2010-09-09 05:14:17 +02:00
|
|
|
<?php
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Injects tokens into the document while parsing for well-formedness.
|
|
|
|
* This enables "formatter-like" functionality such as auto-paragraphing,
|
|
|
|
* smiley-ification and linkification to take place.
|
|
|
|
*
|
|
|
|
* A note on how handlers create changes; this is done by assigning a new
|
|
|
|
* value to the $token reference. These values can take a variety of forms and
|
|
|
|
* are best described HTMLPurifier_Strategy_MakeWellFormed->processToken()
|
|
|
|
* documentation.
|
|
|
|
*
|
|
|
|
* @todo Allow injectors to request a re-run on their output. This
|
|
|
|
* would help if an operation is recursive.
|
|
|
|
*/
|
|
|
|
abstract class HTMLPurifier_Injector
|
|
|
|
{
|
|
|
|
|
|
|
|
/**
|
2016-02-09 11:06:17 +01:00
|
|
|
* Advisory name of injector, this is for friendly error messages.
|
|
|
|
* @type string
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
|
|
|
public $name;
|
|
|
|
|
|
|
|
/**
|
2016-02-09 11:06:17 +01:00
|
|
|
* @type HTMLPurifier_HTMLDefinition
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
|
|
|
protected $htmlDefinition;
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Reference to CurrentNesting variable in Context. This is an array
|
|
|
|
* list of tokens that we are currently "inside"
|
2016-02-09 11:06:17 +01:00
|
|
|
* @type array
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
|
|
|
protected $currentNesting;
|
|
|
|
|
|
|
|
/**
|
2016-02-09 11:06:17 +01:00
|
|
|
* Reference to current token.
|
|
|
|
* @type HTMLPurifier_Token
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
protected $currentToken;
|
2010-09-09 05:14:17 +02:00
|
|
|
|
|
|
|
/**
|
2016-02-09 11:06:17 +01:00
|
|
|
* Reference to InputZipper variable in Context.
|
|
|
|
* @type HTMLPurifier_Zipper
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
protected $inputZipper;
|
2010-09-09 05:14:17 +02:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Array of elements and attributes this injector creates and therefore
|
|
|
|
* need to be allowed by the definition. Takes form of
|
|
|
|
* array('element' => array('attr', 'attr2'), 'element2')
|
2016-02-09 11:06:17 +01:00
|
|
|
* @type array
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
|
|
|
public $needed = array();
|
|
|
|
|
|
|
|
/**
|
2016-02-09 11:06:17 +01:00
|
|
|
* Number of elements to rewind backwards (relative).
|
|
|
|
* @type bool|int
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
protected $rewindOffset = false;
|
2010-09-09 05:14:17 +02:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Rewind to a spot to re-perform processing. This is useful if you
|
|
|
|
* deleted a node, and now need to see if this change affected any
|
|
|
|
* earlier nodes. Rewinding does not affect other injectors, and can
|
|
|
|
* result in infinite loops if not used carefully.
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param bool|int $offset
|
2010-09-09 05:14:17 +02:00
|
|
|
* @warning HTML Purifier will prevent you from fast-forwarding with this
|
|
|
|
* function.
|
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function rewindOffset($offset)
|
|
|
|
{
|
|
|
|
$this->rewindOffset = $offset;
|
2010-09-09 05:14:17 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2016-02-09 11:06:17 +01:00
|
|
|
* Retrieves rewind offset, and then unsets it.
|
|
|
|
* @return bool|int
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function getRewindOffset()
|
|
|
|
{
|
|
|
|
$r = $this->rewindOffset;
|
|
|
|
$this->rewindOffset = false;
|
2010-09-09 05:14:17 +02:00
|
|
|
return $r;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Prepares the injector by giving it the config and context objects:
|
|
|
|
* this allows references to important variables to be made within
|
|
|
|
* the injector. This function also checks if the HTML environment
|
|
|
|
* will work with the Injector (see checkNeeded()).
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param HTMLPurifier_Config $config
|
|
|
|
* @param HTMLPurifier_Context $context
|
|
|
|
* @return bool|string Boolean false if success, string of missing needed element/attribute if failure
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function prepare($config, $context)
|
|
|
|
{
|
2010-09-09 05:14:17 +02:00
|
|
|
$this->htmlDefinition = $config->getHTMLDefinition();
|
|
|
|
// Even though this might fail, some unit tests ignore this and
|
|
|
|
// still test checkNeeded, so be careful. Maybe get rid of that
|
|
|
|
// dependency.
|
|
|
|
$result = $this->checkNeeded($config);
|
2016-02-09 11:06:17 +01:00
|
|
|
if ($result !== false) {
|
|
|
|
return $result;
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
$this->currentNesting =& $context->get('CurrentNesting');
|
2016-02-09 11:06:17 +01:00
|
|
|
$this->currentToken =& $context->get('CurrentToken');
|
|
|
|
$this->inputZipper =& $context->get('InputZipper');
|
2010-09-09 05:14:17 +02:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* This function checks if the HTML environment
|
|
|
|
* will work with the Injector: if p tags are not allowed, the
|
|
|
|
* Auto-Paragraphing injector should not be enabled.
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param HTMLPurifier_Config $config
|
|
|
|
* @return bool|string Boolean false if success, string of missing needed element/attribute if failure
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function checkNeeded($config)
|
|
|
|
{
|
2010-09-09 05:14:17 +02:00
|
|
|
$def = $config->getHTMLDefinition();
|
|
|
|
foreach ($this->needed as $element => $attributes) {
|
2016-02-09 11:06:17 +01:00
|
|
|
if (is_int($element)) {
|
|
|
|
$element = $attributes;
|
|
|
|
}
|
|
|
|
if (!isset($def->info[$element])) {
|
|
|
|
return $element;
|
|
|
|
}
|
|
|
|
if (!is_array($attributes)) {
|
|
|
|
continue;
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
foreach ($attributes as $name) {
|
2016-02-09 11:06:17 +01:00
|
|
|
if (!isset($def->info[$element]->attr[$name])) {
|
|
|
|
return "$element.$name";
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Tests if the context node allows a certain element
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param string $name Name of element to test for
|
|
|
|
* @return bool True if element is allowed, false if it is not
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function allowsElement($name)
|
|
|
|
{
|
2010-09-09 05:14:17 +02:00
|
|
|
if (!empty($this->currentNesting)) {
|
|
|
|
$parent_token = array_pop($this->currentNesting);
|
|
|
|
$this->currentNesting[] = $parent_token;
|
|
|
|
$parent = $this->htmlDefinition->info[$parent_token->name];
|
|
|
|
} else {
|
|
|
|
$parent = $this->htmlDefinition->info_parent_def;
|
|
|
|
}
|
|
|
|
if (!isset($parent->child->elements[$name]) || isset($parent->excludes[$name])) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
// check for exclusion
|
|
|
|
for ($i = count($this->currentNesting) - 2; $i >= 0; $i--) {
|
|
|
|
$node = $this->currentNesting[$i];
|
|
|
|
$def = $this->htmlDefinition->info[$node->name];
|
2016-02-09 11:06:17 +01:00
|
|
|
if (isset($def->excludes[$name])) {
|
|
|
|
return false;
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Iterator function, which starts with the next token and continues until
|
|
|
|
* you reach the end of the input tokens.
|
|
|
|
* @warning Please prevent previous references from interfering with this
|
|
|
|
* functions by setting $i = null beforehand!
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param int $i Current integer index variable for inputTokens
|
|
|
|
* @param HTMLPurifier_Token $current Current token variable.
|
|
|
|
* Do NOT use $token, as that variable is also a reference
|
|
|
|
* @return bool
|
|
|
|
*/
|
|
|
|
protected function forward(&$i, &$current)
|
|
|
|
{
|
|
|
|
if ($i === null) {
|
|
|
|
$i = count($this->inputZipper->back) - 1;
|
|
|
|
} else {
|
|
|
|
$i--;
|
|
|
|
}
|
|
|
|
if ($i < 0) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
$current = $this->inputZipper->back[$i];
|
2010-09-09 05:14:17 +02:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Similar to _forward, but accepts a third parameter $nesting (which
|
|
|
|
* should be initialized at 0) and stops when we hit the end tag
|
|
|
|
* for the node $this->inputIndex starts in.
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param int $i Current integer index variable for inputTokens
|
|
|
|
* @param HTMLPurifier_Token $current Current token variable.
|
|
|
|
* Do NOT use $token, as that variable is also a reference
|
|
|
|
* @param int $nesting
|
|
|
|
* @return bool
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
protected function forwardUntilEndToken(&$i, &$current, &$nesting)
|
|
|
|
{
|
2010-09-09 05:14:17 +02:00
|
|
|
$result = $this->forward($i, $current);
|
2016-02-09 11:06:17 +01:00
|
|
|
if (!$result) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
if ($nesting === null) {
|
|
|
|
$nesting = 0;
|
|
|
|
}
|
|
|
|
if ($current instanceof HTMLPurifier_Token_Start) {
|
|
|
|
$nesting++;
|
|
|
|
} elseif ($current instanceof HTMLPurifier_Token_End) {
|
|
|
|
if ($nesting <= 0) {
|
|
|
|
return false;
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
$nesting--;
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Iterator function, starts with the previous token and continues until
|
|
|
|
* you reach the beginning of input tokens.
|
|
|
|
* @warning Please prevent previous references from interfering with this
|
|
|
|
* functions by setting $i = null beforehand!
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param int $i Current integer index variable for inputTokens
|
|
|
|
* @param HTMLPurifier_Token $current Current token variable.
|
|
|
|
* Do NOT use $token, as that variable is also a reference
|
|
|
|
* @return bool
|
2010-09-09 05:14:17 +02:00
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
protected function backward(&$i, &$current)
|
|
|
|
{
|
|
|
|
if ($i === null) {
|
|
|
|
$i = count($this->inputZipper->front) - 1;
|
|
|
|
} else {
|
|
|
|
$i--;
|
|
|
|
}
|
|
|
|
if ($i < 0) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
$current = $this->inputZipper->front[$i];
|
|
|
|
return true;
|
2010-09-09 05:14:17 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Handler that is called when a text token is processed
|
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function handleText(&$token)
|
|
|
|
{
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Handler that is called when a start or empty token is processed
|
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function handleElement(&$token)
|
|
|
|
{
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Handler that is called when an end token is processed
|
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function handleEnd(&$token)
|
|
|
|
{
|
2010-09-09 05:14:17 +02:00
|
|
|
$this->notifyEnd($token);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Notifier that is called when an end token is processed
|
2016-02-09 11:06:17 +01:00
|
|
|
* @param HTMLPurifier_Token $token Current token variable.
|
2010-09-09 05:14:17 +02:00
|
|
|
* @note This differs from handlers in that the token is read-only
|
|
|
|
* @deprecated
|
|
|
|
*/
|
2016-02-09 11:06:17 +01:00
|
|
|
public function notifyEnd($token)
|
|
|
|
{
|
|
|
|
}
|
2010-09-09 05:14:17 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
// vim: et sw=4 sts=4
|