* \-------^
*
* At this point we loop in order to find all attributes.
@@ -284,8 +285,7 @@ function tln_getnxtag($body, $offset)
* the end of the tag.
*/
$matches = array();
- preg_match('%^(\s*)(>|/>)%s', substr($body, $pos), $matches);
- if (isset($matches[0]) && $matches[0]) {
+ if (preg_match('%^(\s*)(>|/>)%s', substr($body, $pos), $matches)) {
/**
* Yep. So we did.
*/
@@ -301,13 +301,13 @@ function tln_getnxtag($body, $offset)
* There are several types of attributes, with optional
* [:space:] between members.
* Type 1:
- * attrname[:space:]=[:space:]'CDATA'
+ * attrname[:space:]=[:space:]'CDATA'
* Type 2:
- * attrname[:space:]=[:space:]"CDATA"
+ * attrname[:space:]=[:space:]"CDATA"
* Type 3:
- * attr[:space:]=[:space:]CDATA
+ * attr[:space:]=[:space:]CDATA
* Type 4:
- * attrname
+ * attrname
*
* We leave types 1 and 2 the same, type 3 we check for
* '"' and convert to """ if needed, then wrap in
@@ -326,102 +326,99 @@ function tln_getnxtag($body, $offset)
/**
* We arrived at the end of attribute name. Several things possible
* here:
- * '>' means the end of the tag and this is attribute type 4
- * '/' if followed by '>' means the same thing as above
+ * '>' means the end of the tag and this is attribute type 4
+ * '/' if followed by '>' means the same thing as above
* '\s' means a lot of things -- look what it's followed by.
- * anything else means the attribute is invalid.
+ * anything else means the attribute is invalid.
*/
switch ($match) {
- case '/':
- /**
- * This is an xhtml-style tag with a closing / at the
- * end, like so: 
. Check if it's followed
- * by the closing bracket. If not, then this tag is invalid
- */
- if (substr($body, $pos, 2) == '/>') {
- $pos++;
- $tagtype = 3;
- } else {
- $gt = tln_findnxstr($body, $pos, '>');
- $retary = array(false, false, false, $lt, $gt);
- return $retary;
- }
+ case '/':
+ /**
+ * This is an xhtml-style tag with a closing / at the
+ * end, like so: . Check if it's followed
+ * by the closing bracket. If not, then this tag is invalid
+ */
+ if (substr($body, $pos, 2) == '/>') {
+ $pos++;
+ $tagtype = 3;
+ } else {
+ $gt = tln_findnxstr($body, $pos, '>');
+ $retary = array(false, false, false, $lt, $gt);
+ return $retary;
+ }
//intentional fall-through
- case '>':
- $attary{$attname} = '"yes"';
- return array($tagname, $attary, $tagtype, $lt, $pos);
- break;
- default:
- /**
- * Skip whitespace and see what we arrive at.
- */
+ case '>':
+ $attary{$attname} = '"yes"';
+ return array($tagname, $attary, $tagtype, $lt, $pos);
+ break;
+ default:
+ /**
+ * Skip whitespace and see what we arrive at.
+ */
+ $pos = tln_skipspace($body, $pos);
+ $char = substr($body, $pos, 1);
+ /**
+ * Two things are valid here:
+ * '=' means this is attribute type 1 2 or 3.
+ * \w means this was attribute type 4.
+ * anything else we ignore and re-loop. End of tag and
+ * invalid stuff will be caught by our checks at the beginning
+ * of the loop.
+ */
+ if ($char == '=') {
+ $pos++;
$pos = tln_skipspace($body, $pos);
- $char = substr($body, $pos, 1);
/**
- * Two things are valid here:
- * '=' means this is attribute type 1 2 or 3.
- * \w means this was attribute type 4.
- * anything else we ignore and re-loop. End of tag and
- * invalid stuff will be caught by our checks at the beginning
- * of the loop.
+ * Here are 3 possibilities:
+ * "'" attribute type 1
+ * '"' attribute type 2
+ * everything else is the content of tag type 3
*/
- if ($char == '=') {
- $pos++;
- $pos = tln_skipspace($body, $pos);
- /**
- * Here are 3 possibilities:
- * "'" attribute type 1
- * '"' attribute type 2
- * everything else is the content of tag type 3
- */
- $quot = substr($body, $pos, 1);
- if ($quot == '\'') {
+ $quot = substr($body, $pos, 1);
+ if ($quot == '\'') {
$regary = tln_findnxreg($body, $pos + 1, '\'');
- if ($regary == false) {
- return array(false, false, false, $lt, strlen($body));
- }
- list($pos, $attval, $match) = $regary;
- $pos++;
- $attary{$attname} = '\'' . $attval . '\'';
- } else {
- if ($quot == '"') {
- $regary = tln_findnxreg($body, $pos + 1, '\"');
- if ($regary == false) {
- return array(false, false, false, $lt, strlen($body));
- }
- list($pos, $attval, $match) = $regary;
- $pos++;
- $attary{$attname} = '"' . $attval . '"';
- } else {
- /**
- * These are hateful. Look for \s, or >.
- */
- $regary = tln_findnxreg($body, $pos, '[\s>]');
- if ($regary == false) {
- return array(false, false, false, $lt, strlen($body));
- }
- list($pos, $attval, $match) = $regary;
- /**
- * If it's ">" it will be caught at the top.
- */
- $attval = preg_replace('/\"/s', '"', $attval);
- $attary{$attname} = '"' . $attval . '"';
- }
+ if ($regary == false) {
+ return array(false, false, false, $lt, strlen($body));
}
+ list($pos, $attval, $match) = $regary;
+ $pos++;
+ $attary{$attname} = '\'' . $attval . '\'';
+ } elseif ($quot == '"') {
+ $regary = tln_findnxreg($body, $pos + 1, '\"');
+ if ($regary == false) {
+ return array(false, false, false, $lt, strlen($body));
+ }
+ list($pos, $attval, $match) = $regary;
+ $pos++;
+ $attary{$attname} = '"' . $attval . '"';
} else {
- if (preg_match('|[\w/>]|', $char)) {
- /**
- * That was attribute type 4.
- */
- $attary{$attname} = '"yes"';
- } else {
- /**
- * An illegal character. Find next '>' and return.
- */
- $gt = tln_findnxstr($body, $pos, '>');
- return array(false, false, false, $lt, $gt);
+ /**
+ * These are hateful. Look for \s, or >.
+ */
+ $regary = tln_findnxreg($body, $pos, '[\s>]');
+ if ($regary == false) {
+ return array(false, false, false, $lt, strlen($body));
}
+ list($pos, $attval, $match) = $regary;
+ /**
+ * If it's ">" it will be caught at the top.
+ */
+ $attval = preg_replace('/\"/s', '"', $attval);
+ $attary{$attname} = '"' . $attval . '"';
}
+ } elseif (preg_match('|[\w/>]|', $char)) {
+ /**
+ * That was attribute type 4.
+ */
+ $attary{$attname} = '"yes"';
+ } else {
+ /**
+ * An illegal character. Find next '>' and return.
+ */
+ $gt = tln_findnxstr($body, $pos, '>');
+ return array(false, false, false, $lt, $gt);
+ }
+ break;
}
}
/**
@@ -512,14 +509,18 @@ function tln_unspace(&$attvalue)
* @param array $rm_attnames See description for tln_sanitize
* @param array $bad_attvals See description for tln_sanitize
* @param array $add_attr_to_tag See description for tln_sanitize
- * @return Array with modified attributes.
+ * @param string $trans_image_path
+ * @param boolean $block_external_images
+ * @return Array with modified attributes.
*/
function tln_fixatts(
$tagname,
$attary,
$rm_attnames,
$bad_attvals,
- $add_attr_to_tag
+ $add_attr_to_tag,
+ $trans_image_path,
+ $block_external_images
) {
while (list($attname, $attvalue) = each($attary)) {
/**
@@ -538,7 +539,12 @@ function tln_fixatts(
/**
* Remove any backslashes, entities, or extraneous whitespace.
*/
+ $oldattvalue = $attvalue;
tln_defang($attvalue);
+ if ($attname == 'style' && $attvalue !== $oldattvalue) {
+ $attvalue = "idiocy";
+ $attary{$attname} = $attvalue;
+ }
tln_unspace($attvalue);
/**
@@ -560,12 +566,26 @@ function tln_fixatts(
$newvalue = preg_replace($valmatch, $valrepl, $attvalue);
if ($newvalue != $attvalue) {
$attary{$attname} = $newvalue;
+ $attvalue = $newvalue;
}
}
}
}
}
- }
+ if ($attname == 'style') {
+ if (preg_match('/[\0-\37\200-\377]+/', $attvalue)) {
+ $attary{$attname} = '"disallowed character"';
+ }
+ preg_match_all("/url\s*\((.+)\)/si", $attvalue, $aMatch);
+ if (count($aMatch)) {
+ foreach($aMatch[1] as $sMatch) {
+ $urlvalue = $sMatch;
+ tln_fixurl($attname, $urlvalue, $trans_image_path, $block_external_images);
+ $attary{$attname} = str_replace($sMatch, $urlvalue, $attvalue);
+ }
+ }
+ }
+ }
/**
* See if we need to append any attributes to this tag.
*/
@@ -577,6 +597,240 @@ function tln_fixatts(
return $attary;
}
+function tln_fixurl($attname, &$attvalue, $trans_image_path, $block_external_images)
+{
+ $sQuote = '"';
+ $attvalue = trim($attvalue);
+ if ($attvalue && ($attvalue[0] =='"'|| $attvalue[0] == "'")) {
+ // remove the double quotes
+ $sQuote = $attvalue[0];
+ $attvalue = trim(substr($attvalue,1,-1));
+ }
+
+ /**
+ * Replace empty src tags with the blank image. src is only used
+ * for frames, images, and image inputs. Doing a replace should
+ * not affect them working as should be, however it will stop
+ * IE from being kicked off when src for img tags are not set
+ */
+ if ($attvalue == '') {
+ $attvalue = $sQuote . $trans_image_path . $sQuote;
+ } else {
+ // first, disallow 8 bit characters and control characters
+ if (preg_match('/[\0-\37\200-\377]+/',$attvalue)) {
+ switch ($attname) {
+ case 'href':
+ $attvalue = $sQuote . 'http://invalid-stuff-detected.example.com' . $sQuote;
+ break;
+ default:
+ $attvalue = $sQuote . $trans_image_path . $sQuote;
+ break;
+ }
+ } else {
+ $aUrl = parse_url($attvalue);
+ if (isset($aUrl['scheme'])) {
+ switch(strtolower($aUrl['scheme'])) {
+ case 'mailto':
+ case 'http':
+ case 'https':
+ case 'ftp':
+ if ($attname != 'href') {
+ if ($block_external_images == true) {
+ $attvalue = $sQuote . $trans_image_path . $sQuote;
+ } else {
+ if (!isset($aUrl['path'])) {
+ $attvalue = $sQuote . $trans_image_path . $sQuote;
+ }
+ }
+ } else {
+ $attvalue = $sQuote . $attvalue . $sQuote;
+ }
+ break;
+ case 'outbind':
+ $attvalue = $sQuote . $attvalue . $sQuote;
+ break;
+ case 'cid':
+ $attvalue = $sQuote . $attvalue . $sQuote;
+ break;
+ default:
+ $attvalue = $sQuote . $trans_image_path . $sQuote;
+ break;
+ }
+ } else {
+ if (!isset($aUrl['path']) || $aUrl['path'] != $trans_image_path) {
+ $$attvalue = $sQuote . $trans_image_path . $sQuote;
+ }
+ }
+ }
+ }
+}
+
+function tln_fixstyle($body, $pos, $trans_image_path, $block_external_images)
+{
+ $me = 'tln_fixstyle';
+ // workaround for in between comments
+ $iCurrentPos = $pos;
+ $content = '';
+ $sToken = '';
+ $bSucces = false;
+ $bEndTag = false;
+ for ($i=$pos,$iCount=strlen($body);$i<$iCount;++$i) {
+ $char = $body{$i};
+ switch ($char) {
+ case '<':
+ $sToken = $char;
+ break;
+ case '/':
+ if ($sToken == '<') {
+ $sToken .= $char;
+ $bEndTag = true;
+ } else {
+ $content .= $char;
+ }
+ break;
+ case '>':
+ if ($bEndTag) {
+ $sToken .= $char;
+ if (preg_match('/\<\/\s*style\s*\>/i',$sToken,$aMatch)) {
+ $newpos = $i + 1;
+ $bSucces = true;
+ break 2;
+ } else {
+ $content .= $sToken;
+ }
+ $bEndTag = false;
+ } else {
+ $content .= $char;
+ }
+ break;
+ case '!':
+ if ($sToken == '<') {
+ // possible comment
+ if (isset($body{$i+2}) && substr($body,$i,3) == '!--') {
+ $i = strpos($body,'-->',$i+3);
+ if ($i === false) { // no end comment
+ $i = strlen($body);
+ }
+ $sToken = '';
+ }
+ } else {
+ $content .= $char;
+ }
+ break;
+ default:
+ if ($bEndTag) {
+ $sToken .= $char;
+ } else {
+ $content .= $char;
+ }
+ break;
+ }
+ }
+ if ($bSucces == FALSE){
+ return array(FALSE, strlen($body));
+ }
+
+
+
+ /**
+ * First look for general BODY style declaration, which would be
+ * like so:
+ * body {background: blah-blah}
+ * and change it to .bodyclass so we can just assign it to a
+ */
+ $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content);
+
+ $trans_image_path = $trans_image_path;
+
+ /**
+ * Fix url('blah') declarations.
+ */
+ // $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si",
+ // "url(\\1$trans_image_path\\2)", $content);
+
+ // first check for 8bit sequences and disallowed control characters
+ if (preg_match('/[\16-\37\200-\377]+/',$content)) {
+ $content = '';
+ return array($content, $newpos);
+ }
+
+ // remove @import line
+ $content = preg_replace("/^\s*(@import.*)$/mi","\n\n",$content);
+
+ $content = preg_replace("/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", 'url', $content);
+ preg_match_all("/url\s*\((.+)\)/si",$content,$aMatch);
+ if (count($aMatch)) {
+ $aValue = $aReplace = array();
+ foreach($aMatch[1] as $sMatch) {
+ // url value
+ $urlvalue = $sMatch;
+ tln_fixurl('style',$urlvalue, $trans_image_path, $block_external_images);
+ $aValue[] = $sMatch;
+ $aReplace[] = $urlvalue;
+ }
+ $content = str_replace($aValue,$aReplace,$content);
+ }
+
+ /**
+ * Remove any backslashes, entities, and extraneous whitespace.
+ */
+ $contentTemp = $content;
+ tln_defang($contentTemp);
+ tln_unspace($contentTemp);
+
+ $match = Array('/\/\*.*\*\//',
+ '/expression/i',
+ '/behaviou*r/i',
+ '/binding/i',
+ '/include-source/i',
+ '/javascript/i',
+ '/script/i',
+ '/position/i');
+ $replace = Array('','idiocy', 'idiocy', 'idiocy', 'idiocy', 'idiocy', 'idiocy', '');
+ $contentNew = preg_replace($match, $replace, $contentTemp);
+ if ($contentNew !== $contentTemp) {
+ $content = $contentNew;
+ }
+ return array($content, $newpos);
+}
+
+function tln_body2div($attary, $trans_image_path)
+{
+ $me = 'tln_body2div';
+ $divattary = array('class' => "'bodyclass'");
+ $text = '#000000';
+ $has_bgc_stl = $has_txt_stl = false;
+ $styledef = '';
+ if (is_array($attary) && sizeof($attary) > 0){
+ foreach ($attary as $attname=>$attvalue){
+ $quotchar = substr($attvalue, 0, 1);
+ $attvalue = str_replace($quotchar, "", $attvalue);
+ switch ($attname){
+ case 'background':
+ $styledef .= "background-image: url('$trans_image_path'); ";
+ break;
+ case 'bgcolor':
+ $has_bgc_stl = true;
+ $styledef .= "background-color: $attvalue; ";
+ break;
+ case 'text':
+ $has_txt_stl = true;
+ $styledef .= "color: $attvalue; ";
+ break;
+ }
+ }
+ // Outlook defines a white bgcolor and no text color. This can lead to
+ // white text on a white bg with certain themes.
+ if ($has_bgc_stl && !$has_txt_stl) {
+ $styledef .= "color: $text; ";
+ }
+ if (strlen($styledef) > 0){
+ $divattary{"style"} = "\"$styledef\"";
+ }
+ }
+ return $divattary;
+}
+
/**
*
* @param string $body The HTML you wish to filter
@@ -587,6 +841,9 @@ function tln_fixatts(
* @param array $rm_attnames see description above
* @param array $bad_attvals see description above
* @param array $add_attr_to_tag see description above
+ * @param string $trans_image_path
+ * @param boolean $block_external_images
+
* @return string Sanitized html safe to show on your pages.
*/
function tln_sanitize(
@@ -597,7 +854,9 @@ function tln_sanitize(
$force_tag_closing,
$rm_attnames,
$bad_attvals,
- $add_attr_to_tag
+ $add_attr_to_tag,
+ $trans_image_path,
+ $block_external_images
) {
/**
* Normalize rm_tags and rm_tags_with_content.
@@ -609,7 +868,7 @@ function tln_sanitize(
/**
* See if tag_list is of tags to remove or tags to allow.
* false means remove these tags
- * true means allow these tags
+ * true means allow these tags
*/
$curpos = 0;
$open_tags = array();
@@ -622,10 +881,32 @@ function tln_sanitize(
$body = preg_replace('/&(\{.*?\};)/si', '&\\1', $body);
while (($curtag = tln_getnxtag($body, $curpos)) != false) {
list($tagname, $attary, $tagtype, $lt, $gt) = $curtag;
- $free_content = substr($body, $curpos, $lt - $curpos);
- if ($skip_content == false) {
+ $free_content = substr($body, $curpos, $lt-$curpos);
+ /**
+ * Take care of