From a621a9d7ecf334c4894ef8f5168eb6208e5ae0e4 Mon Sep 17 00:00:00 2001 From: thomascube <thomas@roundcube.net> Date: Wed, 14 Mar 2012 17:33:05 -0400 Subject: [PATCH] Accept DateTime object as input to format_date() --- program/lib/Mail/mimePart.php | 340 +++++++++++++++++++++++++++++++++++-------------------- 1 files changed, 215 insertions(+), 125 deletions(-) diff --git a/program/lib/Mail/mimePart.php b/program/lib/Mail/mimePart.php index 2181170..a1cf111 100644 --- a/program/lib/Mail/mimePart.php +++ b/program/lib/Mail/mimePart.php @@ -131,6 +131,7 @@ */ var $_eol = "\r\n"; + /** * Constructor. * @@ -141,17 +142,19 @@ * content_type - The content type for this part eg multipart/mixed * encoding - The encoding to use, 7bit, 8bit, * base64, or quoted-printable + * charset - Content character set * cid - Content ID to apply * disposition - Content disposition, inline or attachment - * dfilename - Filename parameter for content disposition + * filename - Filename parameter for content disposition * description - Content description - * charset - Character set to use - * name_encoding - Encoding for attachment name (Content-Type) + * name_encoding - Encoding of the attachment name (Content-Type) * By default filenames are encoded using RFC2231 * Here you can set RFC2047 encoding (quoted-printable * or base64) instead - * filename_encoding - Encoding for attachment filename (Content-Disposition) + * filename_encoding - Encoding of the attachment filename (Content-Disposition) * See 'name_encoding' + * headers_charset - Charset of the headers e.g. filename, description. + * If not set, 'charset' will be used * eol - End of line sequence. Default: "\r\n" * body_file - Location of file with part's body (instead of $body) * @@ -165,14 +168,8 @@ $this->_eol = MAIL_MIMEPART_CRLF; } - $c_type = array(); - $c_disp = array(); foreach ($params as $key => $value) { switch ($key) { - case 'content_type': - $c_type['type'] = $value; - break; - case 'encoding': $this->_encoding = $value; $headers['Content-Transfer-Encoding'] = $value; @@ -182,29 +179,6 @@ $headers['Content-ID'] = '<' . $value . '>'; break; - case 'disposition': - $c_disp['disp'] = $value; - break; - - case 'dfilename': - $c_disp['filename'] = $value; - $c_type['name'] = $value; - break; - - case 'description': - $headers['Content-Description'] = $value; - break; - - case 'charset': - $c_type['charset'] = $value; - $c_disp['charset'] = $value; - break; - - case 'language': - $c_type['language'] = $value; - $c_disp['language'] = $value; - break; - case 'location': $headers['Content-Location'] = $value; break; @@ -212,53 +186,67 @@ case 'body_file': $this->_body_file = $value; break; - } - } - // Content-Type - if (isset($c_type['type'])) { - $headers['Content-Type'] = $c_type['type']; - if (isset($c_type['name'])) { - $headers['Content-Type'] .= ';' . $this->_eol; - $headers['Content-Type'] .= $this->_buildHeaderParam( - 'name', $c_type['name'], - isset($c_type['charset']) ? $c_type['charset'] : 'US-ASCII', - isset($c_type['language']) ? $c_type['language'] : null, - isset($params['name_encoding']) ? $params['name_encoding'] : null - ); + // for backward compatibility + case 'dfilename': + $params['filename'] = $value; + break; } - if (isset($c_type['charset'])) { - $headers['Content-Type'] - .= ';' . $this->_eol . " charset={$c_type['charset']}"; - } - } - - // Content-Disposition - if (isset($c_disp['disp'])) { - $headers['Content-Disposition'] = $c_disp['disp']; - if (isset($c_disp['filename'])) { - $headers['Content-Disposition'] .= ';' . $this->_eol; - $headers['Content-Disposition'] .= $this->_buildHeaderParam( - 'filename', $c_disp['filename'], - isset($c_disp['charset']) ? $c_disp['charset'] : 'US-ASCII', - isset($c_disp['language']) ? $c_disp['language'] : null, - isset($params['filename_encoding']) ? $params['filename_encoding'] : null - ); - } - } - - if (!empty($headers['Content-Description'])) { - $headers['Content-Description'] = $this->encodeHeader( - 'Content-Description', $headers['Content-Description'], - isset($c_type['charset']) ? $c_type['charset'] : 'US-ASCII', - isset($params['name_encoding']) ? $params['name_encoding'] : 'quoted-printable', - $this->_eol - ); } // Default content-type - if (!isset($headers['Content-Type'])) { - $headers['Content-Type'] = 'text/plain'; + if (empty($params['content_type'])) { + $params['content_type'] = 'text/plain'; + } + + // Content-Type + $headers['Content-Type'] = $params['content_type']; + if (!empty($params['charset'])) { + $charset = "charset={$params['charset']}"; + // place charset parameter in the same line, if possible + if ((strlen($headers['Content-Type']) + strlen($charset) + 16) <= 76) { + $headers['Content-Type'] .= '; '; + } else { + $headers['Content-Type'] .= ';' . $this->_eol . ' '; + } + $headers['Content-Type'] .= $charset; + + // Default headers charset + if (!isset($params['headers_charset'])) { + $params['headers_charset'] = $params['charset']; + } + } + if (!empty($params['filename'])) { + $headers['Content-Type'] .= ';' . $this->_eol; + $headers['Content-Type'] .= $this->_buildHeaderParam( + 'name', $params['filename'], + !empty($params['headers_charset']) ? $params['headers_charset'] : 'US-ASCII', + !empty($params['language']) ? $params['language'] : null, + !empty($params['name_encoding']) ? $params['name_encoding'] : null + ); + } + + // Content-Disposition + if (!empty($params['disposition'])) { + $headers['Content-Disposition'] = $params['disposition']; + if (!empty($params['filename'])) { + $headers['Content-Disposition'] .= ';' . $this->_eol; + $headers['Content-Disposition'] .= $this->_buildHeaderParam( + 'filename', $params['filename'], + !empty($params['headers_charset']) ? $params['headers_charset'] : 'US-ASCII', + !empty($params['language']) ? $params['language'] : null, + !empty($params['filename_encoding']) ? $params['filename_encoding'] : null + ); + } + } + + if (!empty($params['description'])) { + $headers['Content-Description'] = $this->encodeHeader( + 'Content-Description', $params['description'], + !empty($params['headers_charset']) ? $params['headers_charset'] : 'US-ASCII', + !empty($params['name_encoding']) ? $params['name_encoding'] : 'quoted-printable', + $this->_eol + ); } // Default encoding @@ -651,8 +639,8 @@ // RFC 2045: // value needs encoding if contains non-ASCII chars or is longer than 78 chars if (!preg_match('#[^\x20-\x7E]#', $value)) { - $token_regexp = '#([^\x21,\x23-\x27,\x2A,\x2B,\x2D' - . ',\x2E,\x30-\x39,\x41-\x5A,\x5E-\x7E])#'; + $token_regexp = '#([^\x21\x23-\x27\x2A\x2B\x2D' + . '\x2E\x30-\x39\x41-\x5A\x5E-\x7E])#'; if (!preg_match($token_regexp, $value)) { // token if (strlen($name) + strlen($value) + 3 <= $maxLength) { @@ -674,7 +662,7 @@ // RFC2231: $encValue = preg_replace_callback( - '/([^\x21,\x23,\x24,\x26,\x2B,\x2D,\x2E,\x30-\x39,\x41-\x5A,\x5E-\x7E])/', + '/([^\x21\x23\x24\x26\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E])/', array($this, '_encodeReplaceCallback'), $value ); $value = "$charset'$language'$encValue"; @@ -818,6 +806,9 @@ // Structured header (make sure addr-spec inside is not encoded) if (!empty($separator)) { + // Simple e-mail address regexp + $email_regexp = '(\S+|("[^\r\n"]+"))@\S+'; + $parts = Mail_mimePart::_explodeQuotedString($separator, $value); $value = ''; @@ -835,12 +826,12 @@ } // let's find phrase (name) and/or addr-spec - if (preg_match('/^<\S+@\S+>$/', $part)) { + if (preg_match('/^<' . $email_regexp . '>$/', $part)) { $value .= $part; - } else if (preg_match('/^\S+@\S+$/', $part)) { + } else if (preg_match('/^' . $email_regexp . '$/', $part)) { // address without brackets and without name $value .= $part; - } else if (preg_match('/<*\S+@\S+>*$/', $part, $matches)) { + } else if (preg_match('/<*' . $email_regexp . '>*$/', $part, $matches)) { // address with name (handle name) $address = $matches[0]; $word = str_replace($address, '', $part); @@ -960,33 +951,39 @@ */ function encodeHeaderValue($value, $charset, $encoding, $prefix_len=0, $eol="\r\n") { - if ($encoding == 'base64') { + // #17311: Use multibyte aware method (requires mbstring extension) + if ($result = Mail_mimePart::encodeMB($value, $charset, $encoding, $prefix_len, $eol)) { + return $result; + } + + // Generate the header using the specified params and dynamicly + // determine the maximum length of such strings. + // 75 is the value specified in the RFC. + $encoding = $encoding == 'base64' ? 'B' : 'Q'; + $prefix = '=?' . $charset . '?' . $encoding .'?'; + $suffix = '?='; + $maxLength = 75 - strlen($prefix . $suffix); + $maxLength1stLine = $maxLength - $prefix_len; + + if ($encoding == 'B') { // Base64 encode the entire string $value = base64_encode($value); - // Generate the header using the specified params and dynamicly - // determine the maximum length of such strings. - // 75 is the value specified in the RFC. - $prefix = '=?' . $charset . '?B?'; - $suffix = '?='; - $maxLength = 75 - strlen($prefix . $suffix) - 2; - $maxLength1stLine = $maxLength - $prefix_len; - - // We can cut base4 every 4 characters, so the real max + // We can cut base64 every 4 characters, so the real max // we can get must be rounded down. $maxLength = $maxLength - ($maxLength % 4); $maxLength1stLine = $maxLength1stLine - ($maxLength1stLine % 4); $cutpoint = $maxLength1stLine; - $value_out = $value; $output = ''; - while ($value_out) { + + while ($value) { // Split translated string at every $maxLength - $part = substr($value_out, 0, $cutpoint); - $value_out = substr($value_out, $cutpoint); + $part = substr($value, 0, $cutpoint); + $value = substr($value, $cutpoint); $cutpoint = $maxLength; // RFC 2047 specifies that any split header should - // be seperated by a CRLF SPACE. + // be seperated by a CRLF SPACE. if ($output) { $output .= $eol . ' '; } @@ -997,30 +994,19 @@ // quoted-printable encoding has been selected $value = Mail_mimePart::encodeQP($value); - // Generate the header using the specified params and dynamicly - // determine the maximum length of such strings. - // 75 is the value specified in the RFC. - $prefix = '=?' . $charset . '?Q?'; - $suffix = '?='; - $maxLength = 75 - strlen($prefix . $suffix) - 3; - $maxLength1stLine = $maxLength - $prefix_len; - $maxLength = $maxLength - 1; - // This regexp will break QP-encoded text at every $maxLength // but will not break any encoded letters. $reg1st = "|(.{0,$maxLength1stLine}[^\=][^\=])|"; $reg2nd = "|(.{0,$maxLength}[^\=][^\=])|"; - $value_out = $value; - $realMax = $maxLength1stLine + strlen($prefix . $suffix); - if (strlen($value_out) >= $realMax) { + if (strlen($value) > $maxLength1stLine) { // Begin with the regexp for the first line. $reg = $reg1st; $output = ''; - while ($value_out) { + while ($value) { // Split translated string at every $maxLength // But make sure not to break any translated chars. - $found = preg_match($reg, $value_out, $matches); + $found = preg_match($reg, $value, $matches); // After this first line, we need to use a different // regexp for the first line. @@ -1032,24 +1018,23 @@ if ($found) { $part = $matches[0]; $len = strlen($matches[0]); - $value_out = substr($value_out, $len); + $value = substr($value, $len); } else { - $part = $value_out; - $value_out = ""; + $part = $value; + $value = ''; } - // RFC 2047 specifies that any split header should + // RFC 2047 specifies that any split header should // be seperated by a CRLF SPACE if ($output) { $output .= $eol . ' '; } $output .= $prefix . $part . $suffix; } - $value_out = $output; + $value = $output; } else { - $value_out = $prefix . $value_out . $suffix; + $value = $prefix . $value . $suffix; } - $value = $value_out; } return $value; @@ -1066,16 +1051,121 @@ */ function encodeQP($str) { - // Replace all special characters used by the encoder - $search = array('=', '_', '?', ' '); - $replace = array('=3D', '=5F', '=3F', '_'); - $str = str_replace($search, $replace, $str); + // Bug #17226 RFC 2047 restricts some characters + // if the word is inside a phrase, permitted chars are only: + // ASCII letters, decimal digits, "!", "*", "+", "-", "/", "=", and "_" - // Replace all extended characters (\x80-xFF) with their - // ASCII values. - return preg_replace_callback( - '/([\x80-\xFF])/', array('Mail_mimePart', '_qpReplaceCallback'), $str + // "=", "_", "?" must be encoded + $regexp = '/([\x22-\x29\x2C\x2E\x3A-\x40\x5B-\x60\x7B-\x7E\x80-\xFF])/'; + $str = preg_replace_callback( + $regexp, array('Mail_mimePart', '_qpReplaceCallback'), $str ); + + return str_replace(' ', '_', $str); + } + + /** + * Encodes the given string using base64 or quoted-printable. + * This method makes sure that encoded-word represents an integral + * number of characters as per RFC2047. + * + * @param string $str String to encode + * @param string $charset Character set name + * @param string $encoding Encoding name (base64 or quoted-printable) + * @param int $prefix_len Prefix length. Default: 0 + * @param string $eol End-of-line sequence. Default: "\r\n" + * + * @return string Encoded string + * @access public + * @since 1.8.0 + */ + function encodeMB($str, $charset, $encoding, $prefix_len=0, $eol="\r\n") + { + if (!function_exists('mb_substr') || !function_exists('mb_strlen')) { + return; + } + + $encoding = $encoding == 'base64' ? 'B' : 'Q'; + // 75 is the value specified in the RFC + $prefix = '=?' . $charset . '?'.$encoding.'?'; + $suffix = '?='; + $maxLength = 75 - strlen($prefix . $suffix); + + // A multi-octet character may not be split across adjacent encoded-words + // So, we'll loop over each character + // mb_stlen() with wrong charset will generate a warning here and return null + $length = mb_strlen($str, $charset); + $result = ''; + $line_length = $prefix_len; + + if ($encoding == 'B') { + // base64 + $start = 0; + $prev = ''; + + for ($i=1; $i<=$length; $i++) { + // See #17311 + $chunk = mb_substr($str, $start, $i-$start, $charset); + $chunk = base64_encode($chunk); + $chunk_len = strlen($chunk); + + if ($line_length + $chunk_len == $maxLength || $i == $length) { + if ($result) { + $result .= "\n"; + } + $result .= $chunk; + $line_length = 0; + $start = $i; + } else if ($line_length + $chunk_len > $maxLength) { + if ($result) { + $result .= "\n"; + } + if ($prev) { + $result .= $prev; + } + $line_length = 0; + $start = $i - 1; + } else { + $prev = $chunk; + } + } + } else { + // quoted-printable + // see encodeQP() + $regexp = '/([\x22-\x29\x2C\x2E\x3A-\x40\x5B-\x60\x7B-\x7E\x80-\xFF])/'; + + for ($i=0; $i<=$length; $i++) { + $char = mb_substr($str, $i, 1, $charset); + // RFC recommends underline (instead of =20) in place of the space + // that's one of the reasons why we're not using iconv_mime_encode() + if ($char == ' ') { + $char = '_'; + $char_len = 1; + } else { + $char = preg_replace_callback( + $regexp, array('Mail_mimePart', '_qpReplaceCallback'), $char + ); + $char_len = strlen($char); + } + + if ($line_length + $char_len > $maxLength) { + if ($result) { + $result .= "\n"; + } + $line_length = 0; + } + + $result .= $char; + $line_length += $char_len; + } + } + + if ($result) { + $result = $prefix + .str_replace("\n", $suffix.$eol.' '.$prefix, $result).$suffix; + } + + return $result; } /** -- Gitblit v1.9.1