diff --git a/tests/HTMLPurifier/AttrValidator_ErrorsTest.php b/tests/HTMLPurifier/AttrValidator_ErrorsTest.php index 08f59c99..8371962b 100644 --- a/tests/HTMLPurifier/AttrValidator_ErrorsTest.php +++ b/tests/HTMLPurifier/AttrValidator_ErrorsTest.php @@ -35,7 +35,7 @@ class HTMLPurifier_AttrValidator_ErrorsTest extends HTMLPurifier_ErrorsHarness $this->invoke($token); } - // to lazy to check for global post and global pre + // too lazy to check for global post and global pre function testAttributeRemoved() { $this->expectErrorCollection(E_ERROR, 'AttrValidator: Attribute removed'); diff --git a/tests/HTMLPurifier/EntityLookupTest.php b/tests/HTMLPurifier/EntityLookupTest.php index f50ee611..5b031503 100644 --- a/tests/HTMLPurifier/EntityLookupTest.php +++ b/tests/HTMLPurifier/EntityLookupTest.php @@ -17,7 +17,7 @@ class HTMLPurifier_EntityLookupTest extends HTMLPurifier_Harness // special char $this->assertIdentical('"', $lookup->table['quot']); $this->assertIdentical('“', $lookup->table['ldquo']); - $this->assertIdentical('<', $lookup->table['lt']); //expressed strangely + $this->assertIdentical('<', $lookup->table['lt']); // expressed strangely in source file // symbol char $this->assertIdentical('θ', $lookup->table['theta']); diff --git a/tests/HTMLPurifier/LexerTest.php b/tests/HTMLPurifier/LexerTest.php index 9388f0ab..b9e00f8e 100644 --- a/tests/HTMLPurifier/LexerTest.php +++ b/tests/HTMLPurifier/LexerTest.php @@ -5,71 +5,95 @@ require_once 'HTMLPurifier/Lexer/DirectLex.php'; class HTMLPurifier_LexerTest extends HTMLPurifier_Harness { - var $Lexer; - var $DirectLex, $PEARSax3, $DOMLex; - var $_entity_lookup; var $_has_pear = false; - var $_has_dom = false; - function setUp() { - $this->Lexer = new HTMLPurifier_Lexer(); - - $this->DirectLex = new HTMLPurifier_Lexer_DirectLex(); - - // E_STRICT = 2048, int used for PHP4 compat - if ( $GLOBALS['HTMLPurifierTest']['PEAR'] && - ((error_reporting() & 2048) != 2048) + function HTMLPurifier_LexerTest() { + parent::HTMLPurifier_Harness(); + // E_STRICT = 2048, int used for PHP4 compat: this check disables + // PEAR if PHP 5 strict mode is on, since the class is not strict safe + if ( + $GLOBALS['HTMLPurifierTest']['PEAR'] && + ((error_reporting() & 2048) != 2048) // ought to be a better way ) { - $this->_has_pear = true; require_once 'HTMLPurifier/Lexer/PEARSax3.php'; - $this->PEARSax3 = new HTMLPurifier_Lexer_PEARSax3(); + $this->_has_pear = true; } - - $this->_has_dom = version_compare(PHP_VERSION, '5', '>='); - if ($this->_has_dom) { - require_once 'HTMLPurifier/Lexer/DOMLex.php'; - $this->DOMLex = new HTMLPurifier_Lexer_DOMLex(); - } - - $this->_entity_lookup = HTMLPurifier_EntityLookup::instance(); - } + // HTMLPurifier_Lexer::create() -------------------------------------------- + function test_create() { - $config = HTMLPurifier_Config::create(array('Core.MaintainLineNumbers' => true)); - $lexer = HTMLPurifier_Lexer::create($config); + $this->config->set('Core', 'MaintainLineNumbers', true); + $lexer = HTMLPurifier_Lexer::create($this->config); $this->assertIsA($lexer, 'HTMLPurifier_Lexer_DirectLex'); } + // HTMLPurifier_Lexer->parseData() ----------------------------------------- + + function assertParseData($input, $expect = true) { + if ($expect === true) $expect = $input; + $lexer = new HTMLPurifier_Lexer(); + $this->assertIdentical($expect, $lexer->parseData($input)); + } + + function test_parseData_plainText() { + $this->assertParseData('asdf'); + } + + function test_parseData_ampersandEntity() { + $this->assertParseData('&', '&'); + } + + function test_parseData_quotEntity() { + $this->assertParseData('"', '"'); + } + + function test_parseData_aposNumericEntity() { + $this->assertParseData(''', "'"); + } + + function test_parseData_aposCompactNumericEntity() { + $this->assertParseData(''', "'"); + } + + function test_parseData_adjacentAmpersandEntities() { + $this->assertParseData('&&&', '&&&'); + } + + function test_parseData_trailingUnescapedAmpersand() { + $this->assertParseData('&&', '&&'); + } + + function test_parseData_internalUnescapedAmpersand() { + $this->assertParseData('Procter & Gamble'); + } + + function test_parseData_improperEntityFaultToleranceTest() { + $this->assertParseData('-'); + } + + // HTMLPurifier_Lexer->extractBody() --------------------------------------- + function assertExtractBody($text, $extract = true) { - $result = $this->Lexer->extractBody($text); + $lexer = new HTMLPurifier_Lexer(); + $result = $lexer->extractBody($text); if ($extract === true) $extract = $text; $this->assertIdentical($extract, $result); } - 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('-')); + function test_extractBody_noBodyTags() { + $this->assertExtractBody('Bold'); } - - function test_extractBody() { - $this->assertExtractBody('Bold'); + function test_extractBody_lowercaseBodyTags() { $this->assertExtractBody('Bold', 'Bold'); + } + + function test_extractBody_uppercaseBodyTags() { $this->assertExtractBody('Bold', 'Bold'); + } + + function test_extractBody_realisticUseCase() { $this->assertExtractBody( ' '); - $this->assertExtractBody('Bold', 'Bold'); - $this->assertExtractBody('asdf'); // not closed, don't accept - } - function test_tokenizeHTML() { - - $input = array(); - $expect = array(); - $sax_expect = array(); - $config = array(); - - $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 bold 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] = '
Totally rad dude. asdf
'; - $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] = ''; - $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] = 'Link to foobar'; - $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] = '
'; - $expect[6] = array( - new HTMLPurifier_Token_Empty('br') - ); - - // [SGML-INVALID] [RECOVERABLE] - $input[7] = ' '; - $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] = ''')) - ); - // 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('') - ); - $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] = ''; - // 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('"') ); - $sax_expect[12] = false; // choked! - - // CDATA sections! - $input[13] = 'can't get me!]]>'; - $expect[13] = array( new HTMLPurifier_Token_Text( - 'You can't 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 '; - $expect[15] = array( new HTMLPurifier_Token_Text($char_theta . ' →') ); - $sax_expect[15] = array( - new HTMLPurifier_Token_Text($char_theta . ' '), - new HTMLPurifier_Token_Text('&'), - new HTMLPurifier_Token_Text('rarr;') - ); - - // test entity resolution in attributes - $input[16] = 'Link'; - $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') - ); - - // 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) ); - - // test weird characters in attributes - $input[18] = '
'; - $expect[18] = array( new HTMLPurifier_Token_Empty('br', array('test' => 'x < 6')) ); - - // test emoticon protection - $input[19] = 'Whoa! <3 That\'s not good >.>'; - $expect[19] = array( - new HTMLPurifier_Token_Start('b'), - new HTMLPurifier_Token_Text('Whoa! '), - new HTMLPurifier_Token_Text('<3 That\'s not good >'), - new HTMLPurifier_Token_Text('.>'), - new HTMLPurifier_Token_End('b'), - ); - $dom_expect[19] = array( - new HTMLPurifier_Token_Start('b'), - new HTMLPurifier_Token_Text('Whoa! <3 That\'s not good >.>'), - new HTMLPurifier_Token_End('b'), - ); - $sax_expect[19] = false; // SAX drops the < character - $config[19] = HTMLPurifier_Config::create(array('Core.AggressivelyFixLt' => true)); - - // test comment parsing with funky characters inside - $input[20] = '
'; - $expect[20] = array( - new HTMLPurifier_Token_Comment(' This >< comment '), - new HTMLPurifier_Token_Empty('br') - ); - $sax_expect[20] = false; - $config[20] = HTMLPurifier_Config::create(array('Core.AggressivelyFixLt' => true)); - - // test comment parsing of missing end - $input[21] = ''; - $expect[23] = array( - new HTMLPurifier_Token_Comment(' This comment < < & ') ); - $sax_expect[23] = false; $config[23] = - HTMLPurifier_Config::create(array('Core.AggressivelyFixLt' => - true)); - - // more DirectLex edge-cases - $input[24] = ''; - $expect[24] = array( - new HTMLPurifier_Token_Start('a', array('href' => '')), - new HTMLPurifier_Token_Text('<">') - ); - $sax_expect[24] = false; - $dom_expect[24] = array( - new HTMLPurifier_Token_Empty('a', array('href' => '><>')) - ); - - // empty tag with attributes - $input[25] = ''; - $expect[25] = array( - new HTMLPurifier_Token_Empty('param', array('name' => 'src', 'value' => 'http://example.com/video.wmv')) - ); - - $default_config = HTMLPurifier_Config::createDefault(); - $default_context = new HTMLPurifier_Context(); - foreach($input as $i => $discard) { - if (!isset($config[$i])) $config[$i] = $default_config; - - $result = $this->DirectLex->tokenizeHTML($input[$i], $config[$i], $default_context); - $this->assertIdentical($expect[$i], $result, 'DirectLexTest '.$i.': %s'); - paintIf($result, $expect[$i] != $result); - - if ($this->_has_pear) { - // assert unless I say otherwise - $sax_result = $this->PEARSax3->tokenizeHTML($input[$i], $config[$i], $default_context); - if (!isset($sax_expect[$i])) { - // by default, assert with normal result - $this->assertIdentical($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 - $this->assertIdentical($sax_expect[$i], $sax_result, 'PEARSax3Test (custom) '.$i.': %s'); - paintIf($sax_result, $sax_expect[$i] != $sax_result); - } - } - - if ($this->_has_dom) { - $dom_result = $this->DOMLex->tokenizeHTML($input[$i], $config[$i], $default_context); - // same structure as SAX - if (!isset($dom_expect[$i])) { - $this->assertIdentical($expect[$i], $dom_result, 'DOMLexTest '.$i.': %s'); - paintIf($dom_result, $expect[$i] != $dom_result); - } elseif ($dom_expect[$i] === false) { - // paintIf($dom_result, $i == NUMBER); - } else { - $this->assertIdentical($dom_expect[$i], $dom_result, 'DOMLexTest (custom) '.$i.': %s'); - paintIf($dom_result, $dom_expect[$i] != $dom_result); - } - } - - } - + function test_extractBody_bodyWithAttributes() { + $this->assertExtractBody('Bold', 'Bold'); } + function test_extractBody_preserveUnclosedBody() { + $this->assertExtractBody('asdf'); // not closed, don't accept + } + + // HTMLPurifier_Lexer->tokenizeHTML() -------------------------------------- + + function assertTokenization($input, $expect, $alt_expect = array()) { + $lexers = array(); + $lexers['DirectLex'] = new HTMLPurifier_Lexer_DirectLex(); + if ($this->_has_pear) $lexers['PEARSax3'] = new HTMLPurifier_Lexer_PEARSax3(); + if (version_compare(PHP_VERSION, "5", ">=") && class_exists('DOMDocument')) { + $lexers['DOMLex'] = new HTMLPurifier_Lexer_DOMLex(); + } + foreach ($lexers as $name => $lexer) { + $result = $lexer->tokenizeHTML($input, $this->config, $this->context); + if (isset($alt_expect[$name])) { + if ($alt_expect[$name] === false) continue; + $this->assertIdentical($result, $alt_expect[$name]); + } else { + $this->assertIdentical($result, $expect); + } + } + } + + function test_tokenizeHTML_emptyInput() { + $this->assertTokenization('', array()); + } + + function test_tokenizeHTML_plainText() { + $this->assertTokenization( + 'This is regular text.', + array( + new HTMLPurifier_Token_Text('This is regular text.') + ) + ); + } + + function test_tokenizeHTML_textAndTags() { + $this->assertTokenization( + 'This is bold text', + 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'), + ) + ); + } + + function test_tokenizeHTML_normalizeCase() { + $this->assertTokenization( + '
Totally rad dude. asdf
', + 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'), + ) + ); + } + + function test_tokenizeHTML_notWellFormed() { + $this->assertTokenization( + '', + 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'), + ), + array( + // DOMLex automatically closes invalid tags + 'DOMLex' => 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'), + ), + ) + ); + } + + function test_tokenizeHTML_whitespaceInTag() { + $this->assertTokenization( + 'Link to foobar', + 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'), + ) + ); + } + + function test_tokenizeHTML_emptyTag() { + $this->assertTokenization( + '
', + array( new HTMLPurifier_Token_Empty('br') ) + ); + } + + function test_tokenizeHTML_comment() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Comment(' Comment ') ) + ); + } + + function test_tokenizeHTML_malformedComment() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Comment(' not so well formed -') ), + array( + 'PEARSax3' => false, // behavior is undefined + ) + ); + } + + function test_tokenizeHTML_unterminatedTag() { + $this->assertTokenization( + ' array( new HTMLPurifier_Token_Empty('a', array('href'=>'')) ), + 'PEARSax3' => array( new HTMLPurifier_Token_Start('a', array('href'=>'')) ), + ) + ); + } + + function test_tokenizeHTML_specialEntities() { + $this->assertTokenization( + '<b>', + array( + new HTMLPurifier_Token_Text('') + ), + array( + // it is possible to configure PEARSax3 to clump nodes together, + // I just don't know how + 'PEARSax3' => array( + new HTMLPurifier_Token_Text('<'), + new HTMLPurifier_Token_Text('b'), + new HTMLPurifier_Token_Text('>'), + ) + ) + ); + } + + function test_tokenizeHTML_earlyQuote() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Empty('a') ), + array( + // we barf on this input + 'DirectLex' => $tokens = array( + new HTMLPurifier_Token_Start('a', array('"' => '')) + ), + 'PEARSax3' => $tokens, + ) + ); + } + + function test_tokenizeHTML_unescapedQuote() { + $this->assertTokenization( + '"', + array( new HTMLPurifier_Token_Text('"') ) + ); + } + + function test_tokenizeHTML_escapedQuote() { + $this->assertTokenization( + '"', + array( new HTMLPurifier_Token_Text('"') ) + ); + } + + function test_tokenizeHTML_cdata() { + $this->assertTokenization( + 'can't get me!]]>', + array( new HTMLPurifier_Token_Text('You can't get me!') ), + array( + // PEAR splits up all of the CDATA + 'PEARSax3' => array( + 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!'), + ), + ) + ); + } + + function test_tokenizeHTML_characterEntity() { + $this->assertTokenization( + 'θ', + array( new HTMLPurifier_Token_Text("\xCE\xB8") ) + ); + } + + function test_tokenizeHTML_characterEntityInCDATA() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Text("→") ), + array( + 'PEARSax3' => array( + new HTMLPurifier_Token_Text('&'), + new HTMLPurifier_Token_Text('rarr;'), + ), + ) + ); + } + + function test_tokenizeHTML_entityInAttribute() { + $this->assertTokenization( + 'Link', + array( + new HTMLPurifier_Token_Start('a',array('href' => 'index.php?title=foo&id=bar')), + new HTMLPurifier_Token_Text('Link'), + new HTMLPurifier_Token_End('a'), + ) + ); + } + + function test_tokenizeHTML_preserveUTF8() { + $this->assertTokenization( + "\xCE\xB8", + array( new HTMLPurifier_Token_Text("\xCE\xB8") ) + ); + } + + function test_tokenizeHTML_specialEntityInAttribute() { + $this->assertTokenization( + '
', + array( new HTMLPurifier_Token_Empty('br', array('test' => 'x < 6')) ) + ); + } + + function test_tokenizeHTML_emoticonProtection() { + $this->config->set('Core', 'AggressivelyFixLt', true); + $this->assertTokenization( + 'Whoa! <3 That\'s not good >.>', + array( + new HTMLPurifier_Token_Start('b'), + new HTMLPurifier_Token_Text('Whoa! '), + new HTMLPurifier_Token_Text('<3 That\'s not good >'), + new HTMLPurifier_Token_Text('.>'), + new HTMLPurifier_Token_End('b') + ), + array( + // text is absorbed together + 'DOMLex' => array( + new HTMLPurifier_Token_Start('b'), + new HTMLPurifier_Token_Text('Whoa! <3 That\'s not good >.>'), + new HTMLPurifier_Token_End('b'), + ), + 'PEARSax3' => false, // totally mangled + ) + ); + } + + function test_tokenizeHTML_commentWithFunkyChars() { + $this->assertTokenization( + '
', + array( + new HTMLPurifier_Token_Comment(' This >< comment '), + new HTMLPurifier_Token_Empty('br'), + ), + array( + 'PEARSax3' => false, + ) + ); + } + + function test_tokenizeHTML_unterminatedComment() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Comment(' This comment < < & ') ), + array( + 'PEARSax3' => false + ) + ); + } + + function test_tokenizeHTML_attributeWithSpecialCharacters() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Empty('a', array('href' => '><>')) ), + array( + 'DirectLex' => array( + new HTMLPurifier_Token_Start('a', array('href' => '')), + new HTMLPurifier_Token_Text('<">'), + ), + 'PEARSax3' => false, + ) + ); + } + + function test_tokenizeHTML_emptyTagWithSlashInAttribute() { + $this->assertTokenization( + '', + array( new HTMLPurifier_Token_Empty('param', array('name' => 'src', 'value' => 'http://example.com/video.wmv')) ) + ); + } + + /* + + function test_tokenizeHTML_() { + $this->assertTokenization( + , + array( + + ) + ); + } + */ + }