | } |
737 | $iLastMatch = $i; |
738 | $j = $i; |
cb718de0 |
739 | if ($htmlsave) { |
740 | $ret .= htmlspecialchars($res[1]); |
741 | } else { |
742 | $ret .= $res[1]; |
743 | } |
098ea084 |
744 | $encoding = ucfirst($res[3]); |
d6f584fc |
745 | |
746 | /* decide about valid decoding */ |
747 | if ($decide && is_conversion_safe($res[2])) { |
8bd0068d |
748 | $utfencode=true; |
749 | $can_be_encoded=true; |
d6f584fc |
750 | } else { |
8bd0068d |
751 | $can_be_encoded=false; |
d6f584fc |
752 | } |
098ea084 |
753 | switch ($encoding) |
754 | { |
8bd0068d |
755 | case 'B': |
756 | $replace = base64_decode($res[4]); |
757 | if ($utfencode) { |
758 | if ($can_be_encoded) { |
759 | /* convert string to different charset, |
760 | * if functions asks for it (usually in compose) |
761 | */ |
bc6c0fba |
762 | $ret .= charset_convert($res[2],$replace,$default_charset,$htmlsave); |
8bd0068d |
763 | } else { |
764 | // convert string to html codes in order to display it |
765 | $ret .= charset_decode($res[2],$replace); |
766 | } |
fab65ca9 |
767 | } else { |
8bd0068d |
768 | if ($htmlsave) { |
769 | $replace = htmlspecialchars($replace); |
770 | } |
771 | $ret.= $replace; |
fab65ca9 |
772 | } |
8bd0068d |
773 | break; |
774 | case 'Q': |
775 | $replace = str_replace('_', ' ', $res[4]); |
776 | $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))', |
777 | $replace); |
778 | if ($utfencode) { |
779 | if ($can_be_encoded) { |
780 | /* convert string to different charset, |
781 | * if functions asks for it (usually in compose) |
782 | */ |
bc6c0fba |
783 | $replace = charset_convert($res[2], $replace,$default_charset,$htmlsave); |
8bd0068d |
784 | } else { |
785 | // convert string to html codes in order to display it |
786 | $replace = charset_decode($res[2], $replace); |
787 | } |
788 | } else { |
789 | if ($htmlsave) { |
790 | $replace = htmlspecialchars($replace); |
791 | } |
098ea084 |
792 | } |
8bd0068d |
793 | $ret .= $replace; |
794 | break; |
795 | default: |
796 | break; |
79e07c7e |
797 | } |
098ea084 |
798 | $chunk = $res[5]; |
799 | $encoded = true; |
08b7f7cc |
800 | } |
801 | if (!$encoded) { |
802 | if ($htmlsave) { |
9f7f68c3 |
803 | $ret .= ' '; |
08b7f7cc |
804 | } else { |
805 | $ret .= ' '; |
da2415c1 |
806 | } |
08b7f7cc |
807 | } |
dc3d13a7 |
808 | |
809 | if (!$encoded && $htmlsave) { |
810 | $ret .= htmlspecialchars($chunk); |
811 | } else { |
812 | $ret .= $chunk; |
813 | } |
098ea084 |
814 | ++$i; |
815 | } |
fd81e884 |
816 | /* remove the first added space */ |
817 | if ($ret) { |
818 | if ($htmlsave) { |
9f7f68c3 |
819 | $ret = substr($ret,5); |
fd81e884 |
820 | } else { |
821 | $ret = substr($ret,1); |
822 | } |
823 | } |
da2415c1 |
824 | |
08b7f7cc |
825 | return $ret; |
451f74a2 |
826 | } |
827 | |
9f7f68c3 |
828 | /** |
a24cf710 |
829 | * Encodes header |
8bd0068d |
830 | * |
a24cf710 |
831 | * Function uses XTRA_CODE _encodeheader function, if such function exists. |
a24cf710 |
832 | * |
758a7889 |
833 | * Function uses Q encoding by default and encodes a string according to RFC |
834 | * 1522 for use in headers if it contains 8-bit characters or anything that |
a24cf710 |
835 | * looks like it should be encoded. |
8bd0068d |
836 | * |
758a7889 |
837 | * Function switches to B encoding and encodeHeaderBase64() function, if |
838 | * string is 8bit and multibyte character set supported by mbstring extension |
839 | * is used. It can cause E_USER_NOTICE errors, if interface is used with |
f270a6eb |
840 | * multibyte character set unsupported by mbstring extension. |
841 | * |
8bd0068d |
842 | * @param string $string header string, that has to be encoded |
843 | * @return string quoted-printable encoded string |
f270a6eb |
844 | * @todo make $mb_charsets system wide constant |
8bd0068d |
845 | */ |
451f74a2 |
846 | function encodeHeader ($string) { |
6fbd125b |
847 | global $default_charset, $languages, $squirrelmail_language; |
83be314a |
848 | |
10dec454 |
849 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
8bd0068d |
850 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) { |
33a55f5a |
851 | return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string); |
83be314a |
852 | } |
793cc001 |
853 | |
a24cf710 |
854 | // Use B encoding for multibyte charsets |
f270a6eb |
855 | $mb_charsets = array('utf-8','big5','gb2313','euc-kr'); |
856 | if (in_array($default_charset,$mb_charsets) && |
857 | in_array($default_charset,sq_mb_list_encodings()) && |
858 | sq_is8bit($string)) { |
859 | return encodeHeaderBase64($string,$default_charset); |
860 | } elseif (in_array($default_charset,$mb_charsets) && |
861 | sq_is8bit($string) && |
862 | ! in_array($default_charset,sq_mb_list_encodings())) { |
863 | // Add E_USER_NOTICE error here (can cause 'Cannot add header information' warning in compose.php) |
864 | // trigger_error('encodeHeader: Multibyte character set unsupported by mbstring extension.',E_USER_NOTICE); |
865 | } |
a24cf710 |
866 | |
451f74a2 |
867 | // Encode only if the string contains 8-bit characters or =? |
3d8371be |
868 | $j = strlen($string); |
098ea084 |
869 | $max_l = 75 - strlen($default_charset) - 7; |
870 | $aRet = array(); |
451f74a2 |
871 | $ret = ''; |
c96c32f4 |
872 | $iEncStart = $enc_init = false; |
0d53f0f9 |
873 | $cur_l = $iOffset = 0; |
3d8371be |
874 | for($i = 0; $i < $j; ++$i) { |
c96c32f4 |
875 | switch($string{$i}) |
876 | { |
8bd0068d |
877 | case '=': |
878 | case '<': |
879 | case '>': |
880 | case ',': |
881 | case '?': |
882 | case '_': |
883 | if ($iEncStart === false) { |
884 | $iEncStart = $i; |
885 | } |
886 | $cur_l+=3; |
887 | if ($cur_l > ($max_l-2)) { |
888 | /* if there is an stringpart that doesn't need encoding, add it */ |
08b7f7cc |
889 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
890 | $aRet[] = "=?$default_charset?Q?$ret?="; |
891 | $iOffset = $i; |
892 | $cur_l = 0; |
893 | $ret = ''; |
894 | $iEncStart = false; |
08b7f7cc |
895 | } else { |
8bd0068d |
896 | $ret .= sprintf("=%02X",ord($string{$i})); |
c96c32f4 |
897 | } |
8bd0068d |
898 | break; |
899 | case '(': |
900 | case ')': |
901 | if ($iEncStart !== false) { |
08b7f7cc |
902 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
8bd0068d |
903 | $aRet[] = "=?$default_charset?Q?$ret?="; |
c96c32f4 |
904 | $iOffset = $i; |
8bd0068d |
905 | $cur_l = 0; |
906 | $ret = ''; |
907 | $iEncStart = false; |
c96c32f4 |
908 | } |
8bd0068d |
909 | break; |
910 | case ' ': |
c96c32f4 |
911 | if ($iEncStart !== false) { |
098ea084 |
912 | $cur_l++; |
913 | if ($cur_l > $max_l) { |
08b7f7cc |
914 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
c96c32f4 |
915 | $aRet[] = "=?$default_charset?Q?$ret?="; |
c96c32f4 |
916 | $iOffset = $i; |
917 | $cur_l = 0; |
098ea084 |
918 | $ret = ''; |
8bd0068d |
919 | $iEncStart = false; |
08b7f7cc |
920 | } else { |
8bd0068d |
921 | $ret .= '_'; |
c96c32f4 |
922 | } |
3d8371be |
923 | } |
8bd0068d |
924 | break; |
925 | default: |
926 | $k = ord($string{$i}); |
927 | if ($k > 126) { |
928 | if ($iEncStart === false) { |
929 | // do not start encoding in the middle of a string, also take the rest of the word. |
930 | $sLeadString = substr($string,0,$i); |
931 | $aLeadString = explode(' ',$sLeadString); |
932 | $sToBeEncoded = array_pop($aLeadString); |
933 | $iEncStart = $i - strlen($sToBeEncoded); |
934 | $ret .= $sToBeEncoded; |
935 | $cur_l += strlen($sToBeEncoded); |
936 | } |
937 | $cur_l += 3; |
938 | /* first we add the encoded string that reached it's max size */ |
939 | if ($cur_l > ($max_l-2)) { |
940 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
941 | $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */ |
942 | $cur_l = 3; |
943 | $ret = ''; |
944 | $iOffset = $i; |
945 | $iEncStart = $i; |
946 | } |
947 | $enc_init = true; |
948 | $ret .= sprintf("=%02X", $k); |
949 | } else { |
950 | if ($iEncStart !== false) { |
951 | $cur_l++; |
952 | if ($cur_l > $max_l) { |
953 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
954 | $aRet[] = "=?$default_charset?Q?$ret?="; |
955 | $iEncStart = false; |
956 | $iOffset = $i; |
957 | $cur_l = 0; |
958 | $ret = ''; |
959 | } else { |
960 | $ret .= $string{$i}; |
961 | } |
962 | } |
963 | } |
964 | break; |
f7b3ba37 |
965 | } |
451f74a2 |
966 | } |
793cc001 |
967 | |
c96c32f4 |
968 | if ($enc_init) { |
969 | if ($iEncStart !== false) { |
970 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
971 | $aRet[] = "=?$default_charset?Q?$ret?="; |
972 | } else { |
973 | $aRet[] = substr($string,$iOffset); |
974 | } |
975 | $string = implode('',$aRet); |
451f74a2 |
976 | } |
3d8371be |
977 | return $string; |
451f74a2 |
978 | } |
b74ba498 |
979 | |
f270a6eb |
980 | /** |
981 | * Encodes string according to rfc2047 B encoding header formating rules |
982 | * |
758a7889 |
983 | * It is recommended way to encode headers with character sets that store |
f270a6eb |
984 | * symbols in more than one byte. |
985 | * |
986 | * Function requires mbstring support. If required mbstring functions are missing, |
987 | * function returns false and sets E_USER_WARNING level error message. |
988 | * |
758a7889 |
989 | * Minimal requirements - php 4.0.6 with mbstring extension. Please note, |
990 | * that mbstring functions will generate E_WARNING errors, if unsupported |
f270a6eb |
991 | * character set is used. mb_encode_mimeheader function provided by php |
992 | * mbstring extension is not used in order to get better control of header |
993 | * encoding. |
994 | * |
758a7889 |
995 | * Used php code functions - function_exists(), trigger_error(), strlen() |
996 | * (is used with charset names and base64 strings). Used php mbstring |
f270a6eb |
997 | * functions - mb_strlen and mb_substr. |
998 | * |
758a7889 |
999 | * Related documents: rfc 2045 (BASE64 encoding), rfc 2047 (mime header |
f270a6eb |
1000 | * encoding), rfc 2822 (header folding) |
1001 | * |
1002 | * @param string $string header string that must be encoded |
758a7889 |
1003 | * @param string $charset character set. Must be supported by mbstring extension. |
f270a6eb |
1004 | * Use sq_mb_list_encodings() to detect supported charsets. |
1005 | * @return string string encoded according to rfc2047 B encoding formating rules |
1006 | * @since 1.5.1 |
1007 | * @todo First header line can be wrapped to $iMaxLength - $HeaderFieldLength - 1 |
1008 | * @todo Do we want to control max length of header? |
1009 | * @todo Do we want to control EOL (end-of-line) marker? |
1010 | * @todo Do we want to translate error message? |
1011 | */ |
1012 | function encodeHeaderBase64($string,$charset) { |
1013 | /** |
1014 | * Check mbstring function requirements. |
1015 | */ |
1016 | if (! function_exists('mb_strlen') || |
1017 | ! function_exists('mb_substr')) { |
1018 | // set E_USER_WARNING |
1019 | trigger_error('encodeHeaderBase64: Required mbstring functions are missing.',E_USER_WARNING); |
1020 | // return false |
1021 | return false; |
1022 | } |
1023 | |
1024 | // initial return array |
1025 | $aRet = array(); |
1026 | |
1027 | /** |
1028 | * header length = 75 symbols max (same as in encodeHeader) |
1029 | * remove $charset length |
1030 | * remove =? ? ?= (5 chars) |
1031 | * remove 2 more chars (\r\n ?) |
1032 | */ |
1033 | $iMaxLength = 75 - strlen($charset) - 7; |
1034 | |
1035 | // set first character position |
1036 | $iStartCharNum = 0; |
1037 | |
1038 | // loop through all characters. count characters and not bytes. |
1039 | for ($iCharNum=1; $iCharNum<=mb_strlen($string,$charset); $iCharNum++) { |
1040 | // encode string from starting character to current character. |
1041 | $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset)); |
1042 | |
1043 | // Check encoded string length |
1044 | if(strlen($encoded_string)>$iMaxLength) { |
1045 | // if string exceeds max length, reduce number of encoded characters and add encoded string part to array |
1046 | $aRet[] = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum-1,$charset)); |
1047 | |
1048 | // set new starting character |
1049 | $iStartCharNum = $iCharNum-1; |
1050 | |
1051 | // encode last char (in case it is last character in string) |
1052 | $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset)); |
1053 | } // if string is shorter than max length - add next character |
1054 | } |
1055 | |
1056 | // add last encoded string to array |
1057 | $aRet[] = $encoded_string; |
1058 | |
1059 | // set initial return string |
1060 | $sRet = ''; |
1061 | |
1062 | // loop through encoded strings |
1063 | foreach($aRet as $string) { |
1064 | // TODO: Do we want to control EOL (end-of-line) marker |
1065 | if ($sRet!='') $sRet.= " "; |
1066 | |
1067 | // add header tags and encoded string to return string |
1068 | $sRet.= '=?'.$charset.'?B?'.$string.'?='; |
1069 | } |
1070 | |
1071 | return $sRet; |
1072 | } |
1073 | |
691a2d25 |
1074 | /* This function trys to locate the entity_id of a specific mime element */ |
3d8371be |
1075 | function find_ent_id($id, $message) { |
a171b359 |
1076 | for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) { |
1077 | if ($message->entities[$i]->header->type0 == 'multipart') { |
3d8371be |
1078 | $ret = find_ent_id($id, $message->entities[$i]); |
451f74a2 |
1079 | } else { |
3d8371be |
1080 | if (strcasecmp($message->entities[$i]->header->id, $id) == 0) { |
d8cffbab |
1081 | // if (sq_check_save_extension($message->entities[$i])) { |
8bd0068d |
1082 | return $message->entities[$i]->entity_id; |
da2415c1 |
1083 | // } |
c8f5f606 |
1084 | } elseif (!empty($message->entities[$i]->header->parameters['name'])) { |
1085 | /** |
1086 | * This is part of a fix for Outlook Express 6.x generating |
1087 | * cid URLs without creating content-id headers |
1088 | * @@JA - 20050207 |
1089 | */ |
1090 | if (strcasecmp($message->entities[$i]->header->parameters['name'], $id) == 0) { |
1091 | return $message->entities[$i]->entity_id; |
1092 | } |
3d8371be |
1093 | } |
a3daaaf3 |
1094 | } |
451f74a2 |
1095 | } |
3d8371be |
1096 | return $ret; |
451f74a2 |
1097 | } |
a3daaaf3 |
1098 | |
e5e9381a |
1099 | function sq_check_save_extension($message) { |
1100 | $filename = $message->getFilename(); |
1101 | $ext = substr($filename, strrpos($filename,'.')+1); |
1102 | $save_extensions = array('jpg','jpeg','gif','png','bmp'); |
3d8371be |
1103 | return in_array($ext, $save_extensions); |
e5e9381a |
1104 | } |
1105 | |
1106 | |
691a2d25 |
1107 | /** |
8bd0068d |
1108 | ** HTMLFILTER ROUTINES |
1109 | */ |
451f74a2 |
1110 | |
2dd879b8 |
1111 | /** |
0493ed11 |
1112 | * This function checks attribute values for entity-encoded values |
1113 | * and returns them translated into 8-bit strings so we can run |
1114 | * checks on them. |
8bd0068d |
1115 | * |
0493ed11 |
1116 | * @param $attvalue A string to run entity check against. |
1117 | * @return Nothing, modifies a reference value. |
8bd0068d |
1118 | */ |
0493ed11 |
1119 | function sq_defang(&$attvalue){ |
1120 | $me = 'sq_defang'; |
2dd879b8 |
1121 | /** |
0493ed11 |
1122 | * Skip this if there aren't ampersands or backslashes. |
8bd0068d |
1123 | */ |
0493ed11 |
1124 | if (strpos($attvalue, '&') === false |
1125 | && strpos($attvalue, '\\') === false){ |
1126 | return; |
2dd879b8 |
1127 | } |
0493ed11 |
1128 | $m = false; |
1129 | do { |
1130 | $m = false; |
1131 | $m = $m || sq_deent($attvalue, '/\�*(\d+);*/s'); |
1132 | $m = $m || sq_deent($attvalue, '/\�*((\d|[a-f])+);*/si', true); |
1133 | $m = $m || sq_deent($attvalue, '/\\\\(\d+)/s', true); |
1134 | } while ($m == true); |
1135 | $attvalue = stripslashes($attvalue); |
2dd879b8 |
1136 | } |
1137 | |
1138 | /** |
8bd0068d |
1139 | * Kill any tabs, newlines, or carriage returns. Our friends the |
1140 | * makers of the browser with 95% market value decided that it'd |
1141 | * be funny to make "java[tab]script" be just as good as "javascript". |
1142 | * |
1143 | * @param attvalue The attribute value before extraneous spaces removed. |
0493ed11 |
1144 | * @return attvalue Nothing, modifies a reference value. |
8bd0068d |
1145 | */ |
0493ed11 |
1146 | function sq_unspace(&$attvalue){ |
1147 | $me = 'sq_unspace'; |
1148 | if (strcspn($attvalue, "\t\r\n\0 ") != strlen($attvalue)){ |
1149 | $attvalue = str_replace(Array("\t", "\r", "\n", "\0", " "), |
1150 | Array('', '', '', '', ''), $attvalue); |
2dd879b8 |
1151 | } |
2dd879b8 |
1152 | } |
1153 | |
691a2d25 |
1154 | /** |
8bd0068d |
1155 | * This function returns the final tag out of the tag name, an array |
1156 | * of attributes, and the type of the tag. This function is called by |
1157 | * sq_sanitize internally. |
1158 | * |
1159 | * @param $tagname the name of the tag. |
1160 | * @param $attary the array of attributes and their values |
1161 | * @param $tagtype The type of the tag (see in comments). |
1162 | * @return a string with the final tag representation. |
1163 | */ |
691a2d25 |
1164 | function sq_tagprint($tagname, $attary, $tagtype){ |
b583c3e8 |
1165 | $me = 'sq_tagprint'; |
3d8371be |
1166 | |
691a2d25 |
1167 | if ($tagtype == 2){ |
1168 | $fulltag = '</' . $tagname . '>'; |
1169 | } else { |
1170 | $fulltag = '<' . $tagname; |
1171 | if (is_array($attary) && sizeof($attary)){ |
1172 | $atts = Array(); |
1173 | while (list($attname, $attvalue) = each($attary)){ |
1174 | array_push($atts, "$attname=$attvalue"); |
1175 | } |
1176 | $fulltag .= ' ' . join(" ", $atts); |
1177 | } |
1178 | if ($tagtype == 3){ |
b583c3e8 |
1179 | $fulltag .= ' /'; |
691a2d25 |
1180 | } |
b583c3e8 |
1181 | $fulltag .= '>'; |
451f74a2 |
1182 | } |
691a2d25 |
1183 | return $fulltag; |
451f74a2 |
1184 | } |
a3daaaf3 |
1185 | |
691a2d25 |
1186 | /** |
8bd0068d |
1187 | * A small helper function to use with array_walk. Modifies a by-ref |
1188 | * value and makes it lowercase. |
1189 | * |
1190 | * @param $val a value passed by-ref. |
1191 | * @return void since it modifies a by-ref value. |
1192 | */ |
691a2d25 |
1193 | function sq_casenormalize(&$val){ |
1194 | $val = strtolower($val); |
1195 | } |
451f74a2 |
1196 | |
691a2d25 |
1197 | /** |
8bd0068d |
1198 | * This function skips any whitespace from the current position within |
1199 | * a string and to the next non-whitespace value. |
1200 | * |
1201 | * @param $body the string |
1202 | * @param $offset the offset within the string where we should start |
1203 | * looking for the next non-whitespace character. |
1204 | * @return the location within the $body where the next |
1205 | * non-whitespace char is located. |
1206 | */ |
691a2d25 |
1207 | function sq_skipspace($body, $offset){ |
b583c3e8 |
1208 | $me = 'sq_skipspace'; |
3d8371be |
1209 | preg_match('/^(\s*)/s', substr($body, $offset), $matches); |
691a2d25 |
1210 | if (sizeof($matches{1})){ |
1211 | $count = strlen($matches{1}); |
1212 | $offset += $count; |
451f74a2 |
1213 | } |
691a2d25 |
1214 | return $offset; |
451f74a2 |
1215 | } |
a3daaaf3 |
1216 | |
691a2d25 |
1217 | /** |
8bd0068d |
1218 | * This function looks for the next character within a string. It's |
1219 | * really just a glorified "strpos", except it catches if failures |
1220 | * nicely. |
1221 | * |
1222 | * @param $body The string to look for needle in. |
1223 | * @param $offset Start looking from this position. |
1224 | * @param $needle The character/string to look for. |
1225 | * @return location of the next occurance of the needle, or |
1226 | * strlen($body) if needle wasn't found. |
1227 | */ |
691a2d25 |
1228 | function sq_findnxstr($body, $offset, $needle){ |
3d8371be |
1229 | $me = 'sq_findnxstr'; |
691a2d25 |
1230 | $pos = strpos($body, $needle, $offset); |
1231 | if ($pos === FALSE){ |
1232 | $pos = strlen($body); |
451f74a2 |
1233 | } |
691a2d25 |
1234 | return $pos; |
451f74a2 |
1235 | } |
a3daaaf3 |
1236 | |
691a2d25 |
1237 | /** |
8bd0068d |
1238 | * This function takes a PCRE-style regexp and tries to match it |
1239 | * within the string. |
1240 | * |
1241 | * @param $body The string to look for needle in. |
1242 | * @param $offset Start looking from here. |
1243 | * @param $reg A PCRE-style regex to match. |
1244 | * @return Returns a false if no matches found, or an array |
1245 | * with the following members: |
1246 | * - integer with the location of the match within $body |
1247 | * - string with whatever content between offset and the match |
1248 | * - string with whatever it is we matched |
1249 | */ |
691a2d25 |
1250 | function sq_findnxreg($body, $offset, $reg){ |
b583c3e8 |
1251 | $me = 'sq_findnxreg'; |
691a2d25 |
1252 | $matches = Array(); |
1253 | $retarr = Array(); |
7d06541f |
1254 | preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches); |
1255 | if (!isset($matches{0}) || !$matches{0}){ |
691a2d25 |
1256 | $retarr = false; |
1257 | } else { |
1258 | $retarr{0} = $offset + strlen($matches{1}); |
1259 | $retarr{1} = $matches{1}; |
1260 | $retarr{2} = $matches{2}; |
1261 | } |
1262 | return $retarr; |
1263 | } |
a3daaaf3 |
1264 | |
691a2d25 |
1265 | /** |
8bd0068d |
1266 | * This function looks for the next tag. |
1267 | * |
1268 | * @param $body String where to look for the next tag. |
1269 | * @param $offset Start looking from here. |
1270 | * @return false if no more tags exist in the body, or |
1271 | * an array with the following members: |
1272 | * - string with the name of the tag |
1273 | * - array with attributes and their values |
1274 | * - integer with tag type (1, 2, or 3) |
1275 | * - integer where the tag starts (starting "<") |
1276 | * - integer where the tag ends (ending ">") |
1277 | * first three members will be false, if the tag is invalid. |
1278 | */ |
691a2d25 |
1279 | function sq_getnxtag($body, $offset){ |
b583c3e8 |
1280 | $me = 'sq_getnxtag'; |
691a2d25 |
1281 | if ($offset > strlen($body)){ |
1282 | return false; |
1283 | } |
1284 | $lt = sq_findnxstr($body, $offset, "<"); |
1285 | if ($lt == strlen($body)){ |
1286 | return false; |
1287 | } |
1288 | /** |
8bd0068d |
1289 | * We are here: |
1290 | * blah blah <tag attribute="value"> |
1291 | * \---------^ |
1292 | */ |
691a2d25 |
1293 | $pos = sq_skipspace($body, $lt+1); |
1294 | if ($pos >= strlen($body)){ |
1295 | return Array(false, false, false, $lt, strlen($body)); |
1296 | } |
1297 | /** |
8bd0068d |
1298 | * There are 3 kinds of tags: |
1299 | * 1. Opening tag, e.g.: |
1300 | * <a href="blah"> |
1301 | * 2. Closing tag, e.g.: |
1302 | * </a> |
1303 | * 3. XHTML-style content-less tag, e.g.: |
1304 | * <img src="blah" /> |
1305 | */ |
691a2d25 |
1306 | $tagtype = false; |
1307 | switch (substr($body, $pos, 1)){ |
3d8371be |
1308 | case '/': |
1309 | $tagtype = 2; |
1310 | $pos++; |
1311 | break; |
1312 | case '!': |
1313 | /** |
8bd0068d |
1314 | * A comment or an SGML declaration. |
1315 | */ |
3d8371be |
1316 | if (substr($body, $pos+1, 2) == "--"){ |
1317 | $gt = strpos($body, "-->", $pos); |
1318 | if ($gt === false){ |
1319 | $gt = strlen($body); |
1320 | } else { |
1321 | $gt += 2; |
1322 | } |
1323 | return Array(false, false, false, $lt, $gt); |
bb8d0799 |
1324 | } else { |
3d8371be |
1325 | $gt = sq_findnxstr($body, $pos, ">"); |
1326 | return Array(false, false, false, $lt, $gt); |
1327 | } |
1328 | break; |
1329 | default: |
1330 | /** |
8bd0068d |
1331 | * Assume tagtype 1 for now. If it's type 3, we'll switch values |
1332 | * later. |
1333 | */ |
3d8371be |
1334 | $tagtype = 1; |
1335 | break; |
691a2d25 |
1336 | } |
a3daaaf3 |
1337 | |
0493ed11 |
1338 | $tag_start = $pos; |
691a2d25 |
1339 | $tagname = ''; |
1340 | /** |
8bd0068d |
1341 | * Look for next [\W-_], which will indicate the end of the tag name. |
1342 | */ |
691a2d25 |
1343 | $regary = sq_findnxreg($body, $pos, "[^\w\-_]"); |
1344 | if ($regary == false){ |
1345 | return Array(false, false, false, $lt, strlen($body)); |
1346 | } |
1347 | list($pos, $tagname, $match) = $regary; |
1348 | $tagname = strtolower($tagname); |
1349 | |
1350 | /** |
8bd0068d |
1351 | * $match can be either of these: |
1352 | * '>' indicating the end of the tag entirely. |
1353 | * '\s' indicating the end of the tag name. |
1354 | * '/' indicating that this is type-3 xhtml tag. |
1355 | * |
1356 | * Whatever else we find there indicates an invalid tag. |
1357 | */ |
691a2d25 |
1358 | switch ($match){ |
3d8371be |
1359 | case '/': |
691a2d25 |
1360 | /** |
8bd0068d |
1361 | * This is an xhtml-style tag with a closing / at the |
1362 | * end, like so: <img src="blah" />. Check if it's followed |
1363 | * by the closing bracket. If not, then this tag is invalid |
1364 | */ |
3d8371be |
1365 | if (substr($body, $pos, 2) == "/>"){ |
1366 | $pos++; |
1367 | $tagtype = 3; |
1368 | } else { |
1369 | $gt = sq_findnxstr($body, $pos, ">"); |
1370 | $retary = Array(false, false, false, $lt, $gt); |
1371 | return $retary; |
1372 | } |
1373 | case '>': |
1374 | return Array($tagname, false, $tagtype, $lt, $pos); |
1375 | break; |
1376 | default: |
1377 | /** |
8bd0068d |
1378 | * Check if it's whitespace |
1379 | */ |
3d8371be |
1380 | if (!preg_match('/\s/', $match)){ |
1381 | /** |
8bd0068d |
1382 | * This is an invalid tag! Look for the next closing ">". |
1383 | */ |
7d06541f |
1384 | $gt = sq_findnxstr($body, $lt, ">"); |
3d8371be |
1385 | return Array(false, false, false, $lt, $gt); |
1386 | } |
1387 | break; |
691a2d25 |
1388 | } |
3d8371be |
1389 | |
691a2d25 |
1390 | /** |
8bd0068d |
1391 | * At this point we're here: |
1392 | * <tagname attribute='blah'> |
1393 | * \-------^ |
1394 | * |
1395 | * At this point we loop in order to find all attributes. |
1396 | */ |
691a2d25 |
1397 | $attname = ''; |
0493ed11 |
1398 | $atttype = false; |
691a2d25 |
1399 | $attary = Array(); |
1400 | |
1401 | while ($pos <= strlen($body)){ |
1402 | $pos = sq_skipspace($body, $pos); |
1403 | if ($pos == strlen($body)){ |
1404 | /** |
8bd0068d |
1405 | * Non-closed tag. |
1406 | */ |
691a2d25 |
1407 | return Array(false, false, false, $lt, $pos); |
1408 | } |
1409 | /** |
8bd0068d |
1410 | * See if we arrived at a ">" or "/>", which means that we reached |
1411 | * the end of the tag. |
1412 | */ |
691a2d25 |
1413 | $matches = Array(); |
164800ad |
1414 | if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) { |
c828931c |
1415 | /** |
8bd0068d |
1416 | * Yep. So we did. |
1417 | */ |
c828931c |
1418 | $pos += strlen($matches{1}); |
1419 | if ($matches{2} == "/>"){ |
1420 | $tagtype = 3; |
1421 | $pos++; |
1422 | } |
1423 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
1424 | } |
a3daaaf3 |
1425 | |
cca46357 |
1426 | /** |
8bd0068d |
1427 | * There are several types of attributes, with optional |
1428 | * [:space:] between members. |
1429 | * Type 1: |
1430 | * attrname[:space:]=[:space:]'CDATA' |
1431 | * Type 2: |
1432 | * attrname[:space:]=[:space:]"CDATA" |
1433 | * Type 3: |
1434 | * attr[:space:]=[:space:]CDATA |
1435 | * Type 4: |
1436 | * attrname |
1437 | * |
1438 | * We leave types 1 and 2 the same, type 3 we check for |
1439 | * '"' and convert to """ if needed, then wrap in |
1440 | * double quotes. Type 4 we convert into: |
1441 | * attrname="yes". |
1442 | */ |
3f7c623f |
1443 | $regary = sq_findnxreg($body, $pos, "[^:\w\-_]"); |
691a2d25 |
1444 | if ($regary == false){ |
1445 | /** |
8bd0068d |
1446 | * Looks like body ended before the end of tag. |
1447 | */ |
691a2d25 |
1448 | return Array(false, false, false, $lt, strlen($body)); |
cca46357 |
1449 | } |
691a2d25 |
1450 | list($pos, $attname, $match) = $regary; |
1451 | $attname = strtolower($attname); |
1452 | /** |
8bd0068d |
1453 | * We arrived at the end of attribute name. Several things possible |
1454 | * here: |
1455 | * '>' means the end of the tag and this is attribute type 4 |
1456 | * '/' if followed by '>' means the same thing as above |
1457 | * '\s' means a lot of things -- look what it's followed by. |
1458 | * anything else means the attribute is invalid. |
1459 | */ |
691a2d25 |
1460 | switch($match){ |
3d8371be |
1461 | case '/': |
691a2d25 |
1462 | /** |
8bd0068d |
1463 | * This is an xhtml-style tag with a closing / at the |
1464 | * end, like so: <img src="blah" />. Check if it's followed |
1465 | * by the closing bracket. If not, then this tag is invalid |
1466 | */ |
3d8371be |
1467 | if (substr($body, $pos, 2) == "/>"){ |
691a2d25 |
1468 | $pos++; |
3d8371be |
1469 | $tagtype = 3; |
691a2d25 |
1470 | } else { |
3d8371be |
1471 | $gt = sq_findnxstr($body, $pos, ">"); |
1472 | $retary = Array(false, false, false, $lt, $gt); |
1473 | return $retary; |
1474 | } |
1475 | case '>': |
1476 | $attary{$attname} = '"yes"'; |
1477 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
1478 | break; |
1479 | default: |
1480 | /** |
8bd0068d |
1481 | * Skip whitespace and see what we arrive at. |
1482 | */ |
3d8371be |
1483 | $pos = sq_skipspace($body, $pos); |
1484 | $char = substr($body, $pos, 1); |
1485 | /** |
8bd0068d |
1486 | * Two things are valid here: |
1487 | * '=' means this is attribute type 1 2 or 3. |
1488 | * \w means this was attribute type 4. |
1489 | * anything else we ignore and re-loop. End of tag and |
1490 | * invalid stuff will be caught by our checks at the beginning |
1491 | * of the loop. |
1492 | */ |
3d8371be |
1493 | if ($char == "="){ |
1494 | $pos++; |
1495 | $pos = sq_skipspace($body, $pos); |
691a2d25 |
1496 | /** |
8bd0068d |
1497 | * Here are 3 possibilities: |
1498 | * "'" attribute type 1 |
1499 | * '"' attribute type 2 |
1500 | * everything else is the content of tag type 3 |
1501 | */ |
3d8371be |
1502 | $quot = substr($body, $pos, 1); |
1503 | if ($quot == "'"){ |
1504 | $regary = sq_findnxreg($body, $pos+1, "\'"); |
1505 | if ($regary == false){ |
1506 | return Array(false, false, false, $lt, strlen($body)); |
1507 | } |
1508 | list($pos, $attval, $match) = $regary; |
1509 | $pos++; |
1510 | $attary{$attname} = "'" . $attval . "'"; |
1511 | } else if ($quot == '"'){ |
1512 | $regary = sq_findnxreg($body, $pos+1, '\"'); |
1513 | if ($regary == false){ |
1514 | return Array(false, false, false, $lt, strlen($body)); |
1515 | } |
1516 | list($pos, $attval, $match) = $regary; |
1517 | $pos++; |
1518 | $attary{$attname} = '"' . $attval . '"'; |
1519 | } else { |
1520 | /** |
8bd0068d |
1521 | * These are hateful. Look for \s, or >. |
1522 | */ |
3d8371be |
1523 | $regary = sq_findnxreg($body, $pos, "[\s>]"); |
1524 | if ($regary == false){ |
1525 | return Array(false, false, false, $lt, strlen($body)); |
1526 | } |
1527 | list($pos, $attval, $match) = $regary; |
1528 | /** |
8bd0068d |
1529 | * If it's ">" it will be caught at the top. |
1530 | */ |
3d8371be |
1531 | $attval = preg_replace("/\"/s", """, $attval); |
1532 | $attary{$attname} = '"' . $attval . '"'; |
7e235a1a |
1533 | } |
3d8371be |
1534 | } else if (preg_match("|[\w/>]|", $char)) { |
691a2d25 |
1535 | /** |
8bd0068d |
1536 | * That was attribute type 4. |
1537 | */ |
3d8371be |
1538 | $attary{$attname} = '"yes"'; |
1539 | } else { |
1540 | /** |
8bd0068d |
1541 | * An illegal character. Find next '>' and return. |
1542 | */ |
3d8371be |
1543 | $gt = sq_findnxstr($body, $pos, ">"); |
1544 | return Array(false, false, false, $lt, $gt); |
451f74a2 |
1545 | } |
3d8371be |
1546 | break; |
691a2d25 |
1547 | } |
1548 | } |
1549 | /** |
8bd0068d |
1550 | * The fact that we got here indicates that the tag end was never |
1551 | * found. Return invalid tag indication so it gets stripped. |
1552 | */ |
691a2d25 |
1553 | return Array(false, false, false, $lt, strlen($body)); |
1554 | } |
1555 | |
1556 | /** |
0493ed11 |
1557 | * Translates entities into literal values so they can be checked. |
8bd0068d |
1558 | * |
0493ed11 |
1559 | * @param $attvalue the by-ref value to check. |
1560 | * @param $regex the regular expression to check against. |
1561 | * @param $hex whether the entites are hexadecimal. |
1562 | * @return True or False depending on whether there were matches. |
8bd0068d |
1563 | */ |
0493ed11 |
1564 | function sq_deent(&$attvalue, $regex, $hex=false){ |
b583c3e8 |
1565 | $me = 'sq_deent'; |
0493ed11 |
1566 | $ret_match = false; |
1567 | preg_match_all($regex, $attvalue, $matches); |
1568 | if (is_array($matches) && sizeof($matches[0]) > 0){ |
1569 | $repl = Array(); |
1570 | for ($i = 0; $i < sizeof($matches[0]); $i++){ |
1571 | $numval = $matches[1][$i]; |
1572 | if ($hex){ |
1573 | $numval = hexdec($numval); |
a3daaaf3 |
1574 | } |
0493ed11 |
1575 | $repl{$matches[0][$i]} = chr($numval); |
691a2d25 |
1576 | } |
0493ed11 |
1577 | $attvalue = strtr($attvalue, $repl); |
1578 | return true; |
1579 | } else { |
1580 | return false; |
691a2d25 |
1581 | } |
691a2d25 |
1582 | } |
1583 | |
1584 | /** |
8bd0068d |
1585 | * This function runs various checks against the attributes. |
1586 | * |
1587 | * @param $tagname String with the name of the tag. |
1588 | * @param $attary Array with all tag attributes. |
1589 | * @param $rm_attnames See description for sq_sanitize |
1590 | * @param $bad_attvals See description for sq_sanitize |
1591 | * @param $add_attr_to_tag See description for sq_sanitize |
1592 | * @param $message message object |
1593 | * @param $id message id |
1594 | * @return Array with modified attributes. |
1595 | */ |
da2415c1 |
1596 | function sq_fixatts($tagname, |
1597 | $attary, |
691a2d25 |
1598 | $rm_attnames, |
1599 | $bad_attvals, |
1600 | $add_attr_to_tag, |
1601 | $message, |
b3af12ef |
1602 | $id, |
3d8371be |
1603 | $mailbox |
691a2d25 |
1604 | ){ |
b583c3e8 |
1605 | $me = 'sq_fixatts'; |
691a2d25 |
1606 | while (list($attname, $attvalue) = each($attary)){ |
1607 | /** |
8bd0068d |
1608 | * See if this attribute should be removed. |
1609 | */ |
691a2d25 |
1610 | foreach ($rm_attnames as $matchtag=>$matchattrs){ |
1611 | if (preg_match($matchtag, $tagname)){ |
1612 | foreach ($matchattrs as $matchattr){ |
1613 | if (preg_match($matchattr, $attname)){ |
1614 | unset($attary{$attname}); |
1615 | continue; |
1616 | } |
451f74a2 |
1617 | } |
451f74a2 |
1618 | } |
691a2d25 |
1619 | } |
1620 | /** |
8bd0068d |
1621 | * Remove any backslashes, entities, and extraneous whitespace. |
1622 | */ |
0493ed11 |
1623 | sq_defang($attvalue); |
1624 | sq_unspace($attvalue); |
af861a34 |
1625 | |
691a2d25 |
1626 | /** |
8bd0068d |
1627 | * Now let's run checks on the attvalues. |
1628 | * I don't expect anyone to comprehend this. If you do, |
1629 | * get in touch with me so I can drive to where you live and |
1630 | * shake your hand personally. :) |
1631 | */ |
691a2d25 |
1632 | foreach ($bad_attvals as $matchtag=>$matchattrs){ |
1633 | if (preg_match($matchtag, $tagname)){ |
1634 | foreach ($matchattrs as $matchattr=>$valary){ |
1635 | if (preg_match($matchattr, $attname)){ |
1636 | /** |
8bd0068d |
1637 | * There are two arrays in valary. |
1638 | * First is matches. |
1639 | * Second one is replacements |
1640 | */ |
691a2d25 |
1641 | list($valmatch, $valrepl) = $valary; |
da2415c1 |
1642 | $newvalue = |
691a2d25 |
1643 | preg_replace($valmatch, $valrepl, $attvalue); |
1644 | if ($newvalue != $attvalue){ |
1645 | $attary{$attname} = $newvalue; |
1646 | } |
1647 | } |
1648 | } |
451f74a2 |
1649 | } |
a3daaaf3 |
1650 | } |
834a1027 |
1651 | |
834a1027 |
1652 | /** |
1653 | * Replace empty src tags with the blank image. src is only used |
1654 | * for frames, images, and image inputs. Doing a replace should |
1655 | * not affect them working as should be, however it will stop |
1656 | * IE from being kicked off when src for img tags are not set |
1657 | */ |
5a6fde9e |
1658 | if (($attname == 'src') && ($attvalue == '""')) { |
834a1027 |
1659 | $attary{$attname} = '"' . SM_PATH . 'images/blank.png"'; |
1660 | } |
1661 | |
691a2d25 |
1662 | /** |
8bd0068d |
1663 | * Turn cid: urls into http-friendly ones. |
1664 | */ |
691a2d25 |
1665 | if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){ |
b3af12ef |
1666 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
691a2d25 |
1667 | } |
ff940ebc |
1668 | |
1669 | /** |
1670 | * "Hack" fix for Outlook using propriatary outbind:// protocol in img tags. |
1671 | * One day MS might actually make it match something useful, for now, falling |
1672 | * back to using cid2http, so we can grab the blank.png. |
1673 | */ |
1674 | if (preg_match("/^[\'\"]\s*outbind:\/\//si", $attvalue)) { |
1675 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
1676 | } |
1677 | |
a3daaaf3 |
1678 | } |
691a2d25 |
1679 | /** |
8bd0068d |
1680 | * See if we need to append any attributes to this tag. |
1681 | */ |
691a2d25 |
1682 | foreach ($add_attr_to_tag as $matchtag=>$addattary){ |
1683 | if (preg_match($matchtag, $tagname)){ |
1684 | $attary = array_merge($attary, $addattary); |
1685 | } |
1686 | } |
1687 | return $attary; |
451f74a2 |
1688 | } |
a3daaaf3 |
1689 | |
691a2d25 |
1690 | /** |
8bd0068d |
1691 | * This function edits the style definition to make them friendly and |
1692 | * usable in SquirrelMail. |
1693 | * |
1694 | * @param $message the message object |
1695 | * @param $id the message id |
1696 | * @param $content a string with whatever is between <style> and </style> |
1697 | * @param $mailbox the message mailbox |
1698 | * @return a string with edited content. |
1699 | */ |
e60a299a |
1700 | function sq_fixstyle($body, $pos, $message, $id, $mailbox){ |
691a2d25 |
1701 | global $view_unsafe_images; |
b583c3e8 |
1702 | $me = 'sq_fixstyle'; |
7d06541f |
1703 | $ret = sq_findnxreg($body, $pos, '</\s*style\s*>'); |
1704 | if ($ret == FALSE){ |
1705 | return array(FALSE, strlen($body)); |
1706 | } |
1707 | $newpos = $ret[0] + strlen($ret[2]); |
1708 | $content = $ret[1]; |
691a2d25 |
1709 | /** |
0493ed11 |
1710 | * First look for general BODY style declaration, which would be |
1711 | * like so: |
1712 | * body {background: blah-blah} |
1713 | * and change it to .bodyclass so we can just assign it to a <div> |
1714 | */ |
691a2d25 |
1715 | $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content); |
3d8371be |
1716 | $secremoveimg = '../images/' . _("sec_remove_eng.png"); |
691a2d25 |
1717 | /** |
0493ed11 |
1718 | * Fix url('blah') declarations. |
1719 | */ |
1720 | // $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si", |
1721 | // "url(\\1$secremoveimg\\2)", $content); |
1722 | // remove NUL |
1723 | $content = str_replace("\0", "", $content); |
5b4884be |
1724 | // translate ur\l and variations (IE parses that) |
1725 | $content = preg_replace("/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", 'url', $content); |
0493ed11 |
1726 | // NB I insert NUL characters to keep to avoid an infinite loop. They are removed after the loop. |
1727 | while (preg_match("/url\s*\(\s*[\'\"]?([^:]+):(.*)?[\'\"]?\s*\)/si", $content, $matches)) { |
1728 | $sProto = strtolower($matches[1]); |
1729 | switch ($sProto) { |
0493ed11 |
1730 | /** |
758a7889 |
1731 | * Fix url('https*://.*) declarations but only if $view_unsafe_images |
1732 | * is false. |
0493ed11 |
1733 | */ |
758a7889 |
1734 | case 'https': |
1735 | case 'http': |
1736 | if (!$view_unsafe_images){ |
1737 | |
1738 | $sExpr = "/url\s*\(\s*[\'\"]?\s*$sProto*:.*[\'\"]?\s*\)/si"; |
1739 | $content = preg_replace($sExpr, "u\0r\0l(\\1$secremoveimg\\2)", $content); |
1740 | |
1741 | } else { |
1742 | $content = preg_replace('/url/i',"u\0r\0l",$content); |
1743 | } |
1744 | break; |
1745 | /** |
1746 | * Fix urls that refer to cid: |
1747 | */ |
1748 | case 'cid': |
1749 | $cidurl = 'cid:'. $matches[2]; |
1750 | $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox); |
1751 | // escape parentheses that can modify the regular expression |
1752 | $cidurl = str_replace(array('(',')'),array('\\(','\\)'),$cidurl); |
1753 | $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si", |
1754 | "u\0r\0l($httpurl)", $content); |
1755 | break; |
1756 | default: |
1757 | /** |
1758 | * replace url with protocol other then the white list |
1759 | * http,https and cid by an empty string. |
1760 | */ |
1761 | $content = preg_replace("/url\s*\(\s*[\'\"]?([^:]+):(.*)?[\'\"]?\s*\)/si", |
1762 | "", $content); |
1763 | break; |
0493ed11 |
1764 | } |
691a2d25 |
1765 | } |
0493ed11 |
1766 | // remove NUL |
1767 | $content = str_replace("\0", "", $content); |
0493ed11 |
1768 | /** |
1769 | * Remove any backslashes, entities, and extraneous whitespace. |
1770 | */ |
1771 | $contentTemp = $content; |
1772 | sq_defang($contentTemp); |
1773 | sq_unspace($contentTemp); |
a3daaaf3 |
1774 | |
691a2d25 |
1775 | /** |
8bd0068d |
1776 | * Fix stupid css declarations which lead to vulnerabilities |
1777 | * in IE. |
1778 | */ |
5db90261 |
1779 | $match = Array('/\/\*.*\*\//', |
1780 | '/expression/i', |
0493ed11 |
1781 | '/behaviou*r/i', |
1782 | '/binding/i', |
1783 | '/include-source/i'); |
5db90261 |
1784 | $replace = Array('','idiocy', 'idiocy', 'idiocy', 'idiocy'); |
0493ed11 |
1785 | $contentNew = preg_replace($match, $replace, $contentTemp); |
1786 | if ($contentNew !== $contentTemp) { |
1787 | // insecure css declarations are used. From now on we don't care |
1788 | // anymore if the css is destroyed by sq_deent, sq_unspace or sq_unbackslash |
1789 | $content = $contentNew; |
1790 | } |
7d06541f |
1791 | return array($content, $newpos); |
691a2d25 |
1792 | } |
a3daaaf3 |
1793 | |
0493ed11 |
1794 | |
691a2d25 |
1795 | /** |
8bd0068d |
1796 | * This function converts cid: url's into the ones that can be viewed in |
1797 | * the browser. |
1798 | * |
1799 | * @param $message the message object |
1800 | * @param $id the message id |
1801 | * @param $cidurl the cid: url. |
1802 | * @param $mailbox the message mailbox |
1803 | * @return a string with a http-friendly url |
1804 | */ |
b3af12ef |
1805 | function sq_cid2http($message, $id, $cidurl, $mailbox){ |
691a2d25 |
1806 | /** |
8bd0068d |
1807 | * Get rid of quotes. |
1808 | */ |
691a2d25 |
1809 | $quotchar = substr($cidurl, 0, 1); |
2dd879b8 |
1810 | if ($quotchar == '"' || $quotchar == "'"){ |
1811 | $cidurl = str_replace($quotchar, "", $cidurl); |
1812 | } else { |
1813 | $quotchar = ''; |
1814 | } |
691a2d25 |
1815 | $cidurl = substr(trim($cidurl), 4); |
0493ed11 |
1816 | |
1817 | $match_str = '/\{.*?\}\//'; |
1818 | $str_rep = ''; |
1819 | $cidurl = preg_replace($match_str, $str_rep, $cidurl); |
1820 | |
e5e9381a |
1821 | $linkurl = find_ent_id($cidurl, $message); |
1822 | /* in case of non-save cid links $httpurl should be replaced by a sort of |
8bd0068d |
1823 | unsave link image */ |
e5e9381a |
1824 | $httpurl = ''; |
c8f5f606 |
1825 | |
8bd0068d |
1826 | /** |
1827 | * This is part of a fix for Outlook Express 6.x generating |
1828 | * cid URLs without creating content-id headers. These images are |
1829 | * not part of the multipart/related html mail. The html contains |
1830 | * <img src="cid:{some_id}/image_filename.ext"> references to |
1831 | * attached images with as goal to render them inline although |
1832 | * the attachment disposition property is not inline. |
1833 | */ |
c8f5f606 |
1834 | |
1835 | if (empty($linkurl)) { |
1836 | if (preg_match('/{.*}\//', $cidurl)) { |
1837 | $cidurl = preg_replace('/{.*}\//','', $cidurl); |
1838 | if (!empty($cidurl)) { |
1839 | $linkurl = find_ent_id($cidurl, $message); |
1840 | } |
1841 | } |
1842 | } |
f8a1ed5a |
1843 | |
c8f5f606 |
1844 | if (!empty($linkurl)) { |
6b04287c |
1845 | $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&' . |
8bd0068d |
1846 | "passed_id=$id&mailbox=" . urlencode($mailbox) . |
1847 | '&ent_id=' . $linkurl . $quotchar; |
c8f5f606 |
1848 | } else { |
1849 | /** |
1850 | * If we couldn't generate a proper img url, drop in a blank image |
1851 | * instead of sending back empty, otherwise it causes unusual behaviour |
1852 | */ |
bc017c1d |
1853 | $httpurl = $quotchar . SM_PATH . 'images/blank.png' . $quotchar; |
e5e9381a |
1854 | } |
f8a1ed5a |
1855 | |
691a2d25 |
1856 | return $httpurl; |
1857 | } |
1858 | |
1859 | /** |
8bd0068d |
1860 | * This function changes the <body> tag into a <div> tag since we |
1861 | * can't really have a body-within-body. |
1862 | * |
1863 | * @param $attary an array of attributes and values of <body> |
1864 | * @param $mailbox mailbox we're currently reading (for cid2http) |
1865 | * @param $message current message (for cid2http) |
1866 | * @param $id current message id (for cid2http) |
1867 | * @return a modified array of attributes to be set for <div> |
1868 | */ |
2dd879b8 |
1869 | function sq_body2div($attary, $mailbox, $message, $id){ |
b583c3e8 |
1870 | $me = 'sq_body2div'; |
3d8371be |
1871 | $divattary = Array('class' => "'bodyclass'"); |
b583c3e8 |
1872 | $text = '#000000'; |
c189a963 |
1873 | $has_bgc_stl = $has_txt_stl = false; |
b583c3e8 |
1874 | $styledef = ''; |
691a2d25 |
1875 | if (is_array($attary) && sizeof($attary) > 0){ |
1876 | foreach ($attary as $attname=>$attvalue){ |
1877 | $quotchar = substr($attvalue, 0, 1); |
1878 | $attvalue = str_replace($quotchar, "", $attvalue); |
1879 | switch ($attname){ |
3d8371be |
1880 | case 'background': |
8bd0068d |
1881 | $attvalue = sq_cid2http($message, $id, $attvalue, $mailbox); |
3d8371be |
1882 | $styledef .= "background-image: url('$attvalue'); "; |
1883 | break; |
1884 | case 'bgcolor': |
c189a963 |
1885 | $has_bgc_stl = true; |
3d8371be |
1886 | $styledef .= "background-color: $attvalue; "; |
1887 | break; |
1888 | case 'text': |
c189a963 |
1889 | $has_txt_stl = true; |
3d8371be |
1890 | $styledef .= "color: $attvalue; "; |
1891 | break; |
691a2d25 |
1892 | } |
a3daaaf3 |
1893 | } |
c189a963 |
1894 | // Outlook defines a white bgcolor and no text color. This can lead to |
1895 | // white text on a white bg with certain themes. |
1896 | if ($has_bgc_stl && !$has_txt_stl) { |
1897 | $styledef .= "color: $text; "; |
1898 | } |
691a2d25 |
1899 | if (strlen($styledef) > 0){ |
1900 | $divattary{"style"} = "\"$styledef\""; |
1901 | } |
1902 | } |
1903 | return $divattary; |
1904 | } |
a3daaaf3 |
1905 | |
691a2d25 |
1906 | /** |
8bd0068d |
1907 | * This is the main function and the one you should actually be calling. |
1908 | * There are several variables you should be aware of an which need |
1909 | * special description. |
1910 | * |
1911 | * Since the description is quite lengthy, see it here: |
1912 | * http://linux.duke.edu/projects/mini/htmlfilter/ |
1913 | * |
1914 | * @param $body the string with HTML you wish to filter |
1915 | * @param $tag_list see description above |
1916 | * @param $rm_tags_with_content see description above |
1917 | * @param $self_closing_tags see description above |
1918 | * @param $force_tag_closing see description above |
1919 | * @param $rm_attnames see description above |
1920 | * @param $bad_attvals see description above |
1921 | * @param $add_attr_to_tag see description above |
1922 | * @param $message message object |
1923 | * @param $id message id |
1924 | * @return sanitized html safe to show on your pages. |
1925 | */ |
da2415c1 |
1926 | function sq_sanitize($body, |
8bd0068d |
1927 | $tag_list, |
1928 | $rm_tags_with_content, |
1929 | $self_closing_tags, |
1930 | $force_tag_closing, |
1931 | $rm_attnames, |
1932 | $bad_attvals, |
1933 | $add_attr_to_tag, |
1934 | $message, |
1935 | $id, |
1936 | $mailbox |
1937 | ){ |
b583c3e8 |
1938 | $me = 'sq_sanitize'; |
7d06541f |
1939 | $rm_tags = array_shift($tag_list); |
691a2d25 |
1940 | /** |
8bd0068d |
1941 | * Normalize rm_tags and rm_tags_with_content. |
1942 | */ |
7d06541f |
1943 | @array_walk($tag_list, 'sq_casenormalize'); |
691a2d25 |
1944 | @array_walk($rm_tags_with_content, 'sq_casenormalize'); |
1945 | @array_walk($self_closing_tags, 'sq_casenormalize'); |
1946 | /** |
8bd0068d |
1947 | * See if tag_list is of tags to remove or tags to allow. |
1948 | * false means remove these tags |
1949 | * true means allow these tags |
1950 | */ |
691a2d25 |
1951 | $curpos = 0; |
1952 | $open_tags = Array(); |
2dd879b8 |
1953 | $trusted = "\n<!-- begin sanitized html -->\n"; |
691a2d25 |
1954 | $skip_content = false; |
bb8d0799 |
1955 | /** |
8bd0068d |
1956 | * Take care of netscape's stupid javascript entities like |
1957 | * &{alert('boo')}; |
1958 | */ |
bb8d0799 |
1959 | $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body); |
691a2d25 |
1960 | |
7d06541f |
1961 | while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){ |
691a2d25 |
1962 | list($tagname, $attary, $tagtype, $lt, $gt) = $curtag; |
1963 | $free_content = substr($body, $curpos, $lt-$curpos); |
1964 | /** |
8bd0068d |
1965 | * Take care of <style> |
1966 | */ |
7d06541f |
1967 | if ($tagname == "style" && $tagtype == 1){ |
da2415c1 |
1968 | list($free_content, $curpos) = |
e60a299a |
1969 | sq_fixstyle($body, $gt+1, $message, $id, $mailbox); |
7d06541f |
1970 | if ($free_content != FALSE){ |
1971 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
1972 | $trusted .= $free_content; |
1973 | $trusted .= sq_tagprint($tagname, false, 2); |
1974 | } |
1975 | continue; |
691a2d25 |
1976 | } |
1977 | if ($skip_content == false){ |
1978 | $trusted .= $free_content; |
691a2d25 |
1979 | } |
1980 | if ($tagname != FALSE){ |
1981 | if ($tagtype == 2){ |
1982 | if ($skip_content == $tagname){ |
1983 | /** |
8bd0068d |
1984 | * Got to the end of tag we needed to remove. |
1985 | */ |
691a2d25 |
1986 | $tagname = false; |
1987 | $skip_content = false; |
1988 | } else { |
1989 | if ($skip_content == false){ |
c828931c |
1990 | if ($tagname == "body"){ |
1991 | $tagname = "div"; |
2dd879b8 |
1992 | } |
da2415c1 |
1993 | if (isset($open_tags{$tagname}) && |
8bd0068d |
1994 | $open_tags{$tagname} > 0){ |
2dd879b8 |
1995 | $open_tags{$tagname}--; |
691a2d25 |
1996 | } else { |
2dd879b8 |
1997 | $tagname = false; |
691a2d25 |
1998 | } |
691a2d25 |
1999 | } |
2000 | } |
2001 | } else { |
2002 | /** |
8bd0068d |
2003 | * $rm_tags_with_content |
2004 | */ |
691a2d25 |
2005 | if ($skip_content == false){ |
2006 | /** |
8bd0068d |
2007 | * See if this is a self-closing type and change |
2008 | * tagtype appropriately. |
2009 | */ |
691a2d25 |
2010 | if ($tagtype == 1 |
8bd0068d |
2011 | && in_array($tagname, $self_closing_tags)){ |
2dd879b8 |
2012 | $tagtype = 3; |
691a2d25 |
2013 | } |
2014 | /** |
8bd0068d |
2015 | * See if we should skip this tag and any content |
2016 | * inside it. |
2017 | */ |
691a2d25 |
2018 | if ($tagtype == 1 && |
8bd0068d |
2019 | in_array($tagname, $rm_tags_with_content)){ |
691a2d25 |
2020 | $skip_content = $tagname; |
2021 | } else { |
da2415c1 |
2022 | if (($rm_tags == false |
8bd0068d |
2023 | && in_array($tagname, $tag_list)) || |
2024 | ($rm_tags == true && |
2025 | !in_array($tagname, $tag_list))){ |
691a2d25 |
2026 | $tagname = false; |
2027 | } else { |
2dd879b8 |
2028 | /** |
8bd0068d |
2029 | * Convert body into div. |
2030 | */ |
2dd879b8 |
2031 | if ($tagname == "body"){ |
2032 | $tagname = "div"; |
da2415c1 |
2033 | $attary = sq_body2div($attary, $mailbox, |
8bd0068d |
2034 | $message, $id); |
2dd879b8 |
2035 | } |
691a2d25 |
2036 | if ($tagtype == 1){ |
2037 | if (isset($open_tags{$tagname})){ |
2038 | $open_tags{$tagname}++; |
2039 | } else { |
2040 | $open_tags{$tagname}=1; |
2041 | } |
2042 | } |
2043 | /** |
8bd0068d |
2044 | * This is where we run other checks. |
2045 | */ |
691a2d25 |
2046 | if (is_array($attary) && sizeof($attary) > 0){ |
2047 | $attary = sq_fixatts($tagname, |
8bd0068d |
2048 | $attary, |
2049 | $rm_attnames, |
2050 | $bad_attvals, |
2051 | $add_attr_to_tag, |
2052 | $message, |
2053 | $id, |
2054 | $mailbox |
2055 | ); |
691a2d25 |
2056 | } |
2057 | } |
2058 | } |
691a2d25 |
2059 | } |
2060 | } |
2061 | if ($tagname != false && $skip_content == false){ |
2062 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
2063 | } |
691a2d25 |
2064 | } |
2065 | $curpos = $gt+1; |
a3daaaf3 |
2066 | } |
691a2d25 |
2067 | $trusted .= substr($body, $curpos, strlen($body)-$curpos); |
2068 | if ($force_tag_closing == true){ |
2069 | foreach ($open_tags as $tagname=>$opentimes){ |
2070 | while ($opentimes > 0){ |
2071 | $trusted .= '</' . $tagname . '>'; |
2072 | $opentimes--; |
2073 | } |
2074 | } |
2075 | $trusted .= "\n"; |
2076 | } |
2077 | $trusted .= "<!-- end sanitized html -->\n"; |
2078 | return $trusted; |
2079 | } |
451f74a2 |
2080 | |
691a2d25 |
2081 | /** |
8bd0068d |
2082 | * This is a wrapper function to call html sanitizing routines. |
2083 | * |
2084 | * @param $body the body of the message |
2085 | * @param $id the id of the message |
c189a963 |
2086 | |
2087 | * @param $message |
2088 | * @param $mailbox |
2089 | * @param boolean $take_mailto_links When TRUE, converts mailto: links |
2090 | * into internal SM compose links |
2091 | * (optional; default = TRUE) |
8bd0068d |
2092 | * @return a string with html safe to display in the browser. |
2093 | */ |
c189a963 |
2094 | function magicHTML($body, $id, $message, $mailbox = 'INBOX', $take_mailto_links =true) { |
2095 | |
202bcbcc |
2096 | // require_once(SM_PATH . 'functions/url_parser.php'); // for $MailTo_PReg_Match |
c189a963 |
2097 | |
691a2d25 |
2098 | global $attachment_common_show_images, $view_unsafe_images, |
8bd0068d |
2099 | $has_unsafe_images; |
691a2d25 |
2100 | /** |
8bd0068d |
2101 | * Don't display attached images in HTML mode. |
2102 | */ |
691a2d25 |
2103 | $attachment_common_show_images = false; |
2104 | $tag_list = Array( |
8bd0068d |
2105 | false, |
2106 | "object", |
2107 | "meta", |
2108 | "html", |
2109 | "head", |
2110 | "base", |
2111 | "link", |
2112 | "frame", |
2113 | "iframe", |
2114 | "plaintext", |
2115 | "marquee" |
2116 | ); |
691a2d25 |
2117 | |
2118 | $rm_tags_with_content = Array( |
8bd0068d |
2119 | "script", |
2120 | "applet", |
2121 | "embed", |
2122 | "title", |
2123 | "frameset", |
0493ed11 |
2124 | "xmp", |
8bd0068d |
2125 | "xml" |
2126 | ); |
691a2d25 |
2127 | |
2128 | $self_closing_tags = Array( |
8bd0068d |
2129 | "img", |
2130 | "br", |
2131 | "hr", |
2132 | "input", |
2133 | "outbind" |
2134 | ); |
691a2d25 |
2135 | |
2dd879b8 |
2136 | $force_tag_closing = true; |
691a2d25 |
2137 | |
2138 | $rm_attnames = Array( |
8bd0068d |
2139 | "/.*/" => |
2140 | Array( |
2141 | "/target/i", |
2142 | "/^on.*/i", |
2143 | "/^dynsrc/i", |
2144 | "/^data.*/i", |
2145 | "/^lowsrc.*/i" |
2146 | ) |
2147 | ); |
691a2d25 |
2148 | |
2149 | $secremoveimg = "../images/" . _("sec_remove_eng.png"); |
2150 | $bad_attvals = Array( |
8bd0068d |
2151 | "/.*/" => |
691a2d25 |
2152 | Array( |
0a6ec9b5 |
2153 | "/^src|background/i" => |
8bd0068d |
2154 | Array( |
691a2d25 |
2155 | Array( |
8bd0068d |
2156 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
2157 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
2158 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
691a2d25 |
2159 | ), |
8bd0068d |
2160 | Array( |
2161 | "\\1$secremoveimg\\2", |
2162 | "\\1$secremoveimg\\2", |
2163 | "\\1$secremoveimg\\2", |
8bd0068d |
2164 | ) |
2165 | ), |
0a6ec9b5 |
2166 | "/^href|action/i" => |
8bd0068d |
2167 | Array( |
0a6ec9b5 |
2168 | Array( |
8bd0068d |
2169 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
2170 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
2171 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
0a6ec9b5 |
2172 | ), |
691a2d25 |
2173 | Array( |
8bd0068d |
2174 | "\\1#\\1", |
2175 | "\\1#\\1", |
2176 | "\\1#\\1" |
02474e43 |
2177 | ) |
8bd0068d |
2178 | ), |
2179 | "/^style/i" => |
2180 | Array( |
2181 | Array( |
5db90261 |
2182 | "/\/\*.*\*\//", |
8bd0068d |
2183 | "/expression/i", |
2184 | "/binding/i", |
2185 | "/behaviou*r/i", |
2186 | "/include-source/i", |
2187 | "/position\s*:\s*absolute/i", |
1d935bc2 |
2188 | "/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", |
8bd0068d |
2189 | "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si", |
2190 | "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si", |
2191 | "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si", |
2192 | "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si" |
2193 | ), |
2194 | Array( |
5db90261 |
2195 | "", |
8bd0068d |
2196 | "idiocy", |
2197 | "idiocy", |
2198 | "idiocy", |
2199 | "idiocy", |
2200 | "", |
5b4884be |
2201 | "url", |
8bd0068d |
2202 | "url(\\1#\\1)", |
2203 | "url(\\1#\\1)", |
2204 | "url(\\1#\\1)", |
8bd0068d |
2205 | "\\1:url(\\2#\\3)" |
2206 | ) |
691a2d25 |
2207 | ) |
8bd0068d |
2208 | ) |
691a2d25 |
2209 | ); |
5262d9a6 |
2210 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
2dd879b8 |
2211 | $view_unsafe_images = false; |
45071bd6 |
2212 | } |
691a2d25 |
2213 | if (!$view_unsafe_images){ |
2214 | /** |
8bd0068d |
2215 | * Remove any references to http/https if view_unsafe_images set |
2216 | * to false. |
2217 | */ |
02474e43 |
2218 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0], |
8bd0068d |
2219 | '/^([\'\"])\s*https*:.*([\'\"])/si'); |
02474e43 |
2220 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1], |
8bd0068d |
2221 | "\\1$secremoveimg\\1"); |
02474e43 |
2222 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[0], |
7ef0b415 |
2223 | '/url\([\'\"]?https?:[^\)]*[\'\"]?\)/si'); |
02474e43 |
2224 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[1], |
8bd0068d |
2225 | "url(\\1$secremoveimg\\1)"); |
691a2d25 |
2226 | } |
451f74a2 |
2227 | |
691a2d25 |
2228 | $add_attr_to_tag = Array( |
8bd0068d |
2229 | "/^a$/i" => |
c25f2fbb |
2230 | Array('target'=>'"_blank"', |
02474e43 |
2231 | 'title'=>'"'._("This external link will open in a new window").'"' |
8bd0068d |
2232 | ) |
2233 | ); |
da2415c1 |
2234 | $trusted = sq_sanitize($body, |
8bd0068d |
2235 | $tag_list, |
2236 | $rm_tags_with_content, |
2237 | $self_closing_tags, |
2238 | $force_tag_closing, |
2239 | $rm_attnames, |
2240 | $bad_attvals, |
2241 | $add_attr_to_tag, |
2242 | $message, |
2243 | $id, |
2244 | $mailbox |
2245 | ); |
f83c60a2 |
2246 | if (preg_match("|$secremoveimg|i", $trusted)){ |
691a2d25 |
2247 | $has_unsafe_images = true; |
da2415c1 |
2248 | } |
c189a963 |
2249 | |
2250 | // we want to parse mailto's in HTML output, change to SM compose links |
2251 | // this is a modified version of code from url_parser.php... but Marc is |
2252 | // right: we need a better filtering implementation; adding this randomly |
2253 | // here is not a great solution |
2254 | // |
2255 | if ($take_mailto_links) { |
2256 | // parseUrl($trusted); // this even parses URLs inside of tags... too aggressive |
2257 | global $MailTo_PReg_Match; |
202bcbcc |
2258 | $MailTo_PReg_Match = '/mailto:' . substr($MailTo_PReg_Match, 1) ; |
c189a963 |
2259 | if ((preg_match_all($MailTo_PReg_Match, $trusted, $regs)) && ($regs[0][0] != '')) { |
2260 | foreach ($regs[0] as $i => $mailto_before) { |
2261 | $mailto_params = $regs[10][$i]; |
2262 | // get rid of any tailing quote since we have to add send_to to the end |
2263 | // |
2264 | if (substr($mailto_before, strlen($mailto_before) - 1) == '"') |
2265 | $mailto_before = substr($mailto_before, 0, strlen($mailto_before) - 1); |
2266 | if (substr($mailto_params, strlen($mailto_params) - 1) == '"') |
2267 | $mailto_params = substr($mailto_params, 0, strlen($mailto_params) - 1); |
2268 | |
2269 | if ($regs[1][$i]) { //if there is an email addr before '?', we need to merge it with the params |
2270 | $to = 'to=' . $regs[1][$i]; |
2271 | if (strpos($mailto_params, 'to=') > -1) //already a 'to=' |
2272 | $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params); |
2273 | else { |
2274 | if ($mailto_params) //already some params, append to them |
2275 | $mailto_params .= '&' . $to; |
2276 | else |
2277 | $mailto_params .= '?' . $to; |
2278 | } |
2279 | } |
2280 | |
2281 | $url_str = preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params); |
2282 | |
2283 | // we'll already have target=_blank, no need to allow comp_in_new |
2284 | // here (which would be a lot more work anyway) |
2285 | // |
2286 | global $compose_new_win; |
2287 | $temp_comp_in_new = $compose_new_win; |
2288 | $compose_new_win = 0; |
2289 | $comp_uri = makeComposeLink('src/compose.php' . $url_str, $mailto_before); |
2290 | $compose_new_win = $temp_comp_in_new; |
2291 | |
2292 | // remove <a href=" and anything after the next quote (we only |
2293 | // need the uri, not the link HTML) in compose uri |
2294 | // |
2295 | $comp_uri = substr($comp_uri, 9); |
2296 | $comp_uri = substr($comp_uri, 0, strpos($comp_uri, '"', 1)); |
2297 | $trusted = str_replace($mailto_before, $comp_uri, $trusted); |
2298 | } |
2299 | } |
2300 | } |
2301 | |
691a2d25 |
2302 | return $trusted; |
451f74a2 |
2303 | } |
a4a70693 |
2304 | |
da2415c1 |
2305 | /** |
8bd0068d |
2306 | * function SendDownloadHeaders - send file to the browser |
2307 | * |
2308 | * Original Source: SM core src/download.php |
2309 | * moved here to make it available to other code, and separate |
2310 | * front end from back end functionality. |
2311 | * |
2312 | * @param string $type0 first half of mime type |
2313 | * @param string $type1 second half of mime type |
2314 | * @param string $filename filename to tell the browser for downloaded file |
2315 | * @param boolean $force whether to force the download dialog to pop |
2316 | * @param optional integer $filesize send the Content-Header and length to the browser |
2317 | * @return void |
2318 | */ |
02474e43 |
2319 | function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) { |
2320 | global $languages, $squirrelmail_language; |
cfffd60b |
2321 | $isIE = $isIE6plus = false; |
02474e43 |
2322 | |
2323 | sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER); |
2324 | |
2325 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false && |
8bd0068d |
2326 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
cfffd60b |
2327 | $isIE = true; |
02474e43 |
2328 | } |
2329 | |
cfffd60b |
2330 | if (preg_match('/compatible; MSIE ([0-9]+)/', $HTTP_USER_AGENT, $match) && |
2331 | ((int)$match[1]) >= 6 && strstr($HTTP_USER_AGENT, 'Opera') === false) { |
2332 | $isIE6plus = true; |
02474e43 |
2333 | } |
2334 | |
2335 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
8bd0068d |
2336 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) { |
02474e43 |
2337 | $filename = |
8bd0068d |
2338 | call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT); |
02474e43 |
2339 | } else { |
2340 | $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename)); |
2341 | } |
2342 | |
2343 | // A Pox on Microsoft and it's Internet Explorer! |
2344 | // |
2345 | // IE has lots of bugs with file downloads. |
2346 | // It also has problems with SSL. Both of these cause problems |
2347 | // for us in this function. |
2348 | // |
2349 | // See this article on Cache Control headers and SSL |
2350 | // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308 |
2351 | // |
2352 | // The best thing you can do for IE is to upgrade to the latest |
2353 | // version |
2354 | //set all the Cache Control Headers for IE |
2355 | if ($isIE) { |
2356 | $filename=rawurlencode($filename); |
2357 | header ("Pragma: public"); |
8bd0068d |
2358 | header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); // HTTP/1.1 |
02474e43 |
2359 | header ("Cache-Control: post-check=0, pre-check=0", false); |
8bd0068d |
2360 | header ("Cache-Control: private"); |
02474e43 |
2361 | |
2362 | //set the inline header for IE, we'll add the attachment header later if we need it |
2363 | header ("Content-Disposition: inline; filename=$filename"); |
2364 | } |
2365 | |
2366 | if (!$force) { |
2367 | // Try to show in browser window |
2368 | header ("Content-Disposition: inline; filename=\"$filename\""); |
2369 | header ("Content-Type: $type0/$type1; name=\"$filename\""); |
2370 | } else { |
2371 | // Try to pop up the "save as" box |
2372 | |
2373 | // IE makes this hard. It pops up 2 save boxes, or none. |
2374 | // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP |
2375 | // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519 |
2376 | // But, according to Microsoft, it is "RFC compliant but doesn't |
2377 | // take into account some deviations that allowed within the |
2378 | // specification." Doesn't that mean RFC non-compliant? |
2379 | // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP |
2380 | |
2381 | // all browsers need the application/octet-stream header for this |
2382 | header ("Content-Type: application/octet-stream; name=\"$filename\""); |
2383 | |
2384 | // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp |
2385 | // Do not have quotes around filename, but that applied to |
2386 | // "attachment"... does it apply to inline too? |
2387 | header ("Content-Disposition: attachment; filename=\"$filename\""); |
2388 | |
cfffd60b |
2389 | if ($isIE && !$isIE6plus) { |
02474e43 |
2390 | // This combination seems to work mostly. IE 5.5 SP 1 has |
2391 | // known issues (see the Microsoft Knowledge Base) |
2392 | |
2393 | // This works for most types, but doesn't work with Word files |
2394 | header ("Content-Type: application/download; name=\"$filename\""); |
2395 | |
2396 | // These are spares, just in case. :-) |
2397 | //header("Content-Type: $type0/$type1; name=\"$filename\""); |
2398 | //header("Content-Type: application/x-msdownload; name=\"$filename\""); |
2399 | //header("Content-Type: application/octet-stream; name=\"$filename\""); |
2400 | } else { |
2401 | // another application/octet-stream forces download for Netscape |
2402 | header ("Content-Type: application/octet-stream; name=\"$filename\""); |
2403 | } |
2404 | } |
2405 | |
2406 | //send the content-length header if the calling function provides it |
2407 | if ($filesize > 0) { |
2408 | header("Content-Length: $filesize"); |
2409 | } |
07c49f57 |
2410 | |
8d863f64 |
2411 | } // end fn SendDownloadHeaders |