L-electron-Rare/moodle
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

MDL-44516 import htmlpurifier 4.6.0

Browse Source
This commit is contained in:
Petr Škoda
2014-03-14 12:12:56 +08:00
parent 2d9a0b8f39
commit 7529d53da0
357 changed files with 31103 additions and 25445 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/htmlpurifier/HTMLPurifier/Strategy/Composite.php
+30 -23
View File
@@ -1,23 +1,30 @@
<?php
/**
* Composite strategy that runs multiple strategies on tokens.
*/
abstract class HTMLPurifier_Strategy_Composite extends HTMLPurifier_Strategy
{
/**
* List of strategies to run tokens through.
*/
protected $strategies = array();
public function execute($tokens, $config, $context) {
foreach ($this->strategies as $strategy) {
$tokens = $strategy->execute($tokens, $config, $context);
}
return $tokens;
}
}
// vim: et sw=4 sts=4
<?php
/**
* Composite strategy that runs multiple strategies on tokens.
*/
abstract class HTMLPurifier_Strategy_Composite extends HTMLPurifier_Strategy
{
/**
* List of strategies to run tokens through.
* @type HTMLPurifier_Strategy[]
*/
protected $strategies = array();
/**
* @param HTMLPurifier_Token[] $tokens
* @param HTMLPurifier_Config $config
* @param HTMLPurifier_Context $context
* @return HTMLPurifier_Token[]
*/
public function execute($tokens, $config, $context)
{
foreach ($this->strategies as $strategy) {
$tokens = $strategy->execute($tokens, $config, $context);
}
return $tokens;
}
}
// vim: et sw=4 sts=4
lib/htmlpurifier/HTMLPurifier/Strategy/Core.php
+17 -18
View File
@@ -1,18 +1,17 @@
<?php
/**
* Core strategy composed of the big four strategies.
*/
class HTMLPurifier_Strategy_Core extends HTMLPurifier_Strategy_Composite
{
public function __construct() {
$this->strategies[] = new HTMLPurifier_Strategy_RemoveForeignElements();
$this->strategies[] = new HTMLPurifier_Strategy_MakeWellFormed();
$this->strategies[] = new HTMLPurifier_Strategy_FixNesting();
$this->strategies[] = new HTMLPurifier_Strategy_ValidateAttributes();
}
}
// vim: et sw=4 sts=4
<?php
/**
* Core strategy composed of the big four strategies.
*/
class HTMLPurifier_Strategy_Core extends HTMLPurifier_Strategy_Composite
{
public function __construct()
{
$this->strategies[] = new HTMLPurifier_Strategy_RemoveForeignElements();
$this->strategies[] = new HTMLPurifier_Strategy_MakeWellFormed();
$this->strategies[] = new HTMLPurifier_Strategy_FixNesting();
$this->strategies[] = new HTMLPurifier_Strategy_ValidateAttributes();
}
}
// vim: et sw=4 sts=4
lib/htmlpurifier/HTMLPurifier/Strategy/FixNesting.php
+181 -346
View File
@@ -1,346 +1,181 @@
<?php
/**
* Takes a well formed list of tokens and fixes their nesting.
*
* HTML elements dictate which elements are allowed to be their children,
* for example, you can't have a p tag in a span tag. Other elements have
* much more rigorous definitions: tables, for instance, require a specific
* order for their elements. There are also constraints not expressible by
* document type definitions, such as the chameleon nature of ins/del
* tags and global child exclusions.
*
* The first major objective of this strategy is to iterate through all the
* nodes (not tokens) of the list of tokens and determine whether or not
* their children conform to the element's definition. If they do not, the
* child definition may optionally supply an amended list of elements that
* is valid or require that the entire node be deleted (and the previous
* node rescanned).
*
* The second objective is to ensure that explicitly excluded elements of
* an element do not appear in its children. Code that accomplishes this
* task is pervasive through the strategy, though the two are distinct tasks
* and could, theoretically, be seperated (although it's not recommended).
*
* @note Whether or not unrecognized children are silently dropped or
* translated into text depends on the child definitions.
*
* @todo Enable nodes to be bubbled out of the structure.
*
* @warning This algorithm (though it may be hard to see) proceeds from
* a top-down fashion. Thus, parents are processed before
* children. This is easy to implement and has a nice effiency
* benefit, in that if a node is removed, we never waste any
* time processing it, but it also means that if a child
* changes in a non-encapsulated way (e.g. it is removed), we
* need to go back and reprocess the parent to see if those
* changes resulted in problems for the parent. See
* [BACKTRACK] for an example of this. In the current
* implementation, this backtracking can only be triggered when
* a node is removed and if that node was the sole node, the
* parent would need to be removed. As such, it is easy to see
* that backtracking only incurs constant overhead. If more
* sophisticated backtracking is implemented, care must be
* taken to avoid nontermination or exponential blowup.
*/
class HTMLPurifier_Strategy_FixNesting extends HTMLPurifier_Strategy
{
public function execute($tokens, $config, $context) {
//####################################################################//
// Pre-processing
// get a copy of the HTML definition
$definition = $config->getHTMLDefinition();
$excludes_enabled = !$config->get('Core.DisableExcludes');
// insert implicit "parent" node, will be removed at end.
// DEFINITION CALL
$parent_name = $definition->info_parent;
array_unshift($tokens, new HTMLPurifier_Token_Start($parent_name));
$tokens[] = new HTMLPurifier_Token_End($parent_name);
// setup the context variable 'IsInline', for chameleon processing
// is 'false' when we are not inline, 'true' when it must always
// be inline, and an integer when it is inline for a certain
// branch of the document tree
$is_inline = $definition->info_parent_def->descendants_are_inline;
$context->register('IsInline', $is_inline);
// setup error collector
$e =& $context->get('ErrorCollector', true);
//####################################################################//
// Loop initialization
// stack that contains the indexes of all parents,
// $stack[count($stack)-1] being the current parent
$stack = array();
// stack that contains all elements that are excluded
// it is organized by parent elements, similar to $stack,
// but it is only populated when an element with exclusions is
// processed, i.e. there won't be empty exclusions.
$exclude_stack = array();
// variable that contains the start token while we are processing
// nodes. This enables error reporting to do its job
$start_token = false;
$context->register('CurrentToken', $start_token);
//####################################################################//
// Loop
// iterate through all start nodes. Determining the start node
// is complicated so it has been omitted from the loop construct
for ($i = 0, $size = count($tokens) ; $i < $size; ) {
//################################################################//
// Gather information on children
// child token accumulator
$child_tokens = array();
// scroll to the end of this node, report number, and collect
// all children
for ($j = $i, $depth = 0; ; $j++) {
if ($tokens[$j] instanceof HTMLPurifier_Token_Start) {
$depth++;
// skip token assignment on first iteration, this is the
// token we currently are on
if ($depth == 1) continue;
} elseif ($tokens[$j] instanceof HTMLPurifier_Token_End) {
$depth--;
// skip token assignment on last iteration, this is the
// end token of the token we're currently on
if ($depth == 0) break;
}
$child_tokens[] = $tokens[$j];
}
// $i is index of start token
// $j is index of end token
$start_token = $tokens[$i]; // to make token available via CurrentToken
//################################################################//
// Gather information on parent
// calculate parent information
if ($count = count($stack)) {
$parent_index = $stack[$count-1];
$parent_name = $tokens[$parent_index]->name;
if ($parent_index == 0) {
$parent_def = $definition->info_parent_def;
} else {
$parent_def = $definition->info[$parent_name];
}
} else {
// processing as if the parent were the "root" node
// unknown info, it won't be used anyway, in the future,
// we may want to enforce one element only (this is
// necessary for HTML Purifier to clean entire documents
$parent_index = $parent_name = $parent_def = null;
}
// calculate context
if ($is_inline === false) {
// check if conditions make it inline
if (!empty($parent_def) && $parent_def->descendants_are_inline) {
$is_inline = $count - 1;
}
} else {
// check if we're out of inline
if ($count === $is_inline) {
$is_inline = false;
}
}
//################################################################//
// Determine whether element is explicitly excluded SGML-style
// determine whether or not element is excluded by checking all
// parent exclusions. The array should not be very large, two
// elements at most.
$excluded = false;
if (!empty($exclude_stack) && $excludes_enabled) {
foreach ($exclude_stack as $lookup) {
if (isset($lookup[$tokens[$i]->name])) {
$excluded = true;
// no need to continue processing
break;
}
}
}
//################################################################//
// Perform child validation
if ($excluded) {
// there is an exclusion, remove the entire node
$result = false;
$excludes = array(); // not used, but good to initialize anyway
} else {
// DEFINITION CALL
if ($i === 0) {
// special processing for the first node
$def = $definition->info_parent_def;
} else {
$def = $definition->info[$tokens[$i]->name];
}
if (!empty($def->child)) {
// have DTD child def validate children
$result = $def->child->validateChildren(
$child_tokens, $config, $context);
} else {
// weird, no child definition, get rid of everything
$result = false;
}
// determine whether or not this element has any exclusions
$excludes = $def->excludes;
}
// $result is now a bool or array
//################################################################//
// Process result by interpreting $result
if ($result === true || $child_tokens === $result) {
// leave the node as is
// register start token as a parental node start
$stack[] = $i;
// register exclusions if there are any
if (!empty($excludes)) $exclude_stack[] = $excludes;
// move cursor to next possible start node
$i++;
} elseif($result === false) {
// remove entire node
if ($e) {
if ($excluded) {
$e->send(E_ERROR, 'Strategy_FixNesting: Node excluded');
} else {
$e->send(E_ERROR, 'Strategy_FixNesting: Node removed');
}
}
// calculate length of inner tokens and current tokens
$length = $j - $i + 1;
// perform removal
array_splice($tokens, $i, $length);
// update size
$size -= $length;
// there is no start token to register,
// current node is now the next possible start node
// unless it turns out that we need to do a double-check
// this is a rought heuristic that covers 100% of HTML's
// cases and 99% of all other cases. A child definition
// that would be tricked by this would be something like:
// ( | a b c) where it's all or nothing. Fortunately,
// our current implementation claims that that case would
// not allow empty, even if it did
if (!$parent_def->child->allow_empty) {
// we need to do a double-check [BACKTRACK]
$i = $parent_index;
array_pop($stack);
}
// PROJECTED OPTIMIZATION: Process all children elements before
// reprocessing parent node.
} else {
// replace node with $result
// calculate length of inner tokens
$length = $j - $i - 1;
if ($e) {
if (empty($result) && $length) {
$e->send(E_ERROR, 'Strategy_FixNesting: Node contents removed');
} else {
$e->send(E_WARNING, 'Strategy_FixNesting: Node reorganized');
}
}
// perform replacement
array_splice($tokens, $i + 1, $length, $result);
// update size
$size -= $length;
$size += count($result);
// register start token as a parental node start
$stack[] = $i;
// register exclusions if there are any
if (!empty($excludes)) $exclude_stack[] = $excludes;
// move cursor to next possible start node
$i++;
}
//################################################################//
// Scroll to next start node
// We assume, at this point, that $i is the index of the token
// that is the first possible new start point for a node.
// Test if the token indeed is a start tag, if not, move forward
// and test again.
$size = count($tokens);
while ($i < $size and !$tokens[$i] instanceof HTMLPurifier_Token_Start) {
if ($tokens[$i] instanceof HTMLPurifier_Token_End) {
// pop a token index off the stack if we ended a node
array_pop($stack);
// pop an exclusion lookup off exclusion stack if
// we ended node and that node had exclusions
if ($i == 0 || $i == $size - 1) {
// use specialized var if it's the super-parent
$s_excludes = $definition->info_parent_def->excludes;
} else {
$s_excludes = $definition->info[$tokens[$i]->name]->excludes;
}
if ($s_excludes) {
array_pop($exclude_stack);
}
}
$i++;
}
}
//####################################################################//
// Post-processing
// remove implicit parent tokens at the beginning and end
array_shift($tokens);
array_pop($tokens);
// remove context variables
$context->destroy('IsInline');
$context->destroy('CurrentToken');
//####################################################################//
// Return
return $tokens;
}
}
// vim: et sw=4 sts=4
<?php
/**
* Takes a well formed list of tokens and fixes their nesting.
*
* HTML elements dictate which elements are allowed to be their children,
* for example, you can't have a p tag in a span tag. Other elements have
* much more rigorous definitions: tables, for instance, require a specific
* order for their elements. There are also constraints not expressible by
* document type definitions, such as the chameleon nature of ins/del
* tags and global child exclusions.
*
* The first major objective of this strategy is to iterate through all
* the nodes and determine whether or not their children conform to the
* element's definition. If they do not, the child definition may
* optionally supply an amended list of elements that is valid or
* require that the entire node be deleted (and the previous node
* rescanned).
*
* The second objective is to ensure that explicitly excluded elements of
* an element do not appear in its children. Code that accomplishes this
* task is pervasive through the strategy, though the two are distinct tasks
* and could, theoretically, be seperated (although it's not recommended).
*
* @note Whether or not unrecognized children are silently dropped or
* translated into text depends on the child definitions.
*
* @todo Enable nodes to be bubbled out of the structure. This is
* easier with our new algorithm.
*/
class HTMLPurifier_Strategy_FixNesting extends HTMLPurifier_Strategy
{
/**
* @param HTMLPurifier_Token[] $tokens
* @param HTMLPurifier_Config $config
* @param HTMLPurifier_Context $context
* @return array|HTMLPurifier_Token[]
*/
public function execute($tokens, $config, $context)
{
//####################################################################//
// Pre-processing
// O(n) pass to convert to a tree, so that we can efficiently
// refer to substrings
$top_node = HTMLPurifier_Arborize::arborize($tokens, $config, $context);
// get a copy of the HTML definition
$definition = $config->getHTMLDefinition();
$excludes_enabled = !$config->get('Core.DisableExcludes');
// setup the context variable 'IsInline', for chameleon processing
// is 'false' when we are not inline, 'true' when it must always
// be inline, and an integer when it is inline for a certain
// branch of the document tree
$is_inline = $definition->info_parent_def->descendants_are_inline;
$context->register('IsInline', $is_inline);
// setup error collector
$e =& $context->get('ErrorCollector', true);
//####################################################################//
// Loop initialization
// stack that contains all elements that are excluded
// it is organized by parent elements, similar to $stack,
// but it is only populated when an element with exclusions is
// processed, i.e. there won't be empty exclusions.
$exclude_stack = array($definition->info_parent_def->excludes);
// variable that contains the start token while we are processing
// nodes. This enables error reporting to do its job
$node = $top_node;
// dummy token
list($token, $d) = $node->toTokenPair();
$context->register('CurrentNode', $node);
$context->register('CurrentToken', $token);
//####################################################################//
// Loop
// We need to implement a post-order traversal iteratively, to
// avoid running into stack space limits. This is pretty tricky
// to reason about, so we just manually stack-ify the recursive
// variant:
//
// function f($node) {
// foreach ($node->children as $child) {
// f($child);
// }
// validate($node);
// }
//
// Thus, we will represent a stack frame as array($node,
// $is_inline, stack of children)
// e.g. array_reverse($node->children) - already processed
// children.
$parent_def = $definition->info_parent_def;
$stack = array(
array($top_node,
$parent_def->descendants_are_inline,
$parent_def->excludes, // exclusions
0)
);
while (!empty($stack)) {
list($node, $is_inline, $excludes, $ix) = array_pop($stack);
// recursive call
$go = false;
$def = empty($stack) ? $definition->info_parent_def : $definition->info[$node->name];
while (isset($node->children[$ix])) {
$child = $node->children[$ix++];
if ($child instanceof HTMLPurifier_Node_Element) {
$go = true;
$stack[] = array($node, $is_inline, $excludes, $ix);
$stack[] = array($child,
// ToDo: I don't think it matters if it's def or
// child_def, but double check this...
$is_inline || $def->descendants_are_inline,
empty($def->excludes) ? $excludes
: array_merge($excludes, $def->excludes),
0);
break;
}
};
if ($go) continue;
list($token, $d) = $node->toTokenPair();
// base case
if ($excludes_enabled && isset($excludes[$node->name])) {
$node->dead = true;
if ($e) $e->send(E_ERROR, 'Strategy_FixNesting: Node excluded');
} else {
// XXX I suppose it would be slightly more efficient to
// avoid the allocation here and have children
// strategies handle it
$children = array();
foreach ($node->children as $child) {
if (!$child->dead) $children[] = $child;
}
$result = $def->child->validateChildren($children, $config, $context);
if ($result === true) {
// nop
$node->children = $children;
} elseif ($result === false) {
$node->dead = true;
if ($e) $e->send(E_ERROR, 'Strategy_FixNesting: Node removed');
} else {
$node->children = $result;
if ($e) {
// XXX This will miss mutations of internal nodes. Perhaps defer to the child validators
if (empty($result) && !empty($children)) {
$e->send(E_ERROR, 'Strategy_FixNesting: Node contents removed');
} else if ($result != $children) {
$e->send(E_WARNING, 'Strategy_FixNesting: Node reorganized');
}
}
}
}
}
//####################################################################//
// Post-processing
// remove context variables
$context->destroy('IsInline');
$context->destroy('CurrentNode');
$context->destroy('CurrentToken');
//####################################################################//
// Return
return HTMLPurifier_Arborize::flatten($node, $config, $context);
}
}
// vim: et sw=4 sts=4
lib/htmlpurifier/HTMLPurifier/Strategy/MakeWellFormed.php
+600 -532
View File
File diff suppressed because it is too large Load Diff
lib/htmlpurifier/HTMLPurifier/Strategy/RemoveForeignElements.php
+207 -188
View File
@@ -1,188 +1,207 @@
<?php
/**
* Removes all unrecognized tags from the list of tokens.
*
* This strategy iterates through all the tokens and removes unrecognized
* tokens. If a token is not recognized but a TagTransform is defined for
* that element, the element will be transformed accordingly.
*/
class HTMLPurifier_Strategy_RemoveForeignElements extends HTMLPurifier_Strategy
{
public function execute($tokens, $config, $context) {
$definition = $config->getHTMLDefinition();
$generator = new HTMLPurifier_Generator($config, $context);
$result = array();
$escape_invalid_tags = $config->get('Core.EscapeInvalidTags');
$remove_invalid_img = $config->get('Core.RemoveInvalidImg');
// currently only used to determine if comments should be kept
$trusted = $config->get('HTML.Trusted');
$comment_lookup = $config->get('HTML.AllowedComments');
$comment_regexp = $config->get('HTML.AllowedCommentsRegexp');
$check_comments = $comment_lookup !== array() || $comment_regexp !== null;
$remove_script_contents = $config->get('Core.RemoveScriptContents');
$hidden_elements = $config->get('Core.HiddenElements');
// remove script contents compatibility
if ($remove_script_contents === true) {
$hidden_elements['script'] = true;
} elseif ($remove_script_contents === false && isset($hidden_elements['script'])) {
unset($hidden_elements['script']);
}
$attr_validator = new HTMLPurifier_AttrValidator();
// removes tokens until it reaches a closing tag with its value
$remove_until = false;
// converts comments into text tokens when this is equal to a tag name
$textify_comments = false;
$token = false;
$context->register('CurrentToken', $token);
$e = false;
if ($config->get('Core.CollectErrors')) {
$e =& $context->get('ErrorCollector');
}
foreach($tokens as $token) {
if ($remove_until) {
if (empty($token->is_tag) || $token->name !== $remove_until) {
continue;
}
}
if (!empty( $token->is_tag )) {
// DEFINITION CALL
// before any processing, try to transform the element
if (
isset($definition->info_tag_transform[$token->name])
) {
$original_name = $token->name;
// there is a transformation for this tag
// DEFINITION CALL
$token = $definition->
info_tag_transform[$token->name]->
transform($token, $config, $context);
if ($e) $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Tag transform', $original_name);
}
if (isset($definition->info[$token->name])) {
// mostly everything's good, but
// we need to make sure required attributes are in order
if (
($token instanceof HTMLPurifier_Token_Start || $token instanceof HTMLPurifier_Token_Empty) &&
$definition->info[$token->name]->required_attr &&
($token->name != 'img' || $remove_invalid_img) // ensure config option still works
) {
$attr_validator->validateToken($token, $config, $context);
$ok = true;
foreach ($definition->info[$token->name]->required_attr as $name) {
if (!isset($token->attr[$name])) {
$ok = false;
break;
}
}
if (!$ok) {
if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Missing required attribute', $name);
continue;
}
$token->armor['ValidateAttributes'] = true;
}
if (isset($hidden_elements[$token->name]) && $token instanceof HTMLPurifier_Token_Start) {
$textify_comments = $token->name;
} elseif ($token->name === $textify_comments && $token instanceof HTMLPurifier_Token_End) {
$textify_comments = false;
}
} elseif ($escape_invalid_tags) {
// invalid tag, generate HTML representation and insert in
if ($e) $e->send(E_WARNING, 'Strategy_RemoveForeignElements: Foreign element to text');
$token = new HTMLPurifier_Token_Text(
$generator->generateFromToken($token)
);
} else {
// check if we need to destroy all of the tag's children
// CAN BE GENERICIZED
if (isset($hidden_elements[$token->name])) {
if ($token instanceof HTMLPurifier_Token_Start) {
$remove_until = $token->name;
} elseif ($token instanceof HTMLPurifier_Token_Empty) {
// do nothing: we're still looking
} else {
$remove_until = false;
}
if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign meta element removed');
} else {
if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign element removed');
}
continue;
}
} elseif ($token instanceof HTMLPurifier_Token_Comment) {
// textify comments in script tags when they are allowed
if ($textify_comments !== false) {
$data = $token->data;
$token = new HTMLPurifier_Token_Text($data);
} elseif ($trusted || $check_comments) {
// always cleanup comments
$trailing_hyphen = false;
if ($e) {
// perform check whether or not there's a trailing hyphen
if (substr($token->data, -1) == '-') {
$trailing_hyphen = true;
}
}
$token->data = rtrim($token->data, '-');
$found_double_hyphen = false;
while (strpos($token->data, '--') !== false) {
$found_double_hyphen = true;
$token->data = str_replace('--', '-', $token->data);
}
if ($trusted || !empty($comment_lookup[trim($token->data)]) || ($comment_regexp !== NULL && preg_match($comment_regexp, trim($token->data)))) {
// OK good
if ($e) {
if ($trailing_hyphen) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Trailing hyphen in comment removed');
}
if ($found_double_hyphen) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Hyphens in comment collapsed');
}
}
} else {
if ($e) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed');
}
continue;
}
} else {
// strip comments
if ($e) $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed');
continue;
}
} elseif ($token instanceof HTMLPurifier_Token_Text) {
} else {
continue;
}
$result[] = $token;
}
if ($remove_until && $e) {
// we removed tokens until the end, throw error
$e->send(E_ERROR, 'Strategy_RemoveForeignElements: Token removed to end', $remove_until);
}
$context->destroy('CurrentToken');
return $result;
}
}
// vim: et sw=4 sts=4
<?php
/**
* Removes all unrecognized tags from the list of tokens.
*
* This strategy iterates through all the tokens and removes unrecognized
* tokens. If a token is not recognized but a TagTransform is defined for
* that element, the element will be transformed accordingly.
*/
class HTMLPurifier_Strategy_RemoveForeignElements extends HTMLPurifier_Strategy
{
/**
* @param HTMLPurifier_Token[] $tokens
* @param HTMLPurifier_Config $config
* @param HTMLPurifier_Context $context
* @return array|HTMLPurifier_Token[]
*/
public function execute($tokens, $config, $context)
{
$definition = $config->getHTMLDefinition();
$generator = new HTMLPurifier_Generator($config, $context);
$result = array();
$escape_invalid_tags = $config->get('Core.EscapeInvalidTags');
$remove_invalid_img = $config->get('Core.RemoveInvalidImg');
// currently only used to determine if comments should be kept
$trusted = $config->get('HTML.Trusted');
$comment_lookup = $config->get('HTML.AllowedComments');
$comment_regexp = $config->get('HTML.AllowedCommentsRegexp');
$check_comments = $comment_lookup !== array() || $comment_regexp !== null;
$remove_script_contents = $config->get('Core.RemoveScriptContents');
$hidden_elements = $config->get('Core.HiddenElements');
// remove script contents compatibility
if ($remove_script_contents === true) {
$hidden_elements['script'] = true;
} elseif ($remove_script_contents === false && isset($hidden_elements['script'])) {
unset($hidden_elements['script']);
}
$attr_validator = new HTMLPurifier_AttrValidator();
// removes tokens until it reaches a closing tag with its value
$remove_until = false;
// converts comments into text tokens when this is equal to a tag name
$textify_comments = false;
$token = false;
$context->register('CurrentToken', $token);
$e = false;
if ($config->get('Core.CollectErrors')) {
$e =& $context->get('ErrorCollector');
}
foreach ($tokens as $token) {
if ($remove_until) {
if (empty($token->is_tag) || $token->name !== $remove_until) {
continue;
}
}
if (!empty($token->is_tag)) {
// DEFINITION CALL
// before any processing, try to transform the element
if (isset($definition->info_tag_transform[$token->name])) {
$original_name = $token->name;
// there is a transformation for this tag
// DEFINITION CALL
$token = $definition->
info_tag_transform[$token->name]->transform($token, $config, $context);
if ($e) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Tag transform', $original_name);
}
}
if (isset($definition->info[$token->name])) {
// mostly everything's good, but
// we need to make sure required attributes are in order
if (($token instanceof HTMLPurifier_Token_Start || $token instanceof HTMLPurifier_Token_Empty) &&
$definition->info[$token->name]->required_attr &&
($token->name != 'img' || $remove_invalid_img) // ensure config option still works
) {
$attr_validator->validateToken($token, $config, $context);
$ok = true;
foreach ($definition->info[$token->name]->required_attr as $name) {
if (!isset($token->attr[$name])) {
$ok = false;
break;
}
}
if (!$ok) {
if ($e) {
$e->send(
E_ERROR,
'Strategy_RemoveForeignElements: Missing required attribute',
$name
);
}
continue;
}
$token->armor['ValidateAttributes'] = true;
}
if (isset($hidden_elements[$token->name]) && $token instanceof HTMLPurifier_Token_Start) {
$textify_comments = $token->name;
} elseif ($token->name === $textify_comments && $token instanceof HTMLPurifier_Token_End) {
$textify_comments = false;
}
} elseif ($escape_invalid_tags) {
// invalid tag, generate HTML representation and insert in
if ($e) {
$e->send(E_WARNING, 'Strategy_RemoveForeignElements: Foreign element to text');
}
$token = new HTMLPurifier_Token_Text(
$generator->generateFromToken($token)
);
} else {
// check if we need to destroy all of the tag's children
// CAN BE GENERICIZED
if (isset($hidden_elements[$token->name])) {
if ($token instanceof HTMLPurifier_Token_Start) {
$remove_until = $token->name;
} elseif ($token instanceof HTMLPurifier_Token_Empty) {
// do nothing: we're still looking
} else {
$remove_until = false;
}
if ($e) {
$e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign meta element removed');
}
} else {
if ($e) {
$e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign element removed');
}
}
continue;
}
} elseif ($token instanceof HTMLPurifier_Token_Comment) {
// textify comments in script tags when they are allowed
if ($textify_comments !== false) {
$data = $token->data;
$token = new HTMLPurifier_Token_Text($data);
} elseif ($trusted || $check_comments) {
// always cleanup comments
$trailing_hyphen = false;
if ($e) {
// perform check whether or not there's a trailing hyphen
if (substr($token->data, -1) == '-') {
$trailing_hyphen = true;
}
}
$token->data = rtrim($token->data, '-');
$found_double_hyphen = false;
while (strpos($token->data, '--') !== false) {
$found_double_hyphen = true;
$token->data = str_replace('--', '-', $token->data);
}
if ($trusted || !empty($comment_lookup[trim($token->data)]) ||
($comment_regexp !== null && preg_match($comment_regexp, trim($token->data)))) {
// OK good
if ($e) {
if ($trailing_hyphen) {
$e->send(
E_NOTICE,
'Strategy_RemoveForeignElements: Trailing hyphen in comment removed'
);
}
if ($found_double_hyphen) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Hyphens in comment collapsed');
}
}
} else {
if ($e) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed');
}
continue;
}
} else {
// strip comments
if ($e) {
$e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed');
}
continue;
}
} elseif ($token instanceof HTMLPurifier_Token_Text) {
} else {
continue;
}
$result[] = $token;
}
if ($remove_until && $e) {
// we removed tokens until the end, throw error
$e->send(E_ERROR, 'Strategy_RemoveForeignElements: Token removed to end', $remove_until);
}
$context->destroy('CurrentToken');
return $result;
}
}
// vim: et sw=4 sts=4
lib/htmlpurifier/HTMLPurifier/Strategy/ValidateAttributes.php
+45 -39
View File
@@ -1,39 +1,45 @@
<?php
/**
* Validate all attributes in the tokens.
*/
class HTMLPurifier_Strategy_ValidateAttributes extends HTMLPurifier_Strategy
{
public function execute($tokens, $config, $context) {
// setup validator
$validator = new HTMLPurifier_AttrValidator();
$token = false;
$context->register('CurrentToken', $token);
foreach ($tokens as $key => $token) {
// only process tokens that have attributes,
// namely start and empty tags
if (!$token instanceof HTMLPurifier_Token_Start && !$token instanceof HTMLPurifier_Token_Empty) continue;
// skip tokens that are armored
if (!empty($token->armor['ValidateAttributes'])) continue;
// note that we have no facilities here for removing tokens
$validator->validateToken($token, $config, $context);
$tokens[$key] = $token; // for PHP 4
}
$context->destroy('CurrentToken');
return $tokens;
}
}
// vim: et sw=4 sts=4
<?php
/**
* Validate all attributes in the tokens.
*/
class HTMLPurifier_Strategy_ValidateAttributes extends HTMLPurifier_Strategy
{
/**
* @param HTMLPurifier_Token[] $tokens
* @param HTMLPurifier_Config $config
* @param HTMLPurifier_Context $context
* @return HTMLPurifier_Token[]
*/
public function execute($tokens, $config, $context)
{
// setup validator
$validator = new HTMLPurifier_AttrValidator();
$token = false;
$context->register('CurrentToken', $token);
foreach ($tokens as $key => $token) {
// only process tokens that have attributes,
// namely start and empty tags
if (!$token instanceof HTMLPurifier_Token_Start && !$token instanceof HTMLPurifier_Token_Empty) {
continue;
}
// skip tokens that are armored
if (!empty($token->armor['ValidateAttributes'])) {
continue;
}
// note that we have no facilities here for removing tokens
$validator->validateToken($token, $config, $context);
}
$context->destroy('CurrentToken');
return $tokens;
}
}
// vim: et sw=4 sts=4