mirror/php-htmlpurifier
1
0
Fork 0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2025-08-06 14:16:32 +02:00
Code Issues Releases Wiki Activity

Rewrite FixNesting implementation to be tree-based.

Browse Source 
This mega-patch rips out the FixNesting implementation and the related
ChildDef components.  The primary algorithmic change is to convert from
use of tokens to tree nodes, which are far more amenable to the style
of processing that FixNesting uses.  Additionally, FixNesting has been
changed to go bottom-up rather than top-down, in order to avoid needing
to implement backtracking.

This patch simplifies a good deal of the relevant logic, since we no
longer need to continually recalculate the nesting structure when
processing things.  However, the conversion to the alternate format
incurs some overhead, so for small inputs these changes are not a win.
One possibility to greatly reduce the constant factors here is to switch
to entirely using libxml's representation, and never serializing tokens;
this would require one to rewrite injectors, however.

The iterative post-order traversal in FixNesting is a bit subtle, but
we have essentially reified the stack and continuations.

We've removed support for %Core.EscapeInvalidChildren.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
This commit is contained in:
Edward Z. Yang
2013-10-20 22:18:59 -07:00
parent b3640e1af6
commit 0767bbc12d
22 changed files with 358 additions and 698 deletions
Download Patch File Download Diff File Expand all files Collapse all files

71
library/HTMLPurifier/ChildDef/Required.php
View File

@@ -50,30 +50,24 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
public $type = 'required';
/**
* @param array $tokens_of_children
* @param array $children
* @param HTMLPurifier_Config $config
* @param HTMLPurifier_Context $context
* @return array
*/
public function validateChildren($tokens_of_children, $config, $context)
public function validateChildren($children, $config, $context)
{
// Flag for subclasses
$this->whitespace = false;
// if there are no tokens, delete parent node
if (empty($tokens_of_children)) {
if (empty($children)) {
return false;
}
// the new set of children
$result = array();
// current depth into the nest
$nesting = 0;
// whether or not we're deleting a node
$is_deleting = false;
// whether or not parsed character data is allowed
// this controls whether or not we silently drop a tag
// or generate escaped HTML from it
@@ -82,51 +76,33 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
// a little sanity check to make sure it's not ALL whitespace
$all_whitespace = true;
// some configuration
$escape_invalid_children = $config->get('Core.EscapeInvalidChildren');
// generator
$gen = new HTMLPurifier_Generator($config, $context);
foreach ($tokens_of_children as $token) {
if (!empty($token->is_whitespace)) {
$result[] = $token;
$stack = array_reverse($children);
while (!empty($stack)) {
$node = array_pop($stack);
if (!empty($node->is_whitespace)) {
$result[] = $node;
continue;
}
$all_whitespace = false; // phew, we're not talking about whitespace
$is_child = ($nesting == 0);
if ($token instanceof HTMLPurifier_Token_Start) {
$nesting++;
} elseif ($token instanceof HTMLPurifier_Token_End) {
$nesting--;
}
if ($is_child) {
$is_deleting = false;
if (!isset($this->elements[$token->name])) {
$is_deleting = true;
if ($pcdata_allowed && $token instanceof HTMLPurifier_Token_Text) {
$result[] = $token;
} elseif ($pcdata_allowed && $escape_invalid_children) {
$result[] = new HTMLPurifier_Token_Text(
$gen->generateFromToken($token)
);
if (!isset($this->elements[$node->name])) {
// special case text
// XXX One of these ought to be redundant or something
if ($pcdata_allowed && $node instanceof HTMLPurifier_Node_Text) {
$result[] = $node;
continue;
}
// spill the child contents in
// ToDo: Make configurable
if ($node instanceof HTMLPurifier_Node_Element) {
for ($i = count($node->children) - 1; $i >= 0; $i--) {
$stack[] = $node->children[$i];
}
continue;
}
continue;
}
if (!$is_deleting || ($pcdata_allowed && $token instanceof HTMLPurifier_Token_Text)) {
$result[] = $token;
} elseif ($pcdata_allowed && $escape_invalid_children) {
$result[] =
new HTMLPurifier_Token_Text(
$gen->generateFromToken($token)
);
} else {
// drop silently
}
$result[] = $node;
}
if (empty($result)) {
return false;
@@ -135,9 +111,6 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
$this->whitespace = true;
return false;
}
if ($tokens_of_children == $result) {
return true;
}
return $result;
}
}