| commit | author | age | ||
| 1c4f23 | 1 | <?php |
| A | 2 | |
| 8b92d2 | 3 | /* |
| 1c4f23 | 4 | +-----------------------------------------------------------------------+ |
| A | 5 | | program/include/rcube_mime.php | |
| 6 | | | | |
| 7 | | This file is part of the Roundcube Webmail client | | |
| 8 | | Copyright (C) 2005-2012, The Roundcube Dev Team | | |
| 9 | | Copyright (C) 2011-2012, Kolab Systems AG | | |
| 7fe381 | 10 | | | |
| T | 11 | | Licensed under the GNU General Public License version 3 or | |
| 12 | | any later version with exceptions for skins & plugins. | | |
| 13 | | See the README file for a full license statement. | | |
| 1c4f23 | 14 | | | |
| A | 15 | | PURPOSE: | |
| 16 | | MIME message parsing utilities | | |
| 17 | | | | |
| 18 | +-----------------------------------------------------------------------+ | |
| 19 | | Author: Thomas Bruederli <roundcube@gmail.com> | | |
| 20 | | Author: Aleksander Machniak <alec@alec.pl> | | |
| 21 | +-----------------------------------------------------------------------+ | |
| 22 | */ | |
| 23 | ||
| 24 | ||
| 25 | /** | |
| 26 | * Class for parsing MIME messages | |
| 27 | * | |
| 9ab346 | 28 | * @package Framework |
| AM | 29 | * @subpackage Storage |
| 30 | * @author Thomas Bruederli <roundcube@gmail.com> | |
| 31 | * @author Aleksander Machniak <alec@alec.pl> | |
| 1c4f23 | 32 | */ |
| A | 33 | class rcube_mime |
| 34 | { | |
| eede51 | 35 | private static $default_charset; |
| 1c4f23 | 36 | |
| A | 37 | |
| 38 | /** | |
| 39 | * Object constructor. | |
| 40 | */ | |
| 41 | function __construct($default_charset = null) | |
| 42 | { | |
| eede51 | 43 | self::$default_charset = $default_charset; |
| 0f5dee | 44 | } |
| AM | 45 | |
| 46 | ||
| 47 | /** | |
| 48 | * Returns message/object character set name | |
| 49 | * | |
| 50 | * @return string Characted set name | |
| 51 | */ | |
| 52 | public static function get_charset() | |
| 53 | { | |
| 54 | if (self::$default_charset) { | |
| 55 | return self::$default_charset; | |
| 1c4f23 | 56 | } |
| 0f5dee | 57 | |
| AM | 58 | if ($charset = rcube::get_instance()->config->get('default_charset')) { |
| 59 | return $charset; | |
| 60 | } | |
| 61 | ||
| a92beb | 62 | return RCUBE_CHARSET; |
| 1c4f23 | 63 | } |
| A | 64 | |
| 65 | ||
| 66 | /** | |
| 8b92d2 | 67 | * Parse the given raw message source and return a structure |
| T | 68 | * of rcube_message_part objects. |
| 69 | * | |
| 70 | * It makes use of the PEAR:Mail_mimeDecode library | |
| 71 | * | |
| 72 | * @param string The message source | |
| 73 | * @return object rcube_message_part The message structure | |
| 74 | */ | |
| 75 | public static function parse_message($raw_body) | |
| 76 | { | |
| 77 | $mime = new Mail_mimeDecode($raw_body); | |
| 78 | $struct = $mime->decode(array('include_bodies' => true, 'decode_bodies' => true)); | |
| 79 | return self::structure_part($struct); | |
| 80 | } | |
| 81 | ||
| 82 | ||
| 83 | /** | |
| 84 | * Recursive method to convert a Mail_mimeDecode part into a rcube_message_part object | |
| 85 | * | |
| 86 | * @param object A message part struct | |
| 87 | * @param int Part count | |
| 88 | * @param string Parent MIME ID | |
| 89 | * | |
| 90 | * @return object rcube_message_part | |
| 91 | */ | |
| 92 | private static function structure_part($part, $count=0, $parent='') | |
| 93 | { | |
| 94 | $struct = new rcube_message_part; | |
| 95 | $struct->mime_id = $part->mime_id ? $part->mime_id : (empty($parent) ? (string)$count : "$parent.$count"); | |
| 96 | $struct->headers = $part->headers; | |
| 97 | $struct->ctype_primary = $part->ctype_primary; | |
| 98 | $struct->ctype_secondary = $part->ctype_secondary; | |
| 99 | $struct->mimetype = $part->ctype_primary . '/' . $part->ctype_secondary; | |
| 100 | $struct->ctype_parameters = $part->ctype_parameters; | |
| 101 | ||
| 102 | if ($part->headers['content-transfer-encoding']) | |
| 103 | $struct->encoding = $part->headers['content-transfer-encoding']; | |
| 104 | if ($part->ctype_parameters['charset']) | |
| 105 | $struct->charset = $part->ctype_parameters['charset']; | |
| 106 | ||
| 0f5dee | 107 | $part_charset = $struct->charset ? $struct->charset : self::get_charset(); |
| 8b92d2 | 108 | |
| 77c779 | 109 | // determine filename |
| 8b92d2 | 110 | if (($filename = $part->d_parameters['filename']) || ($filename = $part->ctype_parameters['name'])) { |
| T | 111 | $struct->filename = rcube_mime::decode_mime_string($filename, $part_charset); |
| 112 | } | |
| 113 | ||
| 114 | // copy part body and convert it to UTF-8 if necessary | |
| 77c779 | 115 | $struct->body = $part->ctype_primary == 'text' || !$part->ctype_parameters['charset'] ? rcube_charset::convert($part->body, $part_charset) : $part->body; |
| 8b92d2 | 116 | $struct->size = strlen($part->body); |
| T | 117 | $struct->disposition = $part->disposition; |
| 118 | ||
| 119 | foreach ((array)$part->parts as $child_part) { | |
| 120 | $struct->parts[] = self::structure_part($child_part, ++$count, $struct->mime_id); | |
| 121 | } | |
| 122 | ||
| 123 | return $struct; | |
| 124 | } | |
| 125 | ||
| 126 | ||
| 127 | /** | |
| 1c4f23 | 128 | * Split an address list into a structured array list |
| A | 129 | * |
| 130 | * @param string $input Input string | |
| 131 | * @param int $max List only this number of addresses | |
| 132 | * @param boolean $decode Decode address strings | |
| 133 | * @param string $fallback Fallback charset if none specified | |
| 134 | * | |
| 135 | * @return array Indexed list of addresses | |
| 136 | */ | |
| 137 | static function decode_address_list($input, $max = null, $decode = true, $fallback = null) | |
| 138 | { | |
| 139 | $a = self::parse_address_list($input, $decode, $fallback); | |
| 140 | $out = array(); | |
| 141 | $j = 0; | |
| 142 | ||
| 143 | // Special chars as defined by RFC 822 need to in quoted string (or escaped). | |
| 144 | $special_chars = '[\(\)\<\>\\\.\[\]@,;:"]'; | |
| 145 | ||
| 146 | if (!is_array($a)) | |
| 147 | return $out; | |
| 148 | ||
| 149 | foreach ($a as $val) { | |
| 150 | $j++; | |
| 151 | $address = trim($val['address']); | |
| 152 | $name = trim($val['name']); | |
| 153 | ||
| 154 | if ($name && $address && $name != $address) | |
| 155 | $string = sprintf('%s <%s>', preg_match("/$special_chars/", $name) ? '"'.addcslashes($name, '"').'"' : $name, $address); | |
| 156 | else if ($address) | |
| 157 | $string = $address; | |
| 158 | else if ($name) | |
| 159 | $string = $name; | |
| 160 | ||
| 161 | $out[$j] = array( | |
| 162 | 'name' => $name, | |
| 163 | 'mailto' => $address, | |
| 164 | 'string' => $string | |
| 165 | ); | |
| 166 | ||
| 167 | if ($max && $j==$max) | |
| 168 | break; | |
| 169 | } | |
| 170 | ||
| 171 | return $out; | |
| 172 | } | |
| 173 | ||
| 174 | ||
| 175 | /** | |
| 176 | * Decode a message header value | |
| 177 | * | |
| 178 | * @param string $input Header value | |
| 179 | * @param string $fallback Fallback charset if none specified | |
| 180 | * | |
| 181 | * @return string Decoded string | |
| 182 | */ | |
| 183 | public static function decode_header($input, $fallback = null) | |
| 184 | { | |
| 185 | $str = self::decode_mime_string((string)$input, $fallback); | |
| 186 | ||
| 187 | return $str; | |
| 188 | } | |
| 189 | ||
| 190 | ||
| 191 | /** | |
| 192 | * Decode a mime-encoded string to internal charset | |
| 193 | * | |
| 194 | * @param string $input Header value | |
| 195 | * @param string $fallback Fallback charset if none specified | |
| 196 | * | |
| 197 | * @return string Decoded string | |
| 198 | */ | |
| 199 | public static function decode_mime_string($input, $fallback = null) | |
| 200 | { | |
| 0f5dee | 201 | $default_charset = !empty($fallback) ? $fallback : self::get_charset(); |
| 1c4f23 | 202 | |
| A | 203 | // rfc: all line breaks or other characters not found |
| 204 | // in the Base64 Alphabet must be ignored by decoding software | |
| 205 | // delete all blanks between MIME-lines, differently we can | |
| 206 | // receive unnecessary blanks and broken utf-8 symbols | |
| 207 | $input = preg_replace("/\?=\s+=\?/", '?==?', $input); | |
| 208 | ||
| 209 | // encoded-word regexp | |
| 210 | $re = '/=\?([^?]+)\?([BbQq])\?([^\n]*?)\?=/'; | |
| 211 | ||
| 212 | // Find all RFC2047's encoded words | |
| 213 | if (preg_match_all($re, $input, $matches, PREG_OFFSET_CAPTURE | PREG_SET_ORDER)) { | |
| 214 | // Initialize variables | |
| 215 | $tmp = array(); | |
| 216 | $out = ''; | |
| 217 | $start = 0; | |
| 218 | ||
| 219 | foreach ($matches as $idx => $m) { | |
| 220 | $pos = $m[0][1]; | |
| 221 | $charset = $m[1][0]; | |
| 222 | $encoding = $m[2][0]; | |
| 223 | $text = $m[3][0]; | |
| 224 | $length = strlen($m[0][0]); | |
| 225 | ||
| 226 | // Append everything that is before the text to be decoded | |
| 227 | if ($start != $pos) { | |
| 228 | $substr = substr($input, $start, $pos-$start); | |
| eebd44 | 229 | $out .= rcube_charset::convert($substr, $default_charset); |
| 1c4f23 | 230 | $start = $pos; |
| A | 231 | } |
| 232 | $start += $length; | |
| 233 | ||
| 234 | // Per RFC2047, each string part "MUST represent an integral number | |
| 235 | // of characters . A multi-octet character may not be split across | |
| 236 | // adjacent encoded-words." However, some mailers break this, so we | |
| 237 | // try to handle characters spanned across parts anyway by iterating | |
| 238 | // through and aggregating sequential encoded parts with the same | |
| 239 | // character set and encoding, then perform the decoding on the | |
| 240 | // aggregation as a whole. | |
| 241 | ||
| 242 | $tmp[] = $text; | |
| 243 | if ($next_match = $matches[$idx+1]) { | |
| 244 | if ($next_match[0][1] == $start | |
| 245 | && $next_match[1][0] == $charset | |
| 246 | && $next_match[2][0] == $encoding | |
| 247 | ) { | |
| 248 | continue; | |
| 249 | } | |
| 250 | } | |
| 251 | ||
| 252 | $count = count($tmp); | |
| 253 | $text = ''; | |
| 254 | ||
| 255 | // Decode and join encoded-word's chunks | |
| 256 | if ($encoding == 'B' || $encoding == 'b') { | |
| 257 | // base64 must be decoded a segment at a time | |
| 258 | for ($i=0; $i<$count; $i++) | |
| 259 | $text .= base64_decode($tmp[$i]); | |
| 260 | } | |
| 261 | else { //if ($encoding == 'Q' || $encoding == 'q') { | |
| 262 | // quoted printable can be combined and processed at once | |
| 263 | for ($i=0; $i<$count; $i++) | |
| 264 | $text .= $tmp[$i]; | |
| 265 | ||
| 266 | $text = str_replace('_', ' ', $text); | |
| 267 | $text = quoted_printable_decode($text); | |
| 268 | } | |
| 269 | ||
| eebd44 | 270 | $out .= rcube_charset::convert($text, $charset); |
| 1c4f23 | 271 | $tmp = array(); |
| A | 272 | } |
| 273 | ||
| 274 | // add the last part of the input string | |
| 275 | if ($start != strlen($input)) { | |
| eebd44 | 276 | $out .= rcube_charset::convert(substr($input, $start), $default_charset); |
| 1c4f23 | 277 | } |
| A | 278 | |
| 279 | // return the results | |
| 280 | return $out; | |
| 281 | } | |
| 282 | ||
| 283 | // no encoding information, use fallback | |
| eebd44 | 284 | return rcube_charset::convert($input, $default_charset); |
| 1c4f23 | 285 | } |
| A | 286 | |
| 287 | ||
| 288 | /** | |
| 289 | * Decode a mime part | |
| 290 | * | |
| 291 | * @param string $input Input string | |
| 292 | * @param string $encoding Part encoding | |
| 293 | * @return string Decoded string | |
| 294 | */ | |
| 295 | public static function decode($input, $encoding = '7bit') | |
| 296 | { | |
| 297 | switch (strtolower($encoding)) { | |
| 298 | case 'quoted-printable': | |
| 299 | return quoted_printable_decode($input); | |
| 300 | case 'base64': | |
| 301 | return base64_decode($input); | |
| 302 | case 'x-uuencode': | |
| 303 | case 'x-uue': | |
| 304 | case 'uue': | |
| 305 | case 'uuencode': | |
| 306 | return convert_uudecode($input); | |
| 307 | case '7bit': | |
| 308 | default: | |
| 309 | return $input; | |
| 310 | } | |
| 311 | } | |
| 312 | ||
| 313 | ||
| 314 | /** | |
| 315 | * Split RFC822 header string into an associative array | |
| 316 | * @access private | |
| 317 | */ | |
| 318 | public static function parse_headers($headers) | |
| 319 | { | |
| 320 | $a_headers = array(); | |
| 321 | $headers = preg_replace('/\r?\n(\t| )+/', ' ', $headers); | |
| 322 | $lines = explode("\n", $headers); | |
| 323 | $c = count($lines); | |
| 324 | ||
| 325 | for ($i=0; $i<$c; $i++) { | |
| 326 | if ($p = strpos($lines[$i], ': ')) { | |
| 327 | $field = strtolower(substr($lines[$i], 0, $p)); | |
| 328 | $value = trim(substr($lines[$i], $p+1)); | |
| 329 | if (!empty($value)) | |
| 330 | $a_headers[$field] = $value; | |
| 331 | } | |
| 332 | } | |
| 333 | ||
| 334 | return $a_headers; | |
| 335 | } | |
| 336 | ||
| 337 | ||
| 338 | /** | |
| 339 | * @access private | |
| 340 | */ | |
| 341 | private static function parse_address_list($str, $decode = true, $fallback = null) | |
| 342 | { | |
| 343 | // remove any newlines and carriage returns before | |
| 344 | $str = preg_replace('/\r?\n(\s|\t)?/', ' ', $str); | |
| 345 | ||
| 346 | // extract list items, remove comments | |
| 347 | $str = self::explode_header_string(',;', $str, true); | |
| 348 | $result = array(); | |
| 349 | ||
| 350 | // simplified regexp, supporting quoted local part | |
| 351 | $email_rx = '(\S+|("\s*(?:[^"\f\n\r\t\v\b\s]+\s*)+"))@\S+'; | |
| 352 | ||
| 353 | foreach ($str as $key => $val) { | |
| 354 | $name = ''; | |
| 355 | $address = ''; | |
| 356 | $val = trim($val); | |
| 357 | ||
| 358 | if (preg_match('/(.*)<('.$email_rx.')>$/', $val, $m)) { | |
| 359 | $address = $m[2]; | |
| 360 | $name = trim($m[1]); | |
| 361 | } | |
| 362 | else if (preg_match('/^('.$email_rx.')$/', $val, $m)) { | |
| 363 | $address = $m[1]; | |
| 364 | $name = ''; | |
| 365 | } | |
| 366 | else { | |
| 367 | $name = $val; | |
| 368 | } | |
| 369 | ||
| 370 | // dequote and/or decode name | |
| 371 | if ($name) { | |
| 372 | if ($name[0] == '"' && $name[strlen($name)-1] == '"') { | |
| 373 | $name = substr($name, 1, -1); | |
| 374 | $name = stripslashes($name); | |
| 375 | } | |
| 376 | if ($decode) { | |
| 377 | $name = self::decode_header($name, $fallback); | |
| 378 | } | |
| 379 | } | |
| 380 | ||
| 381 | if (!$address && $name) { | |
| 382 | $address = $name; | |
| 383 | } | |
| 384 | ||
| 385 | if ($address) { | |
| 386 | $result[$key] = array('name' => $name, 'address' => $address); | |
| 387 | } | |
| 388 | } | |
| 389 | ||
| 390 | return $result; | |
| 391 | } | |
| 392 | ||
| 393 | ||
| 394 | /** | |
| 395 | * Explodes header (e.g. address-list) string into array of strings | |
| 396 | * using specified separator characters with proper handling | |
| 397 | * of quoted-strings and comments (RFC2822) | |
| 398 | * | |
| 399 | * @param string $separator String containing separator characters | |
| 400 | * @param string $str Header string | |
| 401 | * @param bool $remove_comments Enable to remove comments | |
| 402 | * | |
| 403 | * @return array Header items | |
| 404 | */ | |
| 405 | public static function explode_header_string($separator, $str, $remove_comments = false) | |
| 406 | { | |
| 407 | $length = strlen($str); | |
| 408 | $result = array(); | |
| 409 | $quoted = false; | |
| 410 | $comment = 0; | |
| 411 | $out = ''; | |
| 412 | ||
| 413 | for ($i=0; $i<$length; $i++) { | |
| 414 | // we're inside a quoted string | |
| 415 | if ($quoted) { | |
| 416 | if ($str[$i] == '"') { | |
| 417 | $quoted = false; | |
| 418 | } | |
| 419 | else if ($str[$i] == "\\") { | |
| 420 | if ($comment <= 0) { | |
| 421 | $out .= "\\"; | |
| 422 | } | |
| 423 | $i++; | |
| 424 | } | |
| 425 | } | |
| 426 | // we are inside a comment string | |
| 427 | else if ($comment > 0) { | |
| 428 | if ($str[$i] == ')') { | |
| 429 | $comment--; | |
| 430 | } | |
| 431 | else if ($str[$i] == '(') { | |
| 432 | $comment++; | |
| 433 | } | |
| 434 | else if ($str[$i] == "\\") { | |
| 435 | $i++; | |
| 436 | } | |
| 437 | continue; | |
| 438 | } | |
| 439 | // separator, add to result array | |
| 440 | else if (strpos($separator, $str[$i]) !== false) { | |
| 441 | if ($out) { | |
| 442 | $result[] = $out; | |
| 443 | } | |
| 444 | $out = ''; | |
| 445 | continue; | |
| 446 | } | |
| 447 | // start of quoted string | |
| 448 | else if ($str[$i] == '"') { | |
| 449 | $quoted = true; | |
| 450 | } | |
| 451 | // start of comment | |
| 452 | else if ($remove_comments && $str[$i] == '(') { | |
| 453 | $comment++; | |
| 454 | } | |
| 455 | ||
| 456 | if ($comment <= 0) { | |
| 457 | $out .= $str[$i]; | |
| 458 | } | |
| 459 | } | |
| 460 | ||
| 461 | if ($out && $comment <= 0) { | |
| 462 | $result[] = $out; | |
| 463 | } | |
| 464 | ||
| 465 | return $result; | |
| 466 | } | |
| 467 | ||
| 468 | ||
| 469 | /** | |
| 470 | * Interpret a format=flowed message body according to RFC 2646 | |
| 471 | * | |
| 472 | * @param string $text Raw body formatted as flowed text | |
| 473 | * | |
| 474 | * @return string Interpreted text with unwrapped lines and stuffed space removed | |
| 475 | */ | |
| 476 | public static function unfold_flowed($text) | |
| 477 | { | |
| 478 | $text = preg_split('/\r?\n/', $text); | |
| 479 | $last = -1; | |
| 480 | $q_level = 0; | |
| 481 | ||
| 482 | foreach ($text as $idx => $line) { | |
| 483 | if ($line[0] == '>' && preg_match('/^(>+\s*)/', $line, $regs)) { | |
| 484 | $q = strlen(str_replace(' ', '', $regs[0])); | |
| 485 | $line = substr($line, strlen($regs[0])); | |
| 486 | ||
| 487 | if ($q == $q_level && $line | |
| 488 | && isset($text[$last]) | |
| 489 | && $text[$last][strlen($text[$last])-1] == ' ' | |
| 490 | ) { | |
| 491 | $text[$last] .= $line; | |
| 492 | unset($text[$idx]); | |
| 493 | } | |
| 494 | else { | |
| 495 | $last = $idx; | |
| 496 | } | |
| 497 | } | |
| 498 | else { | |
| 499 | $q = 0; | |
| 500 | if ($line == '-- ') { | |
| 501 | $last = $idx; | |
| 502 | } | |
| 503 | else { | |
| 504 | // remove space-stuffing | |
| 505 | $line = preg_replace('/^\s/', '', $line); | |
| 506 | ||
| 507 | if (isset($text[$last]) && $line | |
| 508 | && $text[$last] != '-- ' | |
| 509 | && $text[$last][strlen($text[$last])-1] == ' ' | |
| 510 | ) { | |
| 511 | $text[$last] .= $line; | |
| 512 | unset($text[$idx]); | |
| 513 | } | |
| 514 | else { | |
| 515 | $text[$idx] = $line; | |
| 516 | $last = $idx; | |
| 517 | } | |
| 518 | } | |
| 519 | } | |
| 520 | $q_level = $q; | |
| 521 | } | |
| 522 | ||
| 523 | return implode("\r\n", $text); | |
| 524 | } | |
| 525 | ||
| 526 | ||
| 527 | /** | |
| 528 | * Wrap the given text to comply with RFC 2646 | |
| 529 | * | |
| 530 | * @param string $text Text to wrap | |
| 531 | * @param int $length Length | |
| c72a96 | 532 | * @param string $charset Character encoding of $text |
| 1c4f23 | 533 | * |
| A | 534 | * @return string Wrapped text |
| 535 | */ | |
| c72a96 | 536 | public static function format_flowed($text, $length = 72, $charset=null) |
| 1c4f23 | 537 | { |
| A | 538 | $text = preg_split('/\r?\n/', $text); |
| 539 | ||
| 540 | foreach ($text as $idx => $line) { | |
| 541 | if ($line != '-- ') { | |
| 42b8a6 | 542 | if ($line[0] == '>' && preg_match('/^(>+ {0,1})+/', $line, $regs)) { |
| AM | 543 | $level = substr_count($regs[0], '>'); |
| 544 | $prefix = str_repeat('>', $level) . ' '; | |
| 545 | $line = rtrim(substr($line, strlen($regs[0]))); | |
| 546 | $line = $prefix . self::wordwrap($line, $length - $level - 2, " \r\n$prefix", false, $charset); | |
| 1c4f23 | 547 | } |
| A | 548 | else if ($line) { |
| c72a96 | 549 | $line = self::wordwrap(rtrim($line), $length - 2, " \r\n", false, $charset); |
| 1c4f23 | 550 | // space-stuffing |
| A | 551 | $line = preg_replace('/(^|\r\n)(From| |>)/', '\\1 \\2', $line); |
| 552 | } | |
| 553 | ||
| 554 | $text[$idx] = $line; | |
| 555 | } | |
| 556 | } | |
| 557 | ||
| 558 | return implode("\r\n", $text); | |
| 559 | } | |
| 560 | ||
| b6a182 | 561 | |
| A | 562 | /** |
| 563 | * Improved wordwrap function. | |
| 564 | * | |
| 565 | * @param string $string Text to wrap | |
| 566 | * @param int $width Line width | |
| 567 | * @param string $break Line separator | |
| 568 | * @param bool $cut Enable to cut word | |
| c72a96 | 569 | * @param string $charset Charset of $string |
| b6a182 | 570 | * |
| A | 571 | * @return string Text |
| 572 | */ | |
| c72a96 | 573 | public static function wordwrap($string, $width=75, $break="\n", $cut=false, $charset=null) |
| b6a182 | 574 | { |
| a92beb | 575 | if ($charset && function_exists('mb_internal_encoding')) { |
| c72a96 | 576 | mb_internal_encoding($charset); |
| a92beb | 577 | } |
| c72a96 | 578 | |
| TB | 579 | $para = preg_split('/\r?\n/', $string); |
| b6a182 | 580 | $string = ''; |
| A | 581 | |
| 582 | while (count($para)) { | |
| 583 | $line = array_shift($para); | |
| 584 | if ($line[0] == '>') { | |
| 585 | $string .= $line.$break; | |
| 586 | continue; | |
| 587 | } | |
| 588 | ||
| 589 | $list = explode(' ', $line); | |
| 590 | $len = 0; | |
| 591 | while (count($list)) { | |
| 592 | $line = array_shift($list); | |
| 593 | $l = mb_strlen($line); | |
| 594 | $newlen = $len + $l + ($len ? 1 : 0); | |
| 595 | ||
| 596 | if ($newlen <= $width) { | |
| 597 | $string .= ($len ? ' ' : '').$line; | |
| 598 | $len += (1 + $l); | |
| 599 | } | |
| 600 | else { | |
| 601 | if ($l > $width) { | |
| 602 | if ($cut) { | |
| 603 | $start = 0; | |
| 604 | while ($l) { | |
| 605 | $str = mb_substr($line, $start, $width); | |
| 606 | $strlen = mb_strlen($str); | |
| 607 | $string .= ($len ? $break : '').$str; | |
| 608 | $start += $strlen; | |
| 609 | $l -= $strlen; | |
| 610 | $len = $strlen; | |
| 611 | } | |
| 612 | } | |
| 613 | else { | |
| 614 | $string .= ($len ? $break : '').$line; | |
| 615 | if (count($list)) { | |
| 616 | $string .= $break; | |
| 617 | } | |
| 618 | $len = 0; | |
| 619 | } | |
| 620 | } | |
| 621 | else { | |
| 622 | $string .= $break.$line; | |
| 623 | $len = $l; | |
| 624 | } | |
| 625 | } | |
| 626 | } | |
| 627 | ||
| 628 | if (count($para)) { | |
| 629 | $string .= $break; | |
| 630 | } | |
| 631 | } | |
| 632 | ||
| a92beb | 633 | if ($charset && function_exists('mb_internal_encoding')) { |
| AM | 634 | mb_internal_encoding(RCUBE_CHARSET); |
| 635 | } | |
| c72a96 | 636 | |
| b6a182 | 637 | return $string; |
| A | 638 | } |
| 639 | ||
| 640 | ||
| 641 | /** | |
| 642 | * A method to guess the mime_type of an attachment. | |
| 643 | * | |
| 0a8397 | 644 | * @param string $path Path to the file or file contents |
| b6a182 | 645 | * @param string $name File name (with suffix) |
| 0a8397 | 646 | * @param string $failover Mime type supplied for failover |
| TB | 647 | * @param boolean $is_stream Set to True if $path contains file contents |
| 648 | * @param boolean $skip_suffix Set to True if the config/mimetypes.php mappig should be ignored | |
| b6a182 | 649 | * |
| A | 650 | * @return string |
| 651 | * @author Till Klampaeckel <till@php.net> | |
| 652 | * @see http://de2.php.net/manual/en/ref.fileinfo.php | |
| 653 | * @see http://de2.php.net/mime_content_type | |
| 654 | */ | |
| 0a8397 | 655 | public static function file_content_type($path, $name, $failover = 'application/octet-stream', $is_stream = false, $skip_suffix = false) |
| b6a182 | 656 | { |
| A | 657 | $mime_type = null; |
| 658 | $mime_magic = rcube::get_instance()->config->get('mime_magic'); | |
| 0a8397 | 659 | $mime_ext = $skip_suffix ? null : @include(RCUBE_CONFIG_DIR . '/mimetypes.php'); |
| b6a182 | 660 | |
| A | 661 | // use file name suffix with hard-coded mime-type map |
| 662 | if (is_array($mime_ext) && $name) { | |
| 663 | if ($suffix = substr($name, strrpos($name, '.')+1)) { | |
| 664 | $mime_type = $mime_ext[strtolower($suffix)]; | |
| 665 | } | |
| 666 | } | |
| 667 | ||
| 668 | // try fileinfo extension if available | |
| 669 | if (!$mime_type && function_exists('finfo_open')) { | |
| 670 | if ($finfo = finfo_open(FILEINFO_MIME, $mime_magic)) { | |
| 671 | if ($is_stream) | |
| 672 | $mime_type = finfo_buffer($finfo, $path); | |
| 673 | else | |
| 674 | $mime_type = finfo_file($finfo, $path); | |
| 675 | finfo_close($finfo); | |
| 676 | } | |
| 677 | } | |
| 678 | ||
| 679 | // try PHP's mime_content_type | |
| 680 | if (!$mime_type && !$is_stream && function_exists('mime_content_type')) { | |
| 681 | $mime_type = @mime_content_type($path); | |
| 682 | } | |
| 683 | ||
| 684 | // fall back to user-submitted string | |
| 685 | if (!$mime_type) { | |
| 686 | $mime_type = $failover; | |
| 687 | } | |
| 688 | else { | |
| 689 | // Sometimes (PHP-5.3?) content-type contains charset definition, | |
| 690 | // Remove it (#1487122) also "charset=binary" is useless | |
| 691 | $mime_type = array_shift(preg_split('/[; ]/', $mime_type)); | |
| 692 | } | |
| 693 | ||
| 694 | return $mime_type; | |
| 695 | } | |
| 696 | ||
| 697 | ||
| 698 | /** | |
| 0a8397 | 699 | * Get mimetype => file extension mapping |
| TB | 700 | * |
| 701 | * @param string Mime-Type to get extensions for | |
| 702 | * @return array List of extensions matching the given mimetype or a hash array with ext -> mimetype mappings if $mimetype is not given | |
| 703 | */ | |
| 704 | public static function get_mime_extensions($mimetype = null) | |
| 705 | { | |
| 706 | static $mime_types, $mime_extensions; | |
| 707 | ||
| 708 | // return cached data | |
| 709 | if (is_array($mime_types)) { | |
| 710 | return $mimetype ? $mime_types[$mimetype] : $mime_extensions; | |
| 711 | } | |
| 712 | ||
| 713 | // load mapping file | |
| 714 | $file_paths = array(); | |
| 715 | ||
| 716 | if ($mime_types = rcube::get_instance()->config->get('mime_types')) | |
| 717 | $file_paths[] = $mime_types; | |
| 718 | ||
| 719 | // try common locations | |
| 720 | $file_paths[] = '/etc/httpd/mime.types'; | |
| 721 | $file_paths[] = '/etc/httpd2/mime.types'; | |
| 722 | $file_paths[] = '/etc/apache/mime.types'; | |
| 723 | $file_paths[] = '/etc/apache2/mime.types'; | |
| 724 | $file_paths[] = '/usr/local/etc/httpd/conf/mime.types'; | |
| 725 | $file_paths[] = '/usr/local/etc/apache/conf/mime.types'; | |
| 726 | ||
| 727 | foreach ($file_paths as $fp) { | |
| 728 | if (is_readable($fp)) { | |
| 729 | $lines = file($fp, FILE_IGNORE_NEW_LINES); | |
| 730 | break; | |
| 731 | } | |
| 732 | } | |
| 733 | ||
| 734 | $mime_types = $mime_extensions = array(); | |
| 735 | $regex = "/([\w\+\-\.\/]+)\t+([\w\s]+)/i"; | |
| 736 | foreach((array)$lines as $line) { | |
| 737 | // skip comments or mime types w/o any extensions | |
| 738 | if ($line[0] == '#' || !preg_match($regex, $line, $matches)) | |
| 739 | continue; | |
| 740 | ||
| 741 | $mime = $matches[1]; | |
| 742 | foreach (explode(' ', $matches[2]) as $ext) { | |
| 743 | $ext = trim($ext); | |
| 744 | $mime_types[$mime][] = $ext; | |
| 745 | $mime_extensions[$ext] = $mime; | |
| 746 | } | |
| 747 | } | |
| 748 | ||
| 749 | // fallback to some well-known types most important for daily emails | |
| 750 | if (empty($mime_types)) { | |
| 751 | $mime_extensions = @include(RCUBE_CONFIG_DIR . '/mimetypes.php'); | |
| 752 | $mime_extensions += array('gif' => 'image/gif', 'png' => 'image/png', 'jpg' => 'image/jpg', 'jpeg' => 'image/jpeg', 'tif' => 'image/tiff'); | |
| 753 | ||
| 754 | foreach ($mime_extensions as $ext => $mime) | |
| 755 | $mime_types[$mime][] = $ext; | |
| 756 | } | |
| 757 | ||
| 758 | return $mimetype ? $mime_types[$mimetype] : $mime_extensions; | |
| 759 | } | |
| 760 | ||
| 761 | ||
| 762 | /** | |
| b6a182 | 763 | * Detect image type of the given binary data by checking magic numbers. |
| A | 764 | * |
| 765 | * @param string $data Binary file content | |
| 766 | * | |
| 767 | * @return string Detected mime-type or jpeg as fallback | |
| 768 | */ | |
| 769 | public static function image_content_type($data) | |
| 770 | { | |
| 771 | $type = 'jpeg'; | |
| 772 | if (preg_match('/^\x89\x50\x4E\x47/', $data)) $type = 'png'; | |
| 773 | else if (preg_match('/^\x47\x49\x46\x38/', $data)) $type = 'gif'; | |
| 774 | else if (preg_match('/^\x00\x00\x01\x00/', $data)) $type = 'ico'; | |
| 775 | // else if (preg_match('/^\xFF\xD8\xFF\xE0/', $data)) $type = 'jpeg'; | |
| 776 | ||
| 777 | return 'image/' . $type; | |
| 778 | } | |
| 779 | ||
| 1c4f23 | 780 | } |
