From ba6f21caeb405c7e8512a09941fefbc97286e45f Mon Sep 17 00:00:00 2001 From: Aleksander Machniak Date: Wed, 21 Nov 2012 19:52:03 +0100 Subject: Framework files moved to lib/Roundcube --- program/lib/Roundcube/rcube_mime.php | 713 +++++++++++++++++++++++++++++++++++ 1 file changed, 713 insertions(+) create mode 100644 program/lib/Roundcube/rcube_mime.php (limited to 'program/lib/Roundcube/rcube_mime.php') diff --git a/program/lib/Roundcube/rcube_mime.php b/program/lib/Roundcube/rcube_mime.php new file mode 100644 index 000000000..7ce7c90fe --- /dev/null +++ b/program/lib/Roundcube/rcube_mime.php @@ -0,0 +1,713 @@ + | + | Author: Aleksander Machniak | + +-----------------------------------------------------------------------+ +*/ + + +/** + * Class for parsing MIME messages + * + * @package Framework + * @subpackage Storage + * @author Thomas Bruederli + * @author Aleksander Machniak + */ +class rcube_mime +{ + private static $default_charset; + + + /** + * Object constructor. + */ + function __construct($default_charset = null) + { + self::$default_charset = $default_charset; + } + + + /** + * Returns message/object character set name + * + * @return string Characted set name + */ + public static function get_charset() + { + if (self::$default_charset) { + return self::$default_charset; + } + + if ($charset = rcube::get_instance()->config->get('default_charset')) { + return $charset; + } + + return RCMAIL_CHARSET; + } + + + /** + * Parse the given raw message source and return a structure + * of rcube_message_part objects. + * + * It makes use of the PEAR:Mail_mimeDecode library + * + * @param string The message source + * @return object rcube_message_part The message structure + */ + public static function parse_message($raw_body) + { + $mime = new Mail_mimeDecode($raw_body); + $struct = $mime->decode(array('include_bodies' => true, 'decode_bodies' => true)); + return self::structure_part($struct); + } + + + /** + * Recursive method to convert a Mail_mimeDecode part into a rcube_message_part object + * + * @param object A message part struct + * @param int Part count + * @param string Parent MIME ID + * + * @return object rcube_message_part + */ + private static function structure_part($part, $count=0, $parent='') + { + $struct = new rcube_message_part; + $struct->mime_id = $part->mime_id ? $part->mime_id : (empty($parent) ? (string)$count : "$parent.$count"); + $struct->headers = $part->headers; + $struct->ctype_primary = $part->ctype_primary; + $struct->ctype_secondary = $part->ctype_secondary; + $struct->mimetype = $part->ctype_primary . '/' . $part->ctype_secondary; + $struct->ctype_parameters = $part->ctype_parameters; + + if ($part->headers['content-transfer-encoding']) + $struct->encoding = $part->headers['content-transfer-encoding']; + if ($part->ctype_parameters['charset']) + $struct->charset = $part->ctype_parameters['charset']; + + $part_charset = $struct->charset ? $struct->charset : self::get_charset(); + + // determine filename + if (($filename = $part->d_parameters['filename']) || ($filename = $part->ctype_parameters['name'])) { + $struct->filename = rcube_mime::decode_mime_string($filename, $part_charset); + } + + // copy part body and convert it to UTF-8 if necessary + $struct->body = $part->ctype_primary == 'text' || !$part->ctype_parameters['charset'] ? rcube_charset::convert($part->body, $part_charset) : $part->body; + $struct->size = strlen($part->body); + $struct->disposition = $part->disposition; + + foreach ((array)$part->parts as $child_part) { + $struct->parts[] = self::structure_part($child_part, ++$count, $struct->mime_id); + } + + return $struct; + } + + + /** + * Split an address list into a structured array list + * + * @param string $input Input string + * @param int $max List only this number of addresses + * @param boolean $decode Decode address strings + * @param string $fallback Fallback charset if none specified + * + * @return array Indexed list of addresses + */ + static function decode_address_list($input, $max = null, $decode = true, $fallback = null) + { + $a = self::parse_address_list($input, $decode, $fallback); + $out = array(); + $j = 0; + + // Special chars as defined by RFC 822 need to in quoted string (or escaped). + $special_chars = '[\(\)\<\>\\\.\[\]@,;:"]'; + + if (!is_array($a)) + return $out; + + foreach ($a as $val) { + $j++; + $address = trim($val['address']); + $name = trim($val['name']); + + if ($name && $address && $name != $address) + $string = sprintf('%s <%s>', preg_match("/$special_chars/", $name) ? '"'.addcslashes($name, '"').'"' : $name, $address); + else if ($address) + $string = $address; + else if ($name) + $string = $name; + + $out[$j] = array( + 'name' => $name, + 'mailto' => $address, + 'string' => $string + ); + + if ($max && $j==$max) + break; + } + + return $out; + } + + + /** + * Decode a message header value + * + * @param string $input Header value + * @param string $fallback Fallback charset if none specified + * + * @return string Decoded string + */ + public static function decode_header($input, $fallback = null) + { + $str = self::decode_mime_string((string)$input, $fallback); + + return $str; + } + + + /** + * Decode a mime-encoded string to internal charset + * + * @param string $input Header value + * @param string $fallback Fallback charset if none specified + * + * @return string Decoded string + */ + public static function decode_mime_string($input, $fallback = null) + { + $default_charset = !empty($fallback) ? $fallback : self::get_charset(); + + // rfc: all line breaks or other characters not found + // in the Base64 Alphabet must be ignored by decoding software + // delete all blanks between MIME-lines, differently we can + // receive unnecessary blanks and broken utf-8 symbols + $input = preg_replace("/\?=\s+=\?/", '?==?', $input); + + // encoded-word regexp + $re = '/=\?([^?]+)\?([BbQq])\?([^\n]*?)\?=/'; + + // Find all RFC2047's encoded words + if (preg_match_all($re, $input, $matches, PREG_OFFSET_CAPTURE | PREG_SET_ORDER)) { + // Initialize variables + $tmp = array(); + $out = ''; + $start = 0; + + foreach ($matches as $idx => $m) { + $pos = $m[0][1]; + $charset = $m[1][0]; + $encoding = $m[2][0]; + $text = $m[3][0]; + $length = strlen($m[0][0]); + + // Append everything that is before the text to be decoded + if ($start != $pos) { + $substr = substr($input, $start, $pos-$start); + $out .= rcube_charset::convert($substr, $default_charset); + $start = $pos; + } + $start += $length; + + // Per RFC2047, each string part "MUST represent an integral number + // of characters . A multi-octet character may not be split across + // adjacent encoded-words." However, some mailers break this, so we + // try to handle characters spanned across parts anyway by iterating + // through and aggregating sequential encoded parts with the same + // character set and encoding, then perform the decoding on the + // aggregation as a whole. + + $tmp[] = $text; + if ($next_match = $matches[$idx+1]) { + if ($next_match[0][1] == $start + && $next_match[1][0] == $charset + && $next_match[2][0] == $encoding + ) { + continue; + } + } + + $count = count($tmp); + $text = ''; + + // Decode and join encoded-word's chunks + if ($encoding == 'B' || $encoding == 'b') { + // base64 must be decoded a segment at a time + for ($i=0; $i<$count; $i++) + $text .= base64_decode($tmp[$i]); + } + else { //if ($encoding == 'Q' || $encoding == 'q') { + // quoted printable can be combined and processed at once + for ($i=0; $i<$count; $i++) + $text .= $tmp[$i]; + + $text = str_replace('_', ' ', $text); + $text = quoted_printable_decode($text); + } + + $out .= rcube_charset::convert($text, $charset); + $tmp = array(); + } + + // add the last part of the input string + if ($start != strlen($input)) { + $out .= rcube_charset::convert(substr($input, $start), $default_charset); + } + + // return the results + return $out; + } + + // no encoding information, use fallback + return rcube_charset::convert($input, $default_charset); + } + + + /** + * Decode a mime part + * + * @param string $input Input string + * @param string $encoding Part encoding + * @return string Decoded string + */ + public static function decode($input, $encoding = '7bit') + { + switch (strtolower($encoding)) { + case 'quoted-printable': + return quoted_printable_decode($input); + case 'base64': + return base64_decode($input); + case 'x-uuencode': + case 'x-uue': + case 'uue': + case 'uuencode': + return convert_uudecode($input); + case '7bit': + default: + return $input; + } + } + + + /** + * Split RFC822 header string into an associative array + * @access private + */ + public static function parse_headers($headers) + { + $a_headers = array(); + $headers = preg_replace('/\r?\n(\t| )+/', ' ', $headers); + $lines = explode("\n", $headers); + $c = count($lines); + + for ($i=0; $i<$c; $i++) { + if ($p = strpos($lines[$i], ': ')) { + $field = strtolower(substr($lines[$i], 0, $p)); + $value = trim(substr($lines[$i], $p+1)); + if (!empty($value)) + $a_headers[$field] = $value; + } + } + + return $a_headers; + } + + + /** + * @access private + */ + private static function parse_address_list($str, $decode = true, $fallback = null) + { + // remove any newlines and carriage returns before + $str = preg_replace('/\r?\n(\s|\t)?/', ' ', $str); + + // extract list items, remove comments + $str = self::explode_header_string(',;', $str, true); + $result = array(); + + // simplified regexp, supporting quoted local part + $email_rx = '(\S+|("\s*(?:[^"\f\n\r\t\v\b\s]+\s*)+"))@\S+'; + + foreach ($str as $key => $val) { + $name = ''; + $address = ''; + $val = trim($val); + + if (preg_match('/(.*)<('.$email_rx.')>$/', $val, $m)) { + $address = $m[2]; + $name = trim($m[1]); + } + else if (preg_match('/^('.$email_rx.')$/', $val, $m)) { + $address = $m[1]; + $name = ''; + } + else { + $name = $val; + } + + // dequote and/or decode name + if ($name) { + if ($name[0] == '"' && $name[strlen($name)-1] == '"') { + $name = substr($name, 1, -1); + $name = stripslashes($name); + } + if ($decode) { + $name = self::decode_header($name, $fallback); + } + } + + if (!$address && $name) { + $address = $name; + } + + if ($address) { + $result[$key] = array('name' => $name, 'address' => $address); + } + } + + return $result; + } + + + /** + * Explodes header (e.g. address-list) string into array of strings + * using specified separator characters with proper handling + * of quoted-strings and comments (RFC2822) + * + * @param string $separator String containing separator characters + * @param string $str Header string + * @param bool $remove_comments Enable to remove comments + * + * @return array Header items + */ + public static function explode_header_string($separator, $str, $remove_comments = false) + { + $length = strlen($str); + $result = array(); + $quoted = false; + $comment = 0; + $out = ''; + + for ($i=0; $i<$length; $i++) { + // we're inside a quoted string + if ($quoted) { + if ($str[$i] == '"') { + $quoted = false; + } + else if ($str[$i] == "\\") { + if ($comment <= 0) { + $out .= "\\"; + } + $i++; + } + } + // we are inside a comment string + else if ($comment > 0) { + if ($str[$i] == ')') { + $comment--; + } + else if ($str[$i] == '(') { + $comment++; + } + else if ($str[$i] == "\\") { + $i++; + } + continue; + } + // separator, add to result array + else if (strpos($separator, $str[$i]) !== false) { + if ($out) { + $result[] = $out; + } + $out = ''; + continue; + } + // start of quoted string + else if ($str[$i] == '"') { + $quoted = true; + } + // start of comment + else if ($remove_comments && $str[$i] == '(') { + $comment++; + } + + if ($comment <= 0) { + $out .= $str[$i]; + } + } + + if ($out && $comment <= 0) { + $result[] = $out; + } + + return $result; + } + + + /** + * Interpret a format=flowed message body according to RFC 2646 + * + * @param string $text Raw body formatted as flowed text + * + * @return string Interpreted text with unwrapped lines and stuffed space removed + */ + public static function unfold_flowed($text) + { + $text = preg_split('/\r?\n/', $text); + $last = -1; + $q_level = 0; + + foreach ($text as $idx => $line) { + if ($line[0] == '>' && preg_match('/^(>+\s*)/', $line, $regs)) { + $q = strlen(str_replace(' ', '', $regs[0])); + $line = substr($line, strlen($regs[0])); + + if ($q == $q_level && $line + && isset($text[$last]) + && $text[$last][strlen($text[$last])-1] == ' ' + ) { + $text[$last] .= $line; + unset($text[$idx]); + } + else { + $last = $idx; + } + } + else { + $q = 0; + if ($line == '-- ') { + $last = $idx; + } + else { + // remove space-stuffing + $line = preg_replace('/^\s/', '', $line); + + if (isset($text[$last]) && $line + && $text[$last] != '-- ' + && $text[$last][strlen($text[$last])-1] == ' ' + ) { + $text[$last] .= $line; + unset($text[$idx]); + } + else { + $text[$idx] = $line; + $last = $idx; + } + } + } + $q_level = $q; + } + + return implode("\r\n", $text); + } + + + /** + * Wrap the given text to comply with RFC 2646 + * + * @param string $text Text to wrap + * @param int $length Length + * @param string $charset Character encoding of $text + * + * @return string Wrapped text + */ + public static function format_flowed($text, $length = 72, $charset=null) + { + $text = preg_split('/\r?\n/', $text); + + foreach ($text as $idx => $line) { + if ($line != '-- ') { + if ($line[0] == '>' && preg_match('/^(>+ {0,1})+/', $line, $regs)) { + $level = substr_count($regs[0], '>'); + $prefix = str_repeat('>', $level) . ' '; + $line = rtrim(substr($line, strlen($regs[0]))); + $line = $prefix . self::wordwrap($line, $length - $level - 2, " \r\n$prefix", false, $charset); + } + else if ($line) { + $line = self::wordwrap(rtrim($line), $length - 2, " \r\n", false, $charset); + // space-stuffing + $line = preg_replace('/(^|\r\n)(From| |>)/', '\\1 \\2', $line); + } + + $text[$idx] = $line; + } + } + + return implode("\r\n", $text); + } + + + /** + * Improved wordwrap function. + * + * @param string $string Text to wrap + * @param int $width Line width + * @param string $break Line separator + * @param bool $cut Enable to cut word + * @param string $charset Charset of $string + * + * @return string Text + */ + public static function wordwrap($string, $width=75, $break="\n", $cut=false, $charset=null) + { + if ($charset && function_exists('mb_internal_encoding')) + mb_internal_encoding($charset); + + $para = preg_split('/\r?\n/', $string); + $string = ''; + + while (count($para)) { + $line = array_shift($para); + if ($line[0] == '>') { + $string .= $line.$break; + continue; + } + + $list = explode(' ', $line); + $len = 0; + while (count($list)) { + $line = array_shift($list); + $l = mb_strlen($line); + $newlen = $len + $l + ($len ? 1 : 0); + + if ($newlen <= $width) { + $string .= ($len ? ' ' : '').$line; + $len += (1 + $l); + } + else { + if ($l > $width) { + if ($cut) { + $start = 0; + while ($l) { + $str = mb_substr($line, $start, $width); + $strlen = mb_strlen($str); + $string .= ($len ? $break : '').$str; + $start += $strlen; + $l -= $strlen; + $len = $strlen; + } + } + else { + $string .= ($len ? $break : '').$line; + if (count($list)) { + $string .= $break; + } + $len = 0; + } + } + else { + $string .= $break.$line; + $len = $l; + } + } + } + + if (count($para)) { + $string .= $break; + } + } + + if ($charset && function_exists('mb_internal_encoding')) + mb_internal_encoding(RCMAIL_CHARSET); + + return $string; + } + + + /** + * A method to guess the mime_type of an attachment. + * + * @param string $path Path to the file. + * @param string $name File name (with suffix) + * @param string $failover Mime type supplied for failover. + * @param string $is_stream Set to True if $path contains file body + * + * @return string + * @author Till Klampaeckel + * @see http://de2.php.net/manual/en/ref.fileinfo.php + * @see http://de2.php.net/mime_content_type + */ + public static function file_content_type($path, $name, $failover = 'application/octet-stream', $is_stream = false) + { + $mime_type = null; + $mime_magic = rcube::get_instance()->config->get('mime_magic'); + $mime_ext = @include RCMAIL_CONFIG_DIR . '/mimetypes.php'; + + // use file name suffix with hard-coded mime-type map + if (is_array($mime_ext) && $name) { + if ($suffix = substr($name, strrpos($name, '.')+1)) { + $mime_type = $mime_ext[strtolower($suffix)]; + } + } + + // try fileinfo extension if available + if (!$mime_type && function_exists('finfo_open')) { + if ($finfo = finfo_open(FILEINFO_MIME, $mime_magic)) { + if ($is_stream) + $mime_type = finfo_buffer($finfo, $path); + else + $mime_type = finfo_file($finfo, $path); + finfo_close($finfo); + } + } + + // try PHP's mime_content_type + if (!$mime_type && !$is_stream && function_exists('mime_content_type')) { + $mime_type = @mime_content_type($path); + } + + // fall back to user-submitted string + if (!$mime_type) { + $mime_type = $failover; + } + else { + // Sometimes (PHP-5.3?) content-type contains charset definition, + // Remove it (#1487122) also "charset=binary" is useless + $mime_type = array_shift(preg_split('/[; ]/', $mime_type)); + } + + return $mime_type; + } + + + /** + * Detect image type of the given binary data by checking magic numbers. + * + * @param string $data Binary file content + * + * @return string Detected mime-type or jpeg as fallback + */ + public static function image_content_type($data) + { + $type = 'jpeg'; + if (preg_match('/^\x89\x50\x4E\x47/', $data)) $type = 'png'; + else if (preg_match('/^\x47\x49\x46\x38/', $data)) $type = 'gif'; + else if (preg_match('/^\x00\x00\x01\x00/', $data)) $type = 'ico'; + // else if (preg_match('/^\xFF\xD8\xFF\xE0/', $data)) $type = 'jpeg'; + + return 'image/' . $type; + } + +} -- cgit v1.2.3