| | 1 | <?php |
| | 2 | |
| | 3 | /** |
| | 4 | * mime.php |
| | 5 | * |
| | 6 | * This contains the functions necessary to detect and decode MIME |
| | 7 | * messages. |
| | 8 | * |
| | 9 | * @copyright © 1999-2006 The SquirrelMail Project Team |
| | 10 | * @license http://opensource.org/licenses/gpl-license.php GNU Public License |
| | 11 | * @version $Id$ |
| | 12 | * @package squirrelmail |
| | 13 | */ |
| | 14 | |
| | 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 | |
| | 41 | |
| | 42 | /* -------------------------------------------------------------------------- */ |
| | 43 | /* MIME DECODING */ |
| | 44 | /* -------------------------------------------------------------------------- */ |
| | 45 | |
| | 46 | /** |
| | 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 | */ |
| | 53 | function mime_structure ($bodystructure, $flags=array()) { |
| | 54 | |
| | 55 | /* Isolate the body structure and remove beginning and end parenthesis. */ |
| | 56 | $read = trim(substr ($bodystructure, strpos(strtolower($bodystructure), 'bodystructure') + 13)); |
| | 57 | $read = trim(substr ($read, 0, -1)); |
| | 58 | $i = 0; |
| | 59 | $msg = Message::parseStructure($read,$i); |
| | 60 | if (!is_object($msg)) { |
| | 61 | global $color, $mailbox; |
| | 62 | /* removed urldecode because $_GET is auto urldecoded ??? */ |
| | 63 | displayPageHeader( $color, $mailbox ); |
| | 64 | $errormessage = _("SquirrelMail could not decode the bodystructure of the message"); |
| | 65 | $errormessage .= '<br />'._("The bodystructure provided by your IMAP server:").'<br /><br />'; |
| | 66 | $errormessage .= '<pre>' . htmlspecialchars($read) . '</pre>'; |
| | 67 | plain_error_message( $errormessage, $color ); |
| | 68 | echo '</body></html>'; |
| | 69 | exit; |
| | 70 | } |
| | 71 | if (count($flags)) { |
| | 72 | foreach ($flags as $flag) { |
| | 73 | $char = strtoupper($flag{1}); |
| | 74 | switch ($char) { |
| | 75 | case 'S': |
| | 76 | if (strtolower($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 | $msg->is_mdnsent = true; |
| | 98 | } |
| | 99 | break; |
| | 100 | default: |
| | 101 | break; |
| | 102 | } |
| | 103 | } |
| | 104 | } |
| | 105 | // listEntities($msg); |
| | 106 | return $msg; |
| | 107 | } |
| | 108 | |
| | 109 | |
| | 110 | |
| | 111 | /* This starts the parsing of a particular structure. It is called recursively, |
| | 112 | * so it can be passed different structures. It returns an object of type |
| | 113 | * $message. |
| | 114 | * First, it checks to see if it is a multipart message. If it is, then it |
| | 115 | * handles that as it sees is necessary. If it is just a regular entity, |
| | 116 | * then it parses it and adds the necessary header information (by calling out |
| | 117 | * to mime_get_elements() |
| | 118 | */ |
| | 119 | |
| | 120 | function mime_fetch_body($imap_stream, $id, $ent_id=1, $fetch_size=0) { |
| | 121 | /* Do a bit of error correction. If we couldn't find the entity id, just guess |
| | 122 | * that it is the first one. That is usually the case anyway. |
| | 123 | */ |
| | 124 | |
| | 125 | if (!$ent_id) { |
| | 126 | $cmd = "FETCH $id BODY[]"; |
| | 127 | } else { |
| | 128 | $cmd = "FETCH $id BODY[$ent_id]"; |
| | 129 | } |
| | 130 | |
| | 131 | if ($fetch_size!=0) $cmd .= "<0.$fetch_size>"; |
| | 132 | |
| | 133 | $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, TRUE); |
| | 134 | do { |
| | 135 | $topline = trim(array_shift($data)); |
| | 136 | } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ; |
| | 137 | |
| | 138 | $wholemessage = implode('', $data); |
| | 139 | if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) { |
| | 140 | $ret = substr($wholemessage, 0, $regs[1]); |
| | 141 | /* There is some information in the content info header that could be important |
| | 142 | * in order to parse html messages. Let's get them here. |
| | 143 | */ |
| | 144 | // if ($ret{0} == '<') { |
| | 145 | // $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, TRUE); |
| | 146 | // } |
| | 147 | } else if (ereg('"([^"]*)"', $topline, $regs)) { |
| | 148 | $ret = $regs[1]; |
| | 149 | } else { |
| | 150 | global $where, $what, $mailbox, $passed_id, $startMessage; |
| | 151 | $par = 'mailbox=' . urlencode($mailbox) . '&passed_id=' . $passed_id; |
| | 152 | if (isset($where) && isset($what)) { |
| | 153 | $par .= '&where=' . urlencode($where) . '&what=' . urlencode($what); |
| | 154 | } else { |
| | 155 | $par .= '&startMessage=' . $startMessage . '&show_more=0'; |
| | 156 | } |
| | 157 | $par .= '&response=' . urlencode($response) . |
| | 158 | '&message=' . urlencode($message) . |
| | 159 | '&topline=' . urlencode($topline); |
| | 160 | |
| | 161 | echo '<tt><br />' . |
| | 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 />"; |
| | 171 | |
| | 172 | $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, TRUE); |
| | 173 | array_shift($data); |
| | 174 | $wholemessage = implode('', $data); |
| | 175 | |
| | 176 | $ret = $wholemessage; |
| | 177 | } |
| | 178 | return $ret; |
| | 179 | } |
| | 180 | |
| | 181 | function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding, $rStream='php://stdout') { |
| | 182 | |
| | 183 | /* Don't kill the connection if the browser is over a dialup |
| | 184 | * and it would take over 30 seconds to download it. |
| | 185 | * Don't call set_time_limit in safe mode. |
| | 186 | */ |
| | 187 | |
| | 188 | if (!ini_get('safe_mode')) { |
| | 189 | set_time_limit(0); |
| | 190 | } |
| | 191 | /* in case of base64 encoded attachments, do not buffer them. |
| | 192 | Instead, echo the decoded attachment directly to screen */ |
| | 193 | if (strtolower($encoding) == 'base64') { |
| | 194 | if (!$ent_id) { |
| | 195 | $query = "FETCH $id BODY[]"; |
| | 196 | } else { |
| | 197 | $query = "FETCH $id BODY[$ent_id]"; |
| | 198 | } |
| | 199 | sqimap_run_command($imap_stream,$query,true,$response,$message,TRUE,'sqimap_base64_decode',$rStream,true); |
| | 200 | } else { |
| | 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 | } |
| | 207 | } |
| | 208 | |
| | 209 | /* |
| | 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 | */ |
| | 219 | /* |
| | 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 | */ |
| | 223 | |
| | 224 | return; |
| | 225 | } |
| | 226 | |
| | 227 | /* -[ END MIME DECODING ]----------------------------------------------------------- */ |
| | 228 | |
| | 229 | /* This is here for debugging purposes. It will print out a list |
| | 230 | * of all the entity IDs that are in the $message object. |
| | 231 | */ |
| | 232 | function listEntities ($message) { |
| | 233 | if ($message) { |
| | 234 | echo "<tt>" . $message->entity_id . ' : ' . $message->type0 . '/' . $message->type1 . ' parent = '. $message->parent->entity_id. '<br />'; |
| | 235 | for ($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 $msg; |
| | 242 | } |
| | 243 | } |
| | 244 | } |
| | 245 | } |
| | 246 | |
| | 247 | function getPriorityStr($priority) { |
| | 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; |
| | 272 | } |
| | 273 | |
| | 274 | /* returns a $message object for a particular entity id */ |
| | 275 | function getEntity ($message, $ent_id) { |
| | 276 | return $message->getEntity($ent_id); |
| | 277 | } |
| | 278 | |
| | 279 | /* translateText |
| | 280 | * Extracted from strings.php 23/03/2002 |
| | 281 | */ |
| | 282 | |
| | 283 | function translateText(&$body, $wrap_at, $charset) { |
| | 284 | global $where, $what; /* from searching */ |
| | 285 | global $color; /* color theme */ |
| | 286 | |
| | 287 | // require_once(SM_PATH . 'functions/url_parser.php'); |
| | 288 | |
| | 289 | $body_ary = explode("\n", $body); |
| | 290 | for ($i=0; $i < count($body_ary); $i++) { |
| | 291 | $line = $body_ary[$i]; |
| | 292 | if (strlen($line) - 2 >= $wrap_at) { |
| | 293 | sqWordWrap($line, $wrap_at, $charset); |
| | 294 | } |
| | 295 | $line = charset_decode($charset, $line); |
| | 296 | $line = str_replace("\t", ' ', $line); |
| | 297 | |
| | 298 | parseUrl ($line); |
| | 299 | |
| | 300 | $quotes = 0; |
| | 301 | $pos = 0; |
| | 302 | $j = strlen($line); |
| | 303 | |
| | 304 | while ($pos < $j) { |
| | 305 | if ($line[$pos] == ' ') { |
| | 306 | $pos++; |
| | 307 | } else if (strpos($line, '>', $pos) === $pos) { |
| | 308 | $pos += 4; |
| | 309 | $quotes++; |
| | 310 | } else { |
| | 311 | break; |
| | 312 | } |
| | 313 | } |
| | 314 | |
| | 315 | if ($quotes % 2) { |
| | 316 | $line = '<span class="quote1">' . $line . '</span>'; |
| | 317 | } elseif ($quotes) { |
| | 318 | $line = '<span class="quote2">' . $line . '</span>'; |
| | 319 | } |
| | 320 | |
| | 321 | $body_ary[$i] = $line; |
| | 322 | } |
| | 323 | $body = '<pre>' . implode("\n", $body_ary) . '</pre>'; |
| | 324 | } |
| | 325 | |
| | 326 | /** |
| | 327 | * This returns a parsed string called $body. That string can then |
| | 328 | * be displayed as the actual message in the HTML. It contains |
| | 329 | * everything needed, including HTML Tags, Attachments at the |
| | 330 | * bottom, etc. |
| | 331 | * |
| | 332 | * Since 1.2.0 function uses message_body hook. |
| | 333 | * Till 1.3.0 function included output of formatAttachments(). |
| | 334 | * |
| | 335 | * @param resource $imap_stream imap connection resource |
| | 336 | * @param object $message squirrelmail message object |
| | 337 | * @param array $color squirrelmail color theme array |
| | 338 | * @param integer $wrap_at number of characters per line |
| | 339 | * @param string $ent_num (since 1.3.0) message part id |
| | 340 | * @param integer $id (since 1.3.0) message id |
| | 341 | * @param string $mailbox (since 1.3.0) imap folder name |
| | 342 | * @param boolean $clean (since 1.5.1) Do not output stuff that's irrelevant for the printable version. |
| | 343 | * @return string html formated message text |
| | 344 | */ |
| | 345 | function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) { |
| | 346 | /* This if statement checks for the entity to show as the |
| | 347 | * primary message. To add more of them, just put them in the |
| | 348 | * order that is their priority. |
| | 349 | */ |
| | 350 | global $startMessage, $languages, $squirrelmail_language, |
| | 351 | $show_html_default, $sort, $has_unsafe_images, $passed_ent_id, |
| | 352 | $use_iframe, $iframe_height, $download_and_unsafe_link, |
| | 353 | $download_href, $unsafe_image_toggle_href, $unsafe_image_toggle_text; |
| | 354 | |
| | 355 | // workaround for not updated config.php |
| | 356 | if (! isset($use_iframe)) $use_iframe = false; |
| | 357 | |
| | 358 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
| | 359 | $view_unsafe_images = false; |
| | 360 | } |
| | 361 | |
| | 362 | $body = ''; |
| | 363 | $urlmailbox = urlencode($mailbox); |
| | 364 | $body_message = getEntity($message, $ent_num); |
| | 365 | if (($body_message->header->type0 == 'text') || |
| | 366 | ($body_message->header->type0 == 'rfc822')) { |
| | 367 | $body = mime_fetch_body ($imap_stream, $id, $ent_num); |
| | 368 | $body = decodeBody($body, $body_message->header->encoding); |
| | 369 | |
| | 370 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| | 371 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) { |
| | 372 | if (mb_detect_encoding($body) != 'ASCII') { |
| | 373 | $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body); |
| | 374 | } |
| | 375 | } |
| | 376 | $hookResults = do_hook("message_body", $body); |
| | 377 | $body = $hookResults[1]; |
| | 378 | |
| | 379 | /* If there are other types that shouldn't be formatted, add |
| | 380 | * them here. |
| | 381 | */ |
| | 382 | |
| | 383 | if ($body_message->header->type1 == 'html') { |
| | 384 | if ($show_html_default <> 1) { |
| | 385 | $entity_conv = array(' ' => ' ', |
| | 386 | '<p>' => "\n", |
| | 387 | '<P>' => "\n", |
| | 388 | '<br>' => "\n", |
| | 389 | '<BR>' => "\n", |
| | 390 | '<br />' => "\n", |
| | 391 | '<BR />' => "\n", |
| | 392 | '>' => '>', |
| | 393 | '<' => '<'); |
| | 394 | $body = strtr($body, $entity_conv); |
| | 395 | $body = strip_tags($body); |
| | 396 | $body = trim($body); |
| | 397 | translateText($body, $wrap_at, |
| | 398 | $body_message->header->getParameter('charset')); |
| | 399 | } elseif ($use_iframe && ! $clean) { |
| | 400 | // $clean is used to remove iframe in printable view. |
| | 401 | |
| | 402 | /** |
| | 403 | * If we don't add html message between iframe tags, |
| | 404 | * we must detect unsafe images and modify $has_unsafe_images. |
| | 405 | */ |
| | 406 | $html_body = magicHTML($body, $id, $message, $mailbox); |
| | 407 | // Convert character set in order to display html mails in different character set |
| | 408 | $html_body = charset_decode($body_message->header->getParameter('charset'),$html_body,false,true); |
| | 409 | |
| | 410 | // creating iframe url |
| | 411 | $iframeurl=sqm_baseuri().'src/view_html.php?' |
| | 412 | . 'mailbox=' . $urlmailbox |
| | 413 | . '&passed_id=' . $id |
| | 414 | . '&ent_id=' . $ent_num |
| | 415 | . '&view_unsafe_images=' . (int) $view_unsafe_images; |
| | 416 | |
| | 417 | // adding warning message |
| | 418 | $body = html_tag('div',_("Viewing HTML formatted email"),'center'); |
| | 419 | |
| | 420 | /** |
| | 421 | * height can't be set to 100%, because it does not work as expected when |
| | 422 | * iframe is inside the table. Browsers do not create full height objects |
| | 423 | * even when iframe is not nested. Maybe there is some way to get full size |
| | 424 | * with CSS. Tested in firefox 1.02 and opera 7.53 |
| | 425 | * |
| | 426 | * width="100%" does not work as expected, when table width is not set (automatic) |
| | 427 | * |
| | 428 | * tokul: I think <iframe> are safer sandbox than <object>. Objects might |
| | 429 | * need special handling for IE and IE6SP2. |
| | 430 | */ |
| | 431 | $body.= "<div><iframe name=\"message_frame\" width=\"100%\" height=\"$iframe_height\" src=\"$iframeurl\"" |
| | 432 | .' frameborder="1" marginwidth="0" marginheight="0" scrolling="auto">' . "\n"; |
| | 433 | |
| | 434 | // Message for browsers without iframe support |
| | 435 | //$body.= _("Your browser does not support inline frames. |
| | 436 | // You can view HTML formated message by following below link."); |
| | 437 | //$body.= "<br /><a href=\"$iframeurl\">"._("View HTML Message")."</a>"; |
| | 438 | |
| | 439 | // if browser can't render iframe, it renders html message. |
| | 440 | $body.= $html_body; |
| | 441 | |
| | 442 | // close iframe |
| | 443 | $body.="</iframe></div>\n"; |
| | 444 | } else { |
| | 445 | // old way of html rendering |
| | 446 | $body = magicHTML($body, $id, $message, $mailbox); |
| | 447 | /** |
| | 448 | * convert character set. charset_decode does not remove html special chars |
| | 449 | * applied by magicHTML functions and does not sanitize them second time if |
| | 450 | * fourth argument is true. |
| | 451 | */ |
| | 452 | $body = charset_decode($body_message->header->getParameter('charset'),$body,false,true); |
| | 453 | } |
| | 454 | } else { |
| | 455 | translateText($body, $wrap_at, |
| | 456 | $body_message->header->getParameter('charset')); |
| | 457 | } |
| | 458 | |
| | 459 | // if this is the clean display (i.e. printer friendly), stop here. |
| | 460 | if ( $clean ) { |
| | 461 | return $body; |
| | 462 | } |
| | 463 | |
| | 464 | $download_and_unsafe_link = ''; |
| | 465 | |
| | 466 | $link = 'passed_id=' . $id . '&ent_id='.$ent_num. |
| | 467 | '&mailbox=' . $urlmailbox .'&sort=' . $sort . |
| | 468 | '&startMessage=' . $startMessage . '&show_more=0'; |
| | 469 | if (isset($passed_ent_id)) { |
| | 470 | $link .= '&passed_ent_id='.$passed_ent_id; |
| | 471 | } |
| | 472 | $download_href = SM_PATH . 'src/download.php?absolute_dl=true&' . $link; |
| | 473 | $download_and_unsafe_link .= ' | <a href="'. $download_href .'">' . _("Download this as a file") . '</a>'; |
| | 474 | if ($view_unsafe_images) { |
| | 475 | $text = _("Hide Unsafe Images"); |
| | 476 | } else { |
| | 477 | if (isset($has_unsafe_images) && $has_unsafe_images) { |
| | 478 | $link .= '&view_unsafe_images=1'; |
| | 479 | $text = _("View Unsafe Images"); |
| | 480 | } else { |
| | 481 | $text = ''; |
| | 482 | } |
| | 483 | } |
| | 484 | if($text != '') { |
| | 485 | $unsafe_image_toggle_href = SM_PATH . 'src/read_body.php?'.$link; |
| | 486 | $unsafe_image_toggle_text = $text; |
| | 487 | $download_and_unsafe_link .= ' | <a href="'. $unsafe_image_toggle_href .'">' . $text . '</a>'; |
| | 488 | } |
| | 489 | } |
| | 490 | return $body; |
| | 491 | } |
| | 492 | |
| | 493 | /** |
| | 494 | * Displays attachment links and information |
| | 495 | * |
| | 496 | * Since 1.3.0 function is not included in formatBody() call. |
| | 497 | * |
| | 498 | * Since 1.0.2 uses attachment $type0/$type1 hook. |
| | 499 | * Since 1.2.5 uses attachment $type0/* hook. |
| | 500 | * Since 1.5.0 uses attachments_bottom hook. |
| | 501 | * Since 1.5.2 uses templates and does *not* return a value. |
| | 502 | * |
| | 503 | * @param object $message SquirrelMail message object |
| | 504 | * @param array $exclude_id message parts that are not attachments. |
| | 505 | * @param string $mailbox mailbox name |
| | 506 | * @param integer $id message id |
| | 507 | */ |
| | 508 | function formatAttachments($message, $exclude_id, $mailbox, $id) { |
| | 509 | global $where, $what, $startMessage, $color, $passed_ent_id, $base_uri, |
| | 510 | $oTemplate; |
| | 511 | |
| | 512 | $att_ar = $message->getAttachments($exclude_id); |
| | 513 | $urlMailbox = urlencode($mailbox); |
| | 514 | |
| | 515 | $attach = array(); |
| | 516 | foreach ($att_ar as $att) { |
| | 517 | $ent = $att->entity_id; |
| | 518 | $header = $att->header; |
| | 519 | $type0 = strtolower($header->type0); |
| | 520 | $type1 = strtolower($header->type1); |
| | 521 | $name = ''; |
| | 522 | $links = array(); |
| | 523 | $links['download link']['text'] = _("Download"); |
| | 524 | $links['download link']['href'] = $base_uri . |
| | 525 | "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent"; |
| | 526 | |
| | 527 | if ($type0 =='message' && $type1 == 'rfc822') { |
| | 528 | $default_page = $base_uri . 'src/read_body.php'; |
| | 529 | $rfc822_header = $att->rfc822_header; |
| | 530 | $filename = $rfc822_header->subject; |
| | 531 | if (trim( $filename ) == '') { |
| | 532 | $filename = 'untitled-[' . $ent . ']' ; |
| | 533 | } |
| | 534 | $from_o = $rfc822_header->from; |
| | 535 | if (is_object($from_o)) { |
| | 536 | $from_name = decodeHeader($from_o->getAddress(false)); |
| | 537 | } elseif (is_array($from_o) && count($from_o) && is_object($from_o[0])) { |
| | 538 | // something weird happens when a digest message is opened and you return to the digest |
| | 539 | // now the from object is part of an array. Probably the parseHeader call overwrites the info |
| | 540 | // retrieved from the bodystructure in a different way. We need to fix this later. |
| | 541 | // possible starting point, do not fetch header we already have and inspect how |
| | 542 | // the rfc822_header object behaves. |
| | 543 | $from_name = decodeHeader($from_o[0]->getAddress(false)); |
| | 544 | } else { |
| | 545 | $from_name = _("Unknown sender"); |
| | 546 | } |
| | 547 | $description = _("From").': '.$from_name; |
| | 548 | } else { |
| | 549 | $default_page = $base_uri . 'src/download.php'; |
| | 550 | $filename = $att->getFilename(); |
| | 551 | if ($header->description) { |
| | 552 | $description = decodeHeader($header->description); |
| | 553 | } else { |
| | 554 | $description = ''; |
| | 555 | } |
| | 556 | } |
| | 557 | |
| | 558 | $display_filename = $filename; |
| | 559 | if (isset($passed_ent_id)) { |
| | 560 | $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id; |
| | 561 | } else { |
| | 562 | $passed_ent_id_link = ''; |
| | 563 | } |
| | 564 | $defaultlink = $default_page . "?startMessage=$startMessage" |
| | 565 | . "&passed_id=$id&mailbox=$urlMailbox" |
| | 566 | . '&ent_id='.$ent.$passed_ent_id_link; |
| | 567 | if ($where && $what) { |
| | 568 | $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what); |
| | 569 | } |
| | 570 | |
| | 571 | /* This executes the attachment hook with a specific MIME-type. |
| | 572 | * If that doesn't have results, it tries if there's a rule |
| | 573 | * for a more generic type. |
| | 574 | */ |
| | 575 | $hookresults = do_hook("attachment $type0/$type1", $links, |
| | 576 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
| | 577 | $display_filename, $where, $what); |
| | 578 | if(count($hookresults[1]) <= 1) { |
| | 579 | $hookresults = do_hook("attachment $type0/*", $links, |
| | 580 | $startMessage, $id, $urlMailbox, $ent, $defaultlink, |
| | 581 | $display_filename, $where, $what); |
| | 582 | } |
| | 583 | |
| | 584 | $links = $hookresults[1]; |
| | 585 | $defaultlink = $hookresults[6]; |
| | 586 | |
| | 587 | $a = array(); |
| | 588 | $a['Name'] = decodeHeader($display_filename); |
| | 589 | $a['Description'] = $description; |
| | 590 | $a['DefaultHREF'] = $defaultlink; |
| | 591 | $a['DownloadHREF'] = $links['download link']['href']; |
| | 592 | $a['ViewHREF'] = isset($links['attachment_common']) ? $links['attachment_common']['href'] : ''; |
| | 593 | $a['Size'] = $header->size; |
| | 594 | $a['ContentType'] = htmlspecialchars($type0 .'/'. $type1); |
| | 595 | $a['OtherLinks'] = array(); |
| | 596 | foreach ($links as $val) { |
| | 597 | if ($val['text']==_("Download") || $val['text'] == _("View")) |
| | 598 | continue; |
| | 599 | if (empty($val['text']) && empty($val['extra'])) |
| | 600 | continue; |
| | 601 | |
| | 602 | $t = array(); |
| | 603 | $t['HREF'] = $val['href']; |
| | 604 | $t['Text'] = (empty($val['text']) ? '' : $val['text']) . (empty($val['extra']) ? '' : $val['extra']); |
| | 605 | $a['OtherLinks'][] = $t; |
| | 606 | } |
| | 607 | $attach[] = $a; |
| | 608 | |
| | 609 | unset($links); |
| | 610 | } |
| | 611 | |
| | 612 | $oTemplate->assign('attachments', $attach); |
| | 613 | $oTemplate->display('read_attachments.tpl'); |
| | 614 | } |
| | 615 | |
| | 616 | function sqimap_base64_decode(&$string) { |
| | 617 | |
| | 618 | // Base64 encoded data goes in pairs of 4 bytes. To achieve on the |
| | 619 | // fly decoding (to reduce memory usage) you have to check if the |
| | 620 | // data has incomplete pairs |
| | 621 | |
| | 622 | // Remove the noise in order to check if the 4 bytes pairs are complete |
| | 623 | $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string); |
| | 624 | |
| | 625 | $sStringRem = ''; |
| | 626 | $iMod = strlen($string) % 4; |
| | 627 | if ($iMod) { |
| | 628 | $sStringRem = substr($string,-$iMod); |
| | 629 | // Check if $sStringRem contains padding characters |
| | 630 | if (substr($sStringRem,-1) != '=') { |
| | 631 | $string = substr($string,0,-$iMod); |
| | 632 | } else { |
| | 633 | $sStringRem = ''; |
| | 634 | } |
| | 635 | } |
| | 636 | $string = base64_decode($string); |
| | 637 | return $sStringRem; |
| | 638 | } |
| | 639 | |
| | 640 | /** |
| | 641 | * Decodes encoded message body |
| | 642 | * |
| | 643 | * This function decodes the body depending on the encoding type. |
| | 644 | * Currently quoted-printable and base64 encodings are supported. |
| | 645 | * decode_body hook was added to this function in 1.4.2/1.5.0 |
| | 646 | * @param string $body encoded message body |
| | 647 | * @param string $encoding used encoding |
| | 648 | * @return string decoded string |
| | 649 | * @since 1.0 |
| | 650 | */ |
| | 651 | function decodeBody($body, $encoding) { |
| | 652 | |
| | 653 | $body = str_replace("\r\n", "\n", $body); |
| | 654 | $encoding = strtolower($encoding); |
| | 655 | |
| | 656 | $encoding_handler = do_hook_function('decode_body', $encoding); |
| | 657 | |
| | 658 | |
| | 659 | // plugins get first shot at decoding the body |
| | 660 | // |
| | 661 | if (!empty($encoding_handler) && function_exists($encoding_handler)) { |
| | 662 | $body = $encoding_handler('decode', $body); |
| | 663 | |
| | 664 | } elseif ($encoding == 'quoted-printable' || |
| | 665 | $encoding == 'quoted_printable') { |
| | 666 | /** |
| | 667 | * quoted_printable_decode() function is broken in older |
| | 668 | * php versions. Text with \r\n decoding was fixed only |
| | 669 | * in php 4.3.0. Minimal code requirement 4.0.4 + |
| | 670 | * str_replace("\r\n", "\n", $body); call. |
| | 671 | */ |
| | 672 | $body = quoted_printable_decode($body); |
| | 673 | } elseif ($encoding == 'base64') { |
| | 674 | $body = base64_decode($body); |
| | 675 | } |
| | 676 | |
| | 677 | // All other encodings are returned raw. |
| | 678 | return $body; |
| | 679 | } |
| | 680 | |
| | 681 | /** |
| | 682 | * Decodes headers |
| | 683 | * |
| | 684 | * This functions decode strings that is encoded according to |
| | 685 | * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text). |
| | 686 | * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002 |
| | 687 | * |
| | 688 | * @param string $string header string that has to be made readable |
| | 689 | * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true |
| | 690 | * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true |
| | 691 | * @param boolean $decide decide if string can be utfencoded. defaults to false |
| | 692 | * @return string decoded header string |
| | 693 | */ |
| | 694 | function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) { |
| | 695 | global $languages, $squirrelmail_language,$default_charset; |
| | 696 | if (is_array($string)) { |
| | 697 | $string = implode("\n", $string); |
| | 698 | } |
| | 699 | |
| | 700 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| | 701 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) { |
| | 702 | $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string); |
| | 703 | // Do we need to return at this point? |
| | 704 | // return $string; |
| | 705 | } |
| | 706 | $i = 0; |
| | 707 | $iLastMatch = -2; |
| | 708 | $encoded = true; |
| | 709 | |
| | 710 | $aString = explode(' ',$string); |
| | 711 | $ret = ''; |
| | 712 | foreach ($aString as $chunk) { |
| | 713 | if ($encoded && $chunk === '') { |
| | 714 | continue; |
| | 715 | } elseif ($chunk === '') { |
| | 716 | $ret .= ' '; |
| | 717 | continue; |
| | 718 | } |
| | 719 | $encoded = false; |
| | 720 | /* if encoded words are not separated by a linear-space-white we still catch them */ |
| | 721 | $j = $i-1; |
| | 722 | |
| | 723 | while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) { |
| | 724 | /* if the last chunk isn't an encoded string then put back the space, otherwise don't */ |
| | 725 | if ($iLastMatch !== $j) { |
| | 726 | if ($htmlsave) { |
| | 727 | $ret .= ' '; |
| | 728 | } else { |
| | 729 | $ret .= ' '; |
| | 730 | } |
| | 731 | } |
| | 732 | $iLastMatch = $i; |
| | 733 | $j = $i; |
| | 734 | if ($htmlsave) { |
| | 735 | $ret .= htmlspecialchars($res[1]); |
| | 736 | } else { |
| | 737 | $ret .= $res[1]; |
| | 738 | } |
| | 739 | $encoding = ucfirst($res[3]); |
| | 740 | |
| | 741 | /* decide about valid decoding */ |
| | 742 | if ($decide && is_conversion_safe($res[2])) { |
| | 743 | $utfencode=true; |
| | 744 | $can_be_encoded=true; |
| | 745 | } else { |
| | 746 | $can_be_encoded=false; |
| | 747 | } |
| | 748 | switch ($encoding) |
| | 749 | { |
| | 750 | case 'B': |
| | 751 | $replace = base64_decode($res[4]); |
| | 752 | if ($utfencode) { |
| | 753 | if ($can_be_encoded) { |
| | 754 | /* convert string to different charset, |
| | 755 | * if functions asks for it (usually in compose) |
| | 756 | */ |
| | 757 | $ret .= charset_convert($res[2],$replace,$default_charset,$htmlsave); |
| | 758 | } else { |
| | 759 | // convert string to html codes in order to display it |
| | 760 | $ret .= charset_decode($res[2],$replace); |
| | 761 | } |
| | 762 | } else { |
| | 763 | if ($htmlsave) { |
| | 764 | $replace = htmlspecialchars($replace); |
| | 765 | } |
| | 766 | $ret.= $replace; |
| | 767 | } |
| | 768 | break; |
| | 769 | case 'Q': |
| | 770 | $replace = str_replace('_', ' ', $res[4]); |
| | 771 | $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))', |
| | 772 | $replace); |
| | 773 | if ($utfencode) { |
| | 774 | if ($can_be_encoded) { |
| | 775 | /* convert string to different charset, |
| | 776 | * if functions asks for it (usually in compose) |
| | 777 | */ |
| | 778 | $replace = charset_convert($res[2], $replace,$default_charset,$htmlsave); |
| | 779 | } else { |
| | 780 | // convert string to html codes in order to display it |
| | 781 | $replace = charset_decode($res[2], $replace); |
| | 782 | } |
| | 783 | } else { |
| | 784 | if ($htmlsave) { |
| | 785 | $replace = htmlspecialchars($replace); |
| | 786 | } |
| | 787 | } |
| | 788 | $ret .= $replace; |
| | 789 | break; |
| | 790 | default: |
| | 791 | break; |
| | 792 | } |
| | 793 | $chunk = $res[5]; |
| | 794 | $encoded = true; |
| | 795 | } |
| | 796 | if (!$encoded) { |
| | 797 | if ($htmlsave) { |
| | 798 | $ret .= ' '; |
| | 799 | } else { |
| | 800 | $ret .= ' '; |
| | 801 | } |
| | 802 | } |
| | 803 | |
| | 804 | if (!$encoded && $htmlsave) { |
| | 805 | $ret .= htmlspecialchars($chunk); |
| | 806 | } else { |
| | 807 | $ret .= $chunk; |
| | 808 | } |
| | 809 | ++$i; |
| | 810 | } |
| | 811 | /* remove the first added space */ |
| | 812 | if ($ret) { |
| | 813 | if ($htmlsave) { |
| | 814 | $ret = substr($ret,5); |
| | 815 | } else { |
| | 816 | $ret = substr($ret,1); |
| | 817 | } |
| | 818 | } |
| | 819 | |
| | 820 | return $ret; |
| | 821 | } |
| | 822 | |
| | 823 | /** |
| | 824 | * Encodes header |
| | 825 | * |
| | 826 | * Function uses XTRA_CODE _encodeheader function, if such function exists. |
| | 827 | * |
| | 828 | * Function uses Q encoding by default and encodes a string according to RFC |
| | 829 | * 1522 for use in headers if it contains 8-bit characters or anything that |
| | 830 | * looks like it should be encoded. |
| | 831 | * |
| | 832 | * Function switches to B encoding and encodeHeaderBase64() function, if |
| | 833 | * string is 8bit and multibyte character set supported by mbstring extension |
| | 834 | * is used. It can cause E_USER_NOTICE errors, if interface is used with |
| | 835 | * multibyte character set unsupported by mbstring extension. |
| | 836 | * |
| | 837 | * @param string $string header string, that has to be encoded |
| | 838 | * @return string quoted-printable encoded string |
| | 839 | * @todo make $mb_charsets system wide constant |
| | 840 | */ |
| | 841 | function encodeHeader ($string) { |
| | 842 | global $default_charset, $languages, $squirrelmail_language; |
| | 843 | |
| | 844 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| | 845 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) { |
| | 846 | return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string); |
| | 847 | } |
| | 848 | |
| | 849 | // Use B encoding for multibyte charsets |
| | 850 | $mb_charsets = array('utf-8','big5','gb2313','euc-kr'); |
| | 851 | if (in_array($default_charset,$mb_charsets) && |
| | 852 | in_array($default_charset,sq_mb_list_encodings()) && |
| | 853 | sq_is8bit($string)) { |
| | 854 | return encodeHeaderBase64($string,$default_charset); |
| | 855 | } elseif (in_array($default_charset,$mb_charsets) && |
| | 856 | sq_is8bit($string) && |
| | 857 | ! in_array($default_charset,sq_mb_list_encodings())) { |
| | 858 | // Add E_USER_NOTICE error here (can cause 'Cannot add header information' warning in compose.php) |
| | 859 | // trigger_error('encodeHeader: Multibyte character set unsupported by mbstring extension.',E_USER_NOTICE); |
| | 860 | } |
| | 861 | |
| | 862 | // Encode only if the string contains 8-bit characters or =? |
| | 863 | $j = strlen($string); |
| | 864 | $max_l = 75 - strlen($default_charset) - 7; |
| | 865 | $aRet = array(); |
| | 866 | $ret = ''; |
| | 867 | $iEncStart = $enc_init = false; |
| | 868 | $cur_l = $iOffset = 0; |
| | 869 | for($i = 0; $i < $j; ++$i) { |
| | 870 | switch($string{$i}) |
| | 871 | { |
| | 872 | case '=': |
| | 873 | case '<': |
| | 874 | case '>': |
| | 875 | case ',': |
| | 876 | case '?': |
| | 877 | case '_': |
| | 878 | if ($iEncStart === false) { |
| | 879 | $iEncStart = $i; |
| | 880 | } |
| | 881 | $cur_l+=3; |
| | 882 | if ($cur_l > ($max_l-2)) { |
| | 883 | /* if there is an stringpart that doesn't need encoding, add it */ |
| | 884 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| | 885 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| | 886 | $iOffset = $i; |
| | 887 | $cur_l = 0; |
| | 888 | $ret = ''; |
| | 889 | $iEncStart = false; |
| | 890 | } else { |
| | 891 | $ret .= sprintf("=%02X",ord($string{$i})); |
| | 892 | } |
| | 893 | break; |
| | 894 | case '(': |
| | 895 | case ')': |
| | 896 | if ($iEncStart !== false) { |
| | 897 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| | 898 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| | 899 | $iOffset = $i; |
| | 900 | $cur_l = 0; |
| | 901 | $ret = ''; |
| | 902 | $iEncStart = false; |
| | 903 | } |
| | 904 | break; |
| | 905 | case ' ': |
| | 906 | if ($iEncStart !== false) { |
| | 907 | $cur_l++; |
| | 908 | if ($cur_l > $max_l) { |
| | 909 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| | 910 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| | 911 | $iOffset = $i; |
| | 912 | $cur_l = 0; |
| | 913 | $ret = ''; |
| | 914 | $iEncStart = false; |
| | 915 | } else { |
| | 916 | $ret .= '_'; |
| | 917 | } |
| | 918 | } |
| | 919 | break; |
| | 920 | default: |
| | 921 | $k = ord($string{$i}); |
| | 922 | if ($k > 126) { |
| | 923 | if ($iEncStart === false) { |
| | 924 | // do not start encoding in the middle of a string, also take the rest of the word. |
| | 925 | $sLeadString = substr($string,0,$i); |
| | 926 | $aLeadString = explode(' ',$sLeadString); |
| | 927 | $sToBeEncoded = array_pop($aLeadString); |
| | 928 | $iEncStart = $i - strlen($sToBeEncoded); |
| | 929 | $ret .= $sToBeEncoded; |
| | 930 | $cur_l += strlen($sToBeEncoded); |
| | 931 | } |
| | 932 | $cur_l += 3; |
| | 933 | /* first we add the encoded string that reached it's max size */ |
| | 934 | if ($cur_l > ($max_l-2)) { |
| | 935 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| | 936 | $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */ |
| | 937 | $cur_l = 3; |
| | 938 | $ret = ''; |
| | 939 | $iOffset = $i; |
| | 940 | $iEncStart = $i; |
| | 941 | } |
| | 942 | $enc_init = true; |
| | 943 | $ret .= sprintf("=%02X", $k); |
| | 944 | } else { |
| | 945 | if ($iEncStart !== false) { |
| | 946 | $cur_l++; |
| | 947 | if ($cur_l > $max_l) { |
| | 948 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| | 949 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| | 950 | $iEncStart = false; |
| | 951 | $iOffset = $i; |
| | 952 | $cur_l = 0; |
| | 953 | $ret = ''; |
| | 954 | } else { |
| | 955 | $ret .= $string{$i}; |
| | 956 | } |
| | 957 | } |
| | 958 | } |
| | 959 | break; |
| | 960 | } |
| | 961 | } |
| | 962 | |
| | 963 | if ($enc_init) { |
| | 964 | if ($iEncStart !== false) { |
| | 965 | $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset); |
| | 966 | $aRet[] = "=?$default_charset?Q?$ret?="; |
| | 967 | } else { |
| | 968 | $aRet[] = substr($string,$iOffset); |
| | 969 | } |
| | 970 | $string = implode('',$aRet); |
| | 971 | } |
| | 972 | return $string; |
| | 973 | } |
| | 974 | |
| | 975 | /** |
| | 976 | * Encodes string according to rfc2047 B encoding header formating rules |
| | 977 | * |
| | 978 | * It is recommended way to encode headers with character sets that store |
| | 979 | * symbols in more than one byte. |
| | 980 | * |
| | 981 | * Function requires mbstring support. If required mbstring functions are missing, |
| | 982 | * function returns false and sets E_USER_WARNING level error message. |
| | 983 | * |
| | 984 | * Minimal requirements - php 4.0.6 with mbstring extension. Please note, |
| | 985 | * that mbstring functions will generate E_WARNING errors, if unsupported |
| | 986 | * character set is used. mb_encode_mimeheader function provided by php |
| | 987 | * mbstring extension is not used in order to get better control of header |
| | 988 | * encoding. |
| | 989 | * |
| | 990 | * Used php code functions - function_exists(), trigger_error(), strlen() |
| | 991 | * (is used with charset names and base64 strings). Used php mbstring |
| | 992 | * functions - mb_strlen and mb_substr. |
| | 993 | * |
| | 994 | * Related documents: rfc 2045 (BASE64 encoding), rfc 2047 (mime header |
| | 995 | * encoding), rfc 2822 (header folding) |
| | 996 | * |
| | 997 | * @param string $string header string that must be encoded |
| | 998 | * @param string $charset character set. Must be supported by mbstring extension. |
| | 999 | * Use sq_mb_list_encodings() to detect supported charsets. |
| | 1000 | * @return string string encoded according to rfc2047 B encoding formating rules |
| | 1001 | * @since 1.5.1 |
| | 1002 | * @todo First header line can be wrapped to $iMaxLength - $HeaderFieldLength - 1 |
| | 1003 | * @todo Do we want to control max length of header? |
| | 1004 | * @todo Do we want to control EOL (end-of-line) marker? |
| | 1005 | * @todo Do we want to translate error message? |
| | 1006 | */ |
| | 1007 | function encodeHeaderBase64($string,$charset) { |
| | 1008 | /** |
| | 1009 | * Check mbstring function requirements. |
| | 1010 | */ |
| | 1011 | if (! function_exists('mb_strlen') || |
| | 1012 | ! function_exists('mb_substr')) { |
| | 1013 | // set E_USER_WARNING |
| | 1014 | trigger_error('encodeHeaderBase64: Required mbstring functions are missing.',E_USER_WARNING); |
| | 1015 | // return false |
| | 1016 | return false; |
| | 1017 | } |
| | 1018 | |
| | 1019 | // initial return array |
| | 1020 | $aRet = array(); |
| | 1021 | |
| | 1022 | /** |
| | 1023 | * header length = 75 symbols max (same as in encodeHeader) |
| | 1024 | * remove $charset length |
| | 1025 | * remove =? ? ?= (5 chars) |
| | 1026 | * remove 2 more chars (\r\n ?) |
| | 1027 | */ |
| | 1028 | $iMaxLength = 75 - strlen($charset) - 7; |
| | 1029 | |
| | 1030 | // set first character position |
| | 1031 | $iStartCharNum = 0; |
| | 1032 | |
| | 1033 | // loop through all characters. count characters and not bytes. |
| | 1034 | for ($iCharNum=1; $iCharNum<=mb_strlen($string,$charset); $iCharNum++) { |
| | 1035 | // encode string from starting character to current character. |
| | 1036 | $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset)); |
| | 1037 | |
| | 1038 | // Check encoded string length |
| | 1039 | if(strlen($encoded_string)>$iMaxLength) { |
| | 1040 | // if string exceeds max length, reduce number of encoded characters and add encoded string part to array |
| | 1041 | $aRet[] = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum-1,$charset)); |
| | 1042 | |
| | 1043 | // set new starting character |
| | 1044 | $iStartCharNum = $iCharNum-1; |
| | 1045 | |
| | 1046 | // encode last char (in case it is last character in string) |
| | 1047 | $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset)); |
| | 1048 | } // if string is shorter than max length - add next character |
| | 1049 | } |
| | 1050 | |
| | 1051 | // add last encoded string to array |
| | 1052 | $aRet[] = $encoded_string; |
| | 1053 | |
| | 1054 | // set initial return string |
| | 1055 | $sRet = ''; |
| | 1056 | |
| | 1057 | // loop through encoded strings |
| | 1058 | foreach($aRet as $string) { |
| | 1059 | // TODO: Do we want to control EOL (end-of-line) marker |
| | 1060 | if ($sRet!='') $sRet.= " "; |
| | 1061 | |
| | 1062 | // add header tags and encoded string to return string |
| | 1063 | $sRet.= '=?'.$charset.'?B?'.$string.'?='; |
| | 1064 | } |
| | 1065 | |
| | 1066 | return $sRet; |
| | 1067 | } |
| | 1068 | |
| | 1069 | /* This function trys to locate the entity_id of a specific mime element */ |
| | 1070 | function find_ent_id($id, $message) { |
| | 1071 | for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) { |
| | 1072 | if ($message->entities[$i]->header->type0 == 'multipart') { |
| | 1073 | $ret = find_ent_id($id, $message->entities[$i]); |
| | 1074 | } else { |
| | 1075 | if (strcasecmp($message->entities[$i]->header->id, $id) == 0) { |
| | 1076 | // if (sq_check_save_extension($message->entities[$i])) { |
| | 1077 | return $message->entities[$i]->entity_id; |
| | 1078 | // } |
| | 1079 | } elseif (!empty($message->entities[$i]->header->parameters['name'])) { |
| | 1080 | /** |
| | 1081 | * This is part of a fix for Outlook Express 6.x generating |
| | 1082 | * cid URLs without creating content-id headers |
| | 1083 | * @@JA - 20050207 |
| | 1084 | */ |
| | 1085 | if (strcasecmp($message->entities[$i]->header->parameters['name'], $id) == 0) { |
| | 1086 | return $message->entities[$i]->entity_id; |
| | 1087 | } |
| | 1088 | } |
| | 1089 | } |
| | 1090 | } |
| | 1091 | return $ret; |
| | 1092 | } |
| | 1093 | |
| | 1094 | function sq_check_save_extension($message) { |
| | 1095 | $filename = $message->getFilename(); |
| | 1096 | $ext = substr($filename, strrpos($filename,'.')+1); |
| | 1097 | $save_extensions = array('jpg','jpeg','gif','png','bmp'); |
| | 1098 | return in_array($ext, $save_extensions); |
| | 1099 | } |
| | 1100 | |
| | 1101 | |
| | 1102 | /** |
| | 1103 | ** HTMLFILTER ROUTINES |
| | 1104 | */ |
| | 1105 | |
| | 1106 | /** |
| | 1107 | * This function checks attribute values for entity-encoded values |
| | 1108 | * and returns them translated into 8-bit strings so we can run |
| | 1109 | * checks on them. |
| | 1110 | * |
| | 1111 | * @param $attvalue A string to run entity check against. |
| | 1112 | * @return Nothing, modifies a reference value. |
| | 1113 | */ |
| | 1114 | function sq_defang(&$attvalue){ |
| | 1115 | $me = 'sq_defang'; |
| | 1116 | /** |
| | 1117 | * Skip this if there aren't ampersands or backslashes. |
| | 1118 | */ |
| | 1119 | if (strpos($attvalue, '&') === false |
| | 1120 | && strpos($attvalue, '\\') === false){ |
| | 1121 | return; |
| | 1122 | } |
| | 1123 | $m = false; |
| | 1124 | do { |
| | 1125 | $m = false; |
| | 1126 | $m = $m || sq_deent($attvalue, '/\�*(\d+);*/s'); |
| | 1127 | $m = $m || sq_deent($attvalue, '/\�*((\d|[a-f])+);*/si', true); |
| | 1128 | $m = $m || sq_deent($attvalue, '/\\\\(\d+)/s', true); |
| | 1129 | } while ($m == true); |
| | 1130 | $attvalue = stripslashes($attvalue); |
| | 1131 | } |
| | 1132 | |
| | 1133 | /** |
| | 1134 | * Kill any tabs, newlines, or carriage returns. Our friends the |
| | 1135 | * makers of the browser with 95% market value decided that it'd |
| | 1136 | * be funny to make "java[tab]script" be just as good as "javascript". |
| | 1137 | * |
| | 1138 | * @param attvalue The attribute value before extraneous spaces removed. |
| | 1139 | * @return attvalue Nothing, modifies a reference value. |
| | 1140 | */ |
| | 1141 | function sq_unspace(&$attvalue){ |
| | 1142 | $me = 'sq_unspace'; |
| | 1143 | if (strcspn($attvalue, "\t\r\n\0 ") != strlen($attvalue)){ |
| | 1144 | $attvalue = str_replace(Array("\t", "\r", "\n", "\0", " "), |
| | 1145 | Array('', '', '', '', ''), $attvalue); |
| | 1146 | } |
| | 1147 | } |
| | 1148 | |
| | 1149 | /** |
| | 1150 | * This function returns the final tag out of the tag name, an array |
| | 1151 | * of attributes, and the type of the tag. This function is called by |
| | 1152 | * sq_sanitize internally. |
| | 1153 | * |
| | 1154 | * @param $tagname the name of the tag. |
| | 1155 | * @param $attary the array of attributes and their values |
| | 1156 | * @param $tagtype The type of the tag (see in comments). |
| | 1157 | * @return a string with the final tag representation. |
| | 1158 | */ |
| | 1159 | function sq_tagprint($tagname, $attary, $tagtype){ |
| | 1160 | $me = 'sq_tagprint'; |
| | 1161 | |
| | 1162 | if ($tagtype == 2){ |
| | 1163 | $fulltag = '</' . $tagname . '>'; |
| | 1164 | } else { |
| | 1165 | $fulltag = '<' . $tagname; |
| | 1166 | if (is_array($attary) && sizeof($attary)){ |
| | 1167 | $atts = Array(); |
| | 1168 | while (list($attname, $attvalue) = each($attary)){ |
| | 1169 | array_push($atts, "$attname=$attvalue"); |
| | 1170 | } |
| | 1171 | $fulltag .= ' ' . join(" ", $atts); |
| | 1172 | } |
| | 1173 | if ($tagtype == 3){ |
| | 1174 | $fulltag .= ' /'; |
| | 1175 | } |
| | 1176 | $fulltag .= '>'; |
| | 1177 | } |
| | 1178 | return $fulltag; |
| | 1179 | } |
| | 1180 | |
| | 1181 | /** |
| | 1182 | * A small helper function to use with array_walk. Modifies a by-ref |
| | 1183 | * value and makes it lowercase. |
| | 1184 | * |
| | 1185 | * @param $val a value passed by-ref. |
| | 1186 | * @return void since it modifies a by-ref value. |
| | 1187 | */ |
| | 1188 | function sq_casenormalize(&$val){ |
| | 1189 | $val = strtolower($val); |
| | 1190 | } |
| | 1191 | |
| | 1192 | /** |
| | 1193 | * This function skips any whitespace from the current position within |
| | 1194 | * a string and to the next non-whitespace value. |
| | 1195 | * |
| | 1196 | * @param $body the string |
| | 1197 | * @param $offset the offset within the string where we should start |
| | 1198 | * looking for the next non-whitespace character. |
| | 1199 | * @return the location within the $body where the next |
| | 1200 | * non-whitespace char is located. |
| | 1201 | */ |
| | 1202 | function sq_skipspace($body, $offset){ |
| | 1203 | $me = 'sq_skipspace'; |
| | 1204 | preg_match('/^(\s*)/s', substr($body, $offset), $matches); |
| | 1205 | if (sizeof($matches{1})){ |
| | 1206 | $count = strlen($matches{1}); |
| | 1207 | $offset += $count; |
| | 1208 | } |
| | 1209 | return $offset; |
| | 1210 | } |
| | 1211 | |
| | 1212 | /** |
| | 1213 | * This function looks for the next character within a string. It's |
| | 1214 | * really just a glorified "strpos", except it catches if failures |
| | 1215 | * nicely. |
| | 1216 | * |
| | 1217 | * @param $body The string to look for needle in. |
| | 1218 | * @param $offset Start looking from this position. |
| | 1219 | * @param $needle The character/string to look for. |
| | 1220 | * @return location of the next occurance of the needle, or |
| | 1221 | * strlen($body) if needle wasn't found. |
| | 1222 | */ |
| | 1223 | function sq_findnxstr($body, $offset, $needle){ |
| | 1224 | $me = 'sq_findnxstr'; |
| | 1225 | $pos = strpos($body, $needle, $offset); |
| | 1226 | if ($pos === FALSE){ |
| | 1227 | $pos = strlen($body); |
| | 1228 | } |
| | 1229 | return $pos; |
| | 1230 | } |
| | 1231 | |
| | 1232 | /** |
| | 1233 | * This function takes a PCRE-style regexp and tries to match it |
| | 1234 | * within the string. |
| | 1235 | * |
| | 1236 | * @param $body The string to look for needle in. |
| | 1237 | * @param $offset Start looking from here. |
| | 1238 | * @param $reg A PCRE-style regex to match. |
| | 1239 | * @return Returns a false if no matches found, or an array |
| | 1240 | * with the following members: |
| | 1241 | * - integer with the location of the match within $body |
| | 1242 | * - string with whatever content between offset and the match |
| | 1243 | * - string with whatever it is we matched |
| | 1244 | */ |
| | 1245 | function sq_findnxreg($body, $offset, $reg){ |
| | 1246 | $me = 'sq_findnxreg'; |
| | 1247 | $matches = Array(); |
| | 1248 | $retarr = Array(); |
| | 1249 | preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches); |
| | 1250 | if (!isset($matches{0}) || !$matches{0}){ |
| | 1251 | $retarr = false; |
| | 1252 | } else { |
| | 1253 | $retarr{0} = $offset + strlen($matches{1}); |
| | 1254 | $retarr{1} = $matches{1}; |
| | 1255 | $retarr{2} = $matches{2}; |
| | 1256 | } |
| | 1257 | return $retarr; |
| | 1258 | } |
| | 1259 | |
| | 1260 | /** |
| | 1261 | * This function looks for the next tag. |
| | 1262 | * |
| | 1263 | * @param $body String where to look for the next tag. |
| | 1264 | * @param $offset Start looking from here. |
| | 1265 | * @return false if no more tags exist in the body, or |
| | 1266 | * an array with the following members: |
| | 1267 | * - string with the name of the tag |
| | 1268 | * - array with attributes and their values |
| | 1269 | * - integer with tag type (1, 2, or 3) |
| | 1270 | * - integer where the tag starts (starting "<") |
| | 1271 | * - integer where the tag ends (ending ">") |
| | 1272 | * first three members will be false, if the tag is invalid. |
| | 1273 | */ |
| | 1274 | function sq_getnxtag($body, $offset){ |
| | 1275 | $me = 'sq_getnxtag'; |
| | 1276 | if ($offset > strlen($body)){ |
| | 1277 | return false; |
| | 1278 | } |
| | 1279 | $lt = sq_findnxstr($body, $offset, "<"); |
| | 1280 | if ($lt == strlen($body)){ |
| | 1281 | return false; |
| | 1282 | } |
| | 1283 | /** |
| | 1284 | * We are here: |
| | 1285 | * blah blah <tag attribute="value"> |
| | 1286 | * \---------^ |
| | 1287 | */ |
| | 1288 | $pos = sq_skipspace($body, $lt+1); |
| | 1289 | if ($pos >= strlen($body)){ |
| | 1290 | return Array(false, false, false, $lt, strlen($body)); |
| | 1291 | } |
| | 1292 | /** |
| | 1293 | * There are 3 kinds of tags: |
| | 1294 | * 1. Opening tag, e.g.: |
| | 1295 | * <a href="blah"> |
| | 1296 | * 2. Closing tag, e.g.: |
| | 1297 | * </a> |
| | 1298 | * 3. XHTML-style content-less tag, e.g.: |
| | 1299 | * <img src="blah" /> |
| | 1300 | */ |
| | 1301 | $tagtype = false; |
| | 1302 | switch (substr($body, $pos, 1)){ |
| | 1303 | case '/': |
| | 1304 | $tagtype = 2; |
| | 1305 | $pos++; |
| | 1306 | break; |
| | 1307 | case '!': |
| | 1308 | /** |
| | 1309 | * A comment or an SGML declaration. |
| | 1310 | */ |
| | 1311 | if (substr($body, $pos+1, 2) == "--"){ |
| | 1312 | $gt = strpos($body, "-->", $pos); |
| | 1313 | if ($gt === false){ |
| | 1314 | $gt = strlen($body); |
| | 1315 | } else { |
| | 1316 | $gt += 2; |
| | 1317 | } |
| | 1318 | return Array(false, false, false, $lt, $gt); |
| | 1319 | } else { |
| | 1320 | $gt = sq_findnxstr($body, $pos, ">"); |
| | 1321 | return Array(false, false, false, $lt, $gt); |
| | 1322 | } |
| | 1323 | break; |
| | 1324 | default: |
| | 1325 | /** |
| | 1326 | * Assume tagtype 1 for now. If it's type 3, we'll switch values |
| | 1327 | * later. |
| | 1328 | */ |
| | 1329 | $tagtype = 1; |
| | 1330 | break; |
| | 1331 | } |
| | 1332 | |
| | 1333 | $tag_start = $pos; |
| | 1334 | $tagname = ''; |
| | 1335 | /** |
| | 1336 | * Look for next [\W-_], which will indicate the end of the tag name. |
| | 1337 | */ |
| | 1338 | $regary = sq_findnxreg($body, $pos, "[^\w\-_]"); |
| | 1339 | if ($regary == false){ |
| | 1340 | return Array(false, false, false, $lt, strlen($body)); |
| | 1341 | } |
| | 1342 | list($pos, $tagname, $match) = $regary; |
| | 1343 | $tagname = strtolower($tagname); |
| | 1344 | |
| | 1345 | /** |
| | 1346 | * $match can be either of these: |
| | 1347 | * '>' indicating the end of the tag entirely. |
| | 1348 | * '\s' indicating the end of the tag name. |
| | 1349 | * '/' indicating that this is type-3 xhtml tag. |
| | 1350 | * |
| | 1351 | * Whatever else we find there indicates an invalid tag. |
| | 1352 | */ |
| | 1353 | switch ($match){ |
| | 1354 | case '/': |
| | 1355 | /** |
| | 1356 | * This is an xhtml-style tag with a closing / at the |
| | 1357 | * end, like so: <img src="blah" />. Check if it's followed |
| | 1358 | * by the closing bracket. If not, then this tag is invalid |
| | 1359 | */ |
| | 1360 | if (substr($body, $pos, 2) == "/>"){ |
| | 1361 | $pos++; |
| | 1362 | $tagtype = 3; |
| | 1363 | } else { |
| | 1364 | $gt = sq_findnxstr($body, $pos, ">"); |
| | 1365 | $retary = Array(false, false, false, $lt, $gt); |
| | 1366 | return $retary; |
| | 1367 | } |
| | 1368 | case '>': |
| | 1369 | return Array($tagname, false, $tagtype, $lt, $pos); |
| | 1370 | break; |
| | 1371 | default: |
| | 1372 | /** |
| | 1373 | * Check if it's whitespace |
| | 1374 | */ |
| | 1375 | if (!preg_match('/\s/', $match)){ |
| | 1376 | /** |
| | 1377 | * This is an invalid tag! Look for the next closing ">". |
| | 1378 | */ |
| | 1379 | $gt = sq_findnxstr($body, $lt, ">"); |
| | 1380 | return Array(false, false, false, $lt, $gt); |
| | 1381 | } |
| | 1382 | break; |
| | 1383 | } |
| | 1384 | |
| | 1385 | /** |
| | 1386 | * At this point we're here: |
| | 1387 | * <tagname attribute='blah'> |
| | 1388 | * \-------^ |
| | 1389 | * |
| | 1390 | * At this point we loop in order to find all attributes. |
| | 1391 | */ |
| | 1392 | $attname = ''; |
| | 1393 | $atttype = false; |
| | 1394 | $attary = Array(); |
| | 1395 | |
| | 1396 | while ($pos <= strlen($body)){ |
| | 1397 | $pos = sq_skipspace($body, $pos); |
| | 1398 | if ($pos == strlen($body)){ |
| | 1399 | /** |
| | 1400 | * Non-closed tag. |
| | 1401 | */ |
| | 1402 | return Array(false, false, false, $lt, $pos); |
| | 1403 | } |
| | 1404 | /** |
| | 1405 | * See if we arrived at a ">" or "/>", which means that we reached |
| | 1406 | * the end of the tag. |
| | 1407 | */ |
| | 1408 | $matches = Array(); |
| | 1409 | if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) { |
| | 1410 | /** |
| | 1411 | * Yep. So we did. |
| | 1412 | */ |
| | 1413 | $pos += strlen($matches{1}); |
| | 1414 | if ($matches{2} == "/>"){ |
| | 1415 | $tagtype = 3; |
| | 1416 | $pos++; |
| | 1417 | } |
| | 1418 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
| | 1419 | } |
| | 1420 | |
| | 1421 | /** |
| | 1422 | * There are several types of attributes, with optional |
| | 1423 | * [:space:] between members. |
| | 1424 | * Type 1: |
| | 1425 | * attrname[:space:]=[:space:]'CDATA' |
| | 1426 | * Type 2: |
| | 1427 | * attrname[:space:]=[:space:]"CDATA" |
| | 1428 | * Type 3: |
| | 1429 | * attr[:space:]=[:space:]CDATA |
| | 1430 | * Type 4: |
| | 1431 | * attrname |
| | 1432 | * |
| | 1433 | * We leave types 1 and 2 the same, type 3 we check for |
| | 1434 | * '"' and convert to """ if needed, then wrap in |
| | 1435 | * double quotes. Type 4 we convert into: |
| | 1436 | * attrname="yes". |
| | 1437 | */ |
| | 1438 | $regary = sq_findnxreg($body, $pos, "[^:\w\-_]"); |
| | 1439 | if ($regary == false){ |
| | 1440 | /** |
| | 1441 | * Looks like body ended before the end of tag. |
| | 1442 | */ |
| | 1443 | return Array(false, false, false, $lt, strlen($body)); |
| | 1444 | } |
| | 1445 | list($pos, $attname, $match) = $regary; |
| | 1446 | $attname = strtolower($attname); |
| | 1447 | /** |
| | 1448 | * We arrived at the end of attribute name. Several things possible |
| | 1449 | * here: |
| | 1450 | * '>' means the end of the tag and this is attribute type 4 |
| | 1451 | * '/' if followed by '>' means the same thing as above |
| | 1452 | * '\s' means a lot of things -- look what it's followed by. |
| | 1453 | * anything else means the attribute is invalid. |
| | 1454 | */ |
| | 1455 | switch($match){ |
| | 1456 | case '/': |
| | 1457 | /** |
| | 1458 | * This is an xhtml-style tag with a closing / at the |
| | 1459 | * end, like so: <img src="blah" />. Check if it's followed |
| | 1460 | * by the closing bracket. If not, then this tag is invalid |
| | 1461 | */ |
| | 1462 | if (substr($body, $pos, 2) == "/>"){ |
| | 1463 | $pos++; |
| | 1464 | $tagtype = 3; |
| | 1465 | } else { |
| | 1466 | $gt = sq_findnxstr($body, $pos, ">"); |
| | 1467 | $retary = Array(false, false, false, $lt, $gt); |
| | 1468 | return $retary; |
| | 1469 | } |
| | 1470 | case '>': |
| | 1471 | $attary{$attname} = '"yes"'; |
| | 1472 | return Array($tagname, $attary, $tagtype, $lt, $pos); |
| | 1473 | break; |
| | 1474 | default: |
| | 1475 | /** |
| | 1476 | * Skip whitespace and see what we arrive at. |
| | 1477 | */ |
| | 1478 | $pos = sq_skipspace($body, $pos); |
| | 1479 | $char = substr($body, $pos, 1); |
| | 1480 | /** |
| | 1481 | * Two things are valid here: |
| | 1482 | * '=' means this is attribute type 1 2 or 3. |
| | 1483 | * \w means this was attribute type 4. |
| | 1484 | * anything else we ignore and re-loop. End of tag and |
| | 1485 | * invalid stuff will be caught by our checks at the beginning |
| | 1486 | * of the loop. |
| | 1487 | */ |
| | 1488 | if ($char == "="){ |
| | 1489 | $pos++; |
| | 1490 | $pos = sq_skipspace($body, $pos); |
| | 1491 | /** |
| | 1492 | * Here are 3 possibilities: |
| | 1493 | * "'" attribute type 1 |
| | 1494 | * '"' attribute type 2 |
| | 1495 | * everything else is the content of tag type 3 |
| | 1496 | */ |
| | 1497 | $quot = substr($body, $pos, 1); |
| | 1498 | if ($quot == "'"){ |
| | 1499 | $regary = sq_findnxreg($body, $pos+1, "\'"); |
| | 1500 | if ($regary == false){ |
| | 1501 | return Array(false, false, false, $lt, strlen($body)); |
| | 1502 | } |
| | 1503 | list($pos, $attval, $match) = $regary; |
| | 1504 | $pos++; |
| | 1505 | $attary{$attname} = "'" . $attval . "'"; |
| | 1506 | } else if ($quot == '"'){ |
| | 1507 | $regary = sq_findnxreg($body, $pos+1, '\"'); |
| | 1508 | if ($regary == false){ |
| | 1509 | return Array(false, false, false, $lt, strlen($body)); |
| | 1510 | } |
| | 1511 | list($pos, $attval, $match) = $regary; |
| | 1512 | $pos++; |
| | 1513 | $attary{$attname} = '"' . $attval . '"'; |
| | 1514 | } else { |
| | 1515 | /** |
| | 1516 | * These are hateful. Look for \s, or >. |
| | 1517 | */ |
| | 1518 | $regary = sq_findnxreg($body, $pos, "[\s>]"); |
| | 1519 | if ($regary == false){ |
| | 1520 | return Array(false, false, false, $lt, strlen($body)); |
| | 1521 | } |
| | 1522 | list($pos, $attval, $match) = $regary; |
| | 1523 | /** |
| | 1524 | * If it's ">" it will be caught at the top. |
| | 1525 | */ |
| | 1526 | $attval = preg_replace("/\"/s", """, $attval); |
| | 1527 | $attary{$attname} = '"' . $attval . '"'; |
| | 1528 | } |
| | 1529 | } else if (preg_match("|[\w/>]|", $char)) { |
| | 1530 | /** |
| | 1531 | * That was attribute type 4. |
| | 1532 | */ |
| | 1533 | $attary{$attname} = '"yes"'; |
| | 1534 | } else { |
| | 1535 | /** |
| | 1536 | * An illegal character. Find next '>' and return. |
| | 1537 | */ |
| | 1538 | $gt = sq_findnxstr($body, $pos, ">"); |
| | 1539 | return Array(false, false, false, $lt, $gt); |
| | 1540 | } |
| | 1541 | break; |
| | 1542 | } |
| | 1543 | } |
| | 1544 | /** |
| | 1545 | * The fact that we got here indicates that the tag end was never |
| | 1546 | * found. Return invalid tag indication so it gets stripped. |
| | 1547 | */ |
| | 1548 | return Array(false, false, false, $lt, strlen($body)); |
| | 1549 | } |
| | 1550 | |
| | 1551 | /** |
| | 1552 | * Translates entities into literal values so they can be checked. |
| | 1553 | * |
| | 1554 | * @param $attvalue the by-ref value to check. |
| | 1555 | * @param $regex the regular expression to check against. |
| | 1556 | * @param $hex whether the entites are hexadecimal. |
| | 1557 | * @return True or False depending on whether there were matches. |
| | 1558 | */ |
| | 1559 | function sq_deent(&$attvalue, $regex, $hex=false){ |
| | 1560 | $me = 'sq_deent'; |
| | 1561 | $ret_match = false; |
| | 1562 | preg_match_all($regex, $attvalue, $matches); |
| | 1563 | if (is_array($matches) && sizeof($matches[0]) > 0){ |
| | 1564 | $repl = Array(); |
| | 1565 | for ($i = 0; $i < sizeof($matches[0]); $i++){ |
| | 1566 | $numval = $matches[1][$i]; |
| | 1567 | if ($hex){ |
| | 1568 | $numval = hexdec($numval); |
| | 1569 | } |
| | 1570 | $repl{$matches[0][$i]} = chr($numval); |
| | 1571 | } |
| | 1572 | $attvalue = strtr($attvalue, $repl); |
| | 1573 | return true; |
| | 1574 | } else { |
| | 1575 | return false; |
| | 1576 | } |
| | 1577 | } |
| | 1578 | |
| | 1579 | /** |
| | 1580 | * This function runs various checks against the attributes. |
| | 1581 | * |
| | 1582 | * @param $tagname String with the name of the tag. |
| | 1583 | * @param $attary Array with all tag attributes. |
| | 1584 | * @param $rm_attnames See description for sq_sanitize |
| | 1585 | * @param $bad_attvals See description for sq_sanitize |
| | 1586 | * @param $add_attr_to_tag See description for sq_sanitize |
| | 1587 | * @param $message message object |
| | 1588 | * @param $id message id |
| | 1589 | * @return Array with modified attributes. |
| | 1590 | */ |
| | 1591 | function sq_fixatts($tagname, |
| | 1592 | $attary, |
| | 1593 | $rm_attnames, |
| | 1594 | $bad_attvals, |
| | 1595 | $add_attr_to_tag, |
| | 1596 | $message, |
| | 1597 | $id, |
| | 1598 | $mailbox |
| | 1599 | ){ |
| | 1600 | $me = 'sq_fixatts'; |
| | 1601 | while (list($attname, $attvalue) = each($attary)){ |
| | 1602 | /** |
| | 1603 | * See if this attribute should be removed. |
| | 1604 | */ |
| | 1605 | foreach ($rm_attnames as $matchtag=>$matchattrs){ |
| | 1606 | if (preg_match($matchtag, $tagname)){ |
| | 1607 | foreach ($matchattrs as $matchattr){ |
| | 1608 | if (preg_match($matchattr, $attname)){ |
| | 1609 | unset($attary{$attname}); |
| | 1610 | continue; |
| | 1611 | } |
| | 1612 | } |
| | 1613 | } |
| | 1614 | } |
| | 1615 | /** |
| | 1616 | * Remove any backslashes, entities, and extraneous whitespace. |
| | 1617 | */ |
| | 1618 | sq_defang($attvalue); |
| | 1619 | sq_unspace($attvalue); |
| | 1620 | |
| | 1621 | /** |
| | 1622 | * Now let's run checks on the attvalues. |
| | 1623 | * I don't expect anyone to comprehend this. If you do, |
| | 1624 | * get in touch with me so I can drive to where you live and |
| | 1625 | * shake your hand personally. :) |
| | 1626 | */ |
| | 1627 | foreach ($bad_attvals as $matchtag=>$matchattrs){ |
| | 1628 | if (preg_match($matchtag, $tagname)){ |
| | 1629 | foreach ($matchattrs as $matchattr=>$valary){ |
| | 1630 | if (preg_match($matchattr, $attname)){ |
| | 1631 | /** |
| | 1632 | * There are two arrays in valary. |
| | 1633 | * First is matches. |
| | 1634 | * Second one is replacements |
| | 1635 | */ |
| | 1636 | list($valmatch, $valrepl) = $valary; |
| | 1637 | $newvalue = |
| | 1638 | preg_replace($valmatch, $valrepl, $attvalue); |
| | 1639 | if ($newvalue != $attvalue){ |
| | 1640 | $attary{$attname} = $newvalue; |
| | 1641 | } |
| | 1642 | } |
| | 1643 | } |
| | 1644 | } |
| | 1645 | } |
| | 1646 | |
| | 1647 | /** |
| | 1648 | * Replace empty src tags with the blank image. src is only used |
| | 1649 | * for frames, images, and image inputs. Doing a replace should |
| | 1650 | * not affect them working as should be, however it will stop |
| | 1651 | * IE from being kicked off when src for img tags are not set |
| | 1652 | */ |
| | 1653 | if (($attname == 'src') && ($attvalue == '""')) { |
| | 1654 | $attary{$attname} = '"' . SM_PATH . 'images/blank.png"'; |
| | 1655 | } |
| | 1656 | |
| | 1657 | /** |
| | 1658 | * Turn cid: urls into http-friendly ones. |
| | 1659 | */ |
| | 1660 | if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){ |
| | 1661 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
| | 1662 | } |
| | 1663 | |
| | 1664 | /** |
| | 1665 | * "Hack" fix for Outlook using propriatary outbind:// protocol in img tags. |
| | 1666 | * One day MS might actually make it match something useful, for now, falling |
| | 1667 | * back to using cid2http, so we can grab the blank.png. |
| | 1668 | */ |
| | 1669 | if (preg_match("/^[\'\"]\s*outbind:\/\//si", $attvalue)) { |
| | 1670 | $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox); |
| | 1671 | } |
| | 1672 | |
| | 1673 | } |
| | 1674 | /** |
| | 1675 | * See if we need to append any attributes to this tag. |
| | 1676 | */ |
| | 1677 | foreach ($add_attr_to_tag as $matchtag=>$addattary){ |
| | 1678 | if (preg_match($matchtag, $tagname)){ |
| | 1679 | $attary = array_merge($attary, $addattary); |
| | 1680 | } |
| | 1681 | } |
| | 1682 | return $attary; |
| | 1683 | } |
| | 1684 | |
| | 1685 | /** |
| | 1686 | * This function edits the style definition to make them friendly and |
| | 1687 | * usable in SquirrelMail. |
| | 1688 | * |
| | 1689 | * @param $message the message object |
| | 1690 | * @param $id the message id |
| | 1691 | * @param $content a string with whatever is between <style> and </style> |
| | 1692 | * @param $mailbox the message mailbox |
| | 1693 | * @return a string with edited content. |
| | 1694 | */ |
| | 1695 | function sq_fixstyle($body, $pos, $message, $id, $mailbox){ |
| | 1696 | global $view_unsafe_images; |
| | 1697 | $me = 'sq_fixstyle'; |
| | 1698 | $ret = sq_findnxreg($body, $pos, '</\s*style\s*>'); |
| | 1699 | if ($ret == FALSE){ |
| | 1700 | return array(FALSE, strlen($body)); |
| | 1701 | } |
| | 1702 | $newpos = $ret[0] + strlen($ret[2]); |
| | 1703 | $content = $ret[1]; |
| | 1704 | /** |
| | 1705 | * First look for general BODY style declaration, which would be |
| | 1706 | * like so: |
| | 1707 | * body {background: blah-blah} |
| | 1708 | * and change it to .bodyclass so we can just assign it to a <div> |
| | 1709 | */ |
| | 1710 | $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content); |
| | 1711 | $secremoveimg = '../images/' . _("sec_remove_eng.png"); |
| | 1712 | /** |
| | 1713 | * Fix url('blah') declarations. |
| | 1714 | */ |
| | 1715 | // $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si", |
| | 1716 | // "url(\\1$secremoveimg\\2)", $content); |
| | 1717 | // remove NUL |
| | 1718 | $content = str_replace("\0", "", $content); |
| | 1719 | // translate ur\l and variations (IE parses that) |
| | 1720 | $content = preg_replace("/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", 'url', $content); |
| | 1721 | // NB I insert NUL characters to keep to avoid an infinite loop. They are removed after the loop. |
| | 1722 | while (preg_match("/url\s*\(\s*[\'\"]?([^:]+):(.*)?[\'\"]?\s*\)/si", $content, $matches)) { |
| | 1723 | $sProto = strtolower($matches[1]); |
| | 1724 | switch ($sProto) { |
| | 1725 | /** |
| | 1726 | * Fix url('https*://.*) declarations but only if $view_unsafe_images |
| | 1727 | * is false. |
| | 1728 | */ |
| | 1729 | case 'https': |
| | 1730 | case 'http': |
| | 1731 | if (!$view_unsafe_images){ |
| | 1732 | |
| | 1733 | $sExpr = "/url\s*\(\s*[\'\"]?\s*$sProto*:.*[\'\"]?\s*\)/si"; |
| | 1734 | $content = preg_replace($sExpr, "u\0r\0l(\\1$secremoveimg\\2)", $content); |
| | 1735 | |
| | 1736 | } else { |
| | 1737 | $content = preg_replace('/url/i',"u\0r\0l",$content); |
| | 1738 | } |
| | 1739 | break; |
| | 1740 | /** |
| | 1741 | * Fix urls that refer to cid: |
| | 1742 | */ |
| | 1743 | case 'cid': |
| | 1744 | $cidurl = 'cid:'. $matches[2]; |
| | 1745 | $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox); |
| | 1746 | // escape parentheses that can modify the regular expression |
| | 1747 | $cidurl = str_replace(array('(',')'),array('\\(','\\)'),$cidurl); |
| | 1748 | $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si", |
| | 1749 | "u\0r\0l($httpurl)", $content); |
| | 1750 | break; |
| | 1751 | default: |
| | 1752 | /** |
| | 1753 | * replace url with protocol other then the white list |
| | 1754 | * http,https and cid by an empty string. |
| | 1755 | */ |
| | 1756 | $content = preg_replace("/url\s*\(\s*[\'\"]?([^:]+):(.*)?[\'\"]?\s*\)/si", |
| | 1757 | "", $content); |
| | 1758 | break; |
| | 1759 | } |
| | 1760 | } |
| | 1761 | // remove NUL |
| | 1762 | $content = str_replace("\0", "", $content); |
| | 1763 | /** |
| | 1764 | * Remove any backslashes, entities, and extraneous whitespace. |
| | 1765 | */ |
| | 1766 | $contentTemp = $content; |
| | 1767 | sq_defang($contentTemp); |
| | 1768 | sq_unspace($contentTemp); |
| | 1769 | |
| | 1770 | /** |
| | 1771 | * Fix stupid css declarations which lead to vulnerabilities |
| | 1772 | * in IE. |
| | 1773 | */ |
| | 1774 | $match = Array('/\/\*.*\*\//', |
| | 1775 | '/expression/i', |
| | 1776 | '/behaviou*r/i', |
| | 1777 | '/binding/i', |
| | 1778 | '/include-source/i'); |
| | 1779 | $replace = Array('','idiocy', 'idiocy', 'idiocy', 'idiocy'); |
| | 1780 | $contentNew = preg_replace($match, $replace, $contentTemp); |
| | 1781 | if ($contentNew !== $contentTemp) { |
| | 1782 | // insecure css declarations are used. From now on we don't care |
| | 1783 | // anymore if the css is destroyed by sq_deent, sq_unspace or sq_unbackslash |
| | 1784 | $content = $contentNew; |
| | 1785 | } |
| | 1786 | return array($content, $newpos); |
| | 1787 | } |
| | 1788 | |
| | 1789 | |
| | 1790 | /** |
| | 1791 | * This function converts cid: url's into the ones that can be viewed in |
| | 1792 | * the browser. |
| | 1793 | * |
| | 1794 | * @param $message the message object |
| | 1795 | * @param $id the message id |
| | 1796 | * @param $cidurl the cid: url. |
| | 1797 | * @param $mailbox the message mailbox |
| | 1798 | * @return a string with a http-friendly url |
| | 1799 | */ |
| | 1800 | function sq_cid2http($message, $id, $cidurl, $mailbox){ |
| | 1801 | /** |
| | 1802 | * Get rid of quotes. |
| | 1803 | */ |
| | 1804 | $quotchar = substr($cidurl, 0, 1); |
| | 1805 | if ($quotchar == '"' || $quotchar == "'"){ |
| | 1806 | $cidurl = str_replace($quotchar, "", $cidurl); |
| | 1807 | } else { |
| | 1808 | $quotchar = ''; |
| | 1809 | } |
| | 1810 | $cidurl = substr(trim($cidurl), 4); |
| | 1811 | |
| | 1812 | $match_str = '/\{.*?\}\//'; |
| | 1813 | $str_rep = ''; |
| | 1814 | $cidurl = preg_replace($match_str, $str_rep, $cidurl); |
| | 1815 | |
| | 1816 | $linkurl = find_ent_id($cidurl, $message); |
| | 1817 | /* in case of non-save cid links $httpurl should be replaced by a sort of |
| | 1818 | unsave link image */ |
| | 1819 | $httpurl = ''; |
| | 1820 | |
| | 1821 | /** |
| | 1822 | * This is part of a fix for Outlook Express 6.x generating |
| | 1823 | * cid URLs without creating content-id headers. These images are |
| | 1824 | * not part of the multipart/related html mail. The html contains |
| | 1825 | * <img src="cid:{some_id}/image_filename.ext"> references to |
| | 1826 | * attached images with as goal to render them inline although |
| | 1827 | * the attachment disposition property is not inline. |
| | 1828 | */ |
| | 1829 | |
| | 1830 | if (empty($linkurl)) { |
| | 1831 | if (preg_match('/{.*}\//', $cidurl)) { |
| | 1832 | $cidurl = preg_replace('/{.*}\//','', $cidurl); |
| | 1833 | if (!empty($cidurl)) { |
| | 1834 | $linkurl = find_ent_id($cidurl, $message); |
| | 1835 | } |
| | 1836 | } |
| | 1837 | } |
| | 1838 | |
| | 1839 | if (!empty($linkurl)) { |
| | 1840 | $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&' . |
| | 1841 | "passed_id=$id&mailbox=" . urlencode($mailbox) . |
| | 1842 | '&ent_id=' . $linkurl . $quotchar; |
| | 1843 | } else { |
| | 1844 | /** |
| | 1845 | * If we couldn't generate a proper img url, drop in a blank image |
| | 1846 | * instead of sending back empty, otherwise it causes unusual behaviour |
| | 1847 | */ |
| | 1848 | $httpurl = $quotchar . SM_PATH . 'images/blank.png' . $quotchar; |
| | 1849 | } |
| | 1850 | |
| | 1851 | return $httpurl; |
| | 1852 | } |
| | 1853 | |
| | 1854 | /** |
| | 1855 | * This function changes the <body> tag into a <div> tag since we |
| | 1856 | * can't really have a body-within-body. |
| | 1857 | * |
| | 1858 | * @param $attary an array of attributes and values of <body> |
| | 1859 | * @param $mailbox mailbox we're currently reading (for cid2http) |
| | 1860 | * @param $message current message (for cid2http) |
| | 1861 | * @param $id current message id (for cid2http) |
| | 1862 | * @return a modified array of attributes to be set for <div> |
| | 1863 | */ |
| | 1864 | function sq_body2div($attary, $mailbox, $message, $id){ |
| | 1865 | $me = 'sq_body2div'; |
| | 1866 | $divattary = Array('class' => "'bodyclass'"); |
| | 1867 | $text = '#000000'; |
| | 1868 | $has_bgc_stl = $has_txt_stl = false; |
| | 1869 | $styledef = ''; |
| | 1870 | if (is_array($attary) && sizeof($attary) > 0){ |
| | 1871 | foreach ($attary as $attname=>$attvalue){ |
| | 1872 | $quotchar = substr($attvalue, 0, 1); |
| | 1873 | $attvalue = str_replace($quotchar, "", $attvalue); |
| | 1874 | switch ($attname){ |
| | 1875 | case 'background': |
| | 1876 | $attvalue = sq_cid2http($message, $id, $attvalue, $mailbox); |
| | 1877 | $styledef .= "background-image: url('$attvalue'); "; |
| | 1878 | break; |
| | 1879 | case 'bgcolor': |
| | 1880 | $has_bgc_stl = true; |
| | 1881 | $styledef .= "background-color: $attvalue; "; |
| | 1882 | break; |
| | 1883 | case 'text': |
| | 1884 | $has_txt_stl = true; |
| | 1885 | $styledef .= "color: $attvalue; "; |
| | 1886 | break; |
| | 1887 | } |
| | 1888 | } |
| | 1889 | // Outlook defines a white bgcolor and no text color. This can lead to |
| | 1890 | // white text on a white bg with certain themes. |
| | 1891 | if ($has_bgc_stl && !$has_txt_stl) { |
| | 1892 | $styledef .= "color: $text; "; |
| | 1893 | } |
| | 1894 | if (strlen($styledef) > 0){ |
| | 1895 | $divattary{"style"} = "\"$styledef\""; |
| | 1896 | } |
| | 1897 | } |
| | 1898 | return $divattary; |
| | 1899 | } |
| | 1900 | |
| | 1901 | /** |
| | 1902 | * This is the main function and the one you should actually be calling. |
| | 1903 | * There are several variables you should be aware of an which need |
| | 1904 | * special description. |
| | 1905 | * |
| | 1906 | * Since the description is quite lengthy, see it here: |
| | 1907 | * http://linux.duke.edu/projects/mini/htmlfilter/ |
| | 1908 | * |
| | 1909 | * @param $body the string with HTML you wish to filter |
| | 1910 | * @param $tag_list see description above |
| | 1911 | * @param $rm_tags_with_content see description above |
| | 1912 | * @param $self_closing_tags see description above |
| | 1913 | * @param $force_tag_closing see description above |
| | 1914 | * @param $rm_attnames see description above |
| | 1915 | * @param $bad_attvals see description above |
| | 1916 | * @param $add_attr_to_tag see description above |
| | 1917 | * @param $message message object |
| | 1918 | * @param $id message id |
| | 1919 | * @return sanitized html safe to show on your pages. |
| | 1920 | */ |
| | 1921 | function sq_sanitize($body, |
| | 1922 | $tag_list, |
| | 1923 | $rm_tags_with_content, |
| | 1924 | $self_closing_tags, |
| | 1925 | $force_tag_closing, |
| | 1926 | $rm_attnames, |
| | 1927 | $bad_attvals, |
| | 1928 | $add_attr_to_tag, |
| | 1929 | $message, |
| | 1930 | $id, |
| | 1931 | $mailbox |
| | 1932 | ){ |
| | 1933 | $me = 'sq_sanitize'; |
| | 1934 | $rm_tags = array_shift($tag_list); |
| | 1935 | /** |
| | 1936 | * Normalize rm_tags and rm_tags_with_content. |
| | 1937 | */ |
| | 1938 | @array_walk($tag_list, 'sq_casenormalize'); |
| | 1939 | @array_walk($rm_tags_with_content, 'sq_casenormalize'); |
| | 1940 | @array_walk($self_closing_tags, 'sq_casenormalize'); |
| | 1941 | /** |
| | 1942 | * See if tag_list is of tags to remove or tags to allow. |
| | 1943 | * false means remove these tags |
| | 1944 | * true means allow these tags |
| | 1945 | */ |
| | 1946 | $curpos = 0; |
| | 1947 | $open_tags = Array(); |
| | 1948 | $trusted = "\n<!-- begin sanitized html -->\n"; |
| | 1949 | $skip_content = false; |
| | 1950 | /** |
| | 1951 | * Take care of netscape's stupid javascript entities like |
| | 1952 | * &{alert('boo')}; |
| | 1953 | */ |
| | 1954 | $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body); |
| | 1955 | |
| | 1956 | while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){ |
| | 1957 | list($tagname, $attary, $tagtype, $lt, $gt) = $curtag; |
| | 1958 | $free_content = substr($body, $curpos, $lt-$curpos); |
| | 1959 | /** |
| | 1960 | * Take care of <style> |
| | 1961 | */ |
| | 1962 | if ($tagname == "style" && $tagtype == 1){ |
| | 1963 | list($free_content, $curpos) = |
| | 1964 | sq_fixstyle($body, $gt+1, $message, $id, $mailbox); |
| | 1965 | if ($free_content != FALSE){ |
| | 1966 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
| | 1967 | $trusted .= $free_content; |
| | 1968 | $trusted .= sq_tagprint($tagname, false, 2); |
| | 1969 | } |
| | 1970 | continue; |
| | 1971 | } |
| | 1972 | if ($skip_content == false){ |
| | 1973 | $trusted .= $free_content; |
| | 1974 | } |
| | 1975 | if ($tagname != FALSE){ |
| | 1976 | if ($tagtype == 2){ |
| | 1977 | if ($skip_content == $tagname){ |
| | 1978 | /** |
| | 1979 | * Got to the end of tag we needed to remove. |
| | 1980 | */ |
| | 1981 | $tagname = false; |
| | 1982 | $skip_content = false; |
| | 1983 | } else { |
| | 1984 | if ($skip_content == false){ |
| | 1985 | if ($tagname == "body"){ |
| | 1986 | $tagname = "div"; |
| | 1987 | } |
| | 1988 | if (isset($open_tags{$tagname}) && |
| | 1989 | $open_tags{$tagname} > 0){ |
| | 1990 | $open_tags{$tagname}--; |
| | 1991 | } else { |
| | 1992 | $tagname = false; |
| | 1993 | } |
| | 1994 | } |
| | 1995 | } |
| | 1996 | } else { |
| | 1997 | /** |
| | 1998 | * $rm_tags_with_content |
| | 1999 | */ |
| | 2000 | if ($skip_content == false){ |
| | 2001 | /** |
| | 2002 | * See if this is a self-closing type and change |
| | 2003 | * tagtype appropriately. |
| | 2004 | */ |
| | 2005 | if ($tagtype == 1 |
| | 2006 | && in_array($tagname, $self_closing_tags)){ |
| | 2007 | $tagtype = 3; |
| | 2008 | } |
| | 2009 | /** |
| | 2010 | * See if we should skip this tag and any content |
| | 2011 | * inside it. |
| | 2012 | */ |
| | 2013 | if ($tagtype == 1 && |
| | 2014 | in_array($tagname, $rm_tags_with_content)){ |
| | 2015 | $skip_content = $tagname; |
| | 2016 | } else { |
| | 2017 | if (($rm_tags == false |
| | 2018 | && in_array($tagname, $tag_list)) || |
| | 2019 | ($rm_tags == true && |
| | 2020 | !in_array($tagname, $tag_list))){ |
| | 2021 | $tagname = false; |
| | 2022 | } else { |
| | 2023 | /** |
| | 2024 | * Convert body into div. |
| | 2025 | */ |
| | 2026 | if ($tagname == "body"){ |
| | 2027 | $tagname = "div"; |
| | 2028 | $attary = sq_body2div($attary, $mailbox, |
| | 2029 | $message, $id); |
| | 2030 | } |
| | 2031 | if ($tagtype == 1){ |
| | 2032 | if (isset($open_tags{$tagname})){ |
| | 2033 | $open_tags{$tagname}++; |
| | 2034 | } else { |
| | 2035 | $open_tags{$tagname}=1; |
| | 2036 | } |
| | 2037 | } |
| | 2038 | /** |
| | 2039 | * This is where we run other checks. |
| | 2040 | */ |
| | 2041 | if (is_array($attary) && sizeof($attary) > 0){ |
| | 2042 | $attary = sq_fixatts($tagname, |
| | 2043 | $attary, |
| | 2044 | $rm_attnames, |
| | 2045 | $bad_attvals, |
| | 2046 | $add_attr_to_tag, |
| | 2047 | $message, |
| | 2048 | $id, |
| | 2049 | $mailbox |
| | 2050 | ); |
| | 2051 | } |
| | 2052 | } |
| | 2053 | } |
| | 2054 | } |
| | 2055 | } |
| | 2056 | if ($tagname != false && $skip_content == false){ |
| | 2057 | $trusted .= sq_tagprint($tagname, $attary, $tagtype); |
| | 2058 | } |
| | 2059 | } |
| | 2060 | $curpos = $gt+1; |
| | 2061 | } |
| | 2062 | $trusted .= substr($body, $curpos, strlen($body)-$curpos); |
| | 2063 | if ($force_tag_closing == true){ |
| | 2064 | foreach ($open_tags as $tagname=>$opentimes){ |
| | 2065 | while ($opentimes > 0){ |
| | 2066 | $trusted .= '</' . $tagname . '>'; |
| | 2067 | $opentimes--; |
| | 2068 | } |
| | 2069 | } |
| | 2070 | $trusted .= "\n"; |
| | 2071 | } |
| | 2072 | $trusted .= "<!-- end sanitized html -->\n"; |
| | 2073 | return $trusted; |
| | 2074 | } |
| | 2075 | |
| | 2076 | /** |
| | 2077 | * This is a wrapper function to call html sanitizing routines. |
| | 2078 | * |
| | 2079 | * @param $body the body of the message |
| | 2080 | * @param $id the id of the message |
| | 2081 | |
| | 2082 | * @param $message |
| | 2083 | * @param $mailbox |
| | 2084 | * @param boolean $take_mailto_links When TRUE, converts mailto: links |
| | 2085 | * into internal SM compose links |
| | 2086 | * (optional; default = TRUE) |
| | 2087 | * @return a string with html safe to display in the browser. |
| | 2088 | */ |
| | 2089 | function magicHTML($body, $id, $message, $mailbox = 'INBOX', $take_mailto_links =true) { |
| | 2090 | |
| | 2091 | // require_once(SM_PATH . 'functions/url_parser.php'); // for $MailTo_PReg_Match |
| | 2092 | |
| | 2093 | global $attachment_common_show_images, $view_unsafe_images, |
| | 2094 | $has_unsafe_images; |
| | 2095 | /** |
| | 2096 | * Don't display attached images in HTML mode. |
| | 2097 | * |
| | 2098 | * SB: why? |
| | 2099 | */ |
| | 2100 | $attachment_common_show_images = false; |
| | 2101 | $tag_list = Array( |
| | 2102 | false, |
| | 2103 | "object", |
| | 2104 | "meta", |
| | 2105 | "html", |
| | 2106 | "head", |
| | 2107 | "base", |
| | 2108 | "link", |
| | 2109 | "frame", |
| | 2110 | "iframe", |
| | 2111 | "plaintext", |
| | 2112 | "marquee" |
| | 2113 | ); |
| | 2114 | |
| | 2115 | $rm_tags_with_content = Array( |
| | 2116 | "script", |
| | 2117 | "applet", |
| | 2118 | "embed", |
| | 2119 | "title", |
| | 2120 | "frameset", |
| | 2121 | "xmp", |
| | 2122 | "xml" |
| | 2123 | ); |
| | 2124 | |
| | 2125 | $self_closing_tags = Array( |
| | 2126 | "img", |
| | 2127 | "br", |
| | 2128 | "hr", |
| | 2129 | "input", |
| | 2130 | "outbind" |
| | 2131 | ); |
| | 2132 | |
| | 2133 | $force_tag_closing = true; |
| | 2134 | |
| | 2135 | $rm_attnames = Array( |
| | 2136 | "/.*/" => |
| | 2137 | Array( |
| | 2138 | "/target/i", |
| | 2139 | "/^on.*/i", |
| | 2140 | "/^dynsrc/i", |
| | 2141 | "/^data.*/i", |
| | 2142 | "/^lowsrc.*/i" |
| | 2143 | ) |
| | 2144 | ); |
| | 2145 | |
| | 2146 | $secremoveimg = "../images/" . _("sec_remove_eng.png"); |
| | 2147 | $bad_attvals = Array( |
| | 2148 | "/.*/" => |
| | 2149 | Array( |
| | 2150 | "/^src|background/i" => |
| | 2151 | Array( |
| | 2152 | Array( |
| | 2153 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
| | 2154 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
| | 2155 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
| | 2156 | ), |
| | 2157 | Array( |
| | 2158 | "\\1$secremoveimg\\2", |
| | 2159 | "\\1$secremoveimg\\2", |
| | 2160 | "\\1$secremoveimg\\2", |
| | 2161 | ) |
| | 2162 | ), |
| | 2163 | "/^href|action/i" => |
| | 2164 | Array( |
| | 2165 | Array( |
| | 2166 | "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si", |
| | 2167 | "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si", |
| | 2168 | "/^([\'\"])\s*about\s*:.*([\'\"])/si" |
| | 2169 | ), |
| | 2170 | Array( |
| | 2171 | "\\1#\\1", |
| | 2172 | "\\1#\\1", |
| | 2173 | "\\1#\\1" |
| | 2174 | ) |
| | 2175 | ), |
| | 2176 | "/^style/i" => |
| | 2177 | Array( |
| | 2178 | Array( |
| | 2179 | "/\/\*.*\*\//", |
| | 2180 | "/expression/i", |
| | 2181 | "/binding/i", |
| | 2182 | "/behaviou*r/i", |
| | 2183 | "/include-source/i", |
| | 2184 | "/position\s*:\s*absolute/i", |
| | 2185 | "/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", |
| | 2186 | "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si", |
| | 2187 | "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si", |
| | 2188 | "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si", |
| | 2189 | "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si" |
| | 2190 | ), |
| | 2191 | Array( |
| | 2192 | "", |
| | 2193 | "idiocy", |
| | 2194 | "idiocy", |
| | 2195 | "idiocy", |
| | 2196 | "idiocy", |
| | 2197 | "", |
| | 2198 | "url", |
| | 2199 | "url(\\1#\\1)", |
| | 2200 | "url(\\1#\\1)", |
| | 2201 | "url(\\1#\\1)", |
| | 2202 | "\\1:url(\\2#\\3)" |
| | 2203 | ) |
| | 2204 | ) |
| | 2205 | ) |
| | 2206 | ); |
| | 2207 | if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) { |
| | 2208 | $view_unsafe_images = false; |
| | 2209 | } |
| | 2210 | if (!$view_unsafe_images){ |
| | 2211 | /** |
| | 2212 | * Remove any references to http/https if view_unsafe_images set |
| | 2213 | * to false. |
| | 2214 | */ |
| | 2215 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0], |
| | 2216 | '/^([\'\"])\s*https*:.*([\'\"])/si'); |
| | 2217 | array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1], |
| | 2218 | "\\1$secremoveimg\\1"); |
| | 2219 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[0], |
| | 2220 | '/url\([\'\"]?https?:[^\)]*[\'\"]?\)/si'); |
| | 2221 | array_push($bad_attvals{'/.*/'}{'/^style/i'}[1], |
| | 2222 | "url(\\1$secremoveimg\\1)"); |
| | 2223 | } |
| | 2224 | |
| | 2225 | $add_attr_to_tag = Array( |
| | 2226 | "/^a$/i" => |
| | 2227 | Array('target'=>'"_blank"', |
| | 2228 | 'title'=>'"'._("This external link will open in a new window").'"' |
| | 2229 | ) |
| | 2230 | ); |
| | 2231 | $trusted = sq_sanitize($body, |
| | 2232 | $tag_list, |
| | 2233 | $rm_tags_with_content, |
| | 2234 | $self_closing_tags, |
| | 2235 | $force_tag_closing, |
| | 2236 | $rm_attnames, |
| | 2237 | $bad_attvals, |
| | 2238 | $add_attr_to_tag, |
| | 2239 | $message, |
| | 2240 | $id, |
| | 2241 | $mailbox |
| | 2242 | ); |
| | 2243 | if (preg_match("|$secremoveimg|i", $trusted)){ |
| | 2244 | $has_unsafe_images = true; |
| | 2245 | } |
| | 2246 | |
| | 2247 | // we want to parse mailto's in HTML output, change to SM compose links |
| | 2248 | // this is a modified version of code from url_parser.php... but Marc is |
| | 2249 | // right: we need a better filtering implementation; adding this randomly |
| | 2250 | // here is not a great solution |
| | 2251 | // |
| | 2252 | if ($take_mailto_links) { |
| | 2253 | // parseUrl($trusted); // this even parses URLs inside of tags... too aggressive |
| | 2254 | global $MailTo_PReg_Match; |
| | 2255 | $MailTo_PReg_Match = '/mailto:' . substr($MailTo_PReg_Match, 1) ; |
| | 2256 | if ((preg_match_all($MailTo_PReg_Match, $trusted, $regs)) && ($regs[0][0] != '')) { |
| | 2257 | foreach ($regs[0] as $i => $mailto_before) { |
| | 2258 | $mailto_params = $regs[10][$i]; |
| | 2259 | // get rid of any tailing quote since we have to add send_to to the end |
| | 2260 | // |
| | 2261 | if (substr($mailto_before, strlen($mailto_before) - 1) == '"') |
| | 2262 | $mailto_before = substr($mailto_before, 0, strlen($mailto_before) - 1); |
| | 2263 | if (substr($mailto_params, strlen($mailto_params) - 1) == '"') |
| | 2264 | $mailto_params = substr($mailto_params, 0, strlen($mailto_params) - 1); |
| | 2265 | |
| | 2266 | if ($regs[1][$i]) { //if there is an email addr before '?', we need to merge it with the params |
| | 2267 | $to = 'to=' . $regs[1][$i]; |
| | 2268 | if (strpos($mailto_params, 'to=') > -1) //already a 'to=' |
| | 2269 | $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params); |
| | 2270 | else { |
| | 2271 | if ($mailto_params) //already some params, append to them |
| | 2272 | $mailto_params .= '&' . $to; |
| | 2273 | else |
| | 2274 | $mailto_params .= '?' . $to; |
| | 2275 | } |
| | 2276 | } |
| | 2277 | |
| | 2278 | $url_str = preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params); |
| | 2279 | |
| | 2280 | // we'll already have target=_blank, no need to allow comp_in_new |
| | 2281 | // here (which would be a lot more work anyway) |
| | 2282 | // |
| | 2283 | global $compose_new_win; |
| | 2284 | $temp_comp_in_new = $compose_new_win; |
| | 2285 | $compose_new_win = 0; |
| | 2286 | $comp_uri = makeComposeLink('src/compose.php' . $url_str, $mailto_before); |
| | 2287 | $compose_new_win = $temp_comp_in_new; |
| | 2288 | |
| | 2289 | // remove <a href=" and anything after the next quote (we only |
| | 2290 | // need the uri, not the link HTML) in compose uri |
| | 2291 | // |
| | 2292 | $comp_uri = substr($comp_uri, 9); |
| | 2293 | $comp_uri = substr($comp_uri, 0, strpos($comp_uri, '"', 1)); |
| | 2294 | $trusted = str_replace($mailto_before, $comp_uri, $trusted); |
| | 2295 | } |
| | 2296 | } |
| | 2297 | } |
| | 2298 | |
| | 2299 | return $trusted; |
| | 2300 | } |
| | 2301 | |
| | 2302 | /** |
| | 2303 | * function SendDownloadHeaders - send file to the browser |
| | 2304 | * |
| | 2305 | * Original Source: SM core src/download.php |
| | 2306 | * moved here to make it available to other code, and separate |
| | 2307 | * front end from back end functionality. |
| | 2308 | * |
| | 2309 | * @param string $type0 first half of mime type |
| | 2310 | * @param string $type1 second half of mime type |
| | 2311 | * @param string $filename filename to tell the browser for downloaded file |
| | 2312 | * @param boolean $force whether to force the download dialog to pop |
| | 2313 | * @param optional integer $filesize send the Content-Header and length to the browser |
| | 2314 | * @return void |
| | 2315 | */ |
| | 2316 | function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) { |
| | 2317 | global $languages, $squirrelmail_language; |
| | 2318 | $isIE = $isIE6plus = false; |
| | 2319 | |
| | 2320 | sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER); |
| | 2321 | |
| | 2322 | if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false && |
| | 2323 | strstr($HTTP_USER_AGENT, 'Opera') === false) { |
| | 2324 | $isIE = true; |
| | 2325 | } |
| | 2326 | |
| | 2327 | if (preg_match('/compatible; MSIE ([0-9]+)/', $HTTP_USER_AGENT, $match) && |
| | 2328 | ((int)$match[1]) >= 6 && strstr($HTTP_USER_AGENT, 'Opera') === false) { |
| | 2329 | $isIE6plus = true; |
| | 2330 | } |
| | 2331 | |
| | 2332 | if (isset($languages[$squirrelmail_language]['XTRA_CODE']) && |
| | 2333 | function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) { |
| | 2334 | $filename = |
| | 2335 | call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT); |
| | 2336 | } else { |
| | 2337 | $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename)); |
| | 2338 | } |
| | 2339 | |
| | 2340 | // A Pox on Microsoft and it's Internet Explorer! |
| | 2341 | // |
| | 2342 | // IE has lots of bugs with file downloads. |
| | 2343 | // It also has problems with SSL. Both of these cause problems |
| | 2344 | // for us in this function. |
| | 2345 | // |
| | 2346 | // See this article on Cache Control headers and SSL |
| | 2347 | // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308 |
| | 2348 | // |
| | 2349 | // The best thing you can do for IE is to upgrade to the latest |
| | 2350 | // version |
| | 2351 | //set all the Cache Control Headers for IE |
| | 2352 | if ($isIE) { |
| | 2353 | $filename=rawurlencode($filename); |
| | 2354 | header ("Pragma: public"); |
| | 2355 | header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); // HTTP/1.1 |
| | 2356 | header ("Cache-Control: post-check=0, pre-check=0", false); |
| | 2357 | header ("Cache-Control: private"); |
| | 2358 | |
| | 2359 | //set the inline header for IE, we'll add the attachment header later if we need it |
| | 2360 | header ("Content-Disposition: inline; filename=$filename"); |
| | 2361 | } |
| | 2362 | |
| | 2363 | if (!$force) { |
| | 2364 | // Try to show in browser window |
| | 2365 | header ("Content-Disposition: inline; filename=\"$filename\""); |
| | 2366 | header ("Content-Type: $type0/$type1; name=\"$filename\""); |
| | 2367 | } else { |
| | 2368 | // Try to pop up the "save as" box |
| | 2369 | |
| | 2370 | // IE makes this hard. It pops up 2 save boxes, or none. |
| | 2371 | // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP |
| | 2372 | // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519 |
| | 2373 | // But, according to Microsoft, it is "RFC compliant but doesn't |
| | 2374 | // take into account some deviations that allowed within the |
| | 2375 | // specification." Doesn't that mean RFC non-compliant? |
| | 2376 | // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP |
| | 2377 | |
| | 2378 | // all browsers need the application/octet-stream header for this |
| | 2379 | header ("Content-Type: application/octet-stream; name=\"$filename\""); |
| | 2380 | |
| | 2381 | // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp |
| | 2382 | // Do not have quotes around filename, but that applied to |
| | 2383 | // "attachment"... does it apply to inline too? |
| | 2384 | header ("Content-Disposition: attachment; filename=\"$filename\""); |
| | 2385 | |
| | 2386 | if ($isIE && !$isIE6plus) { |
| | 2387 | // This combination seems to work mostly. IE 5.5 SP 1 has |
| | 2388 | // known issues (see the Microsoft Knowledge Base) |
| | 2389 | |
| | 2390 | // This works for most types, but doesn't work with Word files |
| | 2391 | header ("Content-Type: application/download; name=\"$filename\""); |
| | 2392 | |
| | 2393 | // These are spares, just in case. :-) |
| | 2394 | //header("Content-Type: $type0/$type1; name=\"$filename\""); |
| | 2395 | //header("Content-Type: application/x-msdownload; name=\"$filename\""); |
| | 2396 | //header("Content-Type: application/octet-stream; name=\"$filename\""); |
| | 2397 | } else { |
| | 2398 | // another application/octet-stream forces download for Netscape |
| | 2399 | header ("Content-Type: application/octet-stream; name=\"$filename\""); |
| | 2400 | } |
| | 2401 | } |
| | 2402 | |
| | 2403 | //send the content-length header if the calling function provides it |
| | 2404 | if ($filesize > 0) { |
| | 2405 | header("Content-Length: $filesize"); |
| | 2406 | } |
| | 2407 | |
| | 2408 | } // end fn SendDownloadHeaders |
projects / squirrelmail.git / blame_incremental