From 1c4f23d6e58e12f93d8de2c3ae416df575e8ad85 Mon Sep 17 00:00:00 2001 From: alecpl Date: Thu, 5 Jan 2012 11:25:42 +0000 Subject: [PATCH] - Exclude MIME functionality from rcube_imap class into rcube_mime class --- bin/msgexport.sh | 12 +- program/include/rcube_imap.php | 447 ++------------------------- program/include/rcube_message.php | 118 ++------ program/include/rcube_mime.php | 484 ++++++++++++++++++++++++++++++ program/steps/mail/addcontact.inc | 2 +- program/steps/mail/compose.inc | 10 +- program/steps/mail/func.inc | 23 +- program/steps/mail/sendmail.inc | 2 +- program/steps/mail/viewsource.inc | 3 +- tests/maildecode.php | 11 +- 10 files changed, 559 insertions(+), 553 deletions(-) create mode 100644 program/include/rcube_mime.php diff --git a/bin/msgexport.sh b/bin/msgexport.sh index 239d89363..5f9061971 100755 --- a/bin/msgexport.sh +++ b/bin/msgexport.sh @@ -51,17 +51,17 @@ function export_mailbox($mbox, $filename) for ($count = $IMAP->messagecount(), $i=1; $i <= $count; $i++) { $headers = $IMAP->get_headers($i, null, false); - $from = current($IMAP->decode_address_list($headers->from, 1, false)); - + $from = current(rcube_mime::decode_address_list($headers->from, 1, false)); + fwrite($out, sprintf("From %s %s UID %d\n", $from['mailto'], $headers->date, $headers->uid)); fwrite($out, $IMAP->conn->fetchPartHeader($mbox, $i)); fwrite($out, $IMAP->conn->handlePartBody($mbox, $i)); fwrite($out, "\n\n\n"); - + progress_update($i, $count); } vputs("\ncomplete.\n"); - + if ($filename) fclose($out); } @@ -114,7 +114,7 @@ $IMAP = new rcube_imap(null); if ($IMAP->connect($host, $args['user'], $args['pass'], $imap_port, $imap_ssl)) { vputs("IMAP login successful.\n"); - + $filename = null; $mailboxes = $args['mbox'] == '*' ? $IMAP->list_mailboxes(null) : array($args['mbox']); @@ -124,7 +124,7 @@ if ($IMAP->connect($host, $args['user'], $args['pass'], $imap_port, $imap_ssl)) $filename = preg_replace('/\.[a-z0-9]{3,4}$/i', '', $args['file']) . asciiwords($mbox) . '.mbox'; else if ($args['mbox'] == '*') $filename = asciiwords($mbox) . '.mbox'; - + if ($args['mbox'] == '*' && in_array(strtolower($mbox), array('junk','spam','trash'))) continue; diff --git a/program/include/rcube_imap.php b/program/include/rcube_imap.php index 2a2f203e9..badf815cf 100644 --- a/program/include/rcube_imap.php +++ b/program/include/rcube_imap.php @@ -381,28 +381,6 @@ class rcube_imap } - /** - * Forces selection of a mailbox - * - * @param string $mailbox Mailbox/Folder name - * @access public - */ - function select_mailbox($mailbox=null) - { - if (!strlen($mailbox)) { - $mailbox = $this->mailbox; - } - - $selected = $this->conn->select($mailbox); - - if ($selected && $this->mailbox != $mailbox) { - // clear messagecount cache for this mailbox - $this->_clear_messagecount($mailbox); - $this->mailbox = $mailbox; - } - } - - /** * Set internal list page * @@ -428,24 +406,24 @@ class rcube_imap /** - * Save a set of message ids for future message listing methods + * Save a search result for future message listing methods * - * @param string IMAP Search query - * @param rcube_result_index|rcube_result_thread Result set - * @param string Charset of search string - * @param string Sorting field - * @param string True if set is sorted (SORT was used for searching) + * @param array $set Search set, result from rcube_imap::get_search_set(): + * 0 - searching criteria, string + * 1 - search result, rcube_result_index|rcube_result_thread + * 2 - searching character set, string + * 3 - sorting field, string + * 4 - true if sorted, bool */ - function set_search_set($str=null, $msgs=null, $charset=null, $sort_field=null, $sorted=false) + function set_search_set($set) { - if (is_array($str) && $msgs === null) - list($str, $msgs, $charset, $sort_field, $sorted) = $str; - - $this->search_string = $str; - $this->search_set = $msgs; - $this->search_charset = $charset; - $this->search_sort_field = $sort_field; - $this->search_sorted = $sorted; + $set = (array)$set; + + $this->search_string = $set[0]; + $this->search_set = $set[1]; + $this->search_charset = $set[2]; + $this->search_sort_field = $set[3]; + $this->search_sorted = $set[4]; $this->search_threads = is_a($this->search_set, 'rcube_result_thread'); } @@ -453,8 +431,6 @@ class rcube_imap /** * Return the saved search set as hash array * - * @param bool $clone Clone result object - * * @return array Search set */ function get_search_set() @@ -1413,8 +1389,8 @@ class rcube_imap $results = $this->_search_index($mailbox, $str, $charset, $sort_field); - $this->set_search_set($str, $results, $charset, $sort_field, - $this->threading || $this->search_sorted ? true : false); + $this->set_search_set(array($str, $results, $charset, $sort_field, + $this->threading || $this->search_sorted ? true : false)); } @@ -1551,22 +1527,6 @@ class rcube_imap } - /** - * Check if the given message UID is part of the current search set - * - * @param string $msgid Message UID - * - * @return boolean True on match or if no search request is stored - */ - function in_searchset($uid) - { - if (!empty($this->search_string)) { - return $this->search_set->exists($uid); - } - return true; - } - - /** * Return message headers object of a specific message * @@ -1662,7 +1622,7 @@ class rcube_imap return $headers; } - $struct = &$this->_structure_part($structure, 0, '', $headers); + $struct = $this->_structure_part($structure, 0, '', $headers); // don't trust given content-type if (empty($struct->parts) && !empty($headers->ctype)) { @@ -1685,7 +1645,7 @@ class rcube_imap * @param string $parent * @access private */ - function &_structure_part($part, $count=0, $parent='', $mime_headers=null) + private function _structure_part($part, $count=0, $parent='', $mime_headers=null) { $struct = new rcube_message_part; $struct->mime_id = empty($parent) ? (string)$count : "$parent.$count"; @@ -1844,7 +1804,7 @@ class rcube_imap } if (is_string($mime_headers)) - $struct->headers = $this->_parse_headers($mime_headers) + $struct->headers; + $struct->headers = rcube_mime::parse_headers($mime_headers) + $struct->headers; else if (is_object($mime_headers)) $struct->headers = get_object_vars($mime_headers) + $struct->headers; @@ -1988,7 +1948,7 @@ class rcube_imap else $charset = rc_detect_encoding($filename_mime, $this->default_charset); - $part->filename = rcube_imap::decode_mime_string($filename_mime, $charset); + $part->filename = rcube_mime::decode_mime_string($filename_mime, $charset); } else if (!empty($filename_encoded)) { // decode filename according to RFC 2231, Section 4 @@ -2031,7 +1991,7 @@ class rcube_imap * * @return string Message/part body if not printed */ - function &get_message_part($uid, $part=1, $o_part=NULL, $print=NULL, $fp=NULL, $skip_charset_conv=false) + function get_message_part($uid, $part=1, $o_part=NULL, $print=NULL, $fp=NULL, $skip_charset_conv=false) { // get part data if not provided if (!is_object($o_part)) { @@ -2084,7 +2044,7 @@ class rcube_imap * @return string $part Message/part body * @see rcube_imap::get_message_part() */ - function &get_body($uid, $part=1) + function get_body($uid, $part=1) { $headers = $this->get_headers($uid); return rcube_charset_convert($this->get_message_part($uid, $part, NULL), @@ -2100,7 +2060,7 @@ class rcube_imap * * @return string Message source string */ - function &get_raw_body($uid, $fp=null) + function get_raw_body($uid, $fp=null) { return $this->conn->handlePartBody($this->mailbox, $uid, true, null, null, false, $fp); @@ -2113,7 +2073,7 @@ class rcube_imap * @param int $uid Message UID * @return string Message headers string */ - function &get_raw_headers($uid) + function get_raw_headers($uid) { return $this->conn->fetchPartHeader($this->mailbox, $uid, true); } @@ -3621,6 +3581,7 @@ class rcube_imap } } + /** * Getter for messages cache object */ @@ -3637,6 +3598,7 @@ class rcube_imap return $this->mcache; } + /** * Clears the messages cache. * @@ -3651,204 +3613,6 @@ class rcube_imap } - - /* -------------------------------- - * encoding/decoding methods - * --------------------------------*/ - - /** - * 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 $decode Decode address strings - * @return array Indexed list of addresses - */ - function decode_address_list($input, $max=null, $decode=true) - { - $a = $this->_parse_address_list($input, $decode); - $out = array(); - // Special chars as defined by RFC 822 need to in quoted string (or escaped). - $special_chars = '[\(\)\<\>\\\.\[\]@,;:"]'; - - if (!is_array($a)) - return $out; - - $c = count($a); - $j = 0; - - foreach ($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 boolean $remove_quotas Remove quotes if necessary - * @return string Decoded string - */ - function decode_header($input, $remove_quotes=false) - { - $str = rcube_imap::decode_mime_string((string)$input, $this->default_charset); - if ($str[0] == '"' && $remove_quotes) - $str = str_replace('"', '', $str); - - return $str; - } - - - /** - * Decode a mime-encoded string to internal charset - * - * @param string $input Header value - * @param string $fallback Fallback charset if none specified - * - * @return string Decoded string - * @static - */ - public static function decode_mime_string($input, $fallback=null) - { - if (!empty($fallback)) { - $default_charset = $fallback; - } - else { - $default_charset = rcmail::get_instance()->config->get('default_charset', 'ISO-8859-1'); - } - - // 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 $encoding Part encoding - * @return string Decoded string - */ - function mime_decode($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 $input; - } - } - - /* -------------------------------- * private methods * --------------------------------*/ @@ -3937,7 +3701,7 @@ class rcube_imap * * @return int Message UID */ - function id2uid($id, $mailbox = null) + private function id2uid($id, $mailbox = null) { if (!strlen($mailbox)) { $mailbox = $this->mailbox; @@ -4028,161 +3792,6 @@ class rcube_imap } - /** - * Split RFC822 header string into an associative array - * @access private - */ - private function _parse_headers($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 $a_headers; - } - - - /** - * @access private - */ - private function _parse_address_list($str, $decode=true) - { - // 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 = $this->decode_header($name); - } - } - - 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 $remove_comments Enable to remove comments - * - * @return array Header items - */ - static function explode_header_string($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 .= '\\'; - } - $i++; - } - } - // we're inside a comment string - else if ($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; - } - - /** * This is our own debug handler for the IMAP connection * @access public diff --git a/program/include/rcube_message.php b/program/include/rcube_message.php index 633f59be2..d05b57abe 100644 --- a/program/include/rcube_message.php +++ b/program/include/rcube_message.php @@ -42,6 +42,13 @@ class rcube_message * @var rcube_imap */ private $imap; + + /** + * Instance of mime class + * + * @var rcube_mime + */ + private $mime; private $opt = array(); private $inline_parts = array(); private $parse_alternative = false; @@ -63,27 +70,24 @@ class rcube_message * * @param string $uid The message UID. * - * @uses rcmail::get_instance() - * @uses rcube_imap::decode_mime_string() - * @uses self::set_safe() - * * @see self::$app, self::$imap, self::$opt, self::$structure */ function __construct($uid) { - $this->app = rcmail::get_instance(); + $this->uid = $uid; + $this->app = rcmail::get_instance(); $this->imap = $this->app->imap; $this->imap->get_all_headers = true; - $this->uid = $uid; $this->headers = $this->imap->get_message($uid); if (!$this->headers) return; - $this->subject = rcube_imap::decode_mime_string( - $this->headers->subject, $this->headers->charset); - list(, $this->sender) = each($this->imap->decode_address_list($this->headers->from)); + $this->mime = new rcube_mime($this->headers->charset); + + $this->subject = $this->mime->decode_mime_string($this->headers->subject); + list(, $this->sender) = each($this->mime->decode_address_list($this->headers->from, 1)); $this->set_safe((intval($_GET['_safe']) || $_SESSION['safe_messages'][$uid])); $this->opt = array( @@ -115,12 +119,15 @@ class rcube_message */ public function get_header($name, $raw = false) { + if (empty($this->headers)) + return null; + if ($this->headers->$name) $value = $this->headers->$name; else if ($this->headers->others[$name]) $value = $this->headers->others[$name]; - return $raw ? $value : $this->imap->decode_header($value); + return $raw ? $value : $this->mime->decode_header($value); } @@ -654,95 +661,4 @@ class rcube_message return $parts; } - - - /** - * 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 $length Length - * @return string Wrapped text - */ - public static function format_flowed($text, $length = 72) - { - $text = preg_split('/\r?\n/', $text); - - foreach ($text as $idx => $line) { - if ($line != '-- ') { - if ($line[0] == '>' && preg_match('/^(>+)/', $line, $regs)) { - $prefix = $regs[0]; - $level = strlen($prefix); - $line = rtrim(substr($line, $level)); - $line = $prefix . rc_wordwrap($line, $length - $level - 2, " \r\n$prefix "); - } - else if ($line) { - $line = rc_wordwrap(rtrim($line), $length - 2, " \r\n"); - // space-stuffing - $line = preg_replace('/(^|\r\n)(From| |>)/', '\\1 \\2', $line); - } - - $text[$idx] = $line; - } - } - - return implode("\r\n", $text); - } - } diff --git a/program/include/rcube_mime.php b/program/include/rcube_mime.php new file mode 100644 index 000000000..4d97199c3 --- /dev/null +++ b/program/include/rcube_mime.php @@ -0,0 +1,484 @@ + | + | Author: Aleksander Machniak | + +-----------------------------------------------------------------------+ + + $Id$ + +*/ + + +/** + * Class for parsing MIME messages + * + * @package Mail + * @author Thomas Bruederli + * @author Aleksander Machniak + */ +class rcube_mime +{ + private static $default_charset = RCMAIL_CHARSET; + + + /** + * Object constructor. + */ + function __construct($default_charset = null) + { + if ($default_charset) { + self::$default_charset = $default_charset; + } + else { + self::$default_charset = rcmail::get_instance()->config->get('default_charset', RCMAIL_CHARSET); + } + } + + + /** + * 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) ? $fallback : self::$default_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 $length Length + * + * @return string Wrapped text + */ + public static function format_flowed($text, $length = 72) + { + $text = preg_split('/\r?\n/', $text); + + foreach ($text as $idx => $line) { + if ($line != '-- ') { + if ($line[0] == '>' && preg_match('/^(>+)/', $line, $regs)) { + $prefix = $regs[0]; + $level = strlen($prefix); + $line = rtrim(substr($line, $level)); + $line = $prefix . rc_wordwrap($line, $length - $level - 2, " \r\n$prefix "); + } + else if ($line) { + $line = rc_wordwrap(rtrim($line), $length - 2, " \r\n"); + // space-stuffing + $line = preg_replace('/(^|\r\n)(From| |>)/', '\\1 \\2', $line); + } + + $text[$idx] = $line; + } + } + + return implode("\r\n", $text); + } + +} diff --git a/program/steps/mail/addcontact.inc b/program/steps/mail/addcontact.inc index 0120852a9..ab7bfc06f 100644 --- a/program/steps/mail/addcontact.inc +++ b/program/steps/mail/addcontact.inc @@ -37,7 +37,7 @@ if ($abook == null || !is_object($CONTACTS)) { if (!empty($_POST['_address']) && is_object($CONTACTS)) { - $contact_arr = $RCMAIL->imap->decode_address_list(get_input_value('_address', RCUBE_INPUT_POST, true), 1, false); + $contact_arr = rcube_mime::decode_address_list(get_input_value('_address', RCUBE_INPUT_POST, true), 1, false); if (!empty($contact_arr[1]['mailto'])) { $contact = array( diff --git a/program/steps/mail/compose.inc b/program/steps/mail/compose.inc index 855fbab67..3095d2436 100644 --- a/program/steps/mail/compose.inc +++ b/program/steps/mail/compose.inc @@ -268,7 +268,7 @@ else if (count($MESSAGE->identities)) { // extract all recipients of the reply-message if (is_object($MESSAGE->headers) && in_array($compose_mode, array(RCUBE_COMPOSE_REPLY, RCUBE_COMPOSE_FORWARD))) { - $a_to = $RCMAIL->imap->decode_address_list($MESSAGE->headers->to); + $a_to = rcube_mime::decode_address_list($MESSAGE->headers->to, null, true, $MESSAGE->headers->charset); foreach ($a_to as $addr) { if (!empty($addr['mailto'])) { $a_recipients[] = strtolower($addr['mailto']); @@ -277,7 +277,7 @@ else if (count($MESSAGE->identities)) { } if (!empty($MESSAGE->headers->cc)) { - $a_cc = $RCMAIL->imap->decode_address_list($MESSAGE->headers->cc); + $a_cc = rcube_mime::decode_address_list($MESSAGE->headers->cc, null, true, $MESSAGE->headers->charset); foreach ($a_cc as $addr) { if (!empty($addr['mailto'])) { $a_recipients[] = strtolower($addr['mailto']); @@ -420,7 +420,7 @@ foreach ($parts as $header) { // split recipients and put them back together in a unique way if (!empty($fvalue) && in_array($header, array('to', 'cc', 'bcc'))) { - $to_addresses = $RCMAIL->imap->decode_address_list($fvalue, null, $decode_header); + $to_addresses = rcube_mime::decode_address_list($fvalue, null, $decode_header, $MESSAGE->headers->charset); $fvalue = array(); foreach ($to_addresses as $addr_part) { @@ -650,7 +650,7 @@ function rcmail_prepare_message_body() if ($body && $part && $part->ctype_secondary == 'plain' && $part->ctype_parameters['format'] == 'flowed' ) { - $body = rcube_message::unfold_flowed($body); + $body = rcube_mime::unfold_flowed($body); } } } @@ -811,7 +811,7 @@ function rcmail_create_reply_body($body, $bodyIsHtml) global $RCMAIL, $MESSAGE, $LINE_LENGTH; // build reply prefix - $from = array_pop($RCMAIL->imap->decode_address_list($MESSAGE->get_header('from'), 1, false)); + $from = array_pop(rcube_mime::decode_address_list($MESSAGE->get_header('from'), 1, false, $MESSAGE->headers->charset)); $prefix = rcube_label(array( 'name' => 'mailreplyintro', 'vars' => array( diff --git a/program/steps/mail/func.inc b/program/steps/mail/func.inc index f791f7033..e17b19615 100644 --- a/program/steps/mail/func.inc +++ b/program/steps/mail/func.inc @@ -266,14 +266,12 @@ function rcmail_js_message_list($a_headers, $insert_top=FALSE, $a_show_cols=null $a_msg_cols = array(); $a_msg_flags = array(); - $RCMAIL->imap->set_charset(!empty($header->charset) ? $header->charset : $CONFIG['default_charset']); - // format each col; similar as in rcmail_message_list() foreach ($a_show_cols as $col) { if (in_array($col, array('from', 'to', 'cc', 'replyto'))) - $cont = Q(rcmail_address_string($header->$col, 3), 'show'); + $cont = Q(rcmail_address_string($header->$col, 3, false, null, $header->charset), 'show'); else if ($col=='subject') { - $cont = trim($RCMAIL->imap->decode_header($header->$col)); + $cont = trim(rcube_mime::decode_header($header->$col, $header->charset)); if (!$cont) $cont = rcube_label('nosubject'); $cont = Q($cont); } @@ -914,7 +912,7 @@ function rcmail_message_headers($attrib, $headers=NULL) } else if ($hkey == 'replyto') { if ($headers['replyto'] != $headers['from']) - $header_value = rcmail_address_string($value, null, true, $attrib['addicon']); + $header_value = rcmail_address_string($value, null, true, $attrib['addicon'], $headers['charset']); else continue; } @@ -922,19 +920,19 @@ function rcmail_message_headers($attrib, $headers=NULL) if ($headers['mail-replyto'] != $headers['reply-to'] && $headers['reply-to'] != $headers['from'] ) - $header_value = rcmail_address_string($value, null, true, $attrib['addicon']); + $header_value = rcmail_address_string($value, null, true, $attrib['addicon'], $headers['charset']); else continue; } else if ($hkey == 'mail-followup-to') { - $header_value = rcmail_address_string($value, null, true, $attrib['addicon']); + $header_value = rcmail_address_string($value, null, true, $attrib['addicon'], $headers['charset']); } else if (in_array($hkey, array('from', 'to', 'cc', 'bcc'))) - $header_value = rcmail_address_string($value, null, true, $attrib['addicon']); + $header_value = rcmail_address_string($value, null, true, $attrib['addicon'], $headers['charset']); else if ($hkey == 'subject' && empty($value)) $header_value = rcube_label('nosubject'); else - $header_value = trim($RCMAIL->imap->decode_header($value)); + $header_value = trim(rcube_mime::decode_header($value, $headers['charset'])); $output_headers[$hkey] = array( 'title' => rcube_label(preg_replace('/(^mail-|-)/', '', $hkey)), @@ -1260,11 +1258,11 @@ function rcmail_alter_html_link($matches) /** * decode address string and re-format it as HTML links */ -function rcmail_address_string($input, $max=null, $linked=false, $addicon=null) +function rcmail_address_string($input, $max=null, $linked=false, $addicon=null, $default_charset=null) { global $RCMAIL, $PRINT_MODE, $CONFIG; - $a_parts = $RCMAIL->imap->decode_address_list($input); + $a_parts = rcube_mime::decode_address_list($input, null, true, $default_charset); if (!sizeof($a_parts)) return $input; @@ -1483,7 +1481,8 @@ function rcmail_send_mdn($message, &$smtp_error) { $identity = $RCMAIL->user->get_identity(); $sender = format_email_recipient($identity['email'], $identity['name']); - $recipient = array_shift($RCMAIL->imap->decode_address_list($message->headers->mdn_to)); + $recipient = array_shift(rcube_mime::decode_address_list( + $message->headers->mdn_to, 1, true, $message->headers->charset)); $mailto = $recipient['mailto']; $compose = new Mail_mime("\r\n"); diff --git a/program/steps/mail/sendmail.inc b/program/steps/mail/sendmail.inc index 9d4d42562..953c7521a 100644 --- a/program/steps/mail/sendmail.inc +++ b/program/steps/mail/sendmail.inc @@ -520,7 +520,7 @@ else { // compose format=flowed content if enabled if ($flowed = $RCMAIL->config->get('send_format_flowed', true)) - $message_body = rcube_message::format_flowed($message_body, min($LINE_LENGTH+2, 79)); + $message_body = rcube_mime::format_flowed($message_body, min($LINE_LENGTH+2, 79)); else $message_body = rc_wordwrap($message_body, $LINE_LENGTH, "\r\n"); diff --git a/program/steps/mail/viewsource.inc b/program/steps/mail/viewsource.inc index d951fbead..e2b4e1b61 100644 --- a/program/steps/mail/viewsource.inc +++ b/program/steps/mail/viewsource.inc @@ -29,7 +29,8 @@ if ($uid = get_input_value('_uid', RCUBE_INPUT_GET)) header("Content-Type: text/plain; charset={$charset}"); if (!empty($_GET['_save'])) { - $filename = ($headers->subject ? $RCMAIL->imap->decode_header($headers->subject) : 'roundcube') . '.eml'; + $subject = rcube_mime::decode_header($headers->subject, $headers->charset); + $filename = ($subject ? $subject : $RCMAIL->config->get('product_name', 'email')) . '.eml'; $browser = $RCMAIL->output->browser; if ($browser->ie && $browser->ver < 7) diff --git a/tests/maildecode.php b/tests/maildecode.php index 664161cce..4ac499360 100644 --- a/tests/maildecode.php +++ b/tests/maildecode.php @@ -12,14 +12,11 @@ class rcube_test_maildecode extends UnitTestCase function __construct() { $this->UnitTestCase('Mail headers decoding tests'); - - $this->app = rcmail::get_instance(); - $this->app->imap_init(false); } /** * Test decoding of single e-mail address strings - * Uses rcube_imap::decode_address_list() + * Uses rcube_mime::decode_address_list() */ function test_decode_single_address() { @@ -76,7 +73,7 @@ class rcube_test_maildecode extends UnitTestCase ); foreach ($headers as $idx => $header) { - $res = $this->app->imap->decode_address_list($header); + $res = rcube_mime::decode_address_list($header); $this->assertEqual($results[$idx][0], count($res), "Rows number in result for header: " . $header); $this->assertEqual($results[$idx][1], $res[1]['name'], "Name part decoding for header: " . $header); @@ -86,7 +83,7 @@ class rcube_test_maildecode extends UnitTestCase /** * Test decoding of header values - * Uses rcube_imap::decode_mime_string() + * Uses rcube_mime::decode_mime_string() */ function test_header_decode_qp() { @@ -123,7 +120,7 @@ class rcube_test_maildecode extends UnitTestCase ); foreach ($test as $idx => $item) { - $res = $this->app->imap->decode_mime_string($item['in'], 'UTF-8'); + $res = rcube_mime::decode_mime_string($item['in'], 'UTF-8'); $res = quoted_printable_encode($res); $this->assertEqual($item['out'], $res, "Header decoding for: " . $idx);