diff options
Diffstat (limited to '')
-rw-r--r-- | library/vendor/HTMLPurifier/Node.php | 49 | ||||
-rw-r--r-- | library/vendor/HTMLPurifier/Node/Comment.php | 36 | ||||
-rw-r--r-- | library/vendor/HTMLPurifier/Node/Element.php | 59 | ||||
-rw-r--r-- | library/vendor/HTMLPurifier/Node/Text.php | 54 |
4 files changed, 198 insertions, 0 deletions
diff --git a/library/vendor/HTMLPurifier/Node.php b/library/vendor/HTMLPurifier/Node.php new file mode 100644 index 0000000..3995fec --- /dev/null +++ b/library/vendor/HTMLPurifier/Node.php @@ -0,0 +1,49 @@ +<?php + +/** + * Abstract base node class that all others inherit from. + * + * Why do we not use the DOM extension? (1) It is not always available, + * (2) it has funny constraints on the data it can represent, + * whereas we want a maximally flexible representation, and (3) its + * interface is a bit cumbersome. + */ +abstract class HTMLPurifier_Node +{ + /** + * Line number of the start token in the source document + * @type int + */ + public $line; + + /** + * Column number of the start token in the source document. Null if unknown. + * @type int + */ + public $col; + + /** + * Lookup array of processing that this token is exempt from. + * Currently, valid values are "ValidateAttributes". + * @type array + */ + public $armor = array(); + + /** + * When true, this node should be ignored as non-existent. + * + * Who is responsible for ignoring dead nodes? FixNesting is + * responsible for removing them before passing on to child + * validators. + */ + public $dead = false; + + /** + * Returns a pair of start and end tokens, where the end token + * is null if it is not necessary. Does not include children. + * @type array + */ + abstract public function toTokenPair(); +} + +// vim: et sw=4 sts=4 diff --git a/library/vendor/HTMLPurifier/Node/Comment.php b/library/vendor/HTMLPurifier/Node/Comment.php new file mode 100644 index 0000000..38ba193 --- /dev/null +++ b/library/vendor/HTMLPurifier/Node/Comment.php @@ -0,0 +1,36 @@ +<?php + +/** + * Concrete comment node class. + */ +class HTMLPurifier_Node_Comment extends HTMLPurifier_Node +{ + /** + * Character data within comment. + * @type string + */ + public $data; + + /** + * @type bool + */ + public $is_whitespace = true; + + /** + * Transparent constructor. + * + * @param string $data String comment data. + * @param int $line + * @param int $col + */ + public function __construct($data, $line = null, $col = null) + { + $this->data = $data; + $this->line = $line; + $this->col = $col; + } + + public function toTokenPair() { + return array(new HTMLPurifier_Token_Comment($this->data, $this->line, $this->col), null); + } +} diff --git a/library/vendor/HTMLPurifier/Node/Element.php b/library/vendor/HTMLPurifier/Node/Element.php new file mode 100644 index 0000000..6cbf56d --- /dev/null +++ b/library/vendor/HTMLPurifier/Node/Element.php @@ -0,0 +1,59 @@ +<?php + +/** + * Concrete element node class. + */ +class HTMLPurifier_Node_Element extends HTMLPurifier_Node +{ + /** + * 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. + * @type string + */ + public $name; + + /** + * Associative array of the node's attributes. + * @type array + */ + public $attr = array(); + + /** + * List of child elements. + * @type array + */ + public $children = array(); + + /** + * Does this use the <a></a> form or the </a> form, i.e. + * is it a pair of start/end tokens or an empty token. + * @bool + */ + public $empty = false; + + public $endCol = null, $endLine = null, $endArmor = array(); + + public function __construct($name, $attr = array(), $line = null, $col = null, $armor = array()) { + $this->name = $name; + $this->attr = $attr; + $this->line = $line; + $this->col = $col; + $this->armor = $armor; + } + + public function toTokenPair() { + // XXX inefficiency here, normalization is not necessary + if ($this->empty) { + return array(new HTMLPurifier_Token_Empty($this->name, $this->attr, $this->line, $this->col, $this->armor), null); + } else { + $start = new HTMLPurifier_Token_Start($this->name, $this->attr, $this->line, $this->col, $this->armor); + $end = new HTMLPurifier_Token_End($this->name, array(), $this->endLine, $this->endCol, $this->endArmor); + //$end->start = $start; + return array($start, $end); + } + } +} + diff --git a/library/vendor/HTMLPurifier/Node/Text.php b/library/vendor/HTMLPurifier/Node/Text.php new file mode 100644 index 0000000..aec9166 --- /dev/null +++ b/library/vendor/HTMLPurifier/Node/Text.php @@ -0,0 +1,54 @@ +<?php + +/** + * 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. + */ +class HTMLPurifier_Node_Text extends HTMLPurifier_Node +{ + + /** + * PCDATA tag name compatible with DTD, see + * HTMLPurifier_ChildDef_Custom for details. + * @type string + */ + public $name = '#PCDATA'; + + /** + * @type string + */ + public $data; + /**< Parsed character data of text. */ + + /** + * @type bool + */ + public $is_whitespace; + + /**< Bool indicating if node is whitespace. */ + + /** + * Constructor, accepts data and determines if it is whitespace. + * @param string $data String parsed character data. + * @param int $line + * @param int $col + */ + public function __construct($data, $is_whitespace, $line = null, $col = null) + { + $this->data = $data; + $this->is_whitespace = $is_whitespace; + $this->line = $line; + $this->col = $col; + } + + public function toTokenPair() { + return array(new HTMLPurifier_Token_Text($this->data, $this->line, $this->col), null); + } +} + +// vim: et sw=4 sts=4 |