2006-07-23 00:11:03 +00:00
|
|
|
<?php
|
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Defines a set of immutable value object tokens for HTML representation.
|
|
|
|
*
|
|
|
|
* @file
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Abstract base token class that all others inherit from.
|
|
|
|
*/
|
|
|
|
class HTMLPurifier_Token {
|
2007-11-25 02:24:39 +00:00
|
|
|
public $type; /**< Type of node to bypass <tt>is_a()</tt>. */
|
|
|
|
public $line; /**< Line number node was on in source document. Null if unknown. */
|
2006-08-20 20:59:13 +00:00
|
|
|
|
2007-06-20 21:39:28 +00:00
|
|
|
/**
|
|
|
|
* Lookup array of processing that this token is exempt from.
|
2007-06-26 15:07:07 +00:00
|
|
|
* Currently, valid values are "ValidateAttributes" and
|
|
|
|
* "MakeWellFormed_TagClosedError"
|
2007-06-20 21:39:28 +00:00
|
|
|
*/
|
2007-11-25 02:24:39 +00:00
|
|
|
public $armor = array();
|
2007-06-20 21:39:28 +00:00
|
|
|
|
2006-08-20 20:59:13 +00:00
|
|
|
/**
|
|
|
|
* Copies the tag into a new one (clone substitute).
|
|
|
|
* @return Copied token
|
|
|
|
*/
|
2007-11-25 02:24:39 +00:00
|
|
|
public function copy() {
|
2007-06-26 02:49:21 +00:00
|
|
|
return unserialize(serialize($this));
|
2006-08-20 20:59:13 +00:00
|
|
|
}
|
2008-01-19 20:23:01 +00:00
|
|
|
|
|
|
|
public function __get($n) {
|
|
|
|
if ($n === 'type') {
|
|
|
|
trigger_error('Deprecated type property called; use instanceof', E_USER_NOTICE);
|
|
|
|
switch (get_class($this)) {
|
|
|
|
case 'HTMLPurifier_Token_Start': return 'start';
|
|
|
|
case 'HTMLPurifier_Token_Empty': return 'empty';
|
|
|
|
case 'HTMLPurifier_Token_End': return 'end';
|
|
|
|
case 'HTMLPurifier_Token_Text': return 'text';
|
|
|
|
case 'HTMLPurifier_Token_Comment': return 'comment';
|
|
|
|
default: return null;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2006-07-23 13:20:15 +00:00
|
|
|
}
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Abstract class of a tag token (start, end or empty), and its behavior.
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
class HTMLPurifier_Token_Tag extends HTMLPurifier_Token // abstract
|
|
|
|
{
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Static bool marker that indicates the class is a tag.
|
|
|
|
*
|
|
|
|
* This allows us to check objects with <tt>!empty($obj->is_tag)</tt>
|
|
|
|
* without having to use a function call <tt>is_a()</tt>.
|
|
|
|
*/
|
2007-11-25 02:24:39 +00:00
|
|
|
public $is_tag = true;
|
2006-07-23 13:20:15 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* The lower-case name of the tag, like 'a', 'b' or 'blockquote'.
|
|
|
|
*
|
|
|
|
* @note Strictly speaking, XML tags are case sensitive, so we shouldn't
|
|
|
|
* be lower-casing them, but these tokens cater to HTML tags, which are
|
|
|
|
* insensitive.
|
|
|
|
*/
|
2007-11-25 02:24:39 +00:00
|
|
|
public $name;
|
2006-07-23 13:20:15 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Associative array of the tag's attributes.
|
|
|
|
*/
|
2007-11-25 02:24:39 +00:00
|
|
|
public $attr = array();
|
2006-07-23 13:20:15 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Non-overloaded constructor, which lower-cases passed tag name.
|
|
|
|
*
|
2006-12-06 22:29:08 +00:00
|
|
|
* @param $name String name.
|
|
|
|
* @param $attr Associative array of attributes.
|
2006-07-23 13:20:15 +00:00
|
|
|
*/
|
2007-11-29 04:29:51 +00:00
|
|
|
public function __construct($name, $attr = array(), $line = null) {
|
2006-07-23 00:11:03 +00:00
|
|
|
$this->name = ctype_lower($name) ? $name : strtolower($name);
|
2006-12-06 22:29:08 +00:00
|
|
|
foreach ($attr as $key => $value) {
|
2006-08-02 02:43:52 +00:00
|
|
|
// normalization only necessary when key is not lowercase
|
|
|
|
if (!ctype_lower($key)) {
|
|
|
|
$new_key = strtolower($key);
|
2006-12-06 22:29:08 +00:00
|
|
|
if (!isset($attr[$new_key])) {
|
|
|
|
$attr[$new_key] = $attr[$key];
|
2006-08-02 02:43:52 +00:00
|
|
|
}
|
2006-08-05 01:50:13 +00:00
|
|
|
if ($new_key !== $key) {
|
2006-12-06 22:29:08 +00:00
|
|
|
unset($attr[$key]);
|
2006-08-05 01:50:13 +00:00
|
|
|
}
|
2006-08-02 02:43:52 +00:00
|
|
|
}
|
|
|
|
}
|
2006-12-06 22:29:08 +00:00
|
|
|
$this->attr = $attr;
|
2007-06-26 19:33:37 +00:00
|
|
|
$this->line = $line;
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Concrete start token class.
|
|
|
|
*/
|
2008-01-19 20:23:01 +00:00
|
|
|
class HTMLPurifier_Token_Start extends HTMLPurifier_Token_Tag {}
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Concrete empty token class.
|
|
|
|
*/
|
2008-01-19 20:23:01 +00:00
|
|
|
class HTMLPurifier_Token_Empty extends HTMLPurifier_Token_Tag {}
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Concrete end token class.
|
|
|
|
*
|
|
|
|
* @warning This class accepts attributes even though end tags cannot. This
|
|
|
|
* is for optimization reasons, as under normal circumstances, the Lexers
|
|
|
|
* do not pass attributes.
|
|
|
|
*/
|
2008-01-19 20:23:01 +00:00
|
|
|
class HTMLPurifier_Token_End extends HTMLPurifier_Token_Tag {}
|
2006-07-23 00:11:03 +00:00
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Concrete text token class.
|
|
|
|
*
|
|
|
|
* Text tokens comprise of regular parsed character data (PCDATA) and raw
|
|
|
|
* character data (from the CDATA sections). Internally, their
|
|
|
|
* data is parsed with all entities expanded. Surprisingly, the text token
|
|
|
|
* does have a "tag name" called #PCDATA, which is how the DTD represents it
|
|
|
|
* in permissible child nodes.
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
class HTMLPurifier_Token_Text extends HTMLPurifier_Token
|
|
|
|
{
|
2006-07-23 13:20:15 +00:00
|
|
|
|
2007-11-25 02:24:39 +00:00
|
|
|
public $name = '#PCDATA'; /**< PCDATA tag name compatible with DTD. */
|
|
|
|
public $data; /**< Parsed character data of text. */
|
|
|
|
public $is_whitespace; /**< Bool indicating if node is whitespace. */
|
2006-07-23 13:20:15 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Constructor, accepts data and determines if it is whitespace.
|
|
|
|
*
|
|
|
|
* @param $data String parsed character data.
|
|
|
|
*/
|
2007-11-29 04:29:51 +00:00
|
|
|
public function __construct($data, $line = null) {
|
2006-07-23 00:11:03 +00:00
|
|
|
$this->data = $data;
|
2006-07-23 13:20:15 +00:00
|
|
|
$this->is_whitespace = ctype_space($data);
|
2007-06-26 19:33:37 +00:00
|
|
|
$this->line = $line;
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
2006-07-23 13:20:15 +00:00
|
|
|
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
|
|
|
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Concrete comment token class. Generally will be ignored.
|
|
|
|
*/
|
2006-07-23 00:11:03 +00:00
|
|
|
class HTMLPurifier_Token_Comment extends HTMLPurifier_Token
|
|
|
|
{
|
2007-11-25 02:24:39 +00:00
|
|
|
public $data; /**< Character data within comment. */
|
2006-07-23 13:20:15 +00:00
|
|
|
/**
|
|
|
|
* Transparent constructor.
|
|
|
|
*
|
|
|
|
* @param $data String comment data.
|
|
|
|
*/
|
2007-11-29 04:29:51 +00:00
|
|
|
public function __construct($data, $line = null) {
|
2006-07-23 00:11:03 +00:00
|
|
|
$this->data = $data;
|
2007-06-26 19:33:37 +00:00
|
|
|
$this->line = $line;
|
2006-07-23 00:11:03 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|