mirror of
https://github.com/ezyang/htmlpurifier.git
synced 2024-12-23 00:41:52 +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>
139 lines
4.0 KiB
PHP
139 lines
4.0 KiB
PHP
<?php
|
|
|
|
/**
|
|
* General-purpose test-harness that makes testing functions that require
|
|
* configuration and context objects easier when those two parameters are
|
|
* meaningless. See HTMLPurifier_ChildDefTest for a good example of usage.
|
|
*/
|
|
class HTMLPurifier_ComplexHarness extends HTMLPurifier_Harness
|
|
{
|
|
|
|
/**
|
|
* Instance of the object that will execute the method.
|
|
* @type object
|
|
*/
|
|
protected $obj;
|
|
|
|
/**
|
|
* Name of the function to be executed.
|
|
* @type string
|
|
*/
|
|
protected $func;
|
|
|
|
/**
|
|
* Whether or not the method deals in tokens.
|
|
* If set to true, assertResult()
|
|
* will transparently convert HTML to and back from tokens.
|
|
* @type bool
|
|
*/
|
|
protected $to_tokens = false;
|
|
|
|
/**
|
|
* Whether or not the method deals in a node list.
|
|
* If set to true, assertResult() will transparently convert HTML
|
|
* to and back from node.
|
|
* @type bool
|
|
*/
|
|
protected $to_node_list = false;
|
|
|
|
/**
|
|
* Whether or not to convert tokens back into HTML before performing
|
|
* equality check, has no effect on bools.
|
|
* @type bool
|
|
*/
|
|
protected $to_html = false;
|
|
|
|
/**
|
|
* Instance of an HTMLPurifier_Lexer implementation.
|
|
* @type HTMLPurifier_Lexer
|
|
*/
|
|
protected $lexer;
|
|
|
|
public function __construct()
|
|
{
|
|
$this->lexer = new HTMLPurifier_Lexer_DirectLex();
|
|
parent::__construct();
|
|
}
|
|
|
|
/**
|
|
* Asserts a specific result from a one parameter + config/context function
|
|
* @param string $input Input parameter
|
|
* @param bool|string $expect Expectation
|
|
*/
|
|
protected function assertResult($input, $expect = true)
|
|
{
|
|
// $func may cause $input to change, so "clone" another copy
|
|
// to sacrifice
|
|
if ($this->to_node_list && is_string($input)) {
|
|
$input = HTMLPurifier_Arborize::arborize($this->tokenize($temp = $input), $this->config, $this->context)->children;
|
|
$input_c = HTMLPurifier_Arborize::arborize($this->tokenize($temp), $this->config, $this->context)->children;
|
|
} elseif ($this->to_tokens && is_string($input)) {
|
|
$input = $this->tokenize($temp = $input);
|
|
$input_c = $this->tokenize($temp);
|
|
} else {
|
|
$input_c = $input;
|
|
}
|
|
|
|
// call the function
|
|
$func = $this->func;
|
|
$result = $this->obj->$func($input_c, $this->config, $this->context);
|
|
|
|
// test a bool result
|
|
if (is_bool($result)) {
|
|
$this->assertIdentical($expect, $result);
|
|
return;
|
|
} elseif (is_bool($expect)) {
|
|
$expect = $input;
|
|
}
|
|
|
|
if ($this->to_html) {
|
|
if ($this->to_node_list) {
|
|
$result = $this->generateTokens($result);
|
|
if (is_array($expect) && !empty($expect) && $expect[0] instanceof HTMLPurifier_Node) {
|
|
$expect = $this->generateTokens($expect);
|
|
}
|
|
}
|
|
$result = $this->generate($result);
|
|
if (is_array($expect)) {
|
|
$expect = $this->generate($expect);
|
|
}
|
|
}
|
|
$this->assertIdentical($expect, $result);
|
|
|
|
if ($expect !== $result) {
|
|
echo '<pre>' . var_dump($result) . '</pre>';
|
|
}
|
|
|
|
}
|
|
|
|
/**
|
|
* Tokenize HTML into tokens, uses member variables for common variables
|
|
*/
|
|
protected function tokenize($html)
|
|
{
|
|
return $this->lexer->tokenizeHTML($html, $this->config, $this->context);
|
|
}
|
|
|
|
/**
|
|
* Generate textual HTML from tokens
|
|
*/
|
|
protected function generate($tokens)
|
|
{
|
|
$generator = new HTMLPurifier_Generator($this->config, $this->context);
|
|
return $generator->generateFromTokens($tokens);
|
|
}
|
|
|
|
/**
|
|
* Generate tokens from node list
|
|
*/
|
|
protected function generateTokens($children)
|
|
{
|
|
$dummy = new HTMLPurifier_Node_Element("dummy");
|
|
$dummy->children = $children;
|
|
return HTMLPurifier_Arborize::flatten($dummy, $this->context, $this->config);
|
|
}
|
|
|
|
}
|
|
|
|
// vim: et sw=4 sts=4
|