mirrors/htmlpurifier
0
0
Fork 0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2024-11-09 15:28:40 +00:00
Code Releases Activity

Remove PEARSax3 lexer.

Browse Source 
Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
This commit is contained in:
Edward Z. Yang 2012-01-03 20:40:17 +08:00
parent e0354fecd9
commit 94468f3c24
3 changed files with 8 additions and 194 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
NEWS
View File

@ -10,6 +10,7 @@ NEWS ( CHANGELOG and HISTORY ) HTMLPurifier
==========================
4.4.0, unknown release date
# Removed PEARSax3 handler.
# URI.Munge now munges URIs inside the same host that go from https
to http. Reported by Neike Taika-Tessaro.
# Core.EscapeNonASCIICharacters now always transforms entities to

139
library/HTMLPurifier/Lexer/PEARSax3.php
View File

@ -1,139 +0,0 @@
<?php
/**
* Proof-of-concept lexer that uses the PEAR package XML_HTMLSax3 to parse HTML.
*
* 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
* it's UTF-8 aware, but it has some entity parsing trouble (in all areas,
* text and attributes).
*
* 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.
*
* @warning Entity-resolution inside attributes is broken.
*/
class HTMLPurifier_Lexer_PEARSax3 extends HTMLPurifier_Lexer
{
/**
* Internal accumulator array for SAX parsers.
*/
protected $tokens = array();
protected $last_token_was_empty;
private $parent_handler;
private $stack = array();
public function tokenizeHTML($string, $config, $context) {
$this->tokens = array();
$this->last_token_was_empty = false;
$string = $this->normalize($string, $config, $context);
$this->parent_handler = set_error_handler(array($this, 'muteStrictErrorHandler'));
$parser = new XML_HTMLSax3();
$parser->set_object($this);
$parser->set_element_handler('openHandler','closeHandler');
$parser->set_data_handler('dataHandler');
$parser->set_escape_handler('escapeHandler');
// doesn't seem to work correctly for attributes
$parser->set_option('XML_OPTION_ENTITIES_PARSED', 1);
$parser->parse($string);
restore_error_handler();
return $this->tokens;
}
/**
* Open tag event handler, interface is defined by PEAR package.
*/
public function openHandler(&$parser, $name, $attrs, $closed) {
// entities are not resolved in attrs
foreach ($attrs as $key => $attr) {
$attrs[$key] = $this->parseData($attr);
}
if ($closed) {
$this->tokens[] = new HTMLPurifier_Token_Empty($name, $attrs);
$this->last_token_was_empty = true;
} else {
$this->tokens[] = new HTMLPurifier_Token_Start($name, $attrs);
}
$this->stack[] = $name;
return true;
}
/**
* Close tag event handler, interface is defined by PEAR package.
*/
public function closeHandler(&$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->last_token_was_empty) {
$this->last_token_was_empty = false;
return true;
}
$this->tokens[] = new HTMLPurifier_Token_End($name);
if (!empty($this->stack)) array_pop($this->stack);
return true;
}
/**
* Data event handler, interface is defined by PEAR package.
*/
public function dataHandler(&$parser, $data) {
$this->last_token_was_empty = false;
$this->tokens[] = new HTMLPurifier_Token_Text($data);
return true;
}
/**
* Escaped text handler, interface is defined by PEAR package.
*/
public function escapeHandler(&$parser, $data) {
if (strpos($data, '--') === 0) {
// remove trailing and leading double-dashes
$data = substr($data, 2);
if (strlen($data) >= 2 && substr($data, -2) == "--") {
$data = substr($data, 0, -2);
}
if (isset($this->stack[sizeof($this->stack) - 1]) &&
$this->stack[sizeof($this->stack) - 1] == "style") {
$this->tokens[] = new HTMLPurifier_Token_Text($data);
} else {
$this->tokens[] = new HTMLPurifier_Token_Comment($data);
}
$this->last_token_was_empty = false;
}
// CDATA is handled elsewhere, but if it was handled here:
//if (strpos($data, '[CDATA[') === 0) {
// $this->tokens[] = new HTMLPurifier_Token_Text(
// substr($data, 7, strlen($data) - 9) );
//}
return true;
}
/**
* An error handler that mutes strict errors
*/
public function muteStrictErrorHandler($errno, $errstr, $errfile=null, $errline=null, $errcontext=null) {
if ($errno == E_STRICT) return;
return call_user_func($this->parent_handler, $errno, $errstr, $errfile, $errline, $errcontext);
}
}
// vim: et sw=4 sts=4

62
tests/HTMLPurifier/LexerTest.php
View File

@ -7,13 +7,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
public function __construct() {
parent::__construct();
if ($GLOBALS['HTMLPurifierTest']['PEAR'] &&
// PEARSax3 is not maintained and throws loads of DEPRECATED
// errors in PHP 5.3
version_compare(PHP_VERSION, '5.3', '<')) {
require_once 'HTMLPurifier/Lexer/PEARSax3.php';
$this->_has_pear = true;
}
if ($GLOBALS['HTMLPurifierTest']['PH5P']) {
require_once 'HTMLPurifier/Lexer/PH5P.php';
}
@ -158,7 +151,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
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 (class_exists('DOMDocument')) {
$lexers['DOMLex'] = new HTMLPurifier_Lexer_DOMLex();
$lexers['PH5P'] = new HTMLPurifier_Lexer_PH5P();
@ -299,7 +291,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
array(
// I like our behavior better, but it's non-standard
'DOMLex' => array( new HTMLPurifier_Token_Empty('a', array('href'=>'')) ),
'PEARSax3' => array( new HTMLPurifier_Token_Start('a', array('href'=>'')) ),
'PH5P' => false, // total barfing, grabs scaffolding too
)
);
@ -313,12 +304,11 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
),
array(
// some parsers will separate entities out
'PEARSax3' => $split = array(
'PH5P' => array(
new HTMLPurifier_Token_Text('<'),
new HTMLPurifier_Token_Text('b'),
new HTMLPurifier_Token_Text('>'),
),
'PH5P' => $split,
)
);
}
@ -329,10 +319,9 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
array( new HTMLPurifier_Token_Empty('a') ),
array(
// we barf on this input
'DirectLex' => $tokens = array(
'DirectLex' => array(
new HTMLPurifier_Token_Start('a', array('"' => ''))
),
'PEARSax3' => $tokens,
'PH5P' => false, // behavior varies; handle this personally
)
);
@ -363,10 +352,7 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
function test_tokenizeHTML_escapedQuote() {
$this->assertTokenization(
'&quot;',
array( new HTMLPurifier_Token_Text('"') ),
array(
'PEARSax3' => false, // PEAR barfs on this
)
array( new HTMLPurifier_Token_Text('"') )
);
}
@ -375,8 +361,7 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
'<![CDATA[You <b>can&#39;t</b> get me!]]>',
array( new HTMLPurifier_Token_Text('You <b>can&#39;t</b> get me!') ),
array(
// PEAR splits up all of the CDATA
'PEARSax3' => $split = array(
'PH5P' => array(
new HTMLPurifier_Token_Text('You '),
new HTMLPurifier_Token_Text('<'),
new HTMLPurifier_Token_Text('b'),
@ -389,7 +374,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
new HTMLPurifier_Token_Text('>'),
new HTMLPurifier_Token_Text(' get me!'),
),
'PH5P' => $split,
)
);
}
@ -406,11 +390,10 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
'<![CDATA[&rarr;]]>',
array( new HTMLPurifier_Token_Text("&rarr;") ),
array(
'PEARSax3' => $split = array(
'PH5P' => array(
new HTMLPurifier_Token_Text('&'),
new HTMLPurifier_Token_Text('rarr;'),
),
'PH5P' => $split,
)
);
}
@ -457,7 +440,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
new HTMLPurifier_Token_Text('Whoa! <3 That\'s not good >.>'),
new HTMLPurifier_Token_End('b'),
),
'PEARSax3' => false, // totally mangled
'PH5P' => array( // interesting grouping
new HTMLPurifier_Token_Start('b'),
new HTMLPurifier_Token_Text('Whoa! '),
@ -475,9 +457,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
array(
new HTMLPurifier_Token_Comment(' This >< comment '),
new HTMLPurifier_Token_Empty('br'),
),
array(
'PEARSax3' => false,
)
);
}
@ -488,7 +467,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
array( new HTMLPurifier_Token_Comment(' This >< comment') ),
array(
'DOMLex' => false,
'PEARSax3' => false,
'PH5P' => false,
)
);
@ -505,7 +483,6 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
new HTMLPurifier_Token_End('script'),
),
array(
'PEARSax3' => false,
// PH5P, for some reason, bubbles the script to <head>
'PH5P' => false,
)
@ -515,10 +492,7 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
function test_tokenizeHTML_entitiesInComment() {
$this->assertTokenization(
'<!-- This comment < &lt; & -->',
array( new HTMLPurifier_Token_Comment(' This comment < &lt; & ') ),
array(
'PEARSax3' => false
)
array( new HTMLPurifier_Token_Comment(' This comment < &lt; & ') )
);
}
@ -531,8 +505,7 @@ class HTMLPurifier_LexerTest extends HTMLPurifier_Harness
new HTMLPurifier_Token_Start('a', array('href' => '')),
new HTMLPurifier_Token_Text('<'),
new HTMLPurifier_Token_Text('">'),
),
'PEARSax3' => false,
)
)
);
}
@ -595,7 +568,6 @@ div {}
),
array(
'DirectLex' => $alt_expect,
'PEARSax3' => $alt_expect,
)
);
}
@ -615,11 +587,6 @@ div {}
new HTMLPurifier_Token_Text('<3'),
new HTMLPurifier_Token_Empty('br'),
),
'PEARSax3' => array(
// bah too lazy to fix this
new HTMLPurifier_Token_Empty('br'),
new HTMLPurifier_Token_Empty('3<br'),
),
)
);
}
@ -639,12 +606,6 @@ div {}
new HTMLPurifier_Token_Text('<<'),
new HTMLPurifier_Token_End('b'),
),
'PEARSax3' => array(
// also too lazy to fix
new HTMLPurifier_Token_Start('b'),
new HTMLPurifier_Token_Empty('<<'),
new HTMLPurifier_Token_Text('b>'),
),
)
);
}
@ -666,13 +627,6 @@ div {}
new HTMLPurifier_Token_Text('test'),
new HTMLPurifier_Token_End('b'),
),
'PEARSax3' => array(
// totally doing the wrong thing here
new HTMLPurifier_Token_Text(' '),
new HTMLPurifier_Token_Start('b'),
new HTMLPurifier_Token_Text('test'),
new HTMLPurifier_Token_End('b'),
),
)
);
}
@ -694,7 +648,6 @@ div {}
),
array(
'PH5P' => $alt_tokens,
'PEARSax3' => $alt_tokens,
)
);
}
@ -777,7 +730,6 @@ div {}
),
array(
'DirectLex' => $start,
'PEARSax3' => $start,
)
);
}