mimedecode.php

This is the script which used on 10minutemail.com for temporary email.

        if (isset($content_type)) {
            switch (strtolower($content_type['value'])) {
                case 'text/plain':
                    $encoding = isset($content_transfer_encoding) ? $content_transfer_encoding['value'] : '7bit';
                    $this->_include_bodies ? $return->body = ($this->_decode_bodies ? $this->_decodeBody($body, $encoding) : $body) : null;
                    break;

                case 'text/html':
                    $encoding = isset($content_transfer_encoding) ? $content_transfer_encoding['value'] : '7bit';
                    $this->_include_bodies ? $return->body = ($this->_decode_bodies ? $this->_decodeBody($body, $encoding) : $body) : null;
                    break;

                case 'multipart/parallel':
                case 'multipart/appledouble': // Appledouble mail
                case 'multipart/report': // RFC1892
                case 'multipart/signed': // PGP
                case 'multipart/digest':
                case 'multipart/alternative':
                case 'multipart/related':
                case 'multipart/mixed':
                    if(!isset($content_type['other']['boundary'])){
                        $this->_error = 'No boundary found for ' . $content_type['value'] . ' part';
                        return false;
                    }

                    $default_ctype = (strtolower($content_type['value']) === 'multipart/digest') ? 'message/rfc822' : 'text/plain';

                    $parts = $this->_boundarySplit($body, $content_type['other']['boundary']);
                    for ($i = 0; $i < count($parts); $i++) {
                        list($part_header, $part_body) = $this->_splitBodyHeader($parts[$i]);
                        $part = $this->_decode($part_header, $part_body, $default_ctype);
                        if($part === false)
                            $part = $this->raiseError($this->_error);
                        $return->parts[] = $part;
                    }
                    break;

                case 'message/rfc822':
                    $obj = &new Mail_mimeDecode($body);
                    $return->parts[] = $obj->decode(array('include_bodies' => $this->_include_bodies,
					                                      'decode_bodies'  => $this->_decode_bodies,
														  'decode_headers' => $this->_decode_headers));
                    unset($obj);
                    break;

                default:
                    if(!isset($content_transfer_encoding['value']))
                        $content_transfer_encoding['value'] = '7bit';
                    $this->_include_bodies ? $return->body = ($this->_decode_bodies ? $this->_decodeBody($body, $content_transfer_encoding['value']) : $body) : null;
                    break;
            }

        } else {
            $ctype = explode('/', $default_ctype);
            $return->ctype_primary   = $ctype[0];
            $return->ctype_secondary = $ctype[1];
            $this->_include_bodies ? $return->body = ($this->_decode_bodies ? $this->_decodeBody($body) : $body) : null;
        }

        return $return;
    }

    /**
     * Given the output of the above function, this will return an
     * array of references to the parts, indexed by mime number.
     *
     * @param  object $structure   The structure to go through
     * @param  string $mime_number Internal use only.
     * @return array               Mime numbers
     */
    function &getMimeNumbers(&$structure, $no_refs = false, $mime_number = '', $prepend = '')
    {
        $return = array();
        if (!empty($structure->parts)) {
            if ($mime_number != '') {
                $structure->mime_id = $prepend . $mime_number;
                $return[$prepend . $mime_number] = &$structure;
            }
            for ($i = 0; $i < count($structure->parts); $i++) {

            
                if (!empty($structure->headers['content-type']) AND substr(strtolower($structure->headers['content-type']), 0, 8) == 'message/') {
                    $prepend      = $prepend . $mime_number . '.';
                    $_mime_number = '';
                } else {
                    $_mime_number = ($mime_number == '' ? $i + 1 : sprintf('%s.%s', $mime_number, $i + 1));
                }

                $arr = &Mail_mimeDecode::getMimeNumbers($structure->parts[$i], $no_refs, $_mime_number, $prepend);
                foreach ($arr as $key => $val) {
                    $no_refs ? $return[$key] = '' : $return[$key] = &$arr[$key];
                }
            }
        } else {
            if ($mime_number == '') {
                $mime_number = '1';
            }
            $structure->mime_id = $prepend . $mime_number;
            $no_refs ? $return[$prepend . $mime_number] = '' : $return[$prepend . $mime_number] = &$structure;
        }
        
        return $return;
    }

    /**
     * Given a string containing a header and body
     * section, this function will split them (at the first
     * blank line) and return them.
     *
     * @param string Input to split apart
     * @return array Contains header and body section
     * @access private
     */
    function _splitBodyHeader($input)
    {
        if (preg_match("/^(.*?)\r?\n\r?\n(.*)/s", $input, $match)) {
            return array($match[1], $match[2]);
        }
        $this->_error = 'Could not split header and body';
        return false;
    }

    /**
     * Parse headers given in $input and return
     * as assoc array.
     *
     * @param string Headers to parse
     * @return array Contains parsed headers
     * @access private
     */
    function _parseHeaders($input)
    {

        if ($input !== '') {
            // Unfold the input
            $input   = preg_replace("/\r?\n/", "\r\n", $input);
            $input   = preg_replace("/\r\n(\t| )+/", ' ', $input);
            $headers = explode("\r\n", trim($input));

            foreach ($headers as $value) {
                $hdr_name = substr($value, 0, $pos = strpos($value, ':'));
                $hdr_value = substr($value, $pos+1);
                if($hdr_value[0] == ' ')
                    $hdr_value = substr($hdr_value, 1);

                $return[] = array(
                                  'name'  => $hdr_name,
                                  'value' => $this->_decode_headers ? $this->_decodeHeader($hdr_value) : $hdr_value
                                 );
            }
        } else {
            $return = array();
        }

        return $return;
    }

    /**
     * Function to parse a header value,
     * extract first part, and any secondary
     * parts (after ;) This function is not as
     * robust as it could be. Eg. header comments
     * in the wrong place will probably break it.
     *
     * @param string Header value to parse
     * @return array Contains parsed result
     * @access private
     */
    function _parseHeaderValue($input)
    {

        if (($pos = strpos($input, ';')) !== false) {

            $return['value'] = trim(substr($input, 0, $pos));
            $input = trim(substr($input, $pos+1));

            if (strlen($input) > 0) {

                // This splits on a semi-colon, if there's no preceeding backslash
                // Now works with quoted values; had to glue the \; breaks in PHP
                // the regex is already bordering on incomprehensible
                $splitRegex = '/([^;\'"]*[\'"]([^\'"]*([^\'"]*)*)[\'"][^;\'"]*|([^;]+))(;|$)/';
                preg_match_all($splitRegex, $input, $matches);
                $parameters = array();
                for ($i=0; $i<count($matches[0]); $i++) {
                    $param = $matches[0][$i];
                    while (substr($param, -2) == '\;') {
                        $param .= $matches[0][++$i];
                    }
                    $parameters[] = $param;
                }

                for ($i = 0; $i < count($parameters); $i++) {
                    $param_name  = trim(substr($parameters[$i], 0, $pos = strpos($parameters[$i], '=')), "'\";\t\\ ");
                    $param_value = trim(str_replace('\;', ';', substr($parameters[$i], $pos + 1)), "'\";\t\\ ");
                    if ($param_value[0] == '"') {
                        $param_value = substr($param_value, 1, -1);
                    }
                    $return['other'][$param_name] = $param_value;
                    $return['other'][strtolower($param_name)] = $param_value;
                }
            }
        } else {
            $return['value'] = trim($input);
        }

        return $return;
    }

    /**
     * This function splits the input based
     * on the given boundary
     *
     * @param string Input to parse
     * @return array Contains array of resulting mime parts
     * @access private
     */
    function _boundarySplit($input, $boundary)
    {
        $parts = array();

        $bs_possible = substr($boundary, 2, -2);
        $bs_check = '\"' . $bs_possible . '\"';

        if ($boundary == $bs_check) {
            $boundary = $bs_possible;
        }

        $tmp = explode('--' . $boundary, $input);

        for ($i = 1; $i < count($tmp) - 1; $i++) {
            $parts[] = $tmp[$i];
        }

        return $parts;
    }

    /**
     * Given a header, this function will decode it
     * according to RFC2047. Probably not *exactly*
     * conformant, but it does pass all the given
     * examples (in RFC2047).
     *
     * @param string Input header value to decode
     * @return string Decoded header value
     * @access private
     */
    function _decodeHeader($input)
    {
        // Remove white space between encoded-words
        $input = preg_replace('/(=\?[^?]+\?(q|b)\?[^?]*\?=)(\s)+=\?/i', '\1=?', $input);

        // For each encoded-word...
        while (preg_match('/(=\?([^?]+)\?(q|b)\?([^?]*)\?=)/i', $input, $matches)) {

            $encoded  = $matches[1];
            $charset  = $matches[2];
            $encoding = $matches[3];
            $text     = $matches[4];

            switch (strtolower($encoding)) {
                case 'b':
                    $text = base64_decode($text);
                    break;

                case 'q':
                    $text = str_replace('_', ' ', $text);
                    preg_match_all('/=([a-f0-9]{2})/i', $text, $matches);
                    foreach($matches[1] as $value)
                        $text = str_replace('='.$value, chr(hexdec($value)), $text);
                    break;
            }

            $input = str_replace($encoded, $text, $input);
        }

        return $input;
    }

    /**
     * Given a body string and an encoding type,
     * this function will decode and return it.
     *