1
0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2025-08-08 07:06:46 +02:00

Add CDATA support to the Lexers, as well as give PEARSax3 entity replacement.

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@106 48356398-32a2-884e-a903-53898d9a118a
This commit is contained in:
Edward Z. Yang
2006-07-23 23:04:34 +00:00
parent 5ce0ae7056
commit 609977f9f5
6 changed files with 165 additions and 65 deletions

View File

@@ -27,8 +27,14 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
public function tokenizeHTML($string) {
$doc = new DOMDocument();
// preprocess string
$string = '<html><body><div>'.$string.'</div></body></html>';
// replace and escape the CDATA sections, since parsing under HTML
// mode won't get 'em.
$string = $this->escapeCDATA($string);
@$doc->loadHTML($string); // mute all errors, handle it transparently
return $this->tokenizeDOM(
$doc->childNodes->item(1)-> // html
@@ -55,7 +61,8 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
if ( !($node instanceof DOMElement) ) {
if ($node instanceof DOMComment) {
$tokens[] = new HTMLPurifier_Token_Comment($node->data);
} elseif ($node instanceof DOMText) {
} elseif ($node instanceof DOMText ||
$node instanceof DOMCharacterData) {
$tokens[] = new HTMLPurifier_Token_Text($node->data);
}
// quite possibly, the object wasn't handled, that's fine