mirror of
https://github.com/ezyang/htmlpurifier.git
synced 2024-11-14 01:08:41 +00:00
0767bbc12d
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>
87 lines
2.7 KiB
PHP
87 lines
2.7 KiB
PHP
<?php
|
|
|
|
/**
|
|
* Definition for list containers ul and ol.
|
|
*
|
|
* What does this do? The big thing is to handle ol/ul at the top
|
|
* level of list nodes, which should be handled specially by /folding/
|
|
* them into the previous list node. We generally shouldn't ever
|
|
* see other disallowed elements, because the autoclose behavior
|
|
* in MakeWellFormed handles it.
|
|
*/
|
|
class HTMLPurifier_ChildDef_List extends HTMLPurifier_ChildDef
|
|
{
|
|
/**
|
|
* @type string
|
|
*/
|
|
public $type = 'list';
|
|
/**
|
|
* @type array
|
|
*/
|
|
// lying a little bit, so that we can handle ul and ol ourselves
|
|
// XXX: This whole business with 'wrap' is all a bit unsatisfactory
|
|
public $elements = array('li' => true, 'ul' => true, 'ol' => true);
|
|
|
|
/**
|
|
* @param array $children
|
|
* @param HTMLPurifier_Config $config
|
|
* @param HTMLPurifier_Context $context
|
|
* @return array
|
|
*/
|
|
public function validateChildren($children, $config, $context)
|
|
{
|
|
// Flag for subclasses
|
|
$this->whitespace = false;
|
|
|
|
// if there are no tokens, delete parent node
|
|
if (empty($children)) {
|
|
return false;
|
|
}
|
|
|
|
// the new set of children
|
|
$result = array();
|
|
|
|
// a little sanity check to make sure it's not ALL whitespace
|
|
$all_whitespace = true;
|
|
|
|
$current_li = false;
|
|
|
|
foreach ($children as $node) {
|
|
if (!empty($node->is_whitespace)) {
|
|
$result[] = $node;
|
|
continue;
|
|
}
|
|
$all_whitespace = false; // phew, we're not talking about whitespace
|
|
|
|
if ($node->name === 'li') {
|
|
// good
|
|
$current_li = $node;
|
|
$result[] = $node;
|
|
} else {
|
|
// we want to tuck this into the previous li
|
|
// Invariant: we expect the node to be ol/ul
|
|
// ToDo: Make this more robust in the case of not ol/ul
|
|
// by distinguishing between existing li and li created
|
|
// to handle non-list elements; non-list elements should
|
|
// not be appended to an existing li; only li created
|
|
// for non-list. This distinction is not currently made.
|
|
if ($current_li === false) {
|
|
$current_li = new HTMLPurifier_Node_Element('li');
|
|
$result[] = $current_li;
|
|
}
|
|
$current_li->children[] = $node;
|
|
$current_li->empty = false; // XXX fascinating! Check for this error elsewhere ToDo
|
|
}
|
|
}
|
|
if (empty($result)) {
|
|
return false;
|
|
}
|
|
if ($all_whitespace) {
|
|
return false;
|
|
}
|
|
return $result;
|
|
}
|
|
}
|
|
|
|
// vim: et sw=4 sts=4
|