banananetwork
/
roundcubemail
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Fix handling of forwarded messages inside of a TNEF message (#5632)

Browse Source
pull/6256/head
Aleksander Machniak 6 years ago
parent 38bda972c1
commit 8a6a9e86ae
3 changed files with 475 additions and 120 deletions
Show Stats Download Patch File Download Diff File

1
CHANGELOG
Unescape Escape View File

@ -67,6 +67,7 @@ CHANGELOG Roundcube Webmail
- Use 7bit encoding for ISO-2022-* charsets in sent mail (#5640)
- Handle inline images also inside multipart/mixed messages (#5905)
- Allow style tags in HTML editor on composed/reply messages (#5751)
- Fix handling of forwarded messages inside of a TNEF message (#5632)
- Fix bug where attachment size wasn't visible when the filename was too long (#6033)
- Fix checking table columns when there's more schemas/databases in postgres/mysql (#6047)
- Fix css conflicts in user interface and e-mail content (#5891)

2
program/lib/Roundcube/rcube_message.php
Unescape Escape View File

@ -981,7 +981,7 @@ class rcube_message
unset($body);
foreach ($tnef_arr as $pid => $winatt) {
foreach ($tnef_arr['attachments'] as $pid => $winatt) {
$tpart = new rcube_message_part;
$tpart->filename = $this->fix_attachment_name(trim($winatt['name']), $part);

592
program/lib/Roundcube/rcube_tnef_decoder.php
Unescape Escape View File

@ -34,31 +34,54 @@ class rcube_tnef_decoder
const LVL_MESSAGE = 0x01;
const LVL_ATTACHMENT = 0x02;
const ASUBJECT = 0x88004;
const AMCLASS = 0x78008;
const ATTACHDATA = 0x6800f;
const AFROM = 0x08000;
const ASUBJECT = 0x18004;
const AMESSAGEID = 0x18009;
const AFILENAME = 0x18010;
const APARENTID = 0x1800a;
const ACONVERSATIONID = 0x1800b;
const ABODY = 0x2800c;
const ADATESENT = 0x38005;
const ADATERECEIVED = 0x38006;
const ADATEMODIFIED = 0x38020;
const APRIORITY = 0x4800d;
const AOWNER = 0x60000;
const ASENTFOR = 0x60001;
const ASTATUS = 0x68007;
const ATTACHDATA = 0x6800f;
const ATTACHMETAFILE = 0x68011;
const ATTACHCREATEDATE = 0x38012;
const ARENDDATA = 0x69002;
const AMAPIPROPS = 0x69003;
const ARECIPIENTTABLE = 0x69004;
const AMAPIATTRS = 0x69005;
const AOEMCODEPAGE = 0x69007;
const AORIGINALMCLASS = 0x70006;
const AMCLASS = 0x78008;
const AVERSION = 0x89006;
const MAPI_NULL = 0x0001;
const MAPI_SHORT = 0x0002;
const MAPI_INT = 0x0003;
const MAPI_FLOAT = 0x0004;
const MAPI_DOUBLE = 0x0005;
const MAPI_CURRENCY = 0x0006;
const MAPI_APPTIME = 0x0007;
const MAPI_ERROR = 0x000a;
const MAPI_BOOLEAN = 0x000b;
const MAPI_OBJECT = 0x000d;
const MAPI_INT8BYTE = 0x0014;
const MAPI_STRING = 0x001e;
const MAPI_TYPE_UNSET = 0x0000;
const MAPI_NULL = 0x0001;
const MAPI_SHORT = 0x0002;
const MAPI_INT = 0x0003;
const MAPI_FLOAT = 0x0004;
const MAPI_DOUBLE = 0x0005;
const MAPI_CURRENCY = 0x0006;
const MAPI_APPTIME = 0x0007;
const MAPI_ERROR = 0x000a;
const MAPI_BOOLEAN = 0x000b;
const MAPI_OBJECT = 0x000d;
const MAPI_INT8BYTE = 0x0014;
const MAPI_STRING = 0x001e;
const MAPI_UNICODE_STRING = 0x001f;
const MAPI_SYSTIME = 0x0040;
const MAPI_CLSID = 0x0048;
const MAPI_BINARY = 0x0102;
const MAPI_SYSTIME = 0x0040;
const MAPI_CLSID = 0x0048;
const MAPI_BINARY = 0x0102;
const MAPI_BODY = 0x1000;
const MAPI_RTF_COMPRESSED = 0x1009;
const MAPI_BODY_HTML = 0x1013;
const MAPI_NATIVE_BODY = 0x1016;
const MAPI_DISPLAY_NAME = 0x3001;
const MAPI_ADDRTYPE = 0x3002;
@ -72,7 +95,7 @@ class rcube_tnef_decoder
const MAPI_PROVIDER_DLL_NAME = 0x300A;
const MAPI_SEARCH_KEY = 0x300B;
const MAPI_ATTACHMENT_X400_PARAMETERS = 0x3700;
const MAPI_ATTACH_DATA_OBJ = 0x3701;
const MAPI_ATTACH_DATA = 0x3701;
const MAPI_ATTACH_ENCODING = 0x3702;
const MAPI_ATTACH_EXTENSION = 0x3703;
const MAPI_ATTACH_FILENAME = 0x3704;
@ -93,67 +116,85 @@ class rcube_tnef_decoder
const MAPI_NAMED_TYPE_ID = 0x0000;
const MAPI_NAMED_TYPE_STRING = 0x0001;
const MAPI_NAMED_TYPE_NONE = 0xff;
const MAPI_MV_FLAG = 0x1000;
const RTF_UNCOMPRESSED = 0x414c454d;
const RTF_COMPRESSED = 0x75465a4c;
/**
* Decompress the data.
*
* @param string $data The data to decompress.
* @param array $params An array of arguments needed to decompress the
* data.
* @param string $data The data to decompress.
*
* @return mixed The decompressed data.
* @return mixed The decompressed data.
*/
public function decompress($data, $params = array())
public function decompress($data)
{
$out = array();
$attachments = array();
$message = array();
if ($this->_geti($data, 32) == self::SIGNATURE) {
$this->_geti($data, 16);
// Version
$this->_geti($data, 8); // lvl_message
$this->_geti($data, 32); // idTnefVersion
$this->_getx($data, $this->_geti($data, 32));
$this->_geti($data, 16); // checksum
// Codepage
$this->_geti($data, 8);
$this->_geti($data, 32); // idCodepage
$this->_getx($data, $this->_geti($data, 32));
$this->_geti($data, 16); // checksum
while (strlen($data) > 0) {
switch ($this->_geti($data, 8)) {
case self::LVL_MESSAGE:
$this->_decodeMessage($data);
$this->_decodeMessage($data, $message);
break;
case self::LVL_ATTACHMENT:
$this->_decodeAttachment($data, $out);
$this->_decodeAttachment($data, $attachments);
break;
}
}
}
return array_reverse($out);
return array(
'message' => $message,
'attachments' => array_reverse($attachments),
);
}
/**
* TODO
* Pop specified number of bytes from the buffer.
*
* @param string &$data The data string.
* @param integer $bits How many bits to retrieve.
* @param string &$data The data string.
* @param integer $bytes How many bytes to retrieve.
*
* @return TODO
* @return string Extracted data
*/
protected function _getx(&$data, $bits)
protected function _getx(&$data, $bytes)
{
$value = null;
if (strlen($data) >= $bits) {
$value = substr($data, 0, $bits);
$data = substr_replace($data, '', 0, $bits);
if (strlen($data) >= $bytes) {
$value = substr($data, 0, $bytes);
$data = substr($data, $bytes);
}
return $value;
}
/**
* TODO
* Pop specified number of bits from the buffer
*
* @param string &$data The data string.
* @param string &$data The data string.
* @param integer $bits How many bits to retrieve.
*
* @return TODO
* @return int
*/
protected function _geti(&$data, $bits)
{
@ -168,24 +209,24 @@ class rcube_tnef_decoder
if ($bytes >= 4) {
$value += (ord($data[2]) << 16) + (ord($data[3]) << 24);
}
$data = substr_replace($data, '', 0, $bytes);
$data = substr($data, $bytes);
}
return $value;
}
/**
* TODO
* Decode a single attribute
*
* @param string &$data The data string.
* @param string $attribute TODO
* @param string &$data The data string.
*/
protected function _decodeAttribute(&$data, $attribute)
protected function _decodeAttribute(&$data)
{
/* Data. */
// Data.
$value = $this->_getx($data, $this->_geti($data, 32));
/* Checksum. */
// Checksum.
$this->_geti($data, 16);
return $value;
@ -194,18 +235,18 @@ class rcube_tnef_decoder
/**
* TODO
*
* @param string $data The data string.
* @param array &$attachment_data TODO
* @param string $data The data string.
* @param array &result TODO
*/
protected function _extractMapiAttributes($data, &$attachment_data)
protected function _extractMapiAttributes($data, &$result)
{
/* Number of attributes. */
// Number of attributes.
$number = $this->_geti($data, 32);
while ((strlen($data) > 0) && $number--) {
$have_mval = false;
$num_mval = 1;
$named_id = $value = null;
$num_mval = 1;
$value = null;
$attr_type = $this->_geti($data, 16);
$attr_name = $this->_geti($data, 16);
@ -220,16 +261,19 @@ class rcube_tnef_decoder
switch ($named_type) {
case self::MAPI_NAMED_TYPE_ID:
$named_id = $this->_geti($data, 32);
$attr_name = $named_id;
$attr_name = $this->_geti($data, 32);
break;
case self::MAPI_NAMED_TYPE_STRING:
$attr_name = 0x9999;
$idlen = $this->_geti($data, 32);
$datalen = $idlen + ((4 - ($idlen % 4)) % 4);
$named_id = substr($this->_getx($data, $datalen), 0, $idlen);
$idlen = $this->_geti($data, 32);
$name = $this->_getx($data, $idlen + ((4 - ($idlen % 4)) % 4));
// $name = $this->convertString(substr($name, 0, $idlen));
break;
case self::MAPI_NAMED_TYPE_NONE:
default:
continue 2;
}
}
@ -238,8 +282,13 @@ class rcube_tnef_decoder
}
switch ($attr_type) {
case self::MAPI_NULL:
case self::MAPI_TYPE_UNSET:
break;
case self::MAPI_SHORT:
$value = $this->_geti($data, 16);
$this->_geti($data, 16);
break;
case self::MAPI_INT:
@ -270,46 +319,71 @@ class rcube_tnef_decoder
for ($i = 0; $i < $num_vals; $i++) {
$length = $this->_geti($data, 32);
/* Pad to next 4 byte boundary. */
// Pad to next 4 byte boundary.
$datalen = $length + ((4 - ($length % 4)) % 4);
if ($attr_type == self::MAPI_STRING) {
--$length;
}
// Read and truncate to length.
$value = $this->_getx($data, $datalen);
}
/* Read and truncate to length. */
$value = substr($this->_getx($data, $datalen), 0, $length);
if ($attr_type == self::MAPI_UNICODE_STRING) {
$value = $this->convertString($value);
}
break;
}
/* Store any interesting attributes. */
// Store any interesting attributes.
switch ($attr_name) {
case self::MAPI_RTF_COMPRESSED:
$result['type'] = 'application';
$result['subtype'] = 'rtf';
$result['name'] = ($result['name'] ?: 'Untitled') . '.rtf';
$result['stream'] = $this->_decodeRTF($value);
$result['size'] = strlen($result['stream']);
break;
case self::MAPI_BODY:
case self::MAPI_BODY_HTML:
$result['type'] = 'text';
$result['subtype'] = $attr_name == self::MAPI_BODY ? 'plain' : 'html';
$result['name'] = ($result['name'] ?: 'Untitled') . ($attr_name == self::MAPI_BODY ? '.txt' : '.html');
$result['stream'] = $value;
$result['size'] = strlen($value);
break;
case self::MAPI_ATTACH_LONG_FILENAME:
$value = $this->convertString($value);
/* Used in preference to AFILENAME value. */
$attachment_data[0]['name'] = preg_replace('/.*[\/](.*)$/', '\1', $value);
// Used in preference to AFILENAME value.
$result['name'] = trim(preg_replace('/.*[\/](.*)$/', '\1', $value));
break;
case self::MAPI_ATTACH_MIME_TAG:
$value = $this->convertString($value);
/* Is this ever set, and what is format? */
$attachment_data[0]['type'] = preg_replace('/^(.*)\/.*/', '\1', $value);
$attachment_data[0]['subtype'] = preg_replace('/.*\/(.*)$/', '\1', $value);
// Is this ever set, and what is format?
$value = explode('/', $value);
$result['type'] = $value[0];
$result['subtype'] = $value[1];
break;
case self::MAPI_ATTACH_DATA:
$this->_getx($value, 16);
$att = new rcube_tnef_decoder;
$res = $att->decompress($value);
$result = array_merge($result, $res['message']);
break;
}
}
}
/**
* TODO
* Decodes TNEF message attributes
*
* @param string &$data The data string.
* @param string &$data The data string.
* @param array &$message Message data
*/
protected function _decodeMessage(&$data)
protected function _decodeMessage(&$data, &$message)
{
$attribute = $this->_geti($data, 32);
$value = $this->_decodeAttribute($data, $attribute);
$value = $this->_decodeAttribute($data);
switch ($attribute) {
case self::AOEMCODEPAGE:
@ -318,67 +392,63 @@ class rcube_tnef_decoder
$this->codepage = $value[1];
break;
default:
case self::AMCLASS:
$value = trim(str_replace('Microsoft Mail v3.0 ', '', $value));
// Normal message will be that with prefix 'IPM.Microsoft Mail.
break;
case self::ASUBJECT:
$message['name'] = $value;
break;
case self::AMAPIPROPS:
$this->_extractMapiAttributes($value, $message);
break;
}
}
/**
* TODO
* Decodes TNEF attachment attributes
*
* @param string &$data The data string.
* @param array &$attachment_data TODO
* @param string &$data The data string.
* @param array &$attachment Attachments data
*/
protected function _decodeAttachment(&$data, &$attachment_data)
protected function _decodeAttachment(&$data, &$attachment)
{
$attribute = $this->_geti($data, 32);
$size = $this->_geti($data, 32);
$value = $this->_getx($data, $size);
$this->_geti($data, 16); // checksum
switch ($attribute) {
case self::ARENDDATA:
/* Marks start of new attachment. */
$this->_getx($data, $this->_geti($data, 32));
// Add a new default data block to hold details of this
// attachment. Reverse order is easier to handle later!
array_unshift($attachment, array(
'type' => 'application',
'subtype' => 'octet-stream',
'name' => 'unknown',
'stream' => ''
));
/* Checksum */
$this->_geti($data, 16);
/* Add a new default data block to hold details of this
attachment. Reverse order is easier to handle later! */
array_unshift($attachment_data, array('type' => 'application',
'subtype' => 'octet-stream',
'name' => 'unknown',
'stream' => ''));
break;
case self::AFILENAME:
$value = $this->_getx($data, $this->_geti($data, 32));
$value = $this->convertString($value, true);
/* Strip path. */
$attachment_data[0]['name'] = preg_replace('/.*[\/](.*)$/', '\1', $value);
/* Checksum */
$this->_geti($data, 16);
// Strip path
$attachment[0]['name'] = trim(preg_replace('/.*[\/](.*)$/', '\1', $value));
break;
case self::ATTACHDATA:
/* The attachment itself. */
$length = $this->_geti($data, 32);
$attachment_data[0]['size'] = $length;
$attachment_data[0]['stream'] = $this->_getx($data, $length);
/* Checksum */
$this->_geti($data, 16);
// The attachment itself
$attachment[0]['size'] = $size;
$attachment[0]['stream'] = $value;
break;
case self::AMAPIATTRS:
$value = $this->_getx($data, $this->_geti($data, 32));
/* Checksum */
$this->_geti($data, 16);
$this->_extractMapiAttributes($value, $attachment_data);
$this->_extractMapiAttributes($value, $attachment[0]);
break;
default:
$this->_decodeAttribute($data, $attribute);
}
}
@ -387,17 +457,301 @@ class rcube_tnef_decoder
*/
protected function convertString($str, $use_codepage = false)
{
if ($convert && $this->codepage
if ($use_codepage && $this->codepage
&& ($charset = rcube_charset::$windows_codepages[$this->codepage])
) {
$str = rcube_charset::convert($str, $charset);
$str = rcube_charset::convert($str, $charset, RCUBE_CHARSET);
}
else if (($pos = strpos($str, "\0")) !== false && $pos != strlen($str)-1) {
$str = rcube_charset::convert($str, 'UTF-16LE', RCUBE_CHARSET);
}
return trim($str);
}
/**
* Decodes TNEF RTF
*/
protected function _decodeRTF($data)
{
$c_size = $this->_geti($data, 32);
$size = $this->_geti($data, 32);
$magic = $this->_geti($data, 32);
$crc = $this->_geti($data, 32);
if ($magic == self::RTF_COMPRESSED) {
$data = $this->_decompressRTF($data, $size);
}
return $data;
}
/**
* Decompress compressed RTF. Logic taken from Horde.
*/
protected function _decompressRTF($data, $size)
{
$in = $out = $flags = $flag_count = 0;
$uncomp = '';
$preload = "{\\rtf1\\ansi\\mac\\deff0\\deftab720{\\fonttbl;}{\\f0\\fnil \\froman \\fswiss \\fmodern \\fscript \\fdecor MS Sans SerifSymbolArialTimes New RomanCourier{\\colortbl\\red0\\green0\\blue0\n\r\\par \\pard\\plain\\f0\\fs20\\b\\i\\u\\tab\\tx";
$length_preload = strlen($preload);
for ($cnt = 0; $cnt < $length_preload; $cnt++) {
$uncomp .= $preload{$cnt};
++$out;
}
else if (strpos($str, "\0") !== false) {
$str = rcube_charset::convert($str, 'UTF-16LE');
while ($out < ($size + $length_preload)) {
if (($flag_count++ % 8) == 0) {
$flags = ord($data{$in++});
}
else {
$flags = $flags >> 1;
}
if (($flags & 1) != 0) {
$offset = ord($data{$in++});
$length = ord($data{$in++});
$offset = ($offset << 4) | ($length >> 4);
$length = ($length & 0xF) + 2;
$offset = ((int)($out / 4096)) * 4096 + $offset;
if ($offset >= $out) {
$offset -= 4096;
}
$end = $offset + $length;
while ($offset < $end) {
$uncomp.= $uncomp[$offset++];
++$out;
}
}
else {
$uncomp .= $data{$in++};
++$out;
}
}
return substr($uncomp, $length_preload);
}
/**
* Parse RTF data and return the best plaintext representation we can.
* Adapted from:
* http://webcheatsheet.com/php/reading_the_clean_text_from_rtf.php
*
* @param string $text The RTF (uncompressed) text.
*
* @return string The plain text.
*/
public static function rtf2text($text)
{
$document = '';
$stack = array();
$j = -1;
// Read the data character-by- character…
for ($i = 0, $len = strlen($text); $i < $len; $i++) {
$c = $text[$i];
switch ($c) {
case "\\":
// Key Word
$nextChar = $text[$i + 1];
// If it is another backslash or nonbreaking space or hyphen,
// then the character is plain text and add it to the output stream.
if ($nextChar == "\\" && self::_rtfIsPlain($stack[$j])) {
$document .= "\\";
}
elseif ($nextChar == '~' && self::_rtfIsPlain($stack[$j])) {
$document .= ' ';
}
elseif ($nextChar == '_' && self::_rtfIsPlain($stack[$j])) {
$document .= '-';
}
elseif ($nextChar == '*') {
// Add to the stack.
$stack[$j]['*'] = true;
}
elseif ($nextChar == "'") {
// If it is a single quote, read next two characters that
// are the hexadecimal notation of a character we should add
// to the output stream.
$hex = substr($text, $i + 2, 2);
if (self::_rtfIsPlain($stack[$j])) {
$document .= html_entity_decode('&#' . hexdec($hex) .';');
}
//Shift the pointer.
$i += 2;
}
elseif ($nextChar >= 'a' && $nextChar <= 'z' || $nextChar >= 'A' && $nextChar <= 'Z') {
// Since, we’ve found the alphabetic character, the next
// characters are control words and, possibly, some digit
// parameter.
$word = '';
$param = null;
// Start reading characters after the backslash.
for ($k = $i + 1, $m = 0; $k < strlen($text); $k++, $m++) {
$nextChar = $text[$k];
// If the current character is a letter and there were
// no digits before it, then we’re still reading the
// control word. If there were digits, we should stop
// since we reach the end of the control word.
if ($nextChar >= 'a' && $nextChar <= 'z'
|| $nextChar >= 'A' && $nextChar <= 'Z') {
if (!empty($param)) {
break;
}
$word .= $nextChar;
}
elseif ($nextChar >= '0' && $nextChar <= '9') {
// If it is a digit, store the parameter.
$param .= $nextChar;
}
elseif ($nextChar == '-') {
// Since minus sign may occur only before a digit
// parameter, check whether $param is empty.
// Otherwise, we reach the end of the control word.
if (!empty($param)) {
break;
}
$param .= $nextChar;
}
else {
break;
}
}
// Shift the pointer on the number of read characters.
$i += $m - 1;
// Start analyzing.We are interested mostly in control words
$toText = '';
switch (strtolower($word)) {
// If the control word is "u", then its parameter is
// the decimal notation of the Unicode character that
// should be added to the output stream. We need to
// check whether the stack contains \ucN control word.
// If it does, we should remove the N characters from
// the output stream.
case 'u':
$toText .= html_entity_decode('&#x' . dechex($param) .';');
$ucDelta = @$stack[$j]['uc'];
if ($ucDelta > 0) {
$i += $ucDelta;
}
break;
case 'par':
case 'page':
case 'column':
case 'line':
case 'lbr':
$toText .= "\n";
break;
case 'emspace':
case 'enspace':
case 'qmspace':
$toText .= ' ';
break;
case 'tab':
$toText .= "\t";
break;
case 'chdate':
$toText .= date('m.d.Y');
break;
case 'chdpl':
$toText .= date('l, j F Y');
break;
case 'chdpa':
$toText .= date('D, j M Y');
break;
case 'chtime':
$toText .= date('H:i:s');
break;
case 'emdash':
$toText .= html_entity_decode('&mdash;');
break;
case 'endash':
$toText .= html_entity_decode('&ndash;');
break;
case 'bullet':
$toText .= html_entity_decode('&#149;');
break;
case 'lquote':
$toText .= html_entity_decode('&lsquo;');
break;
case 'rquote':
$toText .= html_entity_decode('&rsquo;');
break;
case 'ldblquote':
$toText .= html_entity_decode('&laquo;');
break;
case 'rdblquote':
$toText .= html_entity_decode('&raquo;');
break;
default:
$stack[$j][strtolower($word)] = empty($param) ? true : $param;
break;
}
// Add data to the output stream if required.
if (self::_rtfIsPlain($stack[$j])) {
$document .= $toText;
}
}
$i++;
break;
case '{':
// New subgroup starts, add new stack element and write the data
// from previous stack element to it.
if (!empty($stack[$j])) {
array_push($stack, $stack[$j++]);
}
else {
$j++;
}
break;
case '}':
array_pop($stack);
$j--;
break;
case '\0':
case '\r':
case '\f':
case '\n':
// Junk
break;
default:
// Add other data to the output stream if required.
if (!empty($stack[$j]) && self::_rtfIsPlain($stack[$j])) {
$document .= $c;
}
break;
}
}
$str = rtrim($str, "\0");
return $document;
}
protected static function _rtfIsPlain($s)
{
$notPlain = array('*', 'fonttbl', 'colortbl', 'datastore', 'themedata', 'stylesheet');
for ($i = 0; $i < count($notPlain); $i++) {
if (!empty($s[$notPlain[$i]])) {
return false;
}
}
return $str;
return true;
}
}

Write Preview
Loading…
Cancel
Save