2006-07-23 00:11:03 +00:00
|
|
|
<?php
|
|
|
|
|
|
|
|
|
|
require_once 'HTMLPurifier/Lexer/DirectLex.php';
|
|
|
|
|
|
|
|
|
|
class HTMLPurifier_LexerTest extends UnitTestCase
|
|
|
|
|
{
|
|
|
|
|
|
2006-07-23 23:04:34 +00:00
|
|
|
var $Lexer;
|
2006-07-23 00:11:03 +00:00
|
|
|
var $DirectLex, $PEARSax3, $DOMLex;
|
2006-07-23 21:07:30 +00:00
|
|
|
var $_entity_lookup;
|
2006-08-10 12:41:39 +00:00
|
|
|
var $_has_pear = false;
|
|
|
|
|
var $_has_dom = false;
|
2006-07-23 00:11:03 +00:00
|
|
|
|
|
|
|
|
function setUp() {
|
2006-07-23 23:04:34 +00:00
|
|
|
$this->Lexer = new HTMLPurifier_Lexer();
|
|
|
|
|
|
2006-07-23 00:11:03 +00:00
|
|
|
$this->DirectLex = new HTMLPurifier_Lexer_DirectLex();
|
2006-08-10 12:41:39 +00:00
|
|
|
|
|
|
|
|
if ( $GLOBALS['HTMLPurifierTest']['PEAR'] ) {
|
|
|
|
|
$this->_has_pear = true;
|
|
|
|
|
require_once 'HTMLPurifier/Lexer/PEARSax3.php';
|
|
|
|
|
$this->PEARSax3 = new HTMLPurifier_Lexer_PEARSax3();
|
|
|
|
|
}
|
2006-07-23 00:11:03 +00:00
|
|
|
|
|
|
|
|
$this->_has_dom = version_compare(PHP_VERSION, '5', '>=');
|
|
|
|
|
if ($this->_has_dom) {
|
|
|
|
|
require_once 'HTMLPurifier/Lexer/DOMLex.php';
|
|
|
|
|
$this->DOMLex = new HTMLPurifier_Lexer_DOMLex();
|
|
|
|
|
}
|
|
|
|
|
|
2006-07-23 21:07:30 +00:00
|
|
|
$this->_entity_lookup = HTMLPurifier_EntityLookup::instance();
|
|
|
|
|
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
|
|
|
|
|
2006-08-15 00:53:24 +00:00
|
|
|
function assertExtractBody($text, $extract = true) {
|
|
|
|
|
$result = $this->Lexer->extractBody($text);
|
|
|
|
|
if ($extract === true) $extract = $text;
|
|
|
|
|
$this->assertIdentical($extract, $result);
|
|
|
|
|
}
|
|
|
|
|
|
2006-09-27 02:09:54 +00:00
|
|
|
function test_parseData() {
|
|
|
|
|
$HP =& $this->Lexer;
|
|
|
|
|
|
|
|
|
|
$this->assertIdentical('asdf', $HP->parseData('asdf'));
|
|
|
|
|
$this->assertIdentical('&', $HP->parseData('&'));
|
|
|
|
|
$this->assertIdentical('"', $HP->parseData('"'));
|
|
|
|
|
$this->assertIdentical("'", $HP->parseData('''));
|
|
|
|
|
$this->assertIdentical("'", $HP->parseData('''));
|
|
|
|
|
$this->assertIdentical('&&&', $HP->parseData('&&&'));
|
|
|
|
|
$this->assertIdentical('&&', $HP->parseData('&&')); // [INVALID]
|
|
|
|
|
$this->assertIdentical('Procter & Gamble',
|
|
|
|
|
$HP->parseData('Procter & Gamble')); // [INVALID]
|
|
|
|
|
|
|
|
|
|
// This is not special, thus not converted. Test of fault tolerance,
|
|
|
|
|
// realistically speaking, this should never happen
|
|
|
|
|
$this->assertIdentical('-', $HP->parseData('-'));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
2006-08-15 00:31:12 +00:00
|
|
|
function test_extractBody() {
|
2006-08-15 00:53:24 +00:00
|
|
|
$this->assertExtractBody('<b>Bold</b>');
|
|
|
|
|
$this->assertExtractBody('<html><body><b>Bold</b></body></html>', '<b>Bold</b>');
|
2006-08-15 01:14:39 +00:00
|
|
|
$this->assertExtractBody('<HTML><BODY><B>Bold</B></BODY></HTML>', '<B>Bold</B>');
|
2006-08-15 00:53:24 +00:00
|
|
|
$this->assertExtractBody(
|
|
|
|
|
'<?xml version="1.0"?>
|
|
|
|
|
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
|
|
|
|
|
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
|
|
|
|
|
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
|
|
|
|
|
<head>
|
|
|
|
|
<title>xyz</title>
|
|
|
|
|
</head>
|
|
|
|
|
<body>
|
|
|
|
|
<form method="post" action="whatever1">
|
|
|
|
|
<div>
|
|
|
|
|
<input type="text" name="username" />
|
|
|
|
|
<input type="text" name="password" />
|
|
|
|
|
<input type="submit" />
|
|
|
|
|
</div>
|
|
|
|
|
</form>
|
|
|
|
|
</body>
|
|
|
|
|
</html>',
|
|
|
|
|
'
|
|
|
|
|
<form method="post" action="whatever1">
|
|
|
|
|
<div>
|
|
|
|
|
<input type="text" name="username" />
|
|
|
|
|
<input type="text" name="password" />
|
|
|
|
|
<input type="submit" />
|
|
|
|
|
</div>
|
|
|
|
|
</form>
|
|
|
|
|
');
|
|
|
|
|
$this->assertExtractBody('<html><body bgcolor="#F00"><b>Bold</b></body></html>', '<b>Bold</b>');
|
|
|
|
|
$this->assertExtractBody('<body>asdf'); // not closed, don't accept
|
2006-08-15 00:31:12 +00:00
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
2006-07-23 00:11:03 +00:00
|
|
|
function test_tokenizeHTML() {
|
|
|
|
|
|
|
|
|
|
$input = array();
|
|
|
|
|
$expect = array();
|
|
|
|
|
$sax_expect = array();
|
2006-08-15 00:31:12 +00:00
|
|
|
$config = array();
|
2006-07-23 00:11:03 +00:00
|
|
|
|
|
|
|
|
$input[0] = '';
|
|
|
|
|
$expect[0] = array();
|
|
|
|
|
|
|
|
|
|
$input[1] = 'This is regular text.';
|
|
|
|
|
$expect[1] = array(
|
|
|
|
|
new HTMLPurifier_Token_Text('This is regular text.')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
$input[2] = 'This is <b>bold</b> text';
|
|
|
|
|
$expect[2] = array(
|
|
|
|
|
new HTMLPurifier_Token_Text('This is ')
|
|
|
|
|
,new HTMLPurifier_Token_Start('b', array())
|
|
|
|
|
,new HTMLPurifier_Token_Text('bold')
|
|
|
|
|
,new HTMLPurifier_Token_End('b')
|
|
|
|
|
,new HTMLPurifier_Token_Text(' text')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
$input[3] = '<DIV>Totally rad dude. <b>asdf</b></div>';
|
|
|
|
|
$expect[3] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('DIV', array())
|
|
|
|
|
,new HTMLPurifier_Token_Text('Totally rad dude. ')
|
|
|
|
|
,new HTMLPurifier_Token_Start('b', array())
|
|
|
|
|
,new HTMLPurifier_Token_Text('asdf')
|
|
|
|
|
,new HTMLPurifier_Token_End('b')
|
|
|
|
|
,new HTMLPurifier_Token_End('div')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
// [XML-INVALID]
|
|
|
|
|
$input[4] = '<asdf></asdf><d></d><poOloka><poolasdf><ds></asdf></ASDF>';
|
|
|
|
|
$expect[4] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('asdf')
|
|
|
|
|
,new HTMLPurifier_Token_End('asdf')
|
|
|
|
|
,new HTMLPurifier_Token_Start('d')
|
|
|
|
|
,new HTMLPurifier_Token_End('d')
|
|
|
|
|
,new HTMLPurifier_Token_Start('poOloka')
|
|
|
|
|
,new HTMLPurifier_Token_Start('poolasdf')
|
|
|
|
|
,new HTMLPurifier_Token_Start('ds')
|
|
|
|
|
,new HTMLPurifier_Token_End('asdf')
|
|
|
|
|
,new HTMLPurifier_Token_End('ASDF')
|
|
|
|
|
);
|
|
|
|
|
// DOM is different because it condenses empty tags into REAL empty ones
|
|
|
|
|
// as well as makes it well-formed
|
|
|
|
|
$dom_expect[4] = array(
|
|
|
|
|
new HTMLPurifier_Token_Empty('asdf')
|
|
|
|
|
,new HTMLPurifier_Token_Empty('d')
|
|
|
|
|
,new HTMLPurifier_Token_Start('pooloka')
|
|
|
|
|
,new HTMLPurifier_Token_Start('poolasdf')
|
|
|
|
|
,new HTMLPurifier_Token_Empty('ds')
|
|
|
|
|
,new HTMLPurifier_Token_End('poolasdf')
|
|
|
|
|
,new HTMLPurifier_Token_End('pooloka')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
$input[5] = '<a'."\t".'href="foobar.php"'."\n".'title="foo!">Link to <b id="asdf">foobar</b></a>';
|
|
|
|
|
$expect[5] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('a',array('href'=>'foobar.php','title'=>'foo!'))
|
|
|
|
|
,new HTMLPurifier_Token_Text('Link to ')
|
|
|
|
|
,new HTMLPurifier_Token_Start('b',array('id'=>'asdf'))
|
|
|
|
|
,new HTMLPurifier_Token_Text('foobar')
|
|
|
|
|
,new HTMLPurifier_Token_End('b')
|
|
|
|
|
,new HTMLPurifier_Token_End('a')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
$input[6] = '<br />';
|
|
|
|
|
$expect[6] = array(
|
|
|
|
|
new HTMLPurifier_Token_Empty('br')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
// [SGML-INVALID] [RECOVERABLE]
|
|
|
|
|
$input[7] = '<!-- Comment --> <!-- not so well formed --->';
|
|
|
|
|
$expect[7] = array(
|
|
|
|
|
new HTMLPurifier_Token_Comment(' Comment ')
|
|
|
|
|
,new HTMLPurifier_Token_Text(' ')
|
|
|
|
|
,new HTMLPurifier_Token_Comment(' not so well formed -')
|
|
|
|
|
);
|
|
|
|
|
$sax_expect[7] = false; // we need to figure out proper comment output
|
|
|
|
|
|
|
|
|
|
// [SGML-INVALID]
|
|
|
|
|
$input[8] = '<a href=""';
|
|
|
|
|
$expect[8] = array(
|
|
|
|
|
new HTMLPurifier_Token_Text('<a href=""')
|
|
|
|
|
);
|
|
|
|
|
// SAX parses it into a tag
|
|
|
|
|
$sax_expect[8] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('a', array('href'=>''))
|
|
|
|
|
);
|
|
|
|
|
// DOM parses it into an empty tag
|
|
|
|
|
$dom_expect[8] = array(
|
|
|
|
|
new HTMLPurifier_Token_Empty('a', array('href'=>''))
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
$input[9] = '<b>';
|
|
|
|
|
$expect[9] = array(
|
|
|
|
|
new HTMLPurifier_Token_Text('<b>')
|
|
|
|
|
);
|
|
|
|
|
$sax_expect[9] = array(
|
|
|
|
|
new HTMLPurifier_Token_Text('<')
|
|
|
|
|
,new HTMLPurifier_Token_Text('b')
|
|
|
|
|
,new HTMLPurifier_Token_Text('>')
|
|
|
|
|
);
|
|
|
|
|
// note that SAX can clump text nodes together. We won't be
|
|
|
|
|
// too picky though
|
|
|
|
|
|
|
|
|
|
// [SGML-INVALID]
|
|
|
|
|
$input[10] = '<a "=>';
|
|
|
|
|
// We barf on this, aim for no attributes
|
|
|
|
|
$expect[10] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('a', array('"' => ''))
|
|
|
|
|
);
|
|
|
|
|
// DOM correctly has no attributes, but also closes the tag
|
|
|
|
|
$dom_expect[10] = array(
|
|
|
|
|
new HTMLPurifier_Token_Empty('a')
|
|
|
|
|
);
|
|
|
|
|
// SAX barfs on this
|
|
|
|
|
$sax_expect[10] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('a', array('"' => ''))
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
// [INVALID] [RECOVERABLE]
|
|
|
|
|
$input[11] = '"';
|
|
|
|
|
$expect[11] = array( new HTMLPurifier_Token_Text('"') );
|
|
|
|
|
|
|
|
|
|
// compare with this valid one:
|
|
|
|
|
$input[12] = '"';
|
|
|
|
|
$expect[12] = array( new HTMLPurifier_Token_Text('"') );
|
2006-07-23 21:07:30 +00:00
|
|
|
$sax_expect[12] = false; // choked!
|
|
|
|
|
|
2006-07-23 23:04:34 +00:00
|
|
|
// CDATA sections!
|
|
|
|
|
$input[13] = '<![CDATA[You <b>can't</b> get me!]]>';
|
|
|
|
|
$expect[13] = array( new HTMLPurifier_Token_Text(
|
|
|
|
|
'You <b>can't</b> get me!' // raw
|
|
|
|
|
) );
|
|
|
|
|
$sax_expect[13] = array( // SAX has a seperate call for each entity
|
|
|
|
|
new HTMLPurifier_Token_Text('You '),
|
|
|
|
|
new HTMLPurifier_Token_Text('<'),
|
|
|
|
|
new HTMLPurifier_Token_Text('b'),
|
|
|
|
|
new HTMLPurifier_Token_Text('>'),
|
|
|
|
|
new HTMLPurifier_Token_Text('can'),
|
|
|
|
|
new HTMLPurifier_Token_Text('&'),
|
|
|
|
|
new HTMLPurifier_Token_Text('#39;t'),
|
|
|
|
|
new HTMLPurifier_Token_Text('<'),
|
|
|
|
|
new HTMLPurifier_Token_Text('/b'),
|
|
|
|
|
new HTMLPurifier_Token_Text('>'),
|
|
|
|
|
new HTMLPurifier_Token_Text(' get me!')
|
|
|
|
|
);
|
|
|
|
|
|
|
|
|
|
$char_theta = $this->_entity_lookup->table['theta'];
|
|
|
|
|
$char_rarr = $this->_entity_lookup->table['rarr'];
|
|
|
|
|
|
|
|
|
|
// test entity replacement
|
|
|
|
|
$input[14] = 'θ';
|
|
|
|
|
$expect[14] = array( new HTMLPurifier_Token_Text($char_theta) );
|
|
|
|
|
|
|
|
|
|
// test that entities aren't replaced in CDATA sections
|
|
|
|
|
$input[15] = 'θ <![CDATA[→]]>';
|
|
|
|
|
$expect[15] = array( new HTMLPurifier_Token_Text($char_theta . ' →') );
|
|
|
|
|
$sax_expect[15] = array(
|
|
|
|
|
new HTMLPurifier_Token_Text($char_theta . ' '),
|
|
|
|
|
new HTMLPurifier_Token_Text('&'),
|
|
|
|
|
new HTMLPurifier_Token_Text('rarr;')
|
|
|
|
|
);
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-08-04 02:59:15 +00:00
|
|
|
// test entity resolution in attributes
|
|
|
|
|
$input[16] = '<a href="index.php?title=foo&id=bar">Link</a>';
|
|
|
|
|
$expect[16] = array(
|
|
|
|
|
new HTMLPurifier_Token_Start('a',array('href' => 'index.php?title=foo&id=bar'))
|
|
|
|
|
,new HTMLPurifier_Token_Text('Link')
|
|
|
|
|
,new HTMLPurifier_Token_End('a')
|
|
|
|
|
);
|
|
|
|
|
|
2006-08-14 13:27:18 +00:00
|
|
|
// test that UTF-8 is preserved
|
|
|
|
|
$char_hearts = $this->_entity_lookup->table['hearts'];
|
|
|
|
|
$input[17] = $char_hearts;
|
|
|
|
|
$expect[17] = array( new HTMLPurifier_Token_Text($char_hearts) );
|
|
|
|
|
|
2006-09-27 02:09:54 +00:00
|
|
|
// test weird characters in attributes
|
|
|
|
|
$input[18] = '<br test="x < 6" />';
|
|
|
|
|
$expect[18] = array( new HTMLPurifier_Token_Empty('br', array('test' => 'x < 6')) );
|
|
|
|
|
|
2006-08-15 00:31:12 +00:00
|
|
|
$default_config = HTMLPurifier_Config::createDefault();
|
2006-10-01 20:47:07 +00:00
|
|
|
$default_context = new HTMLPurifier_Context();
|
2006-07-23 00:11:03 +00:00
|
|
|
foreach($input as $i => $discard) {
|
2006-08-15 00:31:12 +00:00
|
|
|
if (!isset($config[$i])) $config[$i] = $default_config;
|
|
|
|
|
|
2006-10-01 20:47:07 +00:00
|
|
|
$result = $this->DirectLex->tokenizeHTML($input[$i], $config[$i], $default_context);
|
2006-07-23 23:04:34 +00:00
|
|
|
$this->assertEqual($expect[$i], $result, 'DirectLexTest '.$i.': %s');
|
2006-07-23 00:11:03 +00:00
|
|
|
paintIf($result, $expect[$i] != $result);
|
|
|
|
|
|
2006-08-10 12:41:39 +00:00
|
|
|
if ($this->_has_pear) {
|
|
|
|
|
// assert unless I say otherwise
|
2006-10-01 20:47:07 +00:00
|
|
|
$sax_result = $this->PEARSax3->tokenizeHTML($input[$i], $config[$i], $default_context);
|
2006-08-10 12:41:39 +00:00
|
|
|
if (!isset($sax_expect[$i])) {
|
|
|
|
|
// by default, assert with normal result
|
|
|
|
|
$this->assertEqual($expect[$i], $sax_result, 'PEARSax3Test '.$i.': %s');
|
|
|
|
|
paintIf($sax_result, $expect[$i] != $sax_result);
|
|
|
|
|
} elseif ($sax_expect[$i] === false) {
|
|
|
|
|
// assertions were turned off, optionally dump
|
|
|
|
|
// paintIf($sax_expect, $i == NUMBER);
|
|
|
|
|
} else {
|
|
|
|
|
// match with a custom SAX result array
|
|
|
|
|
$this->assertEqual($sax_expect[$i], $sax_result, 'PEARSax3Test (custom) '.$i.': %s');
|
|
|
|
|
paintIf($sax_result, $sax_expect[$i] != $sax_result);
|
|
|
|
|
}
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
2006-08-10 12:41:39 +00:00
|
|
|
|
2006-07-23 00:11:03 +00:00
|
|
|
if ($this->_has_dom) {
|
2006-10-01 20:47:07 +00:00
|
|
|
$dom_result = $this->DOMLex->tokenizeHTML($input[$i], $config[$i], $default_context);
|
2006-07-23 00:11:03 +00:00
|
|
|
// same structure as SAX
|
|
|
|
|
if (!isset($dom_expect[$i])) {
|
2006-07-23 23:04:34 +00:00
|
|
|
$this->assertEqual($expect[$i], $dom_result, 'DOMLexTest '.$i.': %s');
|
2006-07-23 00:11:03 +00:00
|
|
|
paintIf($dom_result, $expect[$i] != $dom_result);
|
|
|
|
|
} elseif ($dom_expect[$i] === false) {
|
|
|
|
|
// paintIf($dom_result, $i == NUMBER);
|
|
|
|
|
} else {
|
2006-07-23 23:04:34 +00:00
|
|
|
$this->assertEqual($dom_expect[$i], $dom_result, 'DOMLexTest (custom) '.$i.': %s');
|
2006-07-23 00:11:03 +00:00
|
|
|
paintIf($dom_result, $dom_expect[$i] != $dom_result);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
2006-04-15 01:17:13 +00:00
|
|
|
?>
|
