2006-07-23 00:11:03 +00:00
|
|
|
<?php
|
|
|
|
|
|
|
|
require_once 'XML/HTMLSax3.php'; // PEAR
|
|
|
|
require_once 'HTMLPurifier/Lexer.php';
|
|
|
|
|
2006-07-23 18:56:00 +00:00
|
|
|
/**
|
2006-08-04 02:59:15 +00:00
|
|
|
* Proof-of-concept lexer that uses the PEAR package XML_HTMLSax3 to parse HTML.
|
2006-07-23 18:56:00 +00:00
|
|
|
*
|
|
|
|
* PEAR, not suprisingly, also has a SAX parser for HTML. I don't know
|
|
|
|
* very much about implementation, but it's fairly well written. However, that
|
|
|
|
* abstraction comes at a price: performance. You need to have it installed,
|
|
|
|
* and if the API changes, it might break our adapter. Not sure whether or not
|
2006-08-04 02:59:15 +00:00
|
|
|
* it's UTF-8 aware, but it has some entity parsing trouble (in all areas,
|
|
|
|
* text and attributes).
|
2006-07-23 18:56:00 +00:00
|
|
|
*
|
|
|
|
* Quite personally, I don't recommend using the PEAR class, and the defaults
|
|
|
|
* don't use it. The unit tests do perform the tests on the SAX parser too, but
|
|
|
|
* whatever it does for poorly formed HTML is up to it.
|
|
|
|
*
|
|
|
|
* @todo Generalize so that XML_HTMLSax is also supported.
|
|
|
|
*/
|
|
|
|
|
2006-07-23 00:11:03 +00:00
|
|
|
class HTMLPurifier_Lexer_PEARSax3 extends HTMLPurifier_Lexer
|
|
|
|
{
|
|
|
|
|
2006-07-23 18:56:00 +00:00
|
|
|
/**
|
|
|
|
* Internal accumulator array for SAX parsers.
|
|
|
|
* @protected
|
|
|
|
*/
|
|
|
|
var $tokens = array();
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-08-15 00:31:12 +00:00
|
|
|
function tokenizeHTML($html, $config = null) {
|
|
|
|
if (!$config) $config = HTMLPurifier_Config::createDefault();
|
2006-07-23 23:04:34 +00:00
|
|
|
$html = $this->escapeCDATA($html);
|
2006-08-15 00:53:24 +00:00
|
|
|
if ($config->get('Core', 'AcceptFullDocuments')) {
|
|
|
|
$html = $this->extractBody($html);
|
|
|
|
}
|
2006-07-23 23:04:34 +00:00
|
|
|
$html = $this->substituteNonSpecialEntities($html);
|
2006-07-23 00:11:03 +00:00
|
|
|
$parser=& new XML_HTMLSax3();
|
|
|
|
$parser->set_object($this);
|
|
|
|
$parser->set_element_handler('openHandler','closeHandler');
|
|
|
|
$parser->set_data_handler('dataHandler');
|
|
|
|
$parser->set_escape_handler('escapeHandler');
|
|
|
|
$parser->set_option('XML_OPTION_ENTITIES_PARSED', 1);
|
|
|
|
$parser->parse($html);
|
2006-07-23 18:56:00 +00:00
|
|
|
$tokens = $this->tokens;
|
|
|
|
$this->tokens = array();
|
|
|
|
return $tokens;
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
|
|
|
|
2006-07-23 18:56:00 +00:00
|
|
|
/**
|
|
|
|
* Open tag event handler, interface is defined by PEAR package.
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
function openHandler(&$parser, $name, $attrs, $closed) {
|
|
|
|
if ($closed) {
|
|
|
|
$this->tokens[] = new HTMLPurifier_Token_Empty($name, $attrs);
|
|
|
|
} else {
|
|
|
|
$this->tokens[] = new HTMLPurifier_Token_Start($name, $attrs);
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2006-07-23 18:56:00 +00:00
|
|
|
/**
|
|
|
|
* Close tag event handler, interface is defined by PEAR package.
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
function closeHandler(&$parser, $name) {
|
|
|
|
// HTMLSax3 seems to always send empty tags an extra close tag
|
|
|
|
// check and ignore if you see it:
|
|
|
|
// [TESTME] to make sure it doesn't overreach
|
|
|
|
if ($this->tokens[count($this->tokens)-1]->type == 'empty') {
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
$this->tokens[] = new HTMLPurifier_Token_End($name);
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2006-07-23 18:56:00 +00:00
|
|
|
/**
|
|
|
|
* Data event handler, interface is defined by PEAR package.
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
function dataHandler(&$parser, $data) {
|
|
|
|
$this->tokens[] = new HTMLPurifier_Token_Text($data);
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2006-07-23 18:56:00 +00:00
|
|
|
/**
|
2006-07-23 18:57:13 +00:00
|
|
|
* Escaped text handler, interface is defined by PEAR package.
|
2006-07-23 18:56:00 +00:00
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
function escapeHandler(&$parser, $data) {
|
2006-07-23 23:04:34 +00:00
|
|
|
if (strpos($data, '--') === 0) {
|
2006-07-23 00:11:03 +00:00
|
|
|
$this->tokens[] = new HTMLPurifier_Token_Comment($data);
|
|
|
|
}
|
2006-07-23 23:04:34 +00:00
|
|
|
// CDATA is handled elsewhere, but if it was handled here:
|
|
|
|
//if (strpos($data, '[CDATA[') === 0) {
|
|
|
|
// $this->tokens[] = new HTMLPurifier_Token_Text(
|
|
|
|
// substr($data, 7, strlen($data) - 9) );
|
|
|
|
//}
|
2006-07-23 00:11:03 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2006-07-22 00:28:28 +00:00
|
|
|
?>
|