sanitize: remove doctype properly, add experimental workaround against unnecessary html elements in sanitized data

This commit is contained in:
Andrew Dolgov 2013-10-06 11:08:15 +04:00
parent ae3851b1b5
commit f44d59992e
1 changed files with 14 additions and 2 deletions

View File

@ -2891,7 +2891,7 @@
}
}
$doc->removeChild($doc->firstChild); //remove doctype
$doc->removeChild($doc->doctype); //remove doctype
$doc = strip_harmful_tags($doc, $allowed_elements, $disallowed_attributes);
if ($highlight_words) {
@ -2924,7 +2924,19 @@
}
}
$res = $doc->saveHTML();
$body = $doc->getElementsByTagName("body")->item(0);
if ($body) {
$div = $doc->createElement("div");
foreach ($body->childNodes as $child) {
$div->appendChild($child);
}
$res = $doc->saveXML($div);
} else {
$res = $doc->saveHTML();
}
return $res;
}