mirrors/htmlpurifier
0
0
Fork 0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2025-03-13 10:07:03 +00:00
Code Releases Activity

Release 2.0.0, merged in 1026 to HEAD.

Browse Source 
git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/branches/strict@1179 48356398-32a2-884e-a903-53898d9a118a
This commit is contained in:
Edward Z. Yang 2007-06-21 00:36:12 +00:00
parent c35eb3e95f
commit 0101311193
172 changed files with 7713 additions and 2520 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Doxyfile
View File

@ -4,7 +4,7 @@
# Project related configuration options
#---------------------------------------------------------------------------
PROJECT_NAME = HTML Purifier
PROJECT_NUMBER = 1.6.1
PROJECT_NUMBER = 2.0.0
OUTPUT_DIRECTORY = "C:/Documents and Settings/Edward/My Documents/My Webs/htmlpurifier/docs/doxygen"
CREATE_SUBDIRS = NO
OUTPUT_LANGUAGE = English

2
INSTALL.fr.utf8
View File

@ -17,7 +17,7 @@ ce document pour quelques choses.
1. Compatibilité
HTML Purifier fonctionne dans PHP 4 et PHP 5. PHP 4.3.9 est le dernier
HTML Purifier fonctionne dans PHP 4 et PHP 5. PHP 4.3.2 est le dernier
version que je le testais. Il ne dépend de les autre librairies.
Les extensions optionnel est iconv (en général déjà installer) et

2
VERSION
View File

@ -1 +1 @@
1.6.1
2.0.0

14
WHATSNEW
View File

@ -1,7 +1,7 @@
The 1.6.1 release, code-named 'Ach! We missed something! Run!', completes
HTML Purifier's roster of attribute transformations. It also implements
a number of minor features (such as better font transformations, smarter
HTML parsing, the CSS property 'white-space' and XHTML 1.1), a few bug
fixes (most notably fixed __autoload compatibility issues) and a ton
of refactoring. 1.6 was for things that absolutely could not wait: this
release, developed in a more leisurely pace, fills in the gaps.
HTML Purifier 2.0 is the culmination of two major architectural changes.
The first is Tidy, which enables HTML Purifier to both natively support
deprecated elements and also convert them to standards-compliant
alternatives. The second is the Advanced API, which enables users to
create new elements and attributes with ease. Keeping in line with a
commitment to high quality, there are also five esoteric bug-fixes and a
plethora of subtle improvements that enhance the library.

4
WYSIWYG
View File

@ -16,7 +16,3 @@ trouble. Therein lies the solution:
HTML Purifier is perfect for filtering pure-HTML input from WYSIWYG editors.
Enough said.
There is a proof-of-concept integration of HTML Purifier with the Mantis
bugtracker at http://hp.jpsband.org/mantis/ You can see notes on how
this integration was acheived at http://hp.jpsband.org/mantis_notes.txt

205
configdoc/generate.php
View File

@ -2,216 +2,37 @@
/**
* Generates XML and HTML documents describing configuration.
* @note PHP 5 only!
*/
/*
TODO:
- make XML format richer (see below)
- make XML format richer (see XMLSerializer_ConfigSchema)
- extend XSLT transformation (see the corresponding XSLT file)
- allow generation of packaged docs that can be easily moved
- multipage documentation
- determine how to multilingualize
- factor out code into classes
- add blurbs to ToC
*/
// ---------------------------------------------------------------------------
// Check and configure environment
if (version_compare('5', PHP_VERSION, '>')) exit('Requires PHP 5 or higher.');
error_reporting(E_ALL);
// ---------------------------------------------------------------------------
// Include HTML Purifier library
error_reporting(E_ALL); // probably not possible to use E_STRICT
// load dual-libraries
require_once '../library/HTMLPurifier.auto.php';
// ---------------------------------------------------------------------------
// Setup convenience functions
function appendHTMLDiv($document, $node, $html) {
global $purifier;
$html = $purifier->purify($html);
$dom_html = $document->createDocumentFragment();
$dom_html->appendXML($html);
$dom_div = $document->createElement('div');
$dom_div->setAttribute('xmlns', 'http://www.w3.org/1999/xhtml');
$dom_div->appendChild($dom_html);
$node->appendChild($dom_div);
}
// ---------------------------------------------------------------------------
// Load copies of HTMLPurifier_ConfigDef and HTMLPurifier
require_once 'library/ConfigDoc.auto.php';
$schema = HTMLPurifier_ConfigSchema::instance();
$purifier = new HTMLPurifier();
$style = 'plain'; // use $_GET in the future
$configdoc = new ConfigDoc();
$output = $configdoc->generate($schema, $style);
// ---------------------------------------------------------------------------
// Generate types.xml, a document describing the constraint "type"
$types_document = new DOMDocument('1.0', 'UTF-8');
$types_root = $types_document->createElement('types');
$types_document->appendChild($types_root);
$types_document->formatOutput = true;
foreach ($schema->types as $name => $expanded_name) {
$types_type = $types_document->createElement('type', $expanded_name);
$types_type->setAttribute('id', $name);
$types_root->appendChild($types_type);
}
$types_document->save('types.xml');
// ---------------------------------------------------------------------------
// Generate configdoc.xml, a document documenting configuration directives
$dom_document = new DOMDocument('1.0', 'UTF-8');
$dom_root = $dom_document->createElement('configdoc');
$dom_document->appendChild($dom_root);
$dom_document->formatOutput = true;
// add the name of the application
$dom_root->appendChild($dom_document->createElement('title', 'HTML Purifier'));
/*
TODO for XML format:
- create a definition (DTD or other) once interface stabilizes
*/
foreach($schema->info as $namespace_name => $namespace_info) {
$dom_namespace = $dom_document->createElement('namespace');
$dom_root->appendChild($dom_namespace);
$dom_namespace->setAttribute('id', $namespace_name);
$dom_namespace->appendChild(
$dom_document->createElement('name', $namespace_name)
);
$dom_namespace_description = $dom_document->createElement('description');
$dom_namespace->appendChild($dom_namespace_description);
appendHTMLDiv($dom_document, $dom_namespace_description,
$schema->info_namespace[$namespace_name]->description);
foreach ($namespace_info as $name => $info) {
if ($info->class == 'alias') continue;
$dom_directive = $dom_document->createElement('directive');
$dom_namespace->appendChild($dom_directive);
$dom_directive->setAttribute('id', $namespace_name . '.' . $name);
$dom_directive->appendChild(
$dom_document->createElement('name', $name)
);
$dom_constraints = $dom_document->createElement('constraints');
$dom_directive->appendChild($dom_constraints);
$dom_type = $dom_document->createElement('type', $info->type);
if ($info->allow_null) {
$dom_type->setAttribute('allow-null', 'yes');
}
$dom_constraints->appendChild($dom_type);
if ($info->allowed !== true) {
$dom_allowed = $dom_document->createElement('allowed');
$dom_constraints->appendChild($dom_allowed);
foreach ($info->allowed as $allowed => $bool) {
$dom_allowed->appendChild(
$dom_document->createElement('value', $allowed)
);
}
}
$raw_default = $schema->defaults[$namespace_name][$name];
if (is_bool($raw_default)) {
$default = $raw_default ? 'true' : 'false';
} elseif (is_string($raw_default)) {
$default = "\"$raw_default\"";
} elseif (is_null($raw_default)) {
$default = 'null';
} else {
$default = print_r(
$schema->defaults[$namespace_name][$name], true
);
}
$dom_default = $dom_document->createElement('default', $default);
// remove this once we get a DTD
$dom_default->setAttribute('xml:space', 'preserve');
$dom_constraints->appendChild($dom_default);
$dom_descriptions = $dom_document->createElement('descriptions');
$dom_directive->appendChild($dom_descriptions);
foreach ($info->descriptions as $file => $file_descriptions) {
foreach ($file_descriptions as $line => $description) {
$dom_description = $dom_document->createElement('description');
$dom_description->setAttribute('file', $file);
$dom_description->setAttribute('line', $line);
appendHTMLDiv($dom_document, $dom_description, $description);
$dom_descriptions->appendChild($dom_description);
}
}
}
}
// print_r($dom_document->saveXML());
// save a copy of the raw XML
$dom_document->save('configdoc.xml');
// ---------------------------------------------------------------------------
// Generate final output using XSLT
// load the stylesheet
$xsl_stylesheet_name = 'plain';
$xsl_stylesheet = "styles/$xsl_stylesheet_name.xsl";
$xsl_dom_stylesheet = new DOMDocument();
$xsl_dom_stylesheet->load($xsl_stylesheet);
// setup the XSLT processor
$xsl_processor = new XSLTProcessor();
// perform the transformation
$xsl_processor->importStylesheet($xsl_dom_stylesheet);
$html_output = $xsl_processor->transformToXML($dom_document);
// some slight fudges to preserve backwards compatibility
$html_output = str_replace('/>', ' />', $html_output); // <br /> not <br/>
$html_output = str_replace(' xmlns=""', '', $html_output); // rm unnecessary xmlns
if (class_exists('Tidy')) {
// cleanup output
$config = array(
'indent' => true,
'output-xhtml' => true,
'wrap' => 80
);
$tidy = new Tidy;
$tidy->parseString($html_output, $config, 'utf8');
$tidy->cleanRepair();
$html_output = (string) $tidy;
}
// write it to a file (todo: parse into seperate pages)
file_put_contents("$xsl_stylesheet_name.html", $html_output);
// ---------------------------------------------------------------------------
// Output for instant feedback
// write out
file_put_contents("$style.html", $output);
if (php_sapi_name() != 'cli') {
echo $html_output;
// output = instant feedback
echo $output;
} else {
echo 'Files generated successfully.';
}

10
configdoc/library/ConfigDoc.auto.php Normal file
View File

@ -0,0 +1,10 @@
<?php
/**
* This is a stub include that automatically configures the include path.
*/
set_include_path(dirname(__FILE__) . PATH_SEPARATOR . get_include_path() );
require_once 'ConfigDoc.php';
?>

39
configdoc/library/ConfigDoc.php Normal file
View File

@ -0,0 +1,39 @@
<?php
require_once 'ConfigDoc/HTMLXSLTProcessor.php';
require_once 'ConfigDoc/XMLSerializer/Types.php';
require_once 'ConfigDoc/XMLSerializer/ConfigSchema.php';
class ConfigDoc
{
function generate($schema, $xsl_stylesheet_name = 'plain', $parameters = array()) {
// generate types document, describing type constraints
$types_serializer = new ConfigDoc_XMLSerializer_Types();
$types_document = $types_serializer->serialize($schema);
$types_document->save(dirname(__FILE__) . '/../types.xml'); // only ONE
// generate configdoc.xml, documents configuration directives
$schema_serializer = new ConfigDoc_XMLSerializer_ConfigSchema();
$schema_document = $schema_serializer->serialize($schema);
$schema_document->save('configdoc.xml');
// setup transformation
$xsl_stylesheet = dirname(__FILE__) . "/../styles/$xsl_stylesheet_name.xsl";
$xslt_processor = new ConfigDoc_HTMLXSLTProcessor();
$xslt_processor->setParameters($parameters);
$xslt_processor->importStylesheet($xsl_stylesheet);
return $xslt_processor->transformToHTML($schema_document);
}
/**
* Remove any generated files
*/
function cleanup() {
unlink('configdoc.xml');
}
}
?>

62
configdoc/library/ConfigDoc/HTMLXSLTProcessor.php Normal file
View File

@ -0,0 +1,62 @@
<?php
/**
* Special XSLTProcessor specifically for HTML documents. Loosely
* based off of XSLTProcessor, but not really
*/
class ConfigDoc_HTMLXSLTProcessor
{
protected $xsltProcessor;
public function __construct() {
$this->xsltProcessor = new XSLTProcessor();
}
/**
* Imports stylesheet for processor to use
* @param $xsl XSLT DOM tree, or filename of the XSL transformation
*/
public function importStylesheet($xsl) {
if (is_string($xsl)) {
$xsl_file = $xsl;
$xsl = new DOMDocument();
$xsl->load($xsl_file);
}
return $this->xsltProcessor->importStylesheet($xsl);
}
/**
* Transforms an XML file into HTML based on the stylesheet
* @param $xml XML DOM tree
*/
public function transformToHTML($xml) {
$out = $this->xsltProcessor->transformToXML($xml);
// fudges for HTML backwards compatibility
$out = str_replace('/>', ' />', $out); // <br /> not <br/>
$out = str_replace(' xmlns=""', '', $out); // rm unnecessary xmlns
if (class_exists('Tidy')) {
// cleanup output
$config = array(
'indent' => true,
'output-xhtml' => true,
'wrap' => 80
);
$tidy = new Tidy;
$tidy->parseString($out, $config, 'utf8');
$tidy->cleanRepair();
$out = (string) $tidy;
}
return $out;
}
public function setParameters($options) {
foreach ($options as $name => $value) {
$this->xsltProcessor->setParameter('', $name, $value);
}
}
}
?>

26
configdoc/library/ConfigDoc/XMLSerializer.php Normal file
View File

@ -0,0 +1,26 @@
<?php
/**
* The XMLSerializer hierarchy of classes consist of classes that take
* objects and serialize them into XML, specifically DOM, form; this
* super-class contains convenience functions for those classes.
*/
class ConfigDoc_XMLSerializer
{
protected function appendHTMLDiv($document, $node, $html) {
$purifier = HTMLPurifier::getInstance();
$html = $purifier->purify($html);
$dom_html = $document->createDocumentFragment();
$dom_html->appendXML($html);
$dom_div = $document->createElement('div');
$dom_div->setAttribute('xmlns', 'http://www.w3.org/1999/xhtml');
$dom_div->appendChild($dom_html);
$node->appendChild($dom_div);
}
}
?>

118
configdoc/library/ConfigDoc/XMLSerializer/ConfigSchema.php Normal file
View File

@ -0,0 +1,118 @@
<?php
require_once 'ConfigDoc/XMLSerializer.php';
class ConfigDoc_XMLSerializer_ConfigSchema extends ConfigDoc_XMLSerializer
{
/**
* Serializes a schema into DOM form
* @todo Split into sub-serializers
* @param $schema HTMLPurifier_ConfigSchema to serialize
*/
public function serialize($schema) {
$dom_document = new DOMDocument('1.0', 'UTF-8');
$dom_root = $dom_document->createElement('configdoc');
$dom_document->appendChild($dom_root);
$dom_document->formatOutput = true;
// add the name of the application
$dom_root->appendChild($dom_document->createElement('title', 'HTML Purifier'));
/*
TODO for XML format:
- create a definition (DTD or other) once interface stabilizes
*/
foreach($schema->info as $namespace_name => $namespace_info) {
$dom_namespace = $dom_document->createElement('namespace');
$dom_root->appendChild($dom_namespace);
$dom_namespace->setAttribute('id', $namespace_name);
$dom_namespace->appendChild(
$dom_document->createElement('name', $namespace_name)
);
$dom_namespace_description = $dom_document->createElement('description');
$dom_namespace->appendChild($dom_namespace_description);
$this->appendHTMLDiv($dom_document, $dom_namespace_description,
$schema->info_namespace[$namespace_name]->description);
foreach ($namespace_info as $name => $info) {
if ($info->class == 'alias') continue;
$dom_directive = $dom_document->createElement('directive');
$dom_namespace->appendChild($dom_directive);
$dom_directive->setAttribute('id', $namespace_name . '.' . $name);
$dom_directive->appendChild(
$dom_document->createElement('name', $name)
);
$dom_constraints = $dom_document->createElement('constraints');
$dom_directive->appendChild($dom_constraints);
$dom_type = $dom_document->createElement('type', $info->type);
if ($info->allow_null) {
$dom_type->setAttribute('allow-null', 'yes');
}
$dom_constraints->appendChild($dom_type);
if ($info->allowed !== true) {
$dom_allowed = $dom_document->createElement('allowed');
$dom_constraints->appendChild($dom_allowed);
foreach ($info->allowed as $allowed => $bool) {
$dom_allowed->appendChild(
$dom_document->createElement('value', $allowed)
);
}
}
$raw_default = $schema->defaults[$namespace_name][$name];
if (is_bool($raw_default)) {
$default = $raw_default ? 'true' : 'false';
} elseif (is_string($raw_default)) {
$default = "\"$raw_default\"";
} elseif (is_null($raw_default)) {
$default = 'null';
} else {
$default = print_r(
$schema->defaults[$namespace_name][$name], true
);
}
$dom_default = $dom_document->createElement('default', $default);
// remove this once we get a DTD
$dom_default->setAttribute('xml:space', 'preserve');
$dom_constraints->appendChild($dom_default);
$dom_descriptions = $dom_document->createElement('descriptions');
$dom_directive->appendChild($dom_descriptions);
foreach ($info->descriptions as $file => $file_descriptions) {
foreach ($file_descriptions as $line => $description) {
$dom_description = $dom_document->createElement('description');
// refuse to write $file if it's a full path
if (str_replace('\\', '/', realpath($file)) != $file) {
$dom_description->setAttribute('file', $file);
$dom_description->setAttribute('line', $line);
}
$this->appendHTMLDiv($dom_document, $dom_description, $description);
$dom_descriptions->appendChild($dom_description);
}
}
}
}
return $dom_document;
}
}
?>

27
configdoc/library/ConfigDoc/XMLSerializer/Types.php Normal file
View File

@ -0,0 +1,27 @@
<?php
require_once 'ConfigDoc/XMLSerializer.php';
class ConfigDoc_XMLSerializer_Types extends ConfigDoc_XMLSerializer
{
/**
* Serializes the types in a schema into DOM form
* @param $schema HTMLPurifier_ConfigSchema owner of types to serialize
*/
public function serialize($schema) {
$types_document = new DOMDocument('1.0', 'UTF-8');
$types_root = $types_document->createElement('types');
$types_document->appendChild($types_root);
$types_document->formatOutput = true;
foreach ($schema->types as $name => $expanded_name) {
$types_type = $types_document->createElement('type', $expanded_name);
$types_type->setAttribute('id', $name);
$types_root->appendChild($types_type);
}
return $types_document;
}
}
?>

28
configdoc/styles/plain.xsl
View File

@ -12,19 +12,21 @@
indent = "no"
media-type = "text/html"
/>
<xsl:param name="css" select="'styles/plain.css'"/>
<xsl:param name="title" select="'Configuration Documentation'"/>
<xsl:variable name="typeLookup" select="document('../types.xml')" />
<xsl:template match="/">
<html lang="en" xml:lang="en">
<head>
<title>Configuration Documentation - <xsl:value-of select="/configdoc/title" /></title>
<title><xsl:value-of select="$title" /> - <xsl:value-of select="/configdoc/title" /></title>
<meta http-equiv="Content-Type" content="text/html;charset=UTF-8" />
<link rel="stylesheet" type="text/css" href="styles/plain.css" />
<link rel="stylesheet" type="text/css" href="{$css}" />
</head>
<body>
<div id="library"><xsl:value-of select="/configdoc/title" /></div>
<h1>Configuration Documentation</h1>
<h1><xsl:value-of select="$title" /></h1>
<h2>Table of Contents</h2>
<ul id="toc">
<xsl:apply-templates mode="toc" />
@ -76,15 +78,17 @@
<table class="constraints">
<xsl:apply-templates />
<!-- Calculated other values -->
<tr>
<th>Used by:</th>
<td>
<xsl:for-each select="../descriptions/description">
<xsl:if test="position()&gt;1">, </xsl:if>
<xsl:value-of select="@file" />
</xsl:for-each>
</td>
</tr>
<xsl:if test="../descriptions/description[@file]">
<tr>
<th>Used by:</th>
<td>
<xsl:for-each select="../descriptions/description">
<xsl:if test="position()&gt;1">, </xsl:if>
<xsl:value-of select="@file" />
</xsl:for-each>
</td>
</tr>
</xsl:if>
</table>
</xsl:template>
<xsl:template match="directive//description">

200
docs/dev-advanced-api.html
View File

@ -17,9 +17,12 @@
<div id="home"><a href="http://htmlpurifier.org/">HTML Purifier</a> End-User Documentation</div>
<p>HTML Purifier currently natively supports only a subset of HTML's
allowed elements, attributes, and behavior. This is by design,
but as the user is always right, they'll need some method to overload
these behaviors.</p>
allowed elements, attributes, and behavior; specifically, this subset
is the set of elements that are safe for untrusted users to use.
However, HTML Purifier is often utilized to ensure standards-compliance
from input that is trusted (making it a sort of Tidy substitute),
and often users need to define new elements or attributes. The
advanced API is oriented specifically for these use-cases.</p>
<p>Our goals are to let the user:</p>
@ -27,20 +30,15 @@ these behaviors.</p>
<dt>Select</dt>
<dd><ul>
<li>Doctype</li>
<li>Mode: Lenient / Correctional</li>
<!-- <li>Filterset</li> -->
<li>Elements / Attributes / Modules</li>
<li>Filterset</li>
<li>Tidy</li>
</ul></dd>
<dt>Customize</dt>
<dd><ul>
<li>Attributes</li>
<li>Elements</li>
</ul></dd>
<dt>Internals</dt>
<dd><ul>
<li>Modules / Elements / Attributes / Attribute Types</li>
<li>Filtersets</li>
<li>Doctype</li>
<!--<li>Doctypes</li>-->
</ul></dd>
</dl>
@ -68,136 +66,64 @@ Transitional, however, we really shouldn't be guessing what the user's
doctype is. Fortunantely, people who can't be bothered to set this won't
be bothered when their pages stop validating.</p>
<h3>Selecting Mode</h3>
<p>Within doctypes, there are various <strong>modes</strong> of operation.
These indicate variant behaviors that, while not strictly changing the
allowed set of elements and attributes, definitely affect the output.
Currently, we have two modes, which may be used together:</p>
<dl>
<dt>Lenient</dt>
<dd>
<p>Deprecated elements and attributes will be transformed into
standards-compliant alternatives when explicitly disallowed.</p>
<p>For example, in the XHTML 1.0 Strict doctype, a <code>center</code>
element would be turned into a <code>div</code> with the CSS property
<code>text-align:center;</code>, but in XHTML 1.0 Transitional
the element would be preserved.</p>
<p>This mode is on by default.</p>
</dd>
<dt>Correctional[items to correct]</dt>
<dd>
<p>Deprecated elements and attributes will be transformed into
standards-compliant alternatives whenever possible.
It may have various levels of operation.</p>
<p>Referring back to the previous example, the <code>center</code> element would
be transformed in both cases. However, elements without a
reasonable standards-compliant alternative will be preserved
in their form.</p>
<p>A user may want to correct certain deprecated attributes, but
not others. For example, the <code>bgcolor</code> attribute may be
acceptable, but the <code>center</code> element not; also, possibly,
an HTML Purifier transformation may be buggy, so the user wants
to forgo it. Thus, correctional accepts an array defining which
elements and attributes to cleanup, or no parameter at all, which
means everything gets corrected. This also means that each
correction needs to be given a unique ID that can be referenced
in this manner. (We may also allow globbing, like *.name or a.*
for mass-enabling correction, and subtractive mode, where things
specified stop correction.) This array gets passed into the
constructor of the mode's module.</p>
<p>This mode is on by default.</p>
</dd>
</dl>
<p>A possible call to select modes would be:</p>
<pre>$config->set('HTML', 'Mode', array('correctional', 'lenient'));</pre>
<p>If modes have extra parameters, a hash is necessary:</p>
<pre>$config->set('HTML', 'Mode', array(
'correctional' => 'center,a.name',
'lenient' => true // this one's just boolean
));</pre>
<p>Modes may be specified along with the doctype declaration (we may want
to get a better set of separator characters):</p>
<pre>$config->setDoctype('XHTML Transitional 1.0', '+correctional[center,a.name] -lenient');</pre>
<p>
With regards to the various levels of operation conjectured in the
Correctional mode, this is prompted by the fact that a user may want to
correct certain problems but not others, for example, fix the <code>center</code>
element but not the <code>u</code> element, both of which are deprecated.
Having an integer <q>level</q> will not work very well for such fine
grained tweaking, but an array of specific settings might.</p>
<h3>Selecting Elements / Attributes / Modules</h3>
<p></p>
<p>HTML Purifier will, by default, allow as many elements and attributes
as possible. However, a user may decide to roll their own filterset by
selecting modules, elements and attributes to allow for their own
specific use-case. This can be done using %HTML.Allowed:</p>
<p>If this cookie cutter approach doesn't appeal to a user, they may
decide to roll their own filterset by selecting modules, elements and
attributes to allow.</p>
<pre>$config->set('HTML', 'Allowed', 'a[href|title],em,p,blockquote');</pre>
<p class="technical">This would make use of the same facilities
as a filterset author would use, except that it would go under an
<q>anonymous</q> filterset that would be auto-selected if any of the
relevant module/elements/attribute selection configuration directives were
non-null.</p>
<p class="technical">The directive %HTML.Allowed is a convenience feature
that may be fully expressed with the legacy interface.</p>
<p>In practice, this is the most commonly demanded feature. Most users are
perfectly happy defining a filterset that looks like:</p>
<pre>$config->setAllowedHTML('a[href,title];em;p;blockquote');</pre>
<p class="technical">The directive %HTML.Allowed is a convenience function
that may be fully expressed with the legacy interface, and thus is
given its own setter.</p>
<p>We currently support a separated interface, which also must be preserved:</p>
<p>We currently support another interface from older versions:</p>
<pre>$config->set('HTML', 'AllowedElements', 'a,em,p,blockquote');
$config->set('HTML', 'AllowedAttributes', 'a.href,a.title');</pre>
<p>A user may also choose to allow modules:</p>
<p>A user may also choose to allow modules using a specialized
directive:</p>
<pre>$config->set('HTML', 'AllowedModules', 'Hypertext,Text,Lists'); // or
$config->setAllowedHTML('Hypertext,Text,Lists');</pre>
<pre>$config->set('HTML', 'AllowedModules', 'Hypertext,Text,Lists');</pre>
<p>But it is not expected that this feature will be widely used.</p>
<p class="fixme">The granularity of these modules is too coarse for
the average user (for example, the core module loads everything from
the essential <code>p</code> element to the not-so-safe <code>h1</code>
element). How do we make this still a viable solution? Possible answers
may be sub-modules or module parameters. This may not even be a problem,
considering that most people won't be selecting modules.</p>
<p class="technical">Module selection will work slightly differently
from the other AllowedElements and AllowedAttributes directives by
directly modifying the doctype you are operating in, in the spirit of
XHTML 1.1's modularization. We stop users from shooting themselves in the
foot by mandating the modules in %HTML.CoreModules be used.</p>
<p class="technical">Modules are distinguished from regular elements by the
case of their first letter. While XML distinguishes between and allows
lower and uppercase letters in element names, most well-known XML
languages use only lower-case
lower and uppercase letters in element names, XHTML uses only lower-case
element names for sake of consistency.</p>
<p class="technical">Considering that, internally speaking, as mandated by
the XHTML 1.1 Modularization specification, we have organized our
elements around modules, considerable gymnastics will be needed to
get this sort of functionality working.</p>
<h3>Selecting Tidy</h3>
<p>The name of this segment of functionality is inspired off of Dave
Ragget's program HTML Tidy, which purported to help clean up HTML. In
HTML Purifier, Tidy functionality involves turning unsupported and
deprecated elements into standards-compliant ones, maintaining
backwards compatibility, and enforcing best practices.</p>
<p>This is a complicated feature, and is explained more in depth at
<a href="enduser-tidy.html">the Tidy documentation page</a>.</p>
<!--
<h3>Unified selector</h3>
<p>Because selecting each and every one of these configuration options
is a chore, we may wish to offer a specialized configuration method
for selecting a filterset. Possibility:</p>
<pre>function selectFilter($doctype, $filterset, $mode)</pre>
<pre>function selectFilter($doctype, $filterset, $tidy)</pre>
<p>...which is simply a light wrapper over the individual configuration
calls. A custom config file format or text format could also be adopted.</p>
-->
<h2>Customize</h2>
@ -209,38 +135,34 @@ use-cases.</p>
<p>Note that the functions described here are only available if
a raw copy of <code>HTMLPurifier_HTMLDefinition</code> was retrieved.
<code>addAttribute</code> may work on a processed copy, but for
consistency's sake we will mandate this for everything.</p>
Furthermore, caching may prevent your changes from immediately
being seen: consult <a href="enduser-customize.html">enduser-customize.html</a> on how
to work around this.</p>
<h3>Attributes</h3>
<p>An attribute is bound to an element by a name and has a specific
<code>AttrDef</code> that validates it. Thus, the interface should
be:</p>
<code>AttrDef</code> that validates it. The interface is therefore:</p>
<pre>function addAttribute($element, $attribute, $attribute_def);</pre>
<p>With a use-case that looks like:</p>
<p>Example of the functionality in action:</p>
<pre>$def->addAttribute('a', 'rel', new HTMLPurifier_AttrDef_Enum(array('nofollow')));</pre>
<pre>$def->addAttribute('a', 'rel', 'Enum#nofollow');</pre>
<p>The <code>$attribute_def</code> value can be a little flexible,
to make things simpler. We'll let it also be:</p>
<p>The <code>$attribute_def</code> value is flexible,
to make things simpler. It can be a literal object or:</p>
<ul>
<li>Class name: We'll instantiate it for you</li>
<!--<li>Class name: We'll instantiate it for you</li>
<li>Function name: We'll create an <code>HTMLPurifier_AttrDef_Anonymous</code>
class with that function registered as a callback.</li>
class with that function registered as a callback.</li>-->
<li>String attribute type: We'll use <code>HTMLPurifier_AttrTypes</code>
</li>
<li>String starting with <code>enum(</code>: We'll explode it and stuff it in an
<code>HTMLPurifier_AttrDef_Enum</code> for you.</li>
to resolve it for you. Any data that follows a hash mark (#) will
be used to customize the attribute type: in the example above,
we specify which values for Enum to allow.</li>
</ul>
<p>Making the previous example written as:</p>
<pre>$def->addAttribute('a', 'rel', 'enum(nofollow)');</pre>
<h3>Elements</h3>
<p>An element requires certain information as specified by
@ -255,7 +177,8 @@ the usual things required are:</p>
<p>This suggests an API like this:</p>
<pre>function addElement($element, $type, $content_model, $attributes = array());</pre>
<pre>function addElement($element, $type, $contents,
$attr_collections = array(); $attributes = array());</pre>
<p>Each parameter explained in depth:</p>
@ -264,11 +187,15 @@ the usual things required are:</p>
<dd>Element name, ex. 'label'</dd>
<dt><code>$type</code></dt>
<dd>Content set to register in, ex. 'Inline' or 'Flow'</dd>
<dt><code>$content_model</code></dt>
<dt><code>$contents</code></dt>
<dd>Description of allowed children. This is a merged form of
<code>HTMLPurifier_ElementDef</code>'s member variables
<code>$content_model</code> and <code>$content_model_type</code>,
where the form is <q>Type: Model</q>, ex. 'Optional: Inline'.</dd>
where the form is <q>Type: Model</q>, ex. 'Optional: Inline'.
There are also a number of predefined templates one may use.</dd>
<dt><code>$attr_collections</code></dt>
<dd>Array (or string if only one) of attribute collection(s) to
merge into the attributes array.</dd>
<dt><code>$attributes</code></dt>
<dd>Array of attribute names to attribute definitions, much like
the above-described attribute customization.</dd>
@ -276,11 +203,10 @@ the usual things required are:</p>
<p>A possible usage:</p>
<pre>$def->addElement('font', 'Inline', 'Optional: Inline',
array(0 => array('Common'), 'color' => 'Color'));</pre>
<pre>$def->addElement('font', 'Inline', 'Optional: Inline', 'Common',
array('color' => 'Color'));</pre>
<p>We may want to Common attribute collection inclusion to be added
by default.</p>
<p>See <code>HTMLPurifier/HTMLModule.php</code> for details.</p>
<div id="version">$Id$</div>

791
docs/enduser-customize.html Normal file
View File

@ -0,0 +1,791 @@
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"><head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<meta name="description" content="Tutorial for customizing HTML Purifier's tag and attribute sets." />
<link rel="stylesheet" type="text/css" href="style.css" />
<title>Customize - HTML Purifier</title>
</head><body>
<h1 class="subtitled">Customize!</h1>
<div class="subtitle">HTML Purifier is a Swiss-Army Knife</div>
<div id="filing">Filed under End-User</div>
<div id="index">Return to the <a href="index.html">index</a>.</div>
<div id="home"><a href="http://htmlpurifier.org/">HTML Purifier</a> End-User Documentation</div>
<div id="applicability">
This document covers currently unreleased functionality and
only applies to recent SVN checkouts.
</div>
<p>
You may have heard of the <a href="dev-advanced-api.html">Advanced API</a>.
If you're interested in reading dry prose and boring functional
specifications, feel free to click that link to get a no-nonsense overview
on the Advanced API. For the rest of us, there's this tutorial. By the time
you're finished reading this, you should have a pretty good idea on
how to implement custom tags and attributes that HTML Purifier may not have.
</p>
<h2>Is it necessary?</h2>
<p>
Before we even write any code, it is paramount to consider whether or
not the code we're writing is necessary or not. HTML Purifier, by default,
contains a large set of elements and attributes: large enough so that
<em>any</em> element or attribute in XHTML 1.0 (and its HTML variant)
that can be safely used by the general public is implemented.
</p>
<p>
So what needs to be implemented? (Feel free to skip this section if
you know what you want).
</p>
<h3>XHTML 1.0</h3>
<p>
All of the modules listed below are based off of the
<a href="http://www.w3.org/TR/2001/REC-xhtml-modularization-20010410/abstract_modules.html#sec_5.2.">modularization of
XHTML</a>, which, while technically for XHTML 1.1, is quite a useful
resource.
</p>
<ul>
<li>Structure</li>
<li>Frames</li>
<li>Applets (deprecated)</li>
<li>Forms</li>
<li>Image maps</li>
<li>Objects</li>
<li>Frames</li>
<li>Events</li>
<li>Meta-information</li>
<li>Style sheets</li>
<li>Link (not hypertext)</li>
<li>Base</li>
<li>Name</li>
</ul>
<p>
If you don't recognize it, you probably don't need it. But the curious
can look all of these modules up in the above-mentioned document. Note
that inline scripting comes packaged with HTML Purifier (more on this
later).
</p>
<h3>XHTML 1.1</h3>
<p>
We have not implemented the
<a href="http://www.w3.org/TR/2001/REC-ruby-20010531/">Ruby module</a>,
which defines a set of tags
for publishing short annotations for text, used mostly in Japanese
and Chinese school texts.
</p>
<h3>XHTML 2.0</h3>
<p>
<a href="http://www.w3.org/TR/xhtml2/">XHTML 2.0</a> is still a
working draft, so any elements introduced in the
specification have not been implemented and will not be implemented
until we get a recommendation or proposal. Because XHTML 2.0 is
an entirely new markup language, implementing rules for it will be
no easy task.
</p>
<h3>HTML 5</h3>
<p>
<a href="http://www.whatwg.org/specs/web-apps/current-work/">HTML 5</a>
is a fork of HTML 4.01 by WHATWG, who believed that XHTML 2.0 was headed
in the wrong direction. It too is a working draft, and may change
drastically before publication, but it should be noted that the
<code>canvas</code> tag has been implemented by many browser vendors.
</p>
<h3>Proprietary</h3>
<p>
There are a number of proprietary tags still in the wild. Many of them
have been documented in <a href="ref-proprietary-tags.txt">ref-proprietary-tags.txt</a>,
but there is currently no implementation for any of them.
</p>
<h3>Extensions</h3>
<p>
There are also a number of other XML languages out there that can
be embedded in HTML documents: two of the most popular are MathML and
SVG, and I frequently get requests to implement these. But they are
expansive, comprehensive specifications, and it would take far too long
to implement them <em>correctly</em> (most systems I've seen go as far
as whitelisting tags and no further; come on, what about nesting!)
</p>
<p>
Word of warning: HTML Purifier is currently <em>not</em> namespace
aware.
</p>
<h2>Giving back</h2>
<p>
As you may imagine from the details above (don't be abashed if you didn't
read it all: a glance over would have done), there's quite a bit that
HTML Purifier doesn't implement. Recent architectural changes have
allowed HTML Purifier to implement elements and attributes that are not
safe! Don't worry, they won't be activated unless you set %HTML.Trusted
to true, but they certainly help out users who need to put, say, forms
on their page and don't want to go through the trouble of reading this
and implementing it themself.
</p>
<p>
So any of the above that you implement for your own application could
help out some other poor sap on the other side of the globe. Help us
out, and send back code so that it can be hammered into a module and
released with the core. Any code would be greatly appreciated!
</p>
<h2>And now...</h2>
<p>
Enough philosophical talk, time for some code:
</p>
<pre>$config = HTMLPurifier_Config::createDefault();
$config->set('HTML', 'DefinitionID', 'enduser-customize.html tutorial');
$config->set('HTML', 'DefinitionRev', 1);
$def =& $config->getHTMLDefinition(true);</pre>
<p>
Assuming that HTML Purifier has already been properly loaded (hint:
include <code>HTMLPurifier.auto.php</code>), this code will set up
the environment that you need to start customizing the HTML definition.
What's going on?
</p>
<ul>
<li>
The first three lines are regular configuration code:
<ul>
<li>
%HTML.DefinitionID is set to a unique identifier for your
custom HTML definition. This prevents it from clobbering
other custom definitions on the same installation.
</li>
<li>
%HTML.DefinitionRev is a revision integer of your HTML
definition. Because HTML definitions are cached, you'll need
to increment this whenever you make a change in order to flush
the cache.
</li>
</ul>
</li>
<li>
The fourth line retrieves a raw <code>HTMLPurifier_HTMLDefinition</code>
object that we will be tweaking. If the parameter was removed, we
would be retrieving a fully formed definition object, which is somewhat
useless for customization purposes.
</li>
</ul>
<h3>Broken backwards-compatibility</h3>
<p>
Those of you who have already been twiddling around with the raw
HTML definition object, you'll be noticing that you're getting an error
when you attempt to retrieve the raw definition object without specifying
a DefinitionID. It is vital to caching (see below) that you make a unique
name for your customized definition, so make up something right now and
things will operate again.
</p>
<h2>Turn off caching</h2>
<p>
To make development easier, we're going to temporarily turn off
definition caching:
</p>
<pre>$config = HTMLPurifier_Config::createDefault();
$config->set('HTML', 'DefinitionID', 'enduser-customize.html tutorial');
$config->set('HTML', 'DefinitionRev', 1);
<strong>$config->set('Core', 'DefinitionCache', null); // remove this later!</strong>
$def =& $config->getHTMLDefinition(true);</pre>
<p>
A few things should be mentioned about the caching mechanism before
we move on. For performance reasons, HTML Purifier caches generated
<code>HTMLPurifier_Definition</code> objects in serialized files
stored (by default) in <code>library/HTMLPurifier/DefinitionCache/Serializer</code>.
A lot of processing is done in order to create these objects, so it
makes little sense to repeat the same processing over and over again
whenever HTML Purifier is called.
</p>
<p>
In order to identify a cache entry, HTML Purifier uses three variables:
the library's version number, the value of %HTML.DefinitionRev and
a serial of relevant configuration. Whenever any of these changes,
a new HTML definition is generated. Notice that there is no way
for the definition object to track changes to customizations: here, it
is up to you to supply appropriate information to DefinitionID and
DefinitionRev.
</p>
<h2 id="addAttribute">Add an attribute</h2>
<p>
For this example, we're going to implement the <code>target</code> attribute found
on <code>a</code> elements. To implement an attribute, we have to
ask a few questions:
</p>
<ol>
<li>What element is it found on?</li>
<li>What is its name?</li>
<li>Is it required or optional?</li>
<li>What are valid values for it?</li>
</ol>
<p>
The first three are easy: the element is <code>a</code>, the attribute
is <code>target</code>, and it is not a required attribute. (If it
was required, we'd need to append an asterisk to the attribute name,
you'll see an example of this in the addElement() example).
</p>
<p>
The last question is a little trickier.
Lets allow the special values: _blank, _self, _target and _top.
The form of this is called an <strong>enumeration</strong>, a list of
valid values, although only one can be used at a time. To translate
this into code form, we write:
</p>
<pre>$config = HTMLPurifier_Config::createDefault();
$config->set('HTML', 'DefinitionID', 'enduser-customize.html tutorial');
$config->set('HTML', 'DefinitionRev', 1);
$config->set('Core', 'DefinitionCache', null); // remove this later!
$def =& $config->getHTMLDefinition(true);
<strong>$def->addAttribute('a', 'target', 'Enum#_blank,_self,_target,_top');</strong></pre>
<p>
The <code>Enum#_blank,_self,_target,_top</code> does all the magic.
The string is split into two parts, separated by a hash mark (#):
</p>
<ol>
<li>The first part is the name of what we call an <code>AttrDef</code></li>
<li>The second part is the parameter of the above-mentioned <code>AttrDef</code></li>
</ol>
<p>
If that sounds vague and generic, it's because it is! HTML Purifier defines
an assortment of different attribute types one can use, and each of these
has their own specialized parameter format. Here are some of the more useful
ones:
</p>
<table class="table">
<thead>
<tr>
<th>Type</th>
<th>Format</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<tr>
<th>Enum</th>
<td><em>[s:]</em>value1,value2,...</td>
<td>
Attribute with a number of valid values, one of which may be used. When
s: is present, the enumeration is case sensitive.
</td>
</tr>
<tr>
<th>Bool</th>
<td>attribute_name</td>
<td>
Boolean attribute, with only one valid value: the name
of the attribute.
</td>
</tr>
<tr>
<th>CDATA</th>
<td></td>
<td>
Attribute of arbitrary text. Can also be referred to as <strong>Text</strong>
(the specification makes a semantic distinction between the two).
</td>
</tr>
<tr>
<th>ID</th>
<td></td>
<td>
Attribute that specifies a unique ID
</td>
</tr>
<tr>
<th>Pixels</th>
<td></td>
<td>
Attribute that specifies an integer pixel length
</td>
</tr>
<tr>
<th>Length</th>
<td></td>
<td>
Attribute that specifies a pixel or percentage length
</td>
</tr>
<tr>
<th>NMTOKENS</th>
<td></td>
<td>
Attribute that specifies a number of name tokens, example: the
<code>class</code> attribute
</td>
</tr>
<tr>
<th>URI</th>
<td></td>
<td>
Attribute that specifies a URI, example: the <code>href</code>
attribute
</td>
</tr>
<tr>
<th>Number</th>
<td></td>
<td>
Attribute that specifies an positive integer number
</td>
</tr>
</tbody>
</table>
<p>
For a complete list, consult
<a href="http://htmlpurifier.org/svnroot/htmlpurifier/trunk/library/HTMLPurifier/AttrTypes.php"><code>library/HTMLPurifier/AttrTypes.php</code></a>;
more information on attributes that accept parameters can be found on their
respective includes in
<a href="http://htmlpurifier.org/svnroot/htmlpurifier/trunk/library/HTMLPurifier/AttrDef/"><code>library/HTMLPurifier/AttrDef</code></a>.
</p>
<p>
Sometimes, the restrictive list in AttrTypes just doesn't cut it. Don't
sweat: you can also use a fully instantiated object as the value. The
equivalent, verbose form of the above example is:
</p>
<pre>$config = HTMLPurifier_Config::createDefault();
$config->set('HTML', 'DefinitionID', 'enduser-customize.html tutorial');
$config->set('HTML', 'DefinitionRev', 1);
$config->set('Core', 'DefinitionCache', null); // remove this later!
$def =& $config->getHTMLDefinition(true);
<strong>$def->addAttribute('a', 'target', new HTMLPurifier_AttrDef_Enum(
array('_blank','_self','_target','_top')
));</strong></pre>
<p>
Trust me, you'll learn to love the shorthand.
</p>
<h2>Add an element</h2>
<p>
Adding attributes is really small-fry stuff, though, and it was possible
to add them (albeit a bit more wordy) prior to 2.0. The real gem of
the Advanced API is adding elements. There are five questions to
ask when adding a new element:
</p>
<ol>
<li>What is the element's name?</li>
<li>What content set does this element belong to?</li>
<li>What are the allowed children of this element?</li>
<li>What attributes does the element allow that are general?</li>
<li>What attributes does the element allow that are specific to this element?</li>
</ol>
<p>
It's a mouthful, and you'll be slightly lost if your not familiar with
the HTML specification, so let's explain them step by step.
</p>
<h3>Content set</h3>
<p>
The HTML specification defines two major content sets: Inline
and Block. Each of these
content sets contain a list of elements: Inline contains things like
<code>span</code> and <code>b</code> while Block contains things like
<code>div</code> and <code>blockquote</code>.
</p>
<p>
These content sets amount to a macro mechanism for HTML definition. Most
elements in HTML are organized into one of these two sets, and most
elements in HTML allow elements from one of these sets. If we had
to write each element verbatim into each other element's allowed
children, we would have ridiculously large lists; instead we use
content sets to compactify the declaration.
</p>
<p>
Practically speaking, there are several useful values you can use here:
</p>
<table class="table">
<thead>
<tr>
<th>Content set</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<tr>
<th>Inline</th>
<td>Character level elements, text</td>
</tr>
<tr>
<th>Block</th>
<td>Block-like elements, like paragraphs and lists</td>
</tr>
<tr>
<th><em>false</em></th>
<td>
Any element that doesn't fit into the mold, for example <code>li</code>
or <code>tr</code>
</td>
</tr>
</tbody>
</table>
<p>
By specifying a valid value here, all other elements that use that
content set will also allow your element, without you having to do
anything. If you specify <em>false</em>, you'll have to register
your element manually.
</p>
<h3>Allowed children</h3>
<p>
Allowed children defines the elements that this element can contain.
The allowed values may range from none to a complex regexp depending on
your element.
</p>
<p>
If you've ever taken a look at the HTML DTD's before, you may have
noticed declarations like this:
</p>
<pre>&lt;!ELEMENT LI - O (%flow;)* -- list item --&gt;</pre>
<p>
The <code>(%flow;)*</code> indicates the allowed children of the
<code>li</code> tag: <code>li</code> allows any number of flow
elements as its children. In HTML Purifier, we'd write it like
<code>Flow</code> (here's where the content sets we were
discussing earlier come into play). There are three shorthand content models you
can specify:
</p>
<table class="table">
<thead>
<tr>
<th>Content model</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<tr>
<th>Empty</th>
<td>No children allowed, like <code>br</code> or <code>hr</code></td>
</tr>
<tr>
<th>Inline</th>
<td>Any number of inline elements and text, like <code>span</code></td>
</tr>
<tr>
<th>Flow</th>
<td>Any number of inline elements, block elements and text, like <code>div</code></td>
</tr>
</tbody>
</table>
<p>
This covers 90% of all the cases out there, but what about elements that
break the mold like <code>ul</code>? This guy requires at least one
child, and the only valid children for it are <code>li</code>. The
content model is: <code>Required: li</code>. There are two parts: the
first type determines what <code>ChildDef</code> will be used to validate
content models. The most common values are:
</p>
<table class="table">
<thead>
<tr>
<th>Type</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<tr>
<th>Required</th>
<td>Children must be one or more of the valid elements</td>
</tr>
<tr>
<th>Optional</th>
<td>Children can be any number of the valid elements</td>
</tr>
<tr>
<th>Custom</th>
<td>Children must follow the DTD-style regex</td>
</tr>
</tbody>
</table>
<p>
You can also implement your own <code>ChildDef</code>: this was done
for a few special cases in HTML Purifier such as <code>Chameleon</code>
(for <code>ins</code> and <code>del</code>), <code>StrictBlockquote</code>
and <code>Table</code>.
</p>
<p>
The second part specifies either valid elements or a regular expression.
Valid elements are separated with horizontal bars (|), i.e.
"<code>a | b | c</code>". Use #PCDATA to represent plain text.
Regular expressions are based off of DTD's style:
</p>
<ul>
<li>Parentheses () are used for grouping</li>
<li>Commas (,) separate elements that should come one after another</li>
<li>Horizontal bars (|) indicate one or the other elements should be used</li>
<li>Plus signs (+) are used for a one or more match</li>
<li>Asterisks (*) are used for a zero or more match</li>
<li>Question marks (?) are used for a zero or one match</li>
</ul>
<p>
For example, "<code>a, b?, (c | d), e+, f*</code>" means "In this order,
one <code>a</code> element, at most one <code>b</code> element,
one <code>c</code> or <code>d</code> element (but not both), one or more
<code>e</code> elements, and any number of <code>f</code> elements."
Regex veterans should be able to jump right in, and those not so savvy
can always copy-paste W3C's content model definitions into HTML Purifier
and hope for the best.
</p>
<p>
A word of warning: while the regex format is extremely flexible on
the developer's side, it is
quite unforgiving on the user's side. If the user input does not <em>exactly</em>
match the specification, the entire contents of the element will
be nuked. This is why there is are specific content model types like
Optional and Required: while they could be implemented as <code>Custom:
(valid | elements)*</code>, the custom classes contain special recovery
measures that make sure as much of the user's original content gets
through. HTML Purifier's core, as a rule, does not use Custom.
</p>
<p>
One final note: you can also use Content Sets inside your valid elements
lists or regular expressions. In fact, the three shorthand content models
mentioned above are just that: abbreviations:
</p>
<table class="table">
<thead>
<tr>
<th>Content model</th>
<th>Implementation</th>
</tr>
</thead>
<tbody>
<tr>
<th>Inline</th>
<td>Optional: Inline | #PCDATA</td>
</tr>
<tr>
<th>Flow</th>
<td>Optional: Flow | #PCDATA</td>
</tr>
</tbody>
</table>
<p>
When the definition is compiled, Inline will be replaced with a
horizontal-bar separated list of inline elements. Also, notice that
it does not contain text: you have to specify that yourself.
</p>
<h3>Common attributes</h3>
<p>
Congratulations: you have just gotten over the proverbial hump (Allowed
children). Common attributes is much simpler, and boils down to
one question: does your element have the <code>id</code>, <code>style</code>,
<code>class</code>, <code>title</code> and <code>lang</code> attributes?
If so, you'll want to specify the <code>Common</code> attribute collection,
which contains these five attributes that are found on almost every
HTML element in the specification.
</p>
<p>
There are a few more collections, but they're really edge cases:
</p>
<table class="table">
<thead>
<tr>
<th>Collection</th>
<th>Attributes</th>
</tr>
</thead>
<tbody>
<tr>
<th>I18N</th>
<td><code>lang</code>, possibly <code>xml:lang</code></td>
</tr>
<tr>
<th>Core</th>
<td><code>style</code>, <code>class</code>, <code>id</code> and <code>title</code></td>
</tr>
</tbody>
</table>
<p>
Common is a combination of the above-mentioned collections.
</p>
<h3>Attributes</h3>
<p>
If you didn't read the <a href="#addAttribute">previous section on
adding attributes</a>, read it now. The last parameter is simply
array of attribute names to attribute implementations, in the exact
same format as <code>addAttribute()</code>.
</p>
<h3>Putting it all together</h3>
<p>
We're going to implement <code>form</code>. Before we embark, lets
grab a reference implementation from over at the
<a href="http://www.w3.org/TR/html4/sgml/loosedtd.html">transitional DTD</a>:
</p>
<pre>&lt;!ELEMENT FORM - - (%flow;)* -(FORM) -- interactive form --&gt;
&lt;!ATTLIST FORM
%attrs; -- %coreattrs, %i18n, %events --
action %URI; #REQUIRED -- server-side form handler --
method (GET|POST) GET -- HTTP method used to submit the form--
enctype %ContentType; &quot;application/x-www-form-urlencoded&quot;
accept %ContentTypes; #IMPLIED -- list of MIME types for file upload --
name CDATA #IMPLIED -- name of form for scripting --
onsubmit %Script; #IMPLIED -- the form was submitted --
onreset %Script; #IMPLIED -- the form was reset --
target %FrameTarget; #IMPLIED -- render in this frame --
accept-charset %Charsets; #IMPLIED -- list of supported charsets --
&gt;</pre>
<p>
Juicy! With just this, we can answer four of our five questions:
</p>
<ol>
<li>What is the element's name? <strong>form</strong></li>
<li>What content set does this element belong to? <strong>Block</strong>
(this needs a little sleuthing, I find the easiest way is to search
the DTD for <code>FORM</code> and determine which set it is in.)</li>
<li>What are the allowed children of this element? <strong>One
or more flow elements, but no nested <code>form</code>s</strong></li>
<li>What attributes does the element allow that are general? <strong>Common</strong></li>
<li>What attributes does the element allow that are specific to this element? <strong>A whole bunch, see ATTLIST;
we're going to the vital ones: <code>action</code>, <code>method</code> and <code>name</code></strong></li>
</ol>
<p>
Time for some code:
</p>
<pre>$config = HTMLPurifier_Config::createDefault();
$config->set('HTML', 'DefinitionID', 'enduser-customize.html tutorial');
$config->set('HTML', 'DefinitionRev', 1);
$config->set('Core', 'DefinitionCache', null); // remove this later!
$def =& $config->getHTMLDefinition(true);
$def->addAttribute('a', 'target', new HTMLPurifier_AttrDef_Enum(
array('_blank','_self','_target','_top')
));
<strong>$form =& $def->addElement(
'form', // name
'Block', // content set
'Flow', // allowed children
'Common', // attribute collection
array( // attributes
'action*' => 'URI',
'method' => 'Enum#get|post',
'name' => 'ID'
)
);
$form->excludes = array('form' => true);</strong></pre>
<p>
Each of the parameters corresponds to one of the questions we asked.
Notice that we added an asterisk to the end of the <code>action</code>
attribute to indicate that it is required. If someone specifies a
<code>form</code> without that attribute, the tag will be axed.
Also, the extra line at the end is a special extra declaration that
prevents forms from being nested within each other.
</p>
<p>
And that's all there is to it! Implementing the rest of the form
module is left as an exercise to the user; to see more examples
check the <a href="http://htmlpurifier.org/svnroot/htmlpurifier/trunk/library/HTMLPurifier/HTMLModule/"><code>library/HTMLPurifier/HTMLModule/</code></a> directory
in your local HTML Purifier installation.
</p>
<h2>And beyond...</h2>
<p>
Perceptive users may have realized that, to a certain extent, we
have simply re-implemented the facilities of XML Schema or the
Document Type Definition. What you are seeing here, however, is
not just an XML Schema or Document Type Definition: it is a fully
expressive method of specifying the definition of HTML that is
a portable superset of the capabilities of the two above-mentioned schema
languages. What makes HTMLDefinition so powerful is the fact that
if we don't have an implementation for a content model or an attribute
definition, you can supply it yourself by writing a PHP class.
</p>
<p>
There are many facets of HTMLDefinition beyond the Advanced API I have
walked you through today. To find out more about these, you can
check out these source files:
</p>
<ul>
<li><a href="http://htmlpurifier.org/svnroot/htmlpurifier/trunk/library/HTMLPurifier/HTMLModule.php"><code>library/HTMLPurifier/HTMLModule.php</code></a></li>
<li><a href="http://htmlpurifier.org/svnroot/htmlpurifier/trunk/library/HTMLPurifier/ElementDef.php"><code>library/HTMLPurifier/ElementDef.php</code></a></li>
</ul>
<div id="version">$Id: enduser-tidy.html 1158 2007-06-18 19:26:29Z Edward $</div>
</body></html>

10
docs/enduser-security.txt
View File

@ -8,15 +8,11 @@ to be effective. Things to remember:
1. Character Encoding: see enduser-utf8.html for more info.
2. Doctype: document pending feature completion
Not strictly necessary, actually. More in-depth discussion once we figure
out how to get strict loose mode working.
2. IDs: see enduser-id.html for more info
3. IDs: see enduser-id.html for more info
4. Links: document pending feature completion
3. Links: document pending feature completion
Rudimentary blacklisting, we should also allow only relative URIs. We
need a doc to explain the stuff.
5. CSS: document pending
4. CSS: document pending
Explain which CSS styles we blocked and why.

235
docs/enduser-tidy.html Normal file
View File

@ -0,0 +1,235 @@
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"><head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<meta name="description" content="Tutorial for tweaking HTML Purifier's Tidy-like behavior." />
<link rel="stylesheet" type="text/css" href="style.css" />
<title>Tidy - HTML Purifier</title>
</head><body>
<h1>Tidy</h1>
<div id="filing">Filed under Development</div>
<div id="index">Return to the <a href="index.html">index</a>.</div>
<div id="home"><a href="http://htmlpurifier.org/">HTML Purifier</a> End-User Documentation</div>
<div id="applicability">
This document covers currently unreleased functionality and
only applies to recent SVN checkouts.
</div>
<p>You've probably heard of HTML Tidy, Dave Raggett's little piece
of software that cleans up poorly written HTML. Let me say it straight
out:</p>
<p class="emphasis">This ain't HTML Tidy!</p>
<p>Rather, Tidy stands for a cool set of Tidy-inspired in HTML Purifier
that allows users to submit deprecated elements and attributes and get
valid strict markup back. For example:</p>
<pre>&lt;center&gt;Centered&lt;/center&gt;</pre>
<p>...becomes:</p>
<pre>&lt;div style=&quot;text-align:center;&quot;&gt;Centered&lt;/div&gt;</pre>
<p>...when this particular fix is run on the HTML. This tutorial will give
you down the lowdown of what exactly HTML Purifier will do when Tidy
is on, and how to fine tune this behavior. Once again, <strong>you do
not need Tidy installed on your PHP to use these features!</strong></p>
<h2>What does it do?</h2>
<p>Tidy will do several things to your HTML:</p>
<ul>
<li>Convert deprecated elements and attributes to standards-compliant
alternatives</li>
<li>Enforce XHTML compatibility guidelines and other best practices</li>
<li>Preserve data that would normally be removed as per W3C</li>
</ul>
<h2>What are levels?</h2>
<p>Levels describe how aggressive the Tidy module should be when
cleaning up HTML. There are four levels to pick: none, light, medium
and heavy. Each of these levels has a well-defined set of behavior
associated with it, although it may change depending on your doctype.</p>
<dl>
<dt>light</dt>
<dd>This is the <strong>lenient</strong> level. If a tag or attribute
is about to be removed because it isn't supported by the
doctype, Tidy will step in and change into an alternative that
is supported.</dd>
<dt>medium</dt>
<dd>This is the <strong>correctional</strong> level. At this level,
all the functions of light are performed, as well as some extra,
non-essential best practices enforcement. Changes made on this
level are very benign and are unlikely to cause problems.</dd>
<dt>heavy</dt>
<dd>This is the <strong>aggressive</strong> level. If a tag or
attribute is deprecated, it will be converted into a non-deprecated
version, no ifs ands or buts.</dd>
</dl>
<p>By default, Tidy operates on the <strong>medium</strong> level. You can
change the level of cleaning by setting the %HTML.TidyLevel configuration
directive:</p>
<pre>$config->set('HTML', 'TidyLevel', 'heavy'); // burn baby burn!</pre>
<h2>Is the light level really light?</h2>
<p>It depends on what doctype you're using. If your documents are HTML
4.01 <em>Transitional</em>, HTML Purifier will be lazy
and won't clean up your <code>center</code>
or <code>font</code> tags. But if you're using HTML 4.01 <em>Strict</em>,
HTML Purifier has no choice: it has to convert them, or they will
be nuked out of existence. So while light on Transitional will result
in little to no changes, light on Strict will still result in quite
a lot of fixes.</p>
<p>This is different behavior from 1.6 or before, where deprecated
tags in transitional documents would
always be cleaned up regardless. This is also better behavior.</p>
<h2>My pages look different!</h2>
<p>HTML Purifier is tasked with converting deprecated tags and
attributes to standards-compliant alternatives, which usually
need copious amounts of CSS. It's also not foolproof: sometimes
things do get lost in the translation. This is why when HTML Purifier
can get away with not doing cleaning, it won't; this is why
the default value is <strong>medium</strong> and not heavy.</p>
<p>Fortunately, only a few attributes have problems with the switch
over. They are described below:</p>
<table class="table">
<thead><tr>
<th>Element@Attr</th>
<th>Changes</th>
</tr></thead>
<tbody>
<tr>
<td>caption@align</td>
<td>Firefox supports stuffing the caption on the
left and right side of the table, a feature that
Internet Explorer, understandably, does not have.
When align equals right or left, the text will simply
be aligned on the left or right side.</td>
</tr>
<tr>
<td>img@align</td>
<td>The implementation for align bottom is good, but not
perfect. There are a few pixel differences.</td>
</tr>
<tr>
<td>br@clear</td>
<td>Clear both gets a little wonky in Internet Explorer. Haven't
really been able to figure out why.</td>
</tr>
<tr>
<td>hr@noshade</td>
<td>All browsers implement this slightly differently: we've
chosen to make noshade horizontal rules gray.</td>
</tr>
</tbody>
</table>
<p>There are a few more minor, although irritating, bugs.
Some older browsers support deprecated attributes,
but not CSS. Transformed elements and attributes will look unstyled
to said browsers. Also, CSS precedence is slightly different for
inline styles versus presentational markup. In increasing precedence:</p>
<ol>
<li>Presentational attributes</li>
<li>External style sheets</li>
<li>Inline styling</li>
</ol>
<p>This means that styling that may have been masked by external CSS
declarations will start showing up (a good thing, perhaps). Finally,
if you've turned off the style attribute, almost all of
these transformations will not work. Sorry mates.</p>
<p>You can review the rendering before and after of these transformations
by consulting the <a
href="http://htmlpurifier.org/live/smoketests/attrTransform.php">attrTransform.php
smoketest</a>.</p>
<h2>I like the general idea, but the specifics bug me!</h2>
<p>So you want HTML Purifier to clean up your HTML, but you're not
so happy about the br@clear implementation. That's perfectly fine!
HTML Purifier will make accomodations:</p>
<pre>$config->set('HTML', 'Doctype', 'XHTML 1.0 Transitional');
$config->set('HTML', 'TidyLevel', 'heavy'); // all changes, minus...
<strong>$config->set('HTML', 'TidyRemove', 'br@clear');</strong></pre>
<p>That third line does the magic, removing the br@clear fix
from the module, ensuring that <code>&lt;br clear="both" /&gt;</code>
will pass through unharmed. The reverse is possible too:</p>
<pre>$config->set('HTML', 'Doctype', 'XHTML 1.0 Transitional');
$config->set('HTML', 'TidyLevel', 'none'); // no changes, plus...
<strong>$config->set('HTML', 'TidyAdd', 'p@align');</strong></pre>
<p>In this case, all transformations are shut off, except for the p@align
one, which you found handy.</p>
<p>To find out what the names of fixes you want to turn on or off are,
you'll have to consult the source code, specifically the files in
<code>HTMLPurifier/HTMLModule/Tidy/</code>. There is, however, a
general syntax:</p>
<table class="table">
<thead>
<tr>
<th>Name</th>
<th>Example</th>
<th>Interpretation</th>
</tr>
</thead>
<tbody>
<tr>
<td>element</td>
<td>font</td>
<td>Tag transform for <em>element</em></td>
</tr>
<tr>
<td>element@attr</td>
<td>br@clear</td>
<td>Attribute transform for <em>attr</em> on <em>element</em></td>
</tr>
<tr>
<td>@attr</td>
<td>@lang</td>
<td>Global attribute transform for <em>attr</em></td>
</tr>
<tr>
<td>e#content_model_type</td>
<td>blockquote#content_model_type</td>
<td>Change of child processing implementation for <em>e</em></td>
</tr>
</tbody>
</table>
<h2>So... what's the lowdown?</h2>
<p>The lowdown is, quite frankly, HTML Purifier's default settings are
probably good enough. The next step is to bump the level up to heavy,
and if that still doesn't satisfy your appetite, do some fine tuning.
Other than that, don't worry about it: this all works silently and
effectively in the background.</p>
<div id="version">$Id$</div>
</body></html>

4
docs/examples/basic.php
View File

@ -8,8 +8,8 @@ require_once '../../library/HTMLPurifier.auto.php';
$config = HTMLPurifier_Config::createDefault();
// configuration goes here:
$config->set('Core', 'Encoding', 'ISO-8859-1'); //replace with your encoding
$config->set('Core', 'XHTML', true); // set to false if HTML 4.01
$config->set('Core', 'Encoding', 'UTF-8'); // replace with your encoding
$config->set('HTML', 'Doctype', 'XHTML 1.0 Transitional'); // replace with your doctype
$purifier = new HTMLPurifier($config);

20
docs/index.html
View File

@ -34,6 +34,12 @@ information for casual developers using HTML Purifier.</p>
<dt><a href="enduser-utf8.html">UTF-8: The Secret of Character Encoding</a></dt>
<dd>Describes the rationale for using UTF-8, the ramifications otherwise, and how to make the switch.</dd>
<dt><a href="enduser-tidy.html">Tidy</a></dt>
<dd>Tutorial for tweaking HTML Purifier's Tidy-like behavior.</dd>
<dt><a href="enduser-customize.html">Customize</a></dt>
<dd>Tutorial for customizing HTML Purifier's tag and attribute sets.</dd>
</dl>
<h2>Development</h2>
@ -128,8 +134,8 @@ the code. They may be upgraded to HTML files or stay as TXT scratchpads.</p>
<tr>
<td>Reference</td>
<td><a href="ref-loose-vs-strict.txt">Loose vs.Strict</a></td>
<td>Differences between HTML Strict and Transitional versions.</td>
<td><a href="ref-content-models.txt">Handling Content Model Changes</a></td>
<td>Discusses how to tidy up content model changes using custom ChildDef classes.</td>
</tr>
<tr>
@ -140,14 +146,8 @@ the code. They may be upgraded to HTML files or stay as TXT scratchpads.</p>
<tr>
<td>Reference</td>
<td><a href="ref-strictness.txt">Strictness</a></td>
<td>Short essay on how loose definition isn't really loose.</td>
</tr>
<tr>
<td>Reference</td>
<td><a href="ref-xhtml-1.1.txt">XHTML 1.1</a></td>
<td>What we'd have to do to support XHTML 1.1.</td>
<td><a href="ref-html-modularization.txt">Modularization of HTMLDefinition</a></td>
<td>Provides a high-level overview of the concepts behind HTMLModules.</td>
</tr>
<tr>

31
docs/proposal-config.txt
View File

@ -12,29 +12,10 @@ the documentation in ConfigDef for more information on these namespaces.
Since configuration is dependant on context, internal classes require a
configuration object to be passed as a parameter. (They also require a
Context object).
Context object). A majority of classes do not need the config object,
but for those who do, it is a lifesaver.
In relation to HTMLDefinition and CSSDefinition, there could be a special class
of directives that influence the *construction* of the Definition object.
A theoretical call pattern would look like:
1. Client calls Config->getHTMLDefinition()
2. Config calls HTMLDefinition->createNew(this)
3. HTMLDefinition constructs itself with base configuration
4. HTMLDefinition calls Config->get('HTML')
5. Config returns array of directives
6. HTMLDefinition performs operations and changes specified by directives
7. HTMLPurifier returns constructed definition
8. Config caches definition so it doesn't have to be generated again
9. Config returns definition
You could also override Config's copy of the definition with your own
custom copy, which OVERRIDES all directives. Only the base, vanilla copy
is the Singleton, the object actually interfaced with is a operated-upon
clone of that object. Also, if an update to the directives would update
the definition, you'd have to force reconstruction.
In practice, the pulling directives from the config object are
solely need-based, and the flex points are littered throughout the
setup() function. Some sort of refactoring is likely in order. See
ref-xhtml-1.1.txt for more info.
Definition objects are complex datatypes influenced by their respective
directive namespaces (HTMLDefinition with HTML and CSSDefinition with CSS).
If any of these directives is updated, HTML Purifier forces the definition
to be regenerated.

34
docs/proposal-filter-levels.txt
View File

@ -2,23 +2,16 @@
Filter Levels
When one size *does not* fit all
The more I think about it, the less sense it makes for maintaining one huge
monolithic HTMLDefinition class. There's simply so much variation that
could go into this definition: the set of HTML good for blog entries is
definitely too large for HTML that would be allowed in blog comments. Going
from Transitional to Strict requires changes to the definition.
It makes little sense to constrain users to one set of HTML elements and
attributes and tell them that they are not allowed to mold this in
any fashion. Many users demand to be able to custom-select which elements
and attributes they want. This is fine: because HTML Purifier keeps close
track of what elements are safe to use, there is no way for them to
accidently allow an XSS-able tag.
Allowing users to specify their own whitelists is one step (implemented, btw),
but I have doubts on only doing this. Simply put, the typical programmer is too
lazy to actually go through the trouble of investigating which tags, attributes
and properties to allow. HTMLDefinition makes a big part of what HTMLPurifier
is.
The idea, then, is to setup fundamentally different set of definitions, which
can further be customized using simpler configuration options. Alternatively,
they could be implemented as configuration profiles, which simply load
a set of recommended directives to acheive a desired affect (no simpler
config options though).
However, combing through the HTML spec to make your own whitelist can
be a daunting task. HTML Purifier ought to offer pre-canned filter levels
that amateur users can select based on what they think is their use-case.
Here are some fuzzy levels you could set:
@ -46,6 +39,10 @@ make forbidden element to text transformations desirable (for example, images).
== Element Risk Analysis ==
Although none of the currently supported elements presents a security
threat per-say, some can cause problems for page layouts or be
extremely complicated.
Legend:
[danger level] - regular tags / uncommon tags ~ deprecated tags
[danger level]* - rare tags
@ -114,6 +111,10 @@ Partially presentational - table.cellpadding, table.cellspacing,
== CSS Risk Analysis ==
Currently, there is no support for fine-grained "allowed CSS" specification,
mainly because I'm lazy, partially because no one has asked for it. However,
this will be added eventually.
There are certain CSS elements that are extremely useful inline, but then
as you get to more presentation oriented styling it may not always be
appropriate to inline them.
@ -126,6 +127,7 @@ any CSS properties that are not currently implemented (such as position).
Dangerous, can go outside container - float
Easy to abuse - font-size, font-family (font), width
Colored - background-color (background), border-color (border), color
(see proposal-colors.html)
Dramatic - border, list-style-position (list-style), margin, padding,
text-align, text-indent, text-transform, vertical-align, line-height

48
docs/ref-content-models.txt Normal file
View File

@ -0,0 +1,48 @@
Handling Content Model Changes
1. Context
The distinction between Transitional and Strict document types is somewhat
of an anomaly in the lineage of XHTML document types (following 1.0, no
doctypes do not have flavors: instead, modularization is used to let
document authors vary their elements). This transition is usually quite
straight-forward, as W3C usually deprecates attributes or elements, which
are quite easily handled using tag and attribute transforms.
However, for two elements, <blockquote>, <body> and <address>, W3C elected
to also change the content model. <blockquote> and <body> originally
accepted both inline and block elements, but in the strict doctype they
only allow block elements. With <address>, the situation is inverted:
<p> tags were now forbidden from appearing within this tag.
2. Current situation
Currently, HTML Purifier treats <blockquote> specially during Tidy mode
using a custom ChildDef class StrictBlockquote. StrictBlockquote
operates similarly to Required, except that when it encounters an inline
element, it will wrap it in a block tag (as specified by
%HTML.BlockWrapper, the default is <p>). The naming suggests it can
only be used for <blockquote>s, although it may be possible to
genericize it to work on other cases of this nature (this would be of
little practical application, as no other element in XHTML 1.1 or earlier
has a block-only content model).
Tidy currently contains no custom, lenient implementation for <address>.
If one were to be written, it would likely operate on the principle that,
when a <p> tag were to be encountered, it would be replaced with a
leading and trailing <br /> tag (the contents of <p>, being inline, are
not an issue). There is no prior work with this sort of operation.
3. Outside applicability
There are a number of other elements that contain restrictive content
models, such as <ul> or <span> (the latter is restrictive in that it
does not allow block elements). In the former case, an errant node
is eliminated completely, in the latter case, the text of the node
would is preserved (as the parent node does allow PCDATA). Custom
content model implementations probably are not the best way of handling
these cases, instead, node bubbling should be implemented instead.

41
docs/ref-xhtml-1.1.txt → docs/ref-html-modularization.txt
View File

@ -1,10 +1,8 @@
XHTML 1.1 and HTML Purifier
The Modularization of HTMLDefinition in HTML Purifier
Todo for XHTML 1.1 support <http://www.w3.org/TR/xhtml11/changes.html>
1. Scratch lang entirely in favor of xml:lang
2. Scratch name entirely in favor of id (partially-done)
3. Support Ruby <http://www.w3.org/TR/2001/REC-ruby-20010531/>
1. Support Ruby <http://www.w3.org/TR/2001/REC-ruby-20010531/>
HTML Purifier uses the modularization of XHTML
<http://www.w3.org/TR/xhtml-modularization/> to organize the internals
@ -12,25 +10,10 @@ of HTMLDefinition into a more manageable and extensible fashion. Rather
than have one super-object, HTMLDefinition is split into HTMLModules,
each of which are responsible for defining elements, their attributes,
and other properties (for a more indepth coverage, see
/library/HTMLPurifier/HTMLModule.php's docblock comments).
/library/HTMLPurifier/HTMLModule.php's docblock comments). These modules
are managed by HTMLModuleManager.
The modules that W3C defines and we support are:
* 5.1. Attribute Collections (technically not a module
* 5.2. Core Modules
o 5.2.2. Text Module
o 5.2.3. Hypertext Module
o 5.2.4. List Module
* 5.4. Text Extension Modules
o 5.4.1. Presentation Module
o 5.4.2. Edit Module
o 5.4.3. Bi-directional Text Module
* 5.6. Table Modules
o 5.6.2. Tables Module
* 5.7. Image Module
* 5.18. Style Attribute Module
Modules that we don't support but coul support are:
Modules that we don't support but could support are:
* 5.6. Table Modules
o 5.6.1. Basic Tables Module [?]
@ -38,10 +21,8 @@ Modules that we don't support but coul support are:
* 5.9. Server-side Image Map Module [?]
* 5.12. Target Module [?]
* 5.21. Name Identification Module [deprecated]
* 5.22. Legacy Module [deprecated]
These modules will not be implemented due to their dangerousness or
inapplicability as an XHTML fragment:
These modules would be implemented as "unsafe":
* 5.2. Core Modules
o 5.2.1. Structure Module
@ -64,11 +45,7 @@ of robust tools for handling them (the main problem is that all the
current parsers are usually PHP 5 only and solely-validating, not
correcting).
The abstraction of the HTMLDefinition creation process will also
contribute to a need for a caching system. Cache invalidation would be
difficult, but could be done by comparing the HTML and Attr config
namespaces with a copy that was packaged along with the serialized
HTMLDefinition object.
This system may be generalized and ported over for CSS.
== General Use-Case ==
@ -91,7 +68,7 @@ like this:
<?php
$config = HTMLPurifier_Config::createDefault();
$def =& $config->getHTMLDefinition(true); // reference to raw
unset($def->modules['Hypertext']); // rm ''a'' link
$def->addElement('marquee', 'Block', 'Flow', 'Common');
$purifier = new HTMLPurifier($config);
$purifier->purify($html); // now the definition is finalized
?>
@ -184,4 +161,4 @@ Content sets can be altered using HTMLModule->content_sets, an associative
array of content set names to content set contents. If the content set
already exists, your values are appended on to it (great for, say,
registering the font tag as an inline element), otherwise it is
created. They are substituted into content_model.
created. They are substituted into content_model.

37
docs/ref-loose-vs-strict.txt
View File

@ -1,37 +0,0 @@
Loose versus Strict
Changes from one doctype to another
There are changes. Wow, how insightful. Not everything changed is relevant
to HTML Purifier, though, so let's take a look:
== Major incompatibilities ==
[done] BLOCKQUOTE changes from 'flow' to 'block'
current behavior: inline inner contents should not be nuked, block-ify as necessary
[partially-done] U, S, STRIKE cut
current behavior: removed completely
projected behavior: replace with appropriate inline span + CSS
[done] ADDRESS from potpourri to Inline (removes p tags)
current behavior: block tags silently dropped
ideal behavior: replace tags with something like <br>. (not high priority)
== Things we can loosen up ==
Tags DIR, MENU, CENTER, ISINDEX, FONT, BASEFONT? allowed in loose
current behavior: transform to strict-valid forms
Attributes allowed in loose (see attribute transforms in 'dev-progress.html')
current behavior: projected to transform into strict-valid forms
== Periphery issues ==
A tag's attribute 'target' (for selecting frames) cut
current behavior: not allowed at all
projected behavior: use loose doctype if needed, needs valid values
[done] OL/LI tag's attribute 'start'/'value' (for renumbering lists) cut
current behavior: no substitute, just delete when in strict, allow in loose
Attribute 'name' deprecated in favor of 'id'
current behavior: dropped silently
projected behavior: create proper AttrTransform
[done] PRE tag allows SUB/SUP? (strict dtd comment vs syntax, loose disallows)
current behavior: disallow as usual

4
docs/ref-proprietary-tags.txt
View File

@ -18,5 +18,7 @@ HTML Purifier context.
<listing>, monospace pre-variant (extremely rare)
<plaintext>, escapes all tags to the end of document
<ruby> and friends, (more research needed, appears to be XHTML 1.1 markup)
<xmp>, monospace, replace with pre
These should be put into their own Tidy module, not loaded by default(?). These
all qualify as "lenient" transforms.

37
docs/ref-strictness.txt
View File

@ -1,37 +0,0 @@
Is HTML Purifier Strict or Transitional?
A little bit of helpful guidance
Despite the fact that HTML Purifier professes to support both transitional and
strict HTML, it rejects a lot of attributes and elements that are actually, indeed,
valid. You can investigate progress.html to find out precisely what we
are doing to these *deprecated* attributes.
However, users have found that Strict HTML imposes some quite unreasonable
restrictions on certain things. The start and value attributes in ol and
li (respectively) perhaps are the most contested. There's is currently no
widely supported browser method short of JavaScript that can replace these
two deprecated elements. It behooves us to allow these deprecated
attributes when the output is transitional.
Fortunantely, that's the only real bugger case. The others have near-perfect
CSS equivalents, and were presentational anyway. However, the other question
pops up: should we always convert these to the CSS forms when 1. the spec
allows them anyway and 2. older browsers support them better? After all, the
whole point about CSS is to seperate styling from content, so inline styling
doesn't solve that problem.
It's an icky question, and we'll have to deal with it as more and more
transforms get implemented. As of right now, however, we currently support
these loose-only constructs in loose mode:
- <ul start="1">, <li value="1"> attributes
- <u>, <strike>, <s> tags
- flow children in <blockquote>
- mixed children in <address>
The changed child definitions as well as the ul.start li.value are the most
compelling reasons why loose should be used. We may want offer disabling <u>,
<strike> and <s> by themselves. We may also want to offer no pre-emptive
deprecated conversions. This all must be unified.

23
docs/ref-whatwg.txt
View File

@ -2,8 +2,23 @@
Web Hypertext Application Technology Working Group
WHATWG
I don't think we need to worry about them. Untrusted users shouldn't be
submitting applications, eh? But if some interesting attribute pops up in
their spec, and might be worth supporting, stick it here.
== HTML 5 ==
(none so far, as you can see)
URL: http://www.whatwg.org/specs/web-apps/current-work/
HTML 5 defines a kaboodle of new elements and attributes, as well as
some well-defined, "quirks mode" HTML parsing. Although WHATWG professes
to be targeted towards web applications, many of their semantic additions
would be quite useful in regular documents. Eventually, HTML
Purifier will need to audit their lists and figure out what changes need
to be made. This process is complicated by the fact that the WHATWG
doesn't buy into W3C's modularization of XHTML 1.1: we may need
to remodularize HTML 5 (probably done by section name). No sense in
committing ourselves till the spec stabilizes, though.
More immediately speaking though, however, is the well-defined parsing
behavior that HTML 5 adds. While I have little interest in writing
another DirectLex parser, other parsers like ph5p
<http://jero.net/lab/ph5p/> can be adapted to DOMLex to support much more
flexible HTML parsing (a cool feature I've seen is how they resolve
<b>bold<i>both</b>italic</i>).

3
docs/style.css
View File

@ -25,6 +25,7 @@ h4 {font-family:sans-serif; font-size:0.9em; font-weight:bold; }
.aside {margin-left:2em; font-family:sans-serif; font-size:0.9em; }
blockquote .label {font-weight:bold; font-size:1em; margin:0 0 .1em;
border-bottom:1px solid #CCC;}
.emphasis {font-weight:bold; text-align:center; font-size:1.3em;}
/* A regular table */
.table {border-collapse:collapse; border-bottom:2px solid #888; margin-left:2em; }
@ -66,3 +67,5 @@ q:after {
/* Marks off sections that are lacking. */
.fixme {margin-left:2em; }
.fixme:before {content:"Fix me: "; font-weight:bold; color:#C00; }
#applicability {margin: 1em 5%; font-style:italic;}

61
library/HTMLPurifier.php
View File

@ -22,7 +22,7 @@
*/
/*
HTML Purifier 1.6.1 - Standards Compliant HTML Filtering
HTML Purifier 2.0.0 - Standards Compliant HTML Filtering
Copyright (C) 2006 Edward Z. Yang
This library is free software; you can redistribute it and/or
@ -42,7 +42,7 @@
// almost every class has an undocumented dependency to these, so make sure
// they get included
require_once 'HTMLPurifier/ConfigSchema.php';
require_once 'HTMLPurifier/ConfigSchema.php'; // important
require_once 'HTMLPurifier/Config.php';
require_once 'HTMLPurifier/Context.php';
@ -51,6 +51,23 @@ require_once 'HTMLPurifier/Generator.php';
require_once 'HTMLPurifier/Strategy/Core.php';
require_once 'HTMLPurifier/Encoder.php';
require_once 'HTMLPurifier/LanguageFactory.php';
HTMLPurifier_ConfigSchema::define(
'Core', 'Language', 'en', 'string', '
ISO 639 language code for localizable things in HTML Purifier to use,
which is mainly error reporting. There is currently only an English (en)
translation, so this directive is currently useless.
This directive has been available since 2.0.0.
');
HTMLPurifier_ConfigSchema::define(
'Core', 'CollectErrors', false, 'bool', '
Whether or not to collect errors found while filtering the document. This
is a useful way to give feedback to your users. CURRENTLY NOT IMPLEMENTED.
This directive has been available since 2.0.0.
');
/**
* Main library execution class.
*
@ -64,12 +81,12 @@ require_once 'HTMLPurifier/Encoder.php';
class HTMLPurifier
{
var $version = '1.6.1';
var $version = '2.0.0';
var $config;
var $filters;
var $lexer, $strategy, $generator;
var $strategy, $generator;
/**
* Final HTMLPurifier_Context of last run purification. Might be an array.
@ -89,7 +106,6 @@ class HTMLPurifier
$this->config = HTMLPurifier_Config::create($config);
$this->lexer = HTMLPurifier_Lexer::create();
$this->strategy = new HTMLPurifier_Strategy_Core();
$this->generator = new HTMLPurifier_Generator();
@ -117,7 +133,23 @@ class HTMLPurifier
$config = $config ? HTMLPurifier_Config::create($config) : $this->config;
// implementation is partially environment dependant, partially
// configuration dependant
$lexer = HTMLPurifier_Lexer::create($config);
$context = new HTMLPurifier_Context();
// set up global context variables
if ($config->get('Core', 'CollectErrors')) {
// may get moved out if other facilities use it
$language_factory = HTMLPurifier_LanguageFactory::instance();
$language = $language_factory->create($config->get('Core', 'Language'));
$context->register('Locale', $language);
$error_collector = new HTMLPurifier_ErrorCollector();
$context->register('ErrorCollector', $language);
}
$html = HTMLPurifier_Encoder::convertToUTF8($html, $config, $context);
for ($i = 0, $size = count($this->filters); $i < $size; $i++) {
@ -130,7 +162,7 @@ class HTMLPurifier
// list of tokens
$this->strategy->execute(
// list of un-purified tokens
$this->lexer->tokenizeHTML(
$lexer->tokenizeHTML(
// un-purified HTML
$html, $config, $context
),
@ -164,6 +196,23 @@ class HTMLPurifier
return $array_of_html;
}
/**
* Singleton for enforcing just one HTML Purifier in your system
*/
function &getInstance($prototype = null) {
static $htmlpurifier;
if (!$htmlpurifier || $prototype) {
if (is_a($prototype, 'HTMLPurifier')) {
$htmlpurifier = $prototype;
} elseif ($prototype) {
$htmlpurifier = new HTMLPurifier(HTMLPurifier_Config::create($prototype));
} else {
$htmlpurifier = new HTMLPurifier();
}
}
return $htmlpurifier;
}
}

64
library/HTMLPurifier/AttrCollections.php
View File

@ -1,7 +1,6 @@
<?php
require_once 'HTMLPurifier/AttrTypes.php';
require_once 'HTMLPurifier/AttrDef/Lang.php';
/**
* Defines common attribute collections that modules reference
@ -12,8 +11,6 @@ class HTMLPurifier_AttrCollections
/**
* Associative array of attribute collections, indexed by name
* @note Technically, the composition of these is more complicated,
* but we bypass it using our own excludes property
*/
var $info = array();
@ -25,27 +22,29 @@ class HTMLPurifier_AttrCollections
* @param $modules Hash array of HTMLPurifier_HTMLModule members
*/
function HTMLPurifier_AttrCollections($attr_types, $modules) {
$info =& $this->info;
// load extensions from the modules
foreach ($modules as $module) {
foreach ($module->attr_collections as $coll_i => $coll) {
if (!isset($this->info[$coll_i])) {
$this->info[$coll_i] = array();
}
foreach ($coll as $attr_i => $attr) {
if ($attr_i === 0 && isset($info[$coll_i][$attr_i])) {
if ($attr_i === 0 && isset($this->info[$coll_i][$attr_i])) {
// merge in includes
$info[$coll_i][$attr_i] = array_merge(
$info[$coll_i][$attr_i], $attr);
$this->info[$coll_i][$attr_i] = array_merge(
$this->info[$coll_i][$attr_i], $attr);
continue;
}
$info[$coll_i][$attr_i] = $attr;
$this->info[$coll_i][$attr_i] = $attr;
}
}
}
// perform internal expansions and inclusions
foreach ($info as $name => $attr) {
foreach ($this->info as $name => $attr) {
// merge attribute collections that include others
$this->performInclusions($info[$name]);
$this->performInclusions($this->info[$name]);
// replace string identifiers with actual attribute objects
$this->expandIdentifiers($info[$name], $attr_types);
$this->expandIdentifiers($this->info[$name], $attr_types);
}
}
@ -57,16 +56,20 @@ class HTMLPurifier_AttrCollections
function performInclusions(&$attr) {
if (!isset($attr[0])) return;
$merge = $attr[0];
$seen = array(); // recursion guard
// loop through all the inclusions
for ($i = 0; isset($merge[$i]); $i++) {
if (isset($seen[$merge[$i]])) continue;
$seen[$merge[$i]] = true;
// foreach attribute of the inclusion, copy it over
if (!isset($this->info[$merge[$i]])) continue;
foreach ($this->info[$merge[$i]] as $key => $value) {
if (isset($attr[$key])) continue; // also catches more inclusions
$attr[$key] = $value;
}
if (isset($info[$merge[$i]][0])) {
if (isset($this->info[$merge[$i]][0])) {
// recursion
$merge = array_merge($merge, isset($info[$merge[$i]][0]));
$merge = array_merge($merge, $this->info[$merge[$i]][0]);
}
}
unset($attr[0]);
@ -79,20 +82,47 @@ class HTMLPurifier_AttrCollections
* @param $attr_types HTMLPurifier_AttrTypes instance
*/
function expandIdentifiers(&$attr, $attr_types) {
// because foreach will process new elements we add, make sure we
// skip duplicates
$processed = array();
foreach ($attr as $def_i => $def) {
// skip inclusions
if ($def_i === 0) continue;
if (!is_string($def)) continue;
if (isset($processed[$def_i])) continue;
// determine whether or not attribute is required
if ($required = (strpos($def_i, '*') !== false)) {
// rename the definition
unset($attr[$def_i]);
$def_i = trim($def_i, '*');
$attr[$def_i] = $def;
}
$processed[$def_i] = true;
// if we've already got a literal object, move on
if (is_object($def)) {
// preserve previous required
$attr[$def_i]->required = ($required || $attr[$def_i]->required);
continue;
}
if ($def === false) {
unset($attr[$def_i]);
continue;
}
if (isset($attr_types->info[$def])) {
$attr[$def_i] = $attr_types->info[$def];
if ($t = $attr_types->get($def)) {
$attr[$def_i] = $t;
$attr[$def_i]->required = $required;
} else {
trigger_error('Attempted to reference undefined attribute type', E_USER_ERROR);
unset($attr[$def_i]);
}
}
}
}

24
library/HTMLPurifier/AttrDef.php
View File

@ -14,11 +14,17 @@ class HTMLPurifier_AttrDef
{
/**
* Tells us whether or not an HTML attribute is minimized. Only the
* boolean attribute vapourware would use this.
* Tells us whether or not an HTML attribute is minimized. Has no
* meaning in other contexts.
*/
var $minimized = false;
/**
* Tells us whether or not an HTML attribute is required. Has no
* meaning in other contexts
*/
var $required = false;
/**
* Validates and cleans passed string according to a definition.
*
@ -62,6 +68,20 @@ class HTMLPurifier_AttrDef
$string = str_replace(array("\r", "\t"), ' ', $string);
return $string;
}
/**
* Factory method for creating this class from a string.
* @param $string String construction info
* @return Created AttrDef object corresponding to $string
* @public
*/
function make($string) {
// default implementation, return flyweight of this object
// if overloaded, it is *necessary* for you to clone the
// object (usually by instantiating a new copy) and return that
return $this;
}
}
?>

54
library/HTMLPurifier/AttrDef/CSS/Color.php
View File

@ -2,43 +2,47 @@
require_once 'HTMLPurifier/AttrDef.php';
HTMLPurifier_ConfigSchema::define(
'Core', 'ColorKeywords', array(
'maroon' => '#800000',
'red' => '#FF0000',
'orange' => '#FFA500',
'yellow' => '#FFFF00',
'olive' => '#808000',
'purple' => '#800080',
'fuchsia' => '#FF00FF',
'white' => '#FFFFFF',
'lime' => '#00FF00',
'green' => '#008000',
'navy' => '#000080',
'blue' => '#0000FF',
'aqua' => '#00FFFF',
'teal' => '#008080',
'black' => '#000000',
'silver' => '#C0C0C0',
'gray' => '#808080'
), 'hash', '
Lookup array of color names to six digit hexadecimal number corresponding
to color, with preceding hash mark. Used when parsing colors.
This directive has been available since 2.0.0.
');
/**
* Validates Color as defined by CSS.
*/
class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef
{
/**
* Color keyword lookup table.
* @todo Extend it to include all usually allowed colors.
*/
var $colors = array(
'maroon' => '#800000',
'red' => '#F00',
'orange' => '#FFA500',
'yellow' => '#FF0',
'olive' => '#808000',
'purple' => '#800080',
'fuchsia' => '#F0F',
'white' => '#FFF',
'lime' => '#0F0',
'green' => '#008000',
'navy' => '#000080',
'blue' => '#00F',
'aqua' => '#0FF',
'teal' => '#008080',
'black' => '#000',
'silver' => '#C0C0C0',
'gray' => '#808080'
);
function validate($color, $config, &$context) {
static $colors = null;
if ($colors === null) $colors = $config->get('Core', 'ColorKeywords');
$color = trim($color);
if (!$color) return false;
$lower = strtolower($color);
if (isset($this->colors[$lower])) return $this->colors[$lower];
if (isset($colors[$lower])) return $colors[$lower];
if ($color[0] === '#') {
// hexadecimal handling

23
library/HTMLPurifier/AttrDef/CSS/Font.php
View File

@ -18,18 +18,6 @@ class HTMLPurifier_AttrDef_CSS_Font extends HTMLPurifier_AttrDef
*/
var $info = array();
/**
* System font keywords.
*/
var $system_fonts = array(
'caption' => true,
'icon' => true,
'menu' => true,
'message-box' => true,
'small-caption' => true,
'status-bar' => true
);
function HTMLPurifier_AttrDef_CSS_Font($config) {
$def = $config->getCSSDefinition();
$this->info['font-style'] = $def->info['font-style'];
@ -42,13 +30,22 @@ class HTMLPurifier_AttrDef_CSS_Font extends HTMLPurifier_AttrDef
function validate($string, $config, &$context) {
static $system_fonts = array(
'caption' => true,
'icon' => true,
'menu' => true,
'message-box' => true,
'small-caption' => true,
'status-bar' => true
);
// regular pre-processing
$string = $this->parseCDATA($string);
if ($string === '') return false;
// check if it's one of the keywords
$lowercase_string = strtolower($string);
if (isset($this->system_fonts[$lowercase_string])) {
if (isset($system_fonts[$lowercase_string])) {
return $lowercase_string;
}

22
library/HTMLPurifier/AttrDef/CSS/FontFamily.php
View File

@ -10,19 +10,15 @@ require_once 'HTMLPurifier/AttrDef.php';
class HTMLPurifier_AttrDef_CSS_FontFamily extends HTMLPurifier_AttrDef
{
/**
* Generic font family keywords.
* @protected
*/
var $generic_names = array(
'serif' => true,
'sans-serif' => true,
'monospace' => true,
'fantasy' => true,
'cursive' => true
);
function validate($string, $config, &$context) {
static $generic_names = array(
'serif' => true,
'sans-serif' => true,
'monospace' => true,
'fantasy' => true,
'cursive' => true
);
$string = $this->parseCDATA($string);
// assume that no font names contain commas in them
$fonts = explode(',', $string);
@ -31,7 +27,7 @@ class HTMLPurifier_AttrDef_CSS_FontFamily extends HTMLPurifier_AttrDef
$font = trim($font);
if ($font === '') continue;
// match a generic name
if (isset($this->generic_names[$font])) {
if (isset($generic_names[$font])) {
$final .= $font . ', ';
continue;
}

18
library/HTMLPurifier/AttrDef/CSS/TextDecoration.php
View File

@ -10,23 +10,19 @@ require_once 'HTMLPurifier/AttrDef.php';
class HTMLPurifier_AttrDef_CSS_TextDecoration extends HTMLPurifier_AttrDef
{
/**
* Lookup table of allowed values.
* @protected
*/
var $allowed_values = array(
'line-through' => true,
'overline' => true,
'underline' => true
);
function validate($string, $config, &$context) {
static $allowed_values = array(
'line-through' => true,
'overline' => true,
'underline' => true
);
$string = strtolower($this->parseCDATA($string));
$parts = explode(' ', $string);
$final = '';
foreach ($parts as $part) {
if (isset($this->allowed_values[$part])) {
if (isset($allowed_values[$part])) {
$final .= $part . ' ';
}
}

2
library/HTMLPurifier/AttrDef/CSS/URI.php
View File

@ -29,7 +29,7 @@ class HTMLPurifier_AttrDef_CSS_URI extends HTMLPurifier_AttrDef_URI
if ($uri_string[$new_length] != ')') return false;
$uri = trim(substr($uri_string, 0, $new_length));
if (isset($uri[0]) && ($uri[0] == "'" || $uri[0] == '"')) {
if (!empty($uri) && ($uri[0] == "'" || $uri[0] == '"')) {
$quote = $uri[0];
$new_length = strlen($uri) - 1;
if ($uri[$new_length] !== $quote) return false;

16
library/HTMLPurifier/AttrDef/Enum.php
View File

@ -45,6 +45,22 @@ class HTMLPurifier_AttrDef_Enum extends HTMLPurifier_AttrDef
return $result ? $string : false;
}
/**
* @param $string In form of comma-delimited list of case-insensitive
* valid values. Example: "foo,bar,baz". Prepend "s:" to make
* case sensitive
*/
function make($string) {
if (strlen($string) > 2 && $string[0] == 's' && $string[1] == ':') {
$string = substr($string, 2);
$sensitive = true;
} else {
$sensitive = false;
}
$values = explode(',', $string);
return new HTMLPurifier_AttrDef_Enum($values, $sensitive);
}
}
?>

30
library/HTMLPurifier/AttrDef/HTML/Bool.php Normal file
View File

@ -0,0 +1,30 @@
<?php
require_once 'HTMLPurifier/AttrDef.php';
/**
* Validates a boolean attribute
*/
class HTMLPurifier_AttrDef_HTML_Bool extends HTMLPurifier_AttrDef
{
var $name;
var $minimized = true;
function HTMLPurifier_AttrDef_HTML_Bool($name = false) {$this->name = $name;}
function validate($string, $config, &$context) {
if (empty($string)) return false;
return $this->name;
}
/**
* @param $string Name of attribute
*/
function make($string) {
return new HTMLPurifier_AttrDef_HTML_Bool($string);
}
}
?>

35
library/HTMLPurifier/AttrDef/HTML/Color.php Normal file
View File

@ -0,0 +1,35 @@
<?php
require_once 'HTMLPurifier/AttrDef.php';
require_once 'HTMLPurifier/AttrDef/CSS/Color.php'; // for %Core.ColorKeywords
/**
* Validates a color according to the HTML spec.
*/
class HTMLPurifier_AttrDef_HTML_Color extends HTMLPurifier_AttrDef
{
function validate($string, $config, &$context) {
static $colors = null;
if ($colors === null) $colors = $config->get('Core', 'ColorKeywords');
$string = trim($string);
if (empty($string)) return false;
if (isset($colors[$string])) return $colors[$string];
if ($string[0] === '#') $hex = substr($string, 1);
else $hex = $string;
$length = strlen($hex);
if ($length !== 3 && $length !== 6) return false;
if (!ctype_xdigit($hex)) return false;
if ($length === 3) $hex = $hex[0].$hex[0].$hex[1].$hex[1].$hex[2].$hex[2];
return "#$hex";
}
}
?>

14
library/HTMLPurifier/AttrDef/HTML/LinkTypes.php
View File

@ -26,22 +26,20 @@ HTMLPurifier_ConfigSchema::define(
class HTMLPurifier_AttrDef_HTML_LinkTypes extends HTMLPurifier_AttrDef
{
/** Lookup array of attribute names to configuration name */
var $configLookup = array(
'rel' => 'AllowedRel',
'rev' => 'AllowedRev'
);
/** Name config attribute to pull. */
var $name;
function HTMLPurifier_AttrDef_HTML_LinkTypes($name) {
if (!isset($this->configLookup[$name])) {
$configLookup = array(
'rel' => 'AllowedRel',
'rev' => 'AllowedRev'
);
if (!isset($configLookup[$name])) {
trigger_error('Unrecognized attribute name for link '.
'relationship.', E_USER_ERROR);
return;
}
$this->name = $this->configLookup[$name];
$this->name = $configLookup[$name];
}
function validate($string, $config, &$context) {

15
library/HTMLPurifier/AttrDef/URI.php
View File

@ -93,7 +93,6 @@ class HTMLPurifier_AttrDef_URI extends HTMLPurifier_AttrDef
{
var $host;
var $PercentEncoder;
var $embeds_resource;
/**
@ -101,12 +100,14 @@ class HTMLPurifier_AttrDef_URI extends HTMLPurifier_AttrDef
*/
function HTMLPurifier_AttrDef_URI($embeds_resource = false) {
$this->host = new HTMLPurifier_AttrDef_URI_Host();
$this->PercentEncoder = new HTMLPurifier_PercentEncoder();
$this->embeds_resource = (bool) $embeds_resource;
}
function validate($uri, $config, &$context) {
static $PercentEncoder = null;
if ($PercentEncoder === null) $PercentEncoder = new HTMLPurifier_PercentEncoder();
// We'll write stack-based parsers later, for now, use regexps to
// get things working as fast as possible (irony)
@ -116,7 +117,7 @@ class HTMLPurifier_AttrDef_URI extends HTMLPurifier_AttrDef
$uri = $this->parseCDATA($uri);
// fix up percent-encoding
$uri = $this->PercentEncoder->normalize($uri);
$uri = $PercentEncoder->normalize($uri);
// while it would be nice to use parse_url(), that's specifically
// for HTTP and thus won't work for our generic URI parsing
@ -157,6 +158,14 @@ class HTMLPurifier_AttrDef_URI extends HTMLPurifier_AttrDef
);
}
// something funky weird happened in the registry, abort!
if (!$scheme_obj) {
trigger_error(
'Default scheme object "' . $config->get('URI', 'DefaultScheme') . '" was not readable',
E_USER_WARNING
);
return false;
}
// the URI we're processing embeds_resource a resource in the page, but the URI
// it references cannot be located

16
library/HTMLPurifier/AttrDef/URI/IPv4.php
View File

@ -15,13 +15,10 @@ class HTMLPurifier_AttrDef_URI_IPv4 extends HTMLPurifier_AttrDef
*/
var $ip4;
function HTMLPurifier_AttrDef_URI_IPv4() {
$oct = '(?:25[0-5]|2[0-4][0-9]|1[0-9]{2}|[1-9][0-9]|[0-9])'; // 0-255
$this->ip4 = "(?:{$oct}\\.{$oct}\\.{$oct}\\.{$oct})";
}
function validate($aIP, $config, &$context) {
if (!$this->ip4) $this->_loadRegex();
if (preg_match('#^' . $this->ip4 . '$#s', $aIP))
{
return $aIP;
@ -31,6 +28,15 @@ class HTMLPurifier_AttrDef_URI_IPv4 extends HTMLPurifier_AttrDef
}
/**
* Lazy load function to prevent regex from being stuffed in
* cache.
*/
function _loadRegex() {
$oct = '(?:25[0-5]|2[0-4][0-9]|1[0-9]{2}|[1-9][0-9]|[0-9])'; // 0-255
$this->ip4 = "(?:{$oct}\\.{$oct}\\.{$oct}\\.{$oct})";
}
}
?>

2
library/HTMLPurifier/AttrDef/URI/IPv6.php
View File

@ -13,6 +13,8 @@ class HTMLPurifier_AttrDef_URI_IPv6 extends HTMLPurifier_AttrDef_URI_IPv4
function validate($aIP, $config, &$context) {
if (!$this->ip4) $this->_loadRegex();
$original = $aIP;
$hex = '[0-9a-fA-F]';

6
library/HTMLPurifier/AttrTransform/ImgRequired.php
View File

@ -20,7 +20,10 @@ HTMLPurifier_ConfigSchema::define(
);
/**
* Post-transform that ensures the required attrs of img (alt and src) are set
* Transform that supplies default values for the src and alt attributes
* in img tags, as well as prevents the img tag from being removed
* because of a missing alt tag. This needs to be registered as both
* a pre and post attribute transform.
*/
class HTMLPurifier_AttrTransform_ImgRequired extends HTMLPurifier_AttrTransform
{
@ -29,6 +32,7 @@ class HTMLPurifier_AttrTransform_ImgRequired extends HTMLPurifier_AttrTransform
$src = true;
if (!isset($attr['src'])) {
if ($config->get('Core', 'RemoveInvalidImg')) return $attr;
$attr['src'] = $config->get('Attr', 'DefaultInvalidImage');
$src = false;
}

45
library/HTMLPurifier/AttrTypes.php
View File

@ -1,10 +1,14 @@
<?php
require_once 'HTMLPurifier/AttrDef/Lang.php';
require_once 'HTMLPurifier/AttrDef/Enum.php';
require_once 'HTMLPurifier/AttrDef/HTML/Bool.php';
require_once 'HTMLPurifier/AttrDef/HTML/ID.php';
require_once 'HTMLPurifier/AttrDef/HTML/Length.php';
require_once 'HTMLPurifier/AttrDef/HTML/MultiLength.php';
require_once 'HTMLPurifier/AttrDef/HTML/Nmtokens.php';
require_once 'HTMLPurifier/AttrDef/HTML/Pixels.php';
require_once 'HTMLPurifier/AttrDef/HTML/Color.php';
require_once 'HTMLPurifier/AttrDef/Integer.php';
require_once 'HTMLPurifier/AttrDef/Text.php';
require_once 'HTMLPurifier/AttrDef/URI.php';
@ -16,14 +20,19 @@ class HTMLPurifier_AttrTypes
{
/**
* Lookup array of attribute string identifiers to concrete implementations
* @public
* @protected
*/
var $info = array();
/**
* Constructs the info array
* Constructs the info array, supplying default implementations for attribute
* types.
*/
function HTMLPurifier_AttrTypes() {
// pseudo-types, must be instantiated via shorthand
$this->info['Enum'] = new HTMLPurifier_AttrDef_Enum();
$this->info['Bool'] = new HTMLPurifier_AttrDef_HTML_Bool();
$this->info['CDATA'] = new HTMLPurifier_AttrDef_Text();
$this->info['ID'] = new HTMLPurifier_AttrDef_HTML_ID();
$this->info['Length'] = new HTMLPurifier_AttrDef_HTML_Length();
@ -32,10 +41,42 @@ class HTMLPurifier_AttrTypes
$this->info['Pixels'] = new HTMLPurifier_AttrDef_HTML_Pixels();
$this->info['Text'] = new HTMLPurifier_AttrDef_Text();
$this->info['URI'] = new HTMLPurifier_AttrDef_URI();
$this->info['LanguageCode'] = new HTMLPurifier_AttrDef_Lang();
$this->info['Color'] = new HTMLPurifier_AttrDef_HTML_Color();
// number is really a positive integer (one or more digits)
// FIXME: ^^ not always, see start and value of list items
$this->info['Number'] = new HTMLPurifier_AttrDef_Integer(false, false, true);
}
/**
* Retrieves a type
* @param $type String type name
* @return Object AttrDef for type
*/
function get($type) {
// determine if there is any extra info tacked on
if (strpos($type, '#') !== false) list($type, $string) = explode('#', $type, 2);
else $string = '';
if (!isset($this->info[$type])) {
trigger_error('Cannot retrieve undefined attribute type ' . $type, E_USER_ERROR);
return;
}
return $this->info[$type]->make($string);
}
/**
* Sets a new implementation for a type
* @param $type String type name
* @param $impl Object AttrDef for type
*/
function set($type, $impl) {
$this->info[$type] = $impl;
}
}
?>

105
library/HTMLPurifier/AttrValidator.php Normal file
View File

@ -0,0 +1,105 @@
<?php
class HTMLPurifier_AttrValidator
{
function validateToken($token, &$config, &$context) {
$definition = $config->getHTMLDefinition();
// create alias to global definition array, see also $defs
// DEFINITION CALL
$d_defs = $definition->info_global_attr;
// copy out attributes for easy manipulation
$attr = $token->attr;
// do global transformations (pre)
// nothing currently utilizes this
foreach ($definition->info_attr_transform_pre as $transform) {
$attr = $transform->transform($attr, $config, $context);
}
// do local transformations only applicable to this element (pre)
// ex. <p align="right"> to <p style="text-align:right;">
foreach ($definition->info[$token->name]->attr_transform_pre
as $transform
) {
$attr = $transform->transform($attr, $config, $context);
}
// create alias to this element's attribute definition array, see
// also $d_defs (global attribute definition array)
// DEFINITION CALL
$defs = $definition->info[$token->name]->attr;
// iterate through all the attribute keypairs
// Watch out for name collisions: $key has previously been used
foreach ($attr as $attr_key => $value) {
// call the definition
if ( isset($defs[$attr_key]) ) {
// there is a local definition defined
if ($defs[$attr_key] === false) {
// We've explicitly been told not to allow this element.
// This is usually when there's a global definition
// that must be overridden.
// Theoretically speaking, we could have a
// AttrDef_DenyAll, but this is faster!
$result = false;
} else {
// validate according to the element's definition
$result = $defs[$attr_key]->validate(
$value, $config, $context
);
}
} elseif ( isset($d_defs[$attr_key]) ) {
// there is a global definition defined, validate according
// to the global definition
$result = $d_defs[$attr_key]->validate(
$value, $config, $context
);
} else {
// system never heard of the attribute? DELETE!
$result = false;
}
// put the results into effect
if ($result === false || $result === null) {
// remove the attribute
unset($attr[$attr_key]);
} elseif (is_string($result)) {
// simple substitution
$attr[$attr_key] = $result;
}
// we'd also want slightly more complicated substitution
// involving an array as the return value,
// although we're not sure how colliding attributes would
// resolve (certain ones would be completely overriden,
// others would prepend themselves).
}
// post transforms
// ex. <x lang="fr"> to <x lang="fr" xml:lang="fr">
foreach ($definition->info_attr_transform_post as $transform) {
$attr = $transform->transform($attr, $config, $context);
}
// ex. <bdo> to <bdo dir="ltr">
foreach ($definition->info[$token->name]->attr_transform_post as $transform) {
$attr = $transform->transform($attr, $config, $context);
}
// commit changes
$token->attr = $attr;
return $token;
}
}
?>

17
library/HTMLPurifier/CSSDefinition.php
View File

@ -1,5 +1,7 @@
<?php
require_once 'HTMLPurifier/Definition.php';
require_once 'HTMLPurifier/AttrDef/CSS/Background.php';
require_once 'HTMLPurifier/AttrDef/CSS/BackgroundPosition.php';
require_once 'HTMLPurifier/AttrDef/CSS/Border.php';
@ -15,13 +17,24 @@ require_once 'HTMLPurifier/AttrDef/CSS/TextDecoration.php';
require_once 'HTMLPurifier/AttrDef/CSS/URI.php';
require_once 'HTMLPurifier/AttrDef/Enum.php';
HTMLPurifier_ConfigSchema::define(
'CSS', 'DefinitionRev', 1, 'int', '
<p>
Revision identifier for your custom definition. See
%HTML.DefinitionRev for details. This directive has been available
since 2.0.0.
</p>
');
/**
* Defines allowed CSS attributes and what their values are.
* @see HTMLPurifier_HTMLDefinition
*/
class HTMLPurifier_CSSDefinition
class HTMLPurifier_CSSDefinition extends HTMLPurifier_Definition
{
var $type = 'CSS';
/**
* Assoc array of attribute name to definition object.
*/
@ -30,7 +43,7 @@ class HTMLPurifier_CSSDefinition
/**
* Constructs the info array. The meat of this class.
*/
function setup($config) {
function doSetup($config) {
$this->info['text-align'] = new HTMLPurifier_AttrDef_Enum(
array('left', 'right', 'center', 'justify'), false);

23
library/HTMLPurifier/ChildDef/Custom.php
View File

@ -38,8 +38,21 @@ class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef
if ($raw{0} != '(') {
$raw = "($raw)";
}
$reg = str_replace(',', ',?', $raw);
$reg = preg_replace('/([#a-zA-Z0-9_.-]+)/', '(,?\\0)', $reg);
$el = '[#a-zA-Z0-9_.-]+';
$reg = $raw;
// COMPLICATED! AND MIGHT BE BUGGY! I HAVE NO CLUE WHAT I'M
// DOING! Seriously: if there's problems, please report them.
// setup all elements as parentheticals with leading commas
$reg = preg_replace("/$el/", '(,\\0)', $reg);
// remove commas when they were not solicited
$reg = preg_replace("/([^,(|]\(+),/", '\\1', $reg);
// remove all non-paranthetical commas: they are handled by first regex
$reg = preg_replace("/,\(/", '(', $reg);
$this->_pcre_regex = $reg;
}
function validateChildren($tokens_of_children, $config, &$context) {
@ -60,11 +73,11 @@ class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef
$list_of_children .= $token->name . ',';
}
}
$list_of_children = rtrim($list_of_children, ',');
// add leading comma to deal with stray comma declarations
$list_of_children = ',' . rtrim($list_of_children, ',');
$okay =
preg_match(
'/^'.$this->_pcre_regex.'$/',
'/^,?'.$this->_pcre_regex.'$/',
$list_of_children
);

11
library/HTMLPurifier/ChildDef/Required.php
View File

@ -29,7 +29,6 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
}
}
$this->elements = $elements;
$this->gen = new HTMLPurifier_Generator();
}
var $allow_empty = false;
var $type = 'required';
@ -57,6 +56,12 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
// some configuration
$escape_invalid_children = $config->get('Core', 'EscapeInvalidChildren');
// generator
static $gen = null;
if ($gen === null) {
$gen = new HTMLPurifier_Generator();
}
foreach ($tokens_of_children as $token) {
if (!empty($token->is_whitespace)) {
$result[] = $token;
@ -80,7 +85,7 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
$result[] = $token;
} elseif ($pcdata_allowed && $escape_invalid_children) {
$result[] = new HTMLPurifier_Token_Text(
$this->gen->generateFromToken($token, $config)
$gen->generateFromToken($token, $config)
);
}
continue;
@ -91,7 +96,7 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
} elseif ($pcdata_allowed && $escape_invalid_children) {
$result[] =
new HTMLPurifier_Token_Text(
$this->gen->generateFromToken( $token, $config )
$gen->generateFromToken( $token, $config )
);
} else {
// drop silently

4
library/HTMLPurifier/ChildDef/StrictBlockquote.php
View File

@ -45,8 +45,8 @@ extends HTMLPurifier_ChildDef_Required
if (!$is_inline) {
if (!$depth) {
if (
$token->type == 'text' ||
!isset($this->elements[$token->name])
($token->type == 'text' && !$token->is_whitespace) ||
($token->type != 'text' && !isset($this->elements[$token->name]))
) {
$is_inline = true;
$ret[] = $block_wrap_start;

237
library/HTMLPurifier/Config.php
View File

@ -1,5 +1,28 @@
<?php
require_once 'HTMLPurifier/ConfigSchema.php';
// member variables
require_once 'HTMLPurifier/HTMLDefinition.php';
require_once 'HTMLPurifier/CSSDefinition.php';
require_once 'HTMLPurifier/Doctype.php';
require_once 'HTMLPurifier/DefinitionCacheFactory.php';
// accomodations for versions earlier than 4.3.10 and 5.0.2
// borrowed from PHP_Compat, LGPL licensed, by Aidan Lister <aidan@php.net>
if (!defined('PHP_EOL')) {
switch (strtoupper(substr(PHP_OS, 0, 3))) {
case 'WIN':
define('PHP_EOL', "\r\n");
break;
case 'DAR':
define('PHP_EOL', "\r");
break;
default:
define('PHP_EOL', "\n");
}
}
/**
* Configuration object that triggers customizable behavior.
*
@ -15,6 +38,11 @@
class HTMLPurifier_Config
{
/**
* HTML Purifier's version
*/
var $version = '2.0.0';
/**
* Two-level associative array of configuration directives
*/
@ -26,14 +54,26 @@ class HTMLPurifier_Config
var $def;
/**
* Cached instance of HTMLPurifier_HTMLDefinition
* Indexed array of definitions
*/
var $html_definition;
var $definitions;
/**
* Cached instance of HTMLPurifier_CSSDefinition
* Bool indicator whether or not config is finalized
*/
var $css_definition;
var $finalized = false;
/**
* Bool indicator whether or not to automatically finalize
* the object if a read operation is done
*/
var $autoFinalize = true;
/**
* Namespace indexed array of serials for specific namespaces (see
* getSerial for more info).
*/
var $serials = array();
/**
* @param $definition HTMLPurifier_ConfigSchema that defines what directives
@ -58,6 +98,7 @@ class HTMLPurifier_Config
$ret = HTMLPurifier_Config::createDefault();
if (is_string($config)) $ret->loadIni($config);
elseif (is_array($config)) $ret->loadArray($config);
if (isset($revision)) $ret->revision = $revision;
return $ret;
}
@ -78,13 +119,16 @@ class HTMLPurifier_Config
* @param $key String key
*/
function get($namespace, $key, $from_alias = false) {
if (!$this->finalized && $this->autoFinalize) $this->finalize();
if (!isset($this->def->info[$namespace][$key])) {
trigger_error('Cannot retrieve value of undefined directive',
// can't add % due to SimpleTest bug
trigger_error('Cannot retrieve value of undefined directive ' . htmlspecialchars("$namespace.$key"),
E_USER_WARNING);
return;
}
if ($this->def->info[$namespace][$key]->class == 'alias') {
trigger_error('Cannot get value from aliased directive, use real name',
$d = $this->def->info[$namespace][$key];
trigger_error('Cannot get value from aliased directive, use real name ' . $d->namespace . '.' . $d->name,
E_USER_ERROR);
return;
}
@ -96,14 +140,35 @@ class HTMLPurifier_Config
* @param $namespace String namespace
*/
function getBatch($namespace) {
if (!$this->finalized && $this->autoFinalize) $this->finalize();
if (!isset($this->def->info[$namespace])) {
trigger_error('Cannot retrieve undefined namespace',
trigger_error('Cannot retrieve undefined namespace ' . htmlspecialchars($namespace),
E_USER_WARNING);
return;
}
return $this->conf[$namespace];
}
/**
* Returns a md5 signature of a segment of the configuration object
* that uniquely identifies that particular configuration
* @param $namespace Namespace to get serial for
*/
function getBatchSerial($namespace) {
if (empty($this->serials[$namespace])) {
$this->serials[$namespace] = md5(serialize($this->getBatch($namespace)));
}
return $this->serials[$namespace];
}
/**
* Retrieves all directives, organized by namespace
*/
function getAll() {
if (!$this->finalized && $this->autoFinalize) $this->finalize();
return $this->conf;
}
/**
* Sets a value to configuration.
* @param $namespace String namespace
@ -111,15 +176,16 @@ class HTMLPurifier_Config
* @param $value Mixed value
*/
function set($namespace, $key, $value, $from_alias = false) {
if ($this->isFinalized('Cannot set directive after finalization')) return;
if (!isset($this->def->info[$namespace][$key])) {
trigger_error('Cannot set undefined directive to value',
trigger_error('Cannot set undefined directive ' . htmlspecialchars("$namespace.$key") . ' to value',
E_USER_WARNING);
return;
}
if ($this->def->info[$namespace][$key]->class == 'alias') {
if ($from_alias) {
trigger_error('Double-aliases not allowed, please fix '.
'ConfigSchema bug');
'ConfigSchema bug with' . "$namespace.$key");
}
$this->set($this->def->info[$namespace][$key]->namespace,
$this->def->info[$namespace][$key]->name,
@ -128,7 +194,7 @@ class HTMLPurifier_Config
}
$value = $this->def->validate(
$value,
$this->def->info[$namespace][$key]->type,
$type = $this->def->info[$namespace][$key]->type,
$this->def->info[$namespace][$key]->allow_null
);
if (is_string($value)) {
@ -139,23 +205,36 @@ class HTMLPurifier_Config
if ($this->def->info[$namespace][$key]->allowed !== true) {
// check to see if the value is allowed
if (!isset($this->def->info[$namespace][$key]->allowed[$value])) {
trigger_error('Value not supported', E_USER_WARNING);
trigger_error('Value not supported, valid values are: ' .
$this->_listify($this->def->info[$namespace][$key]->allowed), E_USER_WARNING);
return;
}
}
}
if ($this->def->isError($value)) {
trigger_error('Value is of invalid type', E_USER_WARNING);
trigger_error('Value for ' . "$namespace.$key" . ' is of invalid type, should be ' . $type, E_USER_WARNING);
return;
}
$this->conf[$namespace][$key] = $value;
if ($namespace == 'HTML' || $namespace == 'Attr') {
// reset HTML definition if relevant attributes changed
$this->html_definition = null;
}
if ($namespace == 'CSS') {
$this->css_definition = null;
// reset definitions if the directives they depend on changed
// this is a very costly process, so it's discouraged
// with finalization
if ($namespace == 'HTML' || $namespace == 'CSS') {
$this->definitions[$namespace] = null;
}
$this->serials[$namespace] = false;
}
/**
* Convenience function for error reporting
* @private
*/
function _listify($lookup) {
$list = array();
foreach ($lookup as $name => $b) $list[] = $name;
return implode(', ', $list);
}
/**
@ -164,26 +243,71 @@ class HTMLPurifier_Config
* called before it's been setup, otherwise won't work.
*/
function &getHTMLDefinition($raw = false) {
if (
empty($this->html_definition) || // hasn't ever been setup
($raw && $this->html_definition->setup) // requesting new one
) {
$this->html_definition = new HTMLPurifier_HTMLDefinition($this);
if ($raw) return $this->html_definition; // no setup!
}
if (!$this->html_definition->setup) $this->html_definition->setup();
return $this->html_definition;
return $this->getDefinition('HTML', $raw);
}
/**
* Retrieves reference to the CSS definition
*/
function &getCSSDefinition() {
if ($this->css_definition === null) {
$this->css_definition = new HTMLPurifier_CSSDefinition();
$this->css_definition->setup($this);
function &getCSSDefinition($raw = false) {
return $this->getDefinition('CSS', $raw);
}
/**
* Retrieves a definition
* @param $type Type of definition: HTML, CSS, etc
* @param $raw Whether or not definition should be returned raw
*/
function &getDefinition($type, $raw = false) {
if (!$this->finalized && $this->autoFinalize) $this->finalize();
$factory = HTMLPurifier_DefinitionCacheFactory::instance();
$cache = $factory->create($type, $this);
if (!$raw) {
// see if we can quickly supply a definition
if (!empty($this->definitions[$type])) {
if (!$this->definitions[$type]->setup) {
$this->definitions[$type]->setup($this);
}
return $this->definitions[$type];
}
// memory check missed, try cache
$this->definitions[$type] = $cache->get($this);
if ($this->definitions[$type]) {
// definition in cache, return it
return $this->definitions[$type];
}
} elseif (
!empty($this->definitions[$type]) &&
!$this->definitions[$type]->setup
) {
// raw requested, raw in memory, quick return
return $this->definitions[$type];
}
return $this->css_definition;
// quick checks failed, let's create the object
if ($type == 'HTML') {
$this->definitions[$type] = new HTMLPurifier_HTMLDefinition();
} elseif ($type == 'CSS') {
$this->definitions[$type] = new HTMLPurifier_CSSDefinition();
} else {
trigger_error("Definition of $type type not supported");
$false = false;
return $false;
}
// quick abort if raw
if ($raw) {
if (is_null($this->get($type, 'DefinitionID'))) {
// fatally error out if definition ID not set
trigger_error("Cannot retrieve raw version without specifying %$type.DefinitionID", E_USER_ERROR);
$false = false;
return $false;
}
return $this->definitions[$type];
}
// set it up
$this->definitions[$type]->setup($this);
// save in cache
$cache->set($this->definitions[$type], $this);
return $this->definitions[$type];
}
/**
@ -192,6 +316,7 @@ class HTMLPurifier_Config
* @param $config_array Configuration associative array
*/
function loadArray($config_array) {
if ($this->isFinalized('Cannot load directives after finalization')) return;
foreach ($config_array as $key => $value) {
$key = str_replace('_', '.', $key);
if (strpos($key, '.') !== false) {
@ -208,15 +333,63 @@ class HTMLPurifier_Config
}
}
/**
* Loads configuration values from $_GET/$_POST that were posted
* via ConfigForm
* @param $array $_GET or $_POST array to import
* @param $index Index/name that the config variables are in
* @param $mq_fix Boolean whether or not to enable magic quotes fix
* @static
*/
function loadArrayFromForm($array, $index, $mq_fix = true) {
$array = (isset($array[$index]) && is_array($array[$index])) ? $array[$index] : array();
$mq = get_magic_quotes_gpc() && $mq_fix;
foreach ($array as $key => $value) {
if (!strncmp($key, 'Null_', 5) && !empty($value)) {
unset($array[substr($key, 5)]);
unset($array[$key]);
}
if ($mq) $array[$key] = stripslashes($value);
}
return @HTMLPurifier_Config::create($array);
}
/**
* Loads configuration values from an ini file
* @param $filename Name of ini file
*/
function loadIni($filename) {
if ($this->isFinalized('Cannot load directives after finalization')) return;
$array = parse_ini_file($filename, true);
$this->loadArray($array);
}
/**
* Checks whether or not the configuration object is finalized.
* @param $error String error message, or false for no error
*/
function isFinalized($error = false) {
if ($this->finalized && $error) {
trigger_error($error, E_USER_ERROR);
}
return $this->finalized;
}
/**
* Finalizes configuration only if auto finalize is on and not
* already finalized
*/
function autoFinalize() {
if (!$this->finalized && $this->autoFinalize) $this->finalize();
}
/**
* Finalizes a configuration object, prohibiting further change
*/
function finalize() {
$this->finalized = true;
}
}
?>

14
library/HTMLPurifier/ConfigSchema.php
View File

@ -8,6 +8,7 @@ require_once 'HTMLPurifier/ConfigDef/DirectiveAlias.php';
/**
* Configuration definition, defines directives and their defaults.
* @note If you update this, please update Printer_ConfigForm
* @todo The ability to define things multiple times is confusing and should
* be factored out to its own function named registerDependency() or
* addNote(), where only the namespace.name and an extra descriptions
@ -66,6 +67,8 @@ class HTMLPurifier_ConfigSchema {
$this->defineNamespace('URI', 'Features regarding Uniform Resource Identifiers.');
$this->defineNamespace('HTML', 'Configuration regarding allowed HTML.');
$this->defineNamespace('CSS', 'Configuration regarding allowed CSS.');
$this->defineNamespace('Output', 'Configuration relating to the generation of (X)HTML.');
$this->defineNamespace('Cache', 'Configuration for DefinitionCache and related subclasses.');
$this->defineNamespace('Test', 'Developer testing configuration for our unit tests.');
}
@ -303,6 +306,7 @@ class HTMLPurifier_ConfigSchema {
if ($allow_null && $var === null) return null;
switch ($type) {
case 'mixed':
//if (is_string($var)) $var = unserialize($var);
return $var;
case 'istring':
case 'string':
@ -343,6 +347,16 @@ class HTMLPurifier_ConfigSchema {
$var = explode(',',$var);
// remove spaces
foreach ($var as $i => $j) $var[$i] = trim($j);
if ($type === 'hash') {
// key:value,key2:value2
$nvar = array();
foreach ($var as $keypair) {
$c = explode(':', $keypair, 2);
if (!isset($c[1])) continue;
$nvar[$c[0]] = $c[1];
}
$var = $nvar;
}
}
if (!is_array($var)) break;
$keys = array_keys($var);

2
library/HTMLPurifier/ContentSets.php
View File

@ -6,6 +6,8 @@ require_once 'HTMLPurifier/ChildDef/Empty.php';
require_once 'HTMLPurifier/ChildDef/Required.php';
require_once 'HTMLPurifier/ChildDef/Optional.php';
// NOT UNIT TESTED!!!
class HTMLPurifier_ContentSets
{

41
library/HTMLPurifier/Definition.php Normal file
View File

@ -0,0 +1,41 @@
<?php
/**
* Super-class for definition datatype objects, implements serialization
* functions for the class.
*/
class HTMLPurifier_Definition
{
/**
* Has setup() been called yet?
*/
var $setup = false;
/**
* What type of definition is it?
*/
var $type;
/**
* Sets up the definition object into the final form, something
* not done by the constructor
* @param $config HTMLPurifier_Config instance
*/
function doSetup($config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Setup function that aborts if already setup
* @param $config HTMLPurifier_Config instance
*/
function setup($config) {
if ($this->setup) return;
$this->setup = true;
$this->doSetup($config);
}
}
?>

121
library/HTMLPurifier/DefinitionCache.php Normal file
View File

@ -0,0 +1,121 @@
<?php
require_once 'HTMLPurifier/DefinitionCache/Serializer.php';
require_once 'HTMLPurifier/DefinitionCache/Null.php';
require_once 'HTMLPurifier/DefinitionCache/Decorator.php';
/**
* Abstract class representing Definition cache managers that implements
* useful common methods and is a factory.
* @todo Get some sort of versioning variable so the library can easily
* invalidate the cache with a new version
* @todo Make the test runner cache aware and allow the user to easily
* flush the cache
* @todo Create a separate maintenance file advanced users can use to
* cache their custom HTMLDefinition, which can be loaded
* via a configuration directive
* @todo Implement memcached
*/
class HTMLPurifier_DefinitionCache
{
var $type;
/**
* @param $name Type of definition objects this instance of the
* cache will handle.
*/
function HTMLPurifier_DefinitionCache($type) {
$this->type = $type;
}
/**
* Generates a unique identifier for a particular configuration
* @param Instance of HTMLPurifier_Config
*/
function generateKey($config) {
return $config->version . '-' . // possibly replace with function calls
$config->get($this->type, 'DefinitionRev') . '-' .
$config->getBatchSerial($this->type);
}
/**
* Tests whether or not a key is old with respect to the configuration's
* version and revision number.
* @param $key Key to test
* @param $config Instance of HTMLPurifier_Config to test against
*/
function isOld($key, $config) {
list($version, $revision, $hash) = explode('-', $key, 3);
$compare = version_compare($version, $config->version);
if ($compare > 0) return false;
if ($compare == 0 && $revision >= $config->get($this->type, 'DefinitionRev')) return false;
return true;
}
/**
* Checks if a definition's type jives with the cache's type
* @note Throws an error on failure
* @param $def Definition object to check
* @return Boolean true if good, false if not
*/
function checkDefType($def) {
if ($def->type !== $this->type) {
trigger_error("Cannot use definition of type {$def->type} in cache for {$this->type}");
return false;
}
return true;
}
/**
* Adds a definition object to the cache
*/
function add($def, $config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Unconditionally saves a definition object to the cache
*/
function set($def, $config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Replace an object in the cache
*/
function replace($def, $config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Retrieves a definition object from the cache
*/
function get($config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Removes a definition object to the cache
*/
function remove($config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Clears all objects from cache
*/
function flush($config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
/**
* Clears all expired (older version or revision) objects from cache
*/
function cleanup($config) {
trigger_error('Cannot call abstract method', E_USER_ERROR);
}
}
?>

63
library/HTMLPurifier/DefinitionCache/Decorator.php Normal file
View File

@ -0,0 +1,63 @@
<?php
require_once 'HTMLPurifier/DefinitionCache.php';
require_once 'HTMLPurifier/DefinitionCache/Decorator/Memory.php';
require_once 'HTMLPurifier/DefinitionCache/Decorator/Cleanup.php';
class HTMLPurifier_DefinitionCache_Decorator extends HTMLPurifier_DefinitionCache
{
/**
* Cache object we are decorating
*/
var $cache;
function HTMLPurifier_DefinitionCache_Decorator() {}
/**
* Lazy decorator function
* @param $cache Reference to cache object to decorate
*/
function decorate(&$cache) {
$decorator = $this->copy();
// reference is necessary for mocks in PHP 4
$decorator->cache =& $cache;
$decorator->type = $cache->type;
return $decorator;
}
/**
* Cross-compatible clone substitute
*/
function copy() {
return new HTMLPurifier_DefinitionCache_Decorator();
}
function add($def, $config) {
return $this->cache->add($def, $config);
}
function set($def, $config) {
return $this->cache->set($def, $config);
}
function replace($def, $config) {
return $this->cache->replace($def, $config);
}
function get($config) {
return $this->cache->get($config);
}
function flush($config) {
return $this->cache->flush($config);
}
function cleanup($config) {
return $this->cache->cleanup($config);
}
}
?>

45
library/HTMLPurifier/DefinitionCache/Decorator/Cleanup.php Normal file
View File

@ -0,0 +1,45 @@
<?php
require_once 'HTMLPurifier/DefinitionCache/Decorator.php';
/**
* Definition cache decorator class that cleans up the cache
* whenever there is a cache miss.
*/
class HTMLPurifier_DefinitionCache_Decorator_Cleanup extends
HTMLPurifier_DefinitionCache_Decorator
{
var $name = 'Cleanup';
function copy() {
return new HTMLPurifier_DefinitionCache_Decorator_Cleanup();
}
function add($def, $config) {
$status = parent::add($def, $config);
if (!$status) parent::cleanup($config);
return $status;
}
function set($def, $config) {
$status = parent::set($def, $config);
if (!$status) parent::cleanup($config);
return $status;
}
function replace($def, $config) {
$status = parent::replace($def, $config);
if (!$status) parent::cleanup($config);
return $status;
}
function get($config) {
$ret = parent::get($config);
if (!$ret) parent::cleanup($config);
return $ret;
}
}
?>

48
library/HTMLPurifier/DefinitionCache/Decorator/Memory.php Normal file
View File

@ -0,0 +1,48 @@
<?php
require_once 'HTMLPurifier/DefinitionCache/Decorator.php';
/**
* Definition cache decorator class that saves all cache retrievals
* to PHP's memory; good for unit tests or circumstances where
* there are lots of configuration objects floating around.
*/
class HTMLPurifier_DefinitionCache_Decorator_Memory extends
HTMLPurifier_DefinitionCache_Decorator
{
var $definitions;
var $name = 'Memory';
function copy() {
return new HTMLPurifier_DefinitionCache_Decorator_Memory();
}
function add($def, $config) {
$status = parent::add($def, $config);
if ($status) $this->definitions[$this->generateKey($config)] = $def;
return $status;
}
function set($def, $config) {
$status = parent::set($def, $config);
if ($status) $this->definitions[$this->generateKey($config)] = $def;
return $status;
}
function replace($def, $config) {
$status = parent::replace($def, $config);
if ($status) $this->definitions[$this->generateKey($config)] = $def;
return $status;
}
function get($config) {
$key = $this->generateKey($config);
if (isset($this->definitions[$key])) return $this->definitions[$key];
$this->definitions[$key] = parent::get($config);
return $this->definitions[$key];
}
}
?>

47
library/HTMLPurifier/DefinitionCache/Decorator/Template.php.in Normal file
View File

@ -0,0 +1,47 @@
<?php
require_once 'HTMLPurifier/DefinitionCache/Decorator.php';
/**
* Definition cache decorator template.
*/
class HTMLPurifier_DefinitionCache_Decorator_Template extends
HTMLPurifier_DefinitionCache_Decorator
{
var $name = 'Template'; // replace this
function copy() {
// replace class name with yours
return new HTMLPurifier_DefinitionCache_Decorator_Template();
}
// remove methods you don't need
function add($def, $config) {
return parent::add($def, $config);
}
function set($def, $config) {
return parent::set($def, $config);
}
function replace($def, $config) {
return parent::replace($def, $config);
}
function get($config) {
return parent::get($config);
}
function flush() {
return parent::flush();
}
function cleanup($config) {
return parent::cleanup($config);
}
}
?>

37
library/HTMLPurifier/DefinitionCache/Null.php Normal file
View File

@ -0,0 +1,37 @@
<?php
require_once 'HTMLPurifier/DefinitionCache.php';
/**
* Null cache object to use when no caching is on.
*/
class HTMLPurifier_DefinitionCache_Null extends HTMLPurifier_DefinitionCache
{
function add($def, $config) {
return false;
}
function set($def, $config) {
return false;
}
function replace($def, $config) {
return false;
}
function get($config) {
return false;
}
function flush($config) {
return false;
}
function cleanup($config) {
return false;
}
}
?>

129
library/HTMLPurifier/DefinitionCache/Serializer.php Normal file
View File

@ -0,0 +1,129 @@
<?php
require_once 'HTMLPurifier/DefinitionCache.php';
HTMLPurifier_ConfigSchema::define(
'Cache', 'SerializerPath', null, 'string/null', '
<p>
Absolute path with no trailing slash to store serialized definitions in.
Default is within the
HTML Purifier library inside DefinitionCache/Serializer. This
path must be writable by the webserver. This directive has been
available since 2.0.0.
</p>
');
class HTMLPurifier_DefinitionCache_Serializer extends
HTMLPurifier_DefinitionCache
{
function add($def, $config) {
if (!$this->checkDefType($def)) return;
$file = $this->generateFilePath($config);
if (file_exists($file)) return false;
$this->_prepareDir($config);
return $this->_write($file, serialize($def));
}
function set($def, $config) {
if (!$this->checkDefType($def)) return;
$file = $this->generateFilePath($config);
$this->_prepareDir($config);
return $this->_write($file, serialize($def));
}
function replace($def, $config) {
if (!$this->checkDefType($def)) return;
$file = $this->generateFilePath($config);
if (!file_exists($file)) return false;
$this->_prepareDir($config);
return $this->_write($file, serialize($def));
}
function get($config) {
$file = $this->generateFilePath($config);
if (!file_exists($file)) return false;
return unserialize(file_get_contents($file));
}
function remove($config) {
$file = $this->generateFilePath($config);
if (!file_exists($file)) return false;
return unlink($file);
}
function flush($config) {
$dir = $this->generateDirectoryPath($config);
$dh = opendir($dir);
while (false !== ($filename = readdir($dh))) {
if (empty($filename)) continue;
if ($filename[0] === '.') continue;
unlink($dir . '/' . $filename);
}
}
function cleanup($config) {
$this->_prepareDir($config);
$dir = $this->generateDirectoryPath($config);
$dh = opendir($dir);
while (false !== ($filename = readdir($dh))) {
if (empty($filename)) continue;
if ($filename[0] === '.') continue;
$key = substr($filename, 0, strlen($filename) - 4);
if ($this->isOld($key, $config)) unlink($dir . '/' . $filename);
}
}
/**
* Generates the file path to the serial file corresponding to
* the configuration and definition name
*/
function generateFilePath($config) {
$key = $this->generateKey($config);
return $this->generateDirectoryPath($config) . '/' . $key . '.ser';
}
/**
* Generates the path to the directory contain this cache's serial files
* @note No trailing slash
*/
function generateDirectoryPath($config) {
$base = $config->get('Cache', 'SerializerPath');
$base = is_null($base) ? dirname(__FILE__) . '/Serializer' : $base;
return $base . '/' . $this->type;
}
/**
* Convenience wrapper function for file_put_contents
* @param $file File name to write to
* @param $data Data to write into file
* @return Number of bytes written if success, or false if failure.
*/
function _write($file, $data) {
static $file_put_contents;
if ($file_put_contents === null) {
$file_put_contents = function_exists('file_put_contents');
}
if ($file_put_contents) {
return file_put_contents($file, $data);
}
$fh = fopen($file, 'w');
if (!$fh) return false;
$status = fwrite($fh, $data);
fclose($fh);
return $status;
}
/**
* Prepares the directory that this type stores the serials in
*/
function _prepareDir($config) {
$directory = $this->generateDirectoryPath($config);
if (!is_dir($directory)) {
mkdir($directory);
}
}
}
?>

90
library/HTMLPurifier/DefinitionCacheFactory.php Normal file
View File

@ -0,0 +1,90 @@
<?php
require_once 'HTMLPurifier/DefinitionCache.php';
HTMLPurifier_ConfigSchema::define(
'Core', 'DefinitionCache', 'Serializer', 'string/null', '
This directive defines which method to use when caching definitions,
the complex data-type that makes HTML Purifier tick. Set to null
to disable caching (not recommended, as you will see a definite
performance degradation). This directive has been available since 2.0.0.
');
HTMLPurifier_ConfigSchema::defineAllowedValues(
'Core', 'DefinitionCache', array('Serializer')
);
/**
* Responsible for creating definition caches.
*/
class HTMLPurifier_DefinitionCacheFactory
{
var $caches = array('Serializer' => array());
var $decorators = array();
/**
* Initialize default decorators
*/
function setup() {
$this->addDecorator('Cleanup');
}
/**
* Retrieves an instance of global definition cache factory.
* @static
*/
static function &instance($prototype = null) {
static $instance;
if ($prototype !== null) {
$instance = $prototype;
} elseif ($instance === null || $prototype === true) {
$instance = new HTMLPurifier_DefinitionCacheFactory();
$instance->setup();
}
return $instance;
}
/**
* Factory method that creates a cache object based on configuration
* @param $name Name of definitions handled by cache
* @param $config Instance of HTMLPurifier_Config
*/
function &create($type, $config) {
// only one implementation as for right now, $config will
// be used to determine implementation
$method = $config->get('Core', 'DefinitionCache');
if ($method === null) {
$null = new HTMLPurifier_DefinitionCache_Null($type);
return $null;
}
if (!empty($this->caches[$method][$type])) {
return $this->caches[$method][$type];
}
$cache = new HTMLPurifier_DefinitionCache_Serializer($type);
foreach ($this->decorators as $decorator) {
$new_cache = $decorator->decorate($cache);
// prevent infinite recursion in PHP 4
unset($cache);
$cache = $new_cache;
}
$this->caches[$method][$type] = $cache;
return $this->caches[$method][$type];
}
/**
* Registers a decorator to add to all new cache objects
* @param
*/
function addDecorator($decorator) {
if (is_string($decorator)) {
$class = "HTMLPurifier_DefinitionCache_Decorator_$decorator";
$decorator = new $class;
}
$this->decorators[$decorator->name] = $decorator;
}
}
?>

55
library/HTMLPurifier/Doctype.php Normal file
View File

@ -0,0 +1,55 @@
<?php
/**
* Represents a document type, contains information on which modules
* need to be loaded.
*/
class HTMLPurifier_Doctype
{
/**
* Full name of doctype
*/
var $name;
/**
* List of standard modules (string identifiers or literal objects)
* that this doctype uses
*/
var $modules = array();
/**
* List of modules to use for tidying up code
*/
var $tidyModules = array();
/**
* Is the language derived from XML (i.e. XHTML)?
*/
var $xml = true;
/**
* List of aliases for this doctype
*/
var $aliases = array();
function HTMLPurifier_Doctype($name = null, $xml = true, $modules = array(),
$tidyModules = array(), $aliases = array()
) {
$this->name = $name;
$this->xml = $xml;
$this->modules = $modules;
$this->tidyModules = $tidyModules;
$this->aliases = $aliases;
}
/**
* Clones the doctype, use before resolving modes and the like
*/
function copy() {
return new HTMLPurifier_Doctype(
$this->name, $this->xml, $this->modules, $this->tidyModules, $this->aliases
);
}
}
?>

125
library/HTMLPurifier/DoctypeRegistry.php Normal file
View File

@ -0,0 +1,125 @@
<?php
require_once 'HTMLPurifier/Doctype.php';
// Legacy directives for doctype specification
HTMLPurifier_ConfigSchema::define(
'HTML', 'Strict', false, 'bool',
'Determines whether or not to use Transitional (loose) or Strict rulesets. '.
'This directive is deprecated in favor of %HTML.Doctype. '.
'This directive has been available since 1.3.0.'
);
HTMLPurifier_ConfigSchema::define(
'HTML', 'XHTML', true, 'bool',
'Determines whether or not output is XHTML 1.0 or HTML 4.01 flavor. '.
'This directive is deprecated in favor of %HTML.Doctype. '.
'This directive was available since 1.1.'
);
HTMLPurifier_ConfigSchema::defineAlias('Core', 'XHTML', 'HTML', 'XHTML');
class HTMLPurifier_DoctypeRegistry
{
/**
* Hash of doctype names to doctype objects
* @protected
*/
var $doctypes;
/**
* Lookup table of aliases to real doctype names
* @protected
*/
var $aliases;
/**
* Registers a doctype to the registry
* @note Accepts a fully-formed doctype object, or the
* parameters for constructing a doctype object
* @param $doctype Name of doctype or literal doctype object
* @param $modules Modules doctype will load
* @param $modules_for_modes Modules doctype will load for certain modes
* @param $aliases Alias names for doctype
* @return Reference to registered doctype (usable for further editing)
*/
function &register($doctype, $xml = true, $modules = array(),
$tidy_modules = array(), $aliases = array()
) {
if (!is_array($modules)) $modules = array($modules);
if (!is_array($tidy_modules)) $tidy_modules = array($tidy_modules);
if (!is_array($aliases)) $aliases = array($aliases);
if (!is_object($doctype)) {
$doctype = new HTMLPurifier_Doctype(
$doctype, $xml, $modules, $tidy_modules, $aliases
);
}
$this->doctypes[$doctype->name] =& $doctype;
$name = $doctype->name;
// hookup aliases
foreach ($doctype->aliases as $alias) {
if (isset($this->doctypes[$alias])) continue;
$this->aliases[$alias] = $name;
}
// remove old aliases
if (isset($this->aliases[$name])) unset($this->aliases[$name]);
return $doctype;
}
/**
* Retrieves reference to a doctype of a certain name
* @note This function resolves aliases
* @note When possible, use the more fully-featured make()
* @param $doctype Name of doctype
* @return Reference to doctype object
*/
function &get($doctype) {
if (isset($this->aliases[$doctype])) $doctype = $this->aliases[$doctype];
if (!isset($this->doctypes[$doctype])) {
trigger_error('Doctype ' . htmlspecialchars($doctype) . ' does not exist');
$anon = new HTMLPurifier_Doctype($doctype);
return $anon;
}
return $this->doctypes[$doctype];
}
/**
* Creates a doctype based on a configuration object,
* will perform initialization on the doctype
* @note Use this function to get a copy of doctype that config
* can hold on to (this is necessary in order to tell
* Generator whether or not the current document is XML
* based or not).
*/
function make($config) {
$original_doctype = $this->get($this->getDoctypeFromConfig($config));
$doctype = $original_doctype->copy();
return $doctype;
}
/**
* Retrieves the doctype from the configuration object
*/
function getDoctypeFromConfig($config) {
// recommended test
$doctype = $config->get('HTML', 'Doctype');
if ($doctype !== null) {
return $doctype;
}
// backwards-compatibility
if ($config->get('HTML', 'XHTML')) {
$doctype = 'XHTML 1.0';
} else {
$doctype = 'HTML 4.01';
}
if ($config->get('HTML', 'Strict')) {
$doctype .= ' Strict';
} else {
$doctype .= ' Transitional';
}
return $doctype;
}
}
?>

89
library/HTMLPurifier/ElementDef.php
View File

@ -51,6 +51,8 @@ class HTMLPurifier_ElementDef
* Abstract string representation of internal ChildDef rules. See
* HTMLPurifier_ContentSets for how this is parsed and then transformed
* into an HTMLPurifier_ChildDef.
* @warning This is a temporary variable that is not available after
* being processed by HTMLDefinition
* @public
*/
var $content_model;
@ -58,6 +60,9 @@ class HTMLPurifier_ElementDef
/**
* Value of $child->type, used to determine which ChildDef to use,
* used in combination with $content_model.
* @warning This must be lowercase
* @warning This is a temporary variable that is not available after
* being processed by HTMLDefinition
* @public
*/
var $content_model_type;
@ -78,14 +83,47 @@ class HTMLPurifier_ElementDef
* have to worry about this one.
* @public
*/
var $descendants_are_inline;
var $descendants_are_inline = false;
/**
* List of the names of required attributes this element has. Dynamically
* populated.
* @public
*/
var $required_attr = array();
/**
* Lookup table of tags excluded from all descendants of this tag.
* @note SGML permits exclusions for all descendants, but this is
* not possible with DTDs or XML Schemas. W3C has elected to
* use complicated compositions of content_models to simulate
* exclusion for children, but we go the simpler, SGML-style
* route of flat-out exclusions, which correctly apply to
* all descendants and not just children. Note that the XHTML
* Modularization Abstract Modules are blithely unaware of such
* distinctions.
* @public
*/
var $excludes = array();
/**
* Is this element safe for untrusted users to use?
*/
var $safe;
/**
* Low-level factory constructor for creating new standalone element defs
* @static
*/
static function create($safe, $content_model, $content_model_type, $attr) {
$def = new HTMLPurifier_ElementDef();
$def->safe = (bool) $safe;
$def->content_model = $content_model;
$def->content_model_type = $content_model_type;
$def->attr = $attr;
return $def;
}
/**
* Merges the values of another element definition into this one.
* Values from the new element def take precedence if a value is
@ -99,24 +137,57 @@ class HTMLPurifier_ElementDef
// merge in the includes
// sorry, no way to override an include
foreach ($v as $v2) {
$def->attr[0][] = $v2;
$this->attr[0][] = $v2;
}
continue;
}
if ($v === false) {
if (isset($this->attr[$k])) unset($this->attr[$k]);
continue;
}
$this->attr[$k] = $v;
}
foreach($def->attr_transform_pre as $k => $v) $this->attr_transform_pre[$k] = $v;
foreach($def->attr_transform_post as $k => $v) $this->attr_transform_post[$k] = $v;
foreach($def->auto_close as $k => $v) $this->auto_close[$k] = $v;
foreach($def->excludes as $k => $v) $this->excludes[$k] = $v;
$this->_mergeAssocArray($this->attr_transform_pre, $def->attr_transform_pre);
$this->_mergeAssocArray($this->attr_transform_post, $def->attr_transform_post);
$this->_mergeAssocArray($this->auto_close, $def->auto_close);
$this->_mergeAssocArray($this->excludes, $def->excludes);
if(!empty($def->content_model)) {
$this->content_model .= ' | ' . $def->content_model;
$this->child = false;
}
if(!empty($def->content_model_type)) {
$this->content_model_type = $def->content_model_type;
$this->child = false;
}
if(!is_null($def->child)) $this->child = $def->child;
if(!empty($def->content_model)) $this->content_model .= ' | ' . $def->content_model;
if(!empty($def->content_model_type)) $this->content_model_type = $def->content_model_type;
if(!is_null($def->descendants_are_inline)) $this->descendants_are_inline = $def->descendants_are_inline;
if($def->descendants_are_inline) $this->descendants_are_inline = $def->descendants_are_inline;
if(!is_null($def->safe)) $this->safe = $def->safe;
}
/**
* Merges one array into another, removes values which equal false
* @param $a1 Array by reference that is merged into
* @param $a2 Array that merges into $a1
*/
function _mergeAssocArray(&$a1, $a2) {
foreach ($a2 as $k => $v) {
if ($v === false) {
if (isset($a1[$k])) unset($a1[$k]);
continue;
}
$a1[$k] = $v;
}
}
/**
* Retrieves a copy of the element definition
*/
function copy() {
return unserialize(serialize($this));
}
}
?>

2
library/HTMLPurifier/Encoder.php
View File

@ -1,7 +1,5 @@
<?php
require_once 'HTMLPurifier/EntityLookup.php';
HTMLPurifier_ConfigSchema::define(
'Core', 'Encoding', 'utf-8', 'istring',
'If for some reason you are unable to convert all webpages to UTF-8, '.

5
library/HTMLPurifier/EntityParser.php
View File

@ -24,8 +24,8 @@ class HTMLPurifier_EntityParser
* @protected
*/
var $_substituteEntitiesRegex =
'/&(?:[#]x([a-fA-F0-9]+)|[#]0*(\d+)|([A-Za-z]+));?/';
// 1. hex 2. dec 3. string
'/&(?:[#]x([a-fA-F0-9]+)|[#]0*(\d+)|([A-Za-z_:][A-Za-z0-9.\-_:]*));?/';
// 1. hex 2. dec 3. string (XML style)
/**
@ -97,7 +97,6 @@ class HTMLPurifier_EntityParser
} else {
if (isset($this->_special_ent2dec[$matches[3]])) return $entity;
if (!$this->_entity_lookup) {
require_once 'HTMLPurifier/EntityLookup.php';
$this->_entity_lookup = HTMLPurifier_EntityLookup::instance();
}
if (isset($this->_entity_lookup->table[$matches[3]])) {

73
library/HTMLPurifier/ErrorCollector.php Normal file
View File

@ -0,0 +1,73 @@
<?php
require_once 'HTMLPurifier/Generator.php';
/**
* Error collection class that enables HTML Purifier to report HTML
* problems back to the user
*/
class HTMLPurifier_ErrorCollector
{
var $errors = array();
/**
* Sends an error message to the collector for later use
* @param string Error message text
* @param HTMLPurifier_Token Token that caused error
* @param array Tokens surrounding the offending token above, use true as placeholder
*/
function send($msg, $token, $context_tokens = array(true)) {
$this->errors[] = array($msg, $token, $context_tokens);
}
/**
* Retrieves raw error data for custom formatter to use
* @param List of arrays in format of array(Error message text,
* token that caused error, tokens surrounding token)
*/
function getRaw() {
return $this->errors;
}
/**
* Default HTML formatting implementation for error messages
* @param $config Configuration array, vital for HTML output nature
*/
function getHTMLFormatted($config) {
$generator = new HTMLPurifier_Generator();
$context = new HTMLPurifier_Context();
$generator->generateFromTokens(array(), $config, $context); // initialize
$ret = array();
$errors = $this->errors;
// sort error array by line
if ($config->get('Core', 'MaintainLineNumbers')) {
$lines = array();
foreach ($errors as $error) $lines[] = $error[1]->line;
array_multisort($lines, SORT_ASC, $errors);
}
foreach ($errors as $error) {
$string = $generator->escape($error[0]); // message
if (!empty($error[1]->line)) {
$string .= ' at line ' . $error[1]->line;
}
$string .= ' (<code>';
foreach ($error[2] as $token) {
if ($token !== true) {
$string .= $generator->escape($generator->generateFromToken($token));
} else {
$string .= '<strong>' . $generator->escape($generator->generateFromToken($error[1])) . '</strong>';
}
}
$string .= '</code>)';
$ret[] = $string;
}
return $ret;
}
}
?>

123
library/HTMLPurifier/Generator.php
View File

@ -1,59 +1,65 @@
<?php
require_once 'HTMLPurifier/Lexer.php';
HTMLPurifier_ConfigSchema::define(
'Core', 'CleanUTF8DuringGeneration', false, 'bool',
'When true, HTMLPurifier_Generator will also check all strings it '.
'escapes for UTF-8 well-formedness as a defense in depth measure. '.
'This could cause a considerable performance impact, and is not '.
'strictly necessary due to the fact that the Lexers should have '.
'ensured that all the UTF-8 strings were well-formed. Note that '.
'the configuration value is only read at the beginning of '.
'generateFromTokens.'
);
HTMLPurifier_ConfigSchema::define(
'Core', 'XHTML', true, 'bool',
'Determines whether or not output is XHTML or not. When disabled, HTML '.
'Purifier goes into HTML 4.01 removes XHTML-specific markup constructs, '.
'such as boolean attribute expansion and trailing slashes in empty tags. '.
'This directive was available since 1.1.'
'Output', 'CommentScriptContents', true, 'bool',
'Determines whether or not HTML Purifier should attempt to fix up '.
'the contents of script tags for legacy browsers with comments. This '.
'directive was available since 1.7.'
);
HTMLPurifier_ConfigSchema::defineAlias('Core', 'CommentScriptContents', 'Output', 'CommentScriptContents');
// extension constraints could be factored into ConfigSchema
HTMLPurifier_ConfigSchema::define(
'Core', 'TidyFormat', false, 'bool',
'<p>Determines whether or not to run Tidy on the final output for pretty '.
'formatting reasons, such as indentation and wrap.</p><p>This can greatly '.
'improve readability for editors who are hand-editing the HTML, but is '.
'by no means necessary as HTML Purifier has already fixed all major '.
'errors the HTML may have had. Tidy is a non-default extension, and this directive '.
'will silently fail if Tidy is not available.</p><p>If you are looking to make '.
'the overall look of your page\'s source better, I recommend running Tidy '.
'on the entire page rather than just user-content (after all, the '.
'indentation relative to the containing blocks will be incorrect).</p><p>This '.
'directive was available since 1.1.1.</p>'
'Output', 'TidyFormat', false, 'bool', <<<HTML
<p>
Determines whether or not to run Tidy on the final output for pretty
formatting reasons, such as indentation and wrap.
</p>
<p>
This can greatly improve readability for editors who are hand-editing
the HTML, but is by no means necessary as HTML Purifier has already
fixed all major errors the HTML may have had. Tidy is a non-default
extension, and this directive will silently fail if Tidy is not
available.
</p>
<p>
If you are looking to make the overall look of your page's source
better, I recommend running Tidy on the entire page rather than just
user-content (after all, the indentation relative to the containing
blocks will be incorrect).
</p>
<p>
This directive was available since 1.1.1.
</p>
HTML
);
HTMLPurifier_ConfigSchema::defineAlias('Core', 'TidyFormat', 'Output', 'TidyFormat');
/**
* Generates HTML from tokens.
* @todo Create a configuration-wide instance that all objects retrieve
*/
class HTMLPurifier_Generator
{
/**
* Bool cache of %Core.CleanUTF8DuringGeneration
* @private
*/
var $_clean_utf8 = false;
/**
* Bool cache of %Core.XHTML
* Bool cache of %HTML.XHTML
* @private
*/
var $_xhtml = true;
/**
* Bool cache of %Output.CommentScriptContents
* @private
*/
var $_scriptFix = false;
/**
* Cache of HTMLDefinition
* @private
*/
var $_def;
/**
* Generates HTML from an array of tokens.
* @param $tokens Array of HTMLPurifier_Token
@ -63,13 +69,24 @@ class HTMLPurifier_Generator
function generateFromTokens($tokens, $config, &$context) {
$html = '';
if (!$config) $config = HTMLPurifier_Config::createDefault();
$this->_clean_utf8 = $config->get('Core', 'CleanUTF8DuringGeneration');
$this->_xhtml = $config->get('Core', 'XHTML');
$this->_scriptFix = $config->get('Output', 'CommentScriptContents');
$this->_def = $config->getHTMLDefinition();
$this->_xhtml = $this->_def->doctype->xml;
if (!$tokens) return '';
foreach ($tokens as $token) {
$html .= $this->generateFromToken($token);
for ($i = 0, $size = count($tokens); $i < $size; $i++) {
if ($this->_scriptFix && $tokens[$i]->name === 'script') {
// script special case
$html .= $this->generateFromToken($tokens[$i++]);
$html .= $this->generateScriptFromToken($tokens[$i++]);
while ($tokens[$i]->name != 'script') {
$html .= $this->generateScriptFromToken($tokens[$i++]);
}
}
$html .= $this->generateFromToken($tokens[$i]);
}
if ($config->get('Core', 'TidyFormat') && extension_loaded('tidy')) {
if ($config->get('Output', 'TidyFormat') && extension_loaded('tidy')) {
$tidy_options = array(
'indent'=> true,
@ -104,14 +121,14 @@ class HTMLPurifier_Generator
function generateFromToken($token) {
if (!isset($token->type)) return '';
if ($token->type == 'start') {
$attr = $this->generateAttributes($token->attr);
$attr = $this->generateAttributes($token->attr, $token->name);
return '<' . $token->name . ($attr ? ' ' : '') . $attr . '>';
} elseif ($token->type == 'end') {
return '</' . $token->name . '>';
} elseif ($token->type == 'empty') {
$attr = $this->generateAttributes($token->attr);
$attr = $this->generateAttributes($token->attr, $token->name);
return '<' . $token->name . ($attr ? ' ' : '') . $attr .
( $this->_xhtml ? ' /': '' )
. '>';
@ -125,18 +142,33 @@ class HTMLPurifier_Generator
}
}
/**
* Special case processor for the contents of script tags
* @warning This runs into problems if there's already a literal
* --> somewhere inside the script contents.
*/
function generateScriptFromToken($token) {
if (!$token->type == 'text') return $this->generateFromToken($token);
return '<!--' . PHP_EOL . $token->data . PHP_EOL . '// -->';
// more advanced version:
// return '<!--//--><![CDATA[//><!--' . PHP_EOL . $token->data . PHP_EOL . '//--><!]]>';
}
/**
* Generates attribute declarations from attribute array.
* @param $assoc_array_of_attributes Attribute array
* @return Generate HTML fragment for insertion.
*/
function generateAttributes($assoc_array_of_attributes) {
function generateAttributes($assoc_array_of_attributes, $element) {
$html = '';
foreach ($assoc_array_of_attributes as $key => $value) {
if (!$this->_xhtml) {
// remove namespaced attributes
if (strpos($key, ':') !== false) continue;
// also needed: check for attribute minimization
if (!empty($this->_def->info[$element]->attr[$key]->minimized)) {
$html .= $key . ' ';
continue;
}
}
$html .= $key.'="'.$this->escape($value).'" ';
}
@ -149,7 +181,6 @@ class HTMLPurifier_Generator
* @return String escaped data.
*/
function escape($string) {
if ($this->_clean_utf8) $string = HTMLPurifier_Lexer::cleanUTF8($string);
return htmlspecialchars($string, ENT_COMPAT, 'UTF-8');
}

318
library/HTMLPurifier/HTMLDefinition.php
View File

@ -1,61 +1,132 @@
<?php
// components
require_once 'HTMLPurifier/Definition.php';
require_once 'HTMLPurifier/HTMLModuleManager.php';
// this definition and its modules MUST NOT define configuration directives
// outside of the HTML or Attr namespaces
// will be superceded by more accurate doctype declaration schemes
HTMLPurifier_ConfigSchema::define(
'HTML', 'Strict', false, 'bool',
'Determines whether or not to use Transitional (loose) or Strict rulesets. '.
'This directive has been available since 1.3.0.'
);
'HTML', 'DefinitionID', null, 'string/null', '
<p>
Unique identifier for a custom-built HTML definition. If you edit
the raw version of the HTMLDefinition, introducing changes that the
configuration object does not reflect, you must specify this variable.
If you change your custom edits, you should change this directive, or
clear your cache. Example:
</p>
<pre>
$config = HTMLPurifier_Config::createDefault();
$config->set(\'HTML\', \'DefinitionID\', \'1\');
$def = $config->getHTMLDefinition();
$def->addAttribute(\'a\', \'tabindex\', \'Number\');
</pre>
<p>
In the above example, the configuration is still at the defaults, but
using the advanced API, an extra attribute has been added. The
configuration object normally has no way of knowing that this change
has taken place, so it needs an extra directive: %HTML.DefinitionID.
If someone else attempts to use the default configuration, these two
pieces of code will not clobber each other in the cache, since one has
an extra directive attached to it.
</p>
<p>
This directive has been available since 2.0.0, and in that version or
later you <em>must</em> specify a value to this directive to use the
advanced API features.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'BlockWrapper', 'p', 'string',
'String name of element to wrap inline elements that are inside a block '.
'context. This only occurs in the children of blockquote in strict mode. '.
'Example: by default value, <code>&lt;blockquote&gt;Foo&lt;/blockquote&gt;</code> '.
'would become <code>&lt;blockquote&gt;&lt;p&gt;Foo&lt;/p&gt;&lt;/blockquote&gt;</code>. The '.
'<code>&lt;p&gt;</code> tags can be replaced '.
'with whatever you desire, as long as it is a block level element. '.
'This directive has been available since 1.3.0.'
);
'HTML', 'DefinitionRev', 1, 'int', '
<p>
Revision identifier for your custom definition specified in
%HTML.DefinitionID. This serves the same purpose: uniquely identifying
your custom definition, but this one does so in a chronological
context: revision 3 is more up-to-date then revision 2. Thus, when
this gets incremented, the cache handling is smart enough to clean
up any older revisions of your definition as well as flush the
cache. This directive has been available since 2.0.0.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'Parent', 'div', 'string',
'String name of element that HTML fragment passed to library will be '.
'inserted in. An interesting variation would be using span as the '.
'parent element, meaning that only inline tags would be allowed. '.
'This directive has been available since 1.3.0.'
);
'HTML', 'BlockWrapper', 'p', 'string', '
<p>
String name of element to wrap inline elements that are inside a block
context. This only occurs in the children of blockquote in strict mode.
</p>
<p>
Example: by default value,
<code>&lt;blockquote&gt;Foo&lt;/blockquote&gt;</code> would become
<code>&lt;blockquote&gt;&lt;p&gt;Foo&lt;/p&gt;&lt;/blockquote&gt;</code>.
The <code>&lt;p&gt;</code> tags can be replaced with whatever you desire,
as long as it is a block level element. This directive has been available
since 1.3.0.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'AllowedElements', null, 'lookup/null',
'If HTML Purifier\'s tag set is unsatisfactory for your needs, you '.
'can overload it with your own list of tags to allow. Note that this '.
'method is subtractive: it does its job by taking away from HTML Purifier '.
'usual feature set, so you cannot add a tag that HTML Purifier never '.
'supported in the first place (like embed, form or head). If you change this, you '.
'probably also want to change %HTML.AllowedAttributes. '.
'<strong>Warning:</strong> If another directive conflicts with the '.
'elements here, <em>that</em> directive will win and override. '.
'This directive has been available since 1.3.0.'
);
'HTML', 'Parent', 'div', 'string', '
<p>
String name of element that HTML fragment passed to library will be
inserted in. An interesting variation would be using span as the
parent element, meaning that only inline tags would be allowed.
This directive has been available since 1.3.0.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'AllowedAttributes', null, 'lookup/null',
'IF HTML Purifier\'s attribute set is unsatisfactory, overload it! '.
'The syntax is \'tag.attr\' or \'*.attr\' for the global attributes '.
'(style, id, class, dir, lang, xml:lang).'.
'<strong>Warning:</strong> If another directive conflicts with the '.
'elements here, <em>that</em> directive will win and override. For '.
'example, %HTML.EnableAttrID will take precedence over *.id in this '.
'directive. You must set that directive to true before you can use '.
'IDs at all. This directive has been available since 1.3.0.'
);
'HTML', 'AllowedElements', null, 'lookup/null', '
<p>
If HTML Purifier\'s tag set is unsatisfactory for your needs, you
can overload it with your own list of tags to allow. Note that this
method is subtractive: it does its job by taking away from HTML Purifier
usual feature set, so you cannot add a tag that HTML Purifier never
supported in the first place (like embed, form or head). If you
change this, you probably also want to change %HTML.AllowedAttributes.
</p>
<p>
<strong>Warning:</strong> If another directive conflicts with the
elements here, <em>that</em> directive will win and override.
This directive has been available since 1.3.0.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'AllowedAttributes', null, 'lookup/null', '
<p>
If HTML Purifier\'s attribute set is unsatisfactory, overload it!
The syntax is "tag.attr" or "*.attr" for the global attributes
(style, id, class, dir, lang, xml:lang).
</p>
<p>
<strong>Warning:</strong> If another directive conflicts with the
elements here, <em>that</em> directive will win and override. For
example, %HTML.EnableAttrID will take precedence over *.id in this
directive. You must set that directive to true before you can use
IDs at all. This directive has been available since 1.3.0.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'Allowed', null, 'string/null', '
<p>
This is a convenience directive that rolls the functionality of
%HTML.AllowedElements and %HTML.AllowedAttributes into one directive.
Specify elements and attributes that are allowed using:
<code>element1[attr1|attr2],element2...</code>.
</p>
<p>
<strong>Warning</strong>:
All of the constraints on the component directives are still enforced.
The syntax is a <em>subset</em> of TinyMCE\'s <code>valid_elements</code>
whitelist: directly copy-pasting it here will probably result in
broken whitelists. If %HTML.AllowedElements or %HTML.AllowedAttributes
are set, this directive has no effect.
This directive has been available since 2.0.0.
</p>
');
/**
* Definition of the purified HTML that describes allowed children,
@ -77,10 +148,10 @@ HTMLPurifier_ConfigSchema::define(
* HTMLPurifier_Printer_HTMLDefinition is a notable exception to this
* rule: in the interest of comprehensiveness, it will sniff everything.
*/
class HTMLPurifier_HTMLDefinition
class HTMLPurifier_HTMLDefinition extends HTMLPurifier_Definition
{
/** FULLY-PUBLIC VARIABLES */
// FULLY-PUBLIC VARIABLES ---------------------------------------------
/**
* Associative array of element names to HTMLPurifier_ElementDef
@ -139,50 +210,97 @@ class HTMLPurifier_HTMLDefinition
*/
var $info_content_sets = array();
/**
* Doctype object
*/
var $doctype;
/** PUBLIC BUT INTERNAL VARIABLES */
var $setup = false; /**< Has setup() been called yet? */
var $config; /**< Temporary instance of HTMLPurifier_Config */
// RAW CUSTOMIZATION STUFF --------------------------------------------
/**
* Adds a custom attribute to a pre-existing element
* @param $element_name String element name to add attribute to
* @param $attr_name String name of attribute
* @param $def Attribute definition, can be string or object, see
* HTMLPurifier_AttrTypes for details
*/
function addAttribute($element_name, $attr_name, $def) {
$module =& $this->getAnonymousModule();
$element =& $module->addBlankElement($element_name);
$element->attr[$attr_name] = $def;
}
/**
* Adds a custom element to your HTML definition
* @note See HTMLPurifier_HTMLModule::addElement for detailed
* parameter descriptions.
*/
function addElement($element_name, $type, $contents, $attr_collections, $attributes) {
$module =& $this->getAnonymousModule();
// assume that if the user is calling this, the element
// is safe. This may not be a good idea
$module->addElement($element_name, true, $type, $contents, $attr_collections, $attributes);
}
/**
* Retrieves a reference to the anonymous module, so you can
* bust out advanced features without having to make your own
* module.
*/
function &getAnonymousModule() {
if (!$this->_anonModule) {
$this->_anonModule = new HTMLPurifier_HTMLModule();
$this->_anonModule->name = 'Anonymous';
}
return $this->_anonModule;
}
var $_anonModule;
// PUBLIC BUT INTERNAL VARIABLES --------------------------------------
var $type = 'HTML';
var $manager; /**< Instance of HTMLPurifier_HTMLModuleManager */
/**
* Performs low-cost, preliminary initialization.
* @param $config Instance of HTMLPurifier_Config
*/
function HTMLPurifier_HTMLDefinition(&$config) {
$this->config =& $config;
function HTMLPurifier_HTMLDefinition() {
$this->manager = new HTMLPurifier_HTMLModuleManager();
}
/**
* Processes internals into form usable by HTMLPurifier internals.
* Modifying the definition after calling this function should not
* be done.
*/
function setup() {
// multiple call guard
if ($this->setup) {return;} else {$this->setup = true;}
$this->processModules();
$this->setupConfigStuff();
unset($this->config);
function doSetup($config) {
$this->processModules($config);
$this->setupConfigStuff($config);
unset($this->manager);
// cleanup some of the element definitions
foreach ($this->info as $k => $v) {
unset($this->info[$k]->content_model);
unset($this->info[$k]->content_model_type);
}
}
/**
* Extract out the information from the manager
*/
function processModules() {
function processModules($config) {
$this->manager->setup($this->config);
if ($this->_anonModule) {
// for user specific changes
// this is late-loaded so we don't have to deal with PHP4
// reference wonky-ness
$this->manager->addModule($this->_anonModule);
unset($this->_anonModule);
}
foreach ($this->manager->activeModules as $module) {
$this->manager->setup($config);
$this->doctype = $this->manager->doctype;
foreach ($this->manager->modules as $module) {
foreach($module->info_tag_transform as $k => $v) {
if ($v === false) unset($this->info_tag_transform[$k]);
else $this->info_tag_transform[$k] = $v;
@ -197,7 +315,7 @@ class HTMLPurifier_HTMLDefinition
}
}
$this->info = $this->manager->getElements($this->config);
$this->info = $this->manager->getElements();
$this->info_content_sets = $this->manager->contentSets->lookup;
}
@ -205,9 +323,9 @@ class HTMLPurifier_HTMLDefinition
/**
* Sets up stuff based on config. We need a better way of doing this.
*/
function setupConfigStuff() {
function setupConfigStuff($config) {
$block_wrapper = $this->config->get('HTML', 'BlockWrapper');
$block_wrapper = $config->get('HTML', 'BlockWrapper');
if (isset($this->info_content_sets['Block'][$block_wrapper])) {
$this->info_block_wrapper = $block_wrapper;
} else {
@ -215,24 +333,33 @@ class HTMLPurifier_HTMLDefinition
E_USER_ERROR);
}
$parent = $this->config->get('HTML', 'Parent');
$def = $this->manager->getElement($parent, $this->config);
$parent = $config->get('HTML', 'Parent');
$def = $this->manager->getElement($parent, true);
if ($def) {
$this->info_parent = $parent;
$this->info_parent_def = $def;
} else {
trigger_error('Cannot use unrecognized element as parent.',
E_USER_ERROR);
$this->info_parent_def = $this->manager->getElement(
$this->info_parent, $this->config);
$this->info_parent_def = $this->manager->getElement($this->info_parent, true);
}
// support template text
$support = "(for information on implementing this, see the ".
"support forums) ";
// setup allowed elements, SubtractiveWhitelist module
$allowed_elements = $this->config->get('HTML', 'AllowedElements');
// setup allowed elements
$allowed_elements = $config->get('HTML', 'AllowedElements');
$allowed_attributes = $config->get('HTML', 'AllowedAttributes');
if (!is_array($allowed_elements) && !is_array($allowed_attributes)) {
$allowed = $config->get('HTML', 'Allowed');
if (is_string($allowed)) {
list($allowed_elements, $allowed_attributes) = $this->parseTinyMCEAllowedList($allowed);
}
}
if (is_array($allowed_elements)) {
foreach ($this->info as $name => $d) {
if(!isset($allowed_elements[$name])) unset($this->info[$name]);
@ -240,11 +367,11 @@ class HTMLPurifier_HTMLDefinition
}
// emit errors
foreach ($allowed_elements as $element => $d) {
$element = htmlspecialchars($element);
trigger_error("Element '$element' is not supported $support", E_USER_WARNING);
}
}
$allowed_attributes = $this->config->get('HTML', 'AllowedAttributes');
$allowed_attributes_mutable = $allowed_attributes; // by copy!
if (is_array($allowed_attributes)) {
foreach ($this->info_global_attr as $attr_key => $info) {
@ -271,6 +398,8 @@ class HTMLPurifier_HTMLDefinition
// emit errors
foreach ($allowed_attributes_mutable as $elattr => $d) {
list($element, $attribute) = explode('.', $elattr);
$element = htmlspecialchars($element);
$attribute = htmlspecialchars($attribute);
if ($element == '*') {
trigger_error("Global attribute '$attribute' is not ".
"supported in any elements $support",
@ -284,6 +413,41 @@ class HTMLPurifier_HTMLDefinition
}
/**
* Parses a TinyMCE-flavored Allowed Elements and Attributes list into
* separate lists for processing. Format is element[attr1|attr2],element2...
* @warning Although it's largely drawn from TinyMCE's implementation,
* it is different, and you'll probably have to modify your lists
* @param $list String list to parse
* @param array($allowed_elements, $allowed_attributes)
*/
function parseTinyMCEAllowedList($list) {
$elements = array();
$attributes = array();
$chunks = explode(',', $list);
foreach ($chunks as $chunk) {
// remove TinyMCE element control characters
if (!strpos($chunk, '[')) {
$element = $chunk;
$attr = false;
} else {
list($element, $attr) = explode('[', $chunk);
}
if ($element !== '*') $elements[$element] = true;
if (!$attr) continue;
$attr = substr($attr, 0, strlen($attr) - 1); // remove trailing ]
$attr = explode('|', $attr);
foreach ($attr as $key) {
$attributes["$element.$key"] = true;
}
}
return array($elements, $attributes);
}
}

137
library/HTMLPurifier/HTMLModule.php
View File

@ -16,16 +16,14 @@
class HTMLPurifier_HTMLModule
{
// -- Overloadable ----------------------------------------------------
/**
* Short unique string identifier of the module
*/
var $name;
/**
* Dynamically set integer that specifies when the module was loaded in.
*/
var $order;
/**
* Informally, a list of elements this module changes. Not used in
* any significant way.
@ -99,27 +97,128 @@ class HTMLPurifier_HTMLModule
*/
function getChildDef($def) {return false;}
/**
* Hook method that lets module perform arbitrary operations on
* HTMLPurifier_HTMLDefinition before the module gets processed.
* @param $definition Reference to HTMLDefinition being setup
*/
function preProcess(&$definition) {}
// -- Convenience -----------------------------------------------------
/**
* Hook method that lets module perform arbitrary operations
* on HTMLPurifier_HTMLDefinition after the module gets processed.
* @param $definition Reference to HTMLDefinition being setup
* Convenience function that sets up a new element
* @param $element Name of element to add
* @param $safe Is element safe for untrusted users to use?
* @param $type What content set should element be registered to?
* Set as false to skip this step.
* @param $contents Allowed children in form of:
* "$content_model_type: $content_model"
* @param $attr_includes What attribute collections to register to
* element?
* @param $attr What unique attributes does the element define?
* @note See ElementDef for in-depth descriptions of these parameters.
* @return Reference to created element definition object, so you
* can set advanced parameters
* @protected
*/
function postProcess(&$definition) {}
function &addElement($element, $safe, $type, $contents, $attr_includes = array(), $attr = array()) {
$this->elements[] = $element;
// parse content_model
list($content_model_type, $content_model) = $this->parseContents($contents);
// merge in attribute inclusions
$this->mergeInAttrIncludes($attr, $attr_includes);
// add element to content sets
if ($type) $this->addElementToContentSet($element, $type);
// create element
$this->info[$element] = HTMLPurifier_ElementDef::create(
$safe, $content_model, $content_model_type, $attr
);
// literal object $contents means direct child manipulation
if (!is_string($contents)) $this->info[$element]->child = $contents;
return $this->info[$element];
}
/**
* Hook method that is called when a module gets registered to
* the definition.
* @param $definition Reference to HTMLDefinition being setup
* Convenience function that creates a totally blank, non-standalone
* element.
* @param $element Name of element to create
* @return Reference to created element
*/
function setup(&$definition) {}
function &addBlankElement($element) {
if (!isset($this->info[$element])) {
$this->elements[] = $element;
$this->info[$element] = new HTMLPurifier_ElementDef();
$this->info[$element]->standalone = false;
} else {
trigger_error("Definition for $element already exists in module, cannot redefine");
}
return $this->info[$element];
}
/**
* Convenience function that registers an element to a content set
* @param Element to register
* @param Name content set (warning: case sensitive, usually upper-case
* first letter)
* @protected
*/
function addElementToContentSet($element, $type) {
if (!isset($this->content_sets[$type])) $this->content_sets[$type] = '';
else $this->content_sets[$type] .= ' | ';
$this->content_sets[$type] .= $element;
}
/**
* Convenience function that transforms single-string contents
* into separate content model and content model type
* @param $contents Allowed children in form of:
* "$content_model_type: $content_model"
* @note If contents is an object, an array of two nulls will be
* returned, and the callee needs to take the original $contents
* and use it directly.
*/
function parseContents($contents) {
if (!is_string($contents)) return array(null, null); // defer
switch ($contents) {
// check for shorthand content model forms
case 'Empty':
return array('empty', '');
case 'Inline':
return array('optional', 'Inline | #PCDATA');
case 'Flow':
return array('optional', 'Flow | #PCDATA');
}
list($content_model_type, $content_model) = explode(':', $contents);
$content_model_type = strtolower(trim($content_model_type));
$content_model = trim($content_model);
return array($content_model_type, $content_model);
}
/**
* Convenience function that merges a list of attribute includes into
* an attribute array.
* @param $attr Reference to attr array to modify
* @param $attr_includes Array of includes / string include to merge in
*/
function mergeInAttrIncludes(&$attr, $attr_includes) {
if (!is_array($attr_includes)) {
if (empty($attr_includes)) $attr_includes = array();
else $attr_includes = array($attr_includes);
}
$attr[0] = $attr_includes;
}
/**
* Convenience function that generates a lookup table with boolean
* true as value.
* @param $list List of values to turn into a lookup
* @note You can also pass an arbitrary number of arguments in
* place of the regular argument
* @return Lookup array equivalent of list
*/
function makeLookup($list) {
if (is_string($list)) $list = func_get_args();
$ret = array();
foreach ($list as $value) {
if (is_null($value)) continue;
$ret[$value] = true;
}
return $ret;
}
}
?>

28
library/HTMLPurifier/HTMLModule/Bdo.php
View File

@ -11,30 +11,22 @@ class HTMLPurifier_HTMLModule_Bdo extends HTMLPurifier_HTMLModule
{
var $name = 'Bdo';
var $elements = array('bdo');
var $content_sets = array('Inline' => 'bdo');
var $attr_collections = array(
'I18N' => array('dir' => false)
);
function HTMLPurifier_HTMLModule_Bdo() {
$dir = new HTMLPurifier_AttrDef_Enum(array('ltr','rtl'), false);
$this->attr_collections['I18N']['dir'] = $dir;
$this->info['bdo'] = new HTMLPurifier_ElementDef();
$this->info['bdo']->attr = array(
0 => array('Core', 'Lang'),
'dir' => $dir, // required
// The Abstract Module specification has the attribute
// inclusions wrong for bdo: bdo allows
// xml:lang too (and we'll toss in lang for good measure,
// though it is not allowed for XHTML 1.1, this will
// be managed with a global attribute transform)
$bdo =& $this->addElement(
'bdo', true, 'Inline', 'Inline', array('Core', 'Lang'),
array(
'dir' => 'Enum#ltr,rtl', // required
// The Abstract Module specification has the attribute
// inclusions wrong for bdo: bdo allows Lang
)
);
$this->info['bdo']->content_model = '#PCDATA | Inline';
$this->info['bdo']->content_model_type = 'optional';
// provides fallback behavior if dir's missing (dir is required)
$this->info['bdo']->attr_transform_post['required-dir'] =
new HTMLPurifier_AttrTransform_BdoDir();
$bdo->attr_transform_post['required-dir'] = new HTMLPurifier_AttrTransform_BdoDir();
$this->attr_collections['I18N']['dir'] = 'Enum#ltr,rtl';
}
}

10
library/HTMLPurifier/HTMLModule/CommonAttributes.php
View File

@ -1,5 +1,7 @@
<?php
require_once 'HTMLPurifier/HTMLModule.php';
class HTMLPurifier_HTMLModule_CommonAttributes extends HTMLPurifier_HTMLModule
{
var $name = 'CommonAttributes';
@ -12,9 +14,7 @@ class HTMLPurifier_HTMLModule_CommonAttributes extends HTMLPurifier_HTMLModule
'id' => 'ID',
'title' => 'CDATA',
),
'Lang' => array(
'xml:lang' => false, // see constructor
),
'Lang' => array(),
'I18N' => array(
0 => array('Lang'), // proprietary, for xml:lang/lang
),
@ -22,10 +22,6 @@ class HTMLPurifier_HTMLModule_CommonAttributes extends HTMLPurifier_HTMLModule
0 => array('Core', 'I18N')
)
);
function HTMLPurifier_HTMLModule_CommonAttributes() {
$this->attr_collections['Lang']['xml:lang'] = new HTMLPurifier_AttrDef_Lang();
}
}
?>

32
library/HTMLPurifier/HTMLModule/Edit.php
View File

@ -11,28 +11,24 @@ class HTMLPurifier_HTMLModule_Edit extends HTMLPurifier_HTMLModule
{
var $name = 'Edit';
var $elements = array('del', 'ins');
var $content_sets = array('Inline' => 'del | ins');
function HTMLPurifier_HTMLModule_Edit() {
foreach ($this->elements as $element) {
$this->info[$element] = new HTMLPurifier_ElementDef();
$this->info[$element]->attr = array(
0 => array('Common'),
'cite' => 'URI',
// 'datetime' => 'Datetime' // Datetime not implemented
);
// Inline context ! Block context (exclamation mark is
// separator, see getChildDef for parsing)
$this->info[$element]->content_model =
'#PCDATA | Inline ! #PCDATA | Flow';
// HTML 4.01 specifies that ins/del must not contain block
// elements when used in an inline context, chameleon is
// a complicated workaround to acheive this effect
$this->info[$element]->content_model_type = 'chameleon';
}
$contents = 'Chameleon: #PCDATA | Inline ! #PCDATA | Flow';
$attr = array(
'cite' => 'URI',
// 'datetime' => 'Datetime', // not implemented
);
$this->addElement('del', true, 'Inline', $contents, 'Common', $attr);
$this->addElement('ins', true, 'Inline', $contents, 'Common', $attr);
}
// HTML 4.01 specifies that ins/del must not contain block
// elements when used in an inline context, chameleon is
// a complicated workaround to acheive this effect
// Inline context ! Block context (exclamation mark is
// separator, see getChildDef for parsing)
var $defines_child_def = true;
function getChildDef($def) {
if ($def->content_model_type != 'chameleon') return false;

29
library/HTMLPurifier/HTMLModule/Hypertext.php
View File

@ -10,25 +10,22 @@ class HTMLPurifier_HTMLModule_Hypertext extends HTMLPurifier_HTMLModule
{
var $name = 'Hypertext';
var $elements = array('a');
var $content_sets = array('Inline' => 'a');
function HTMLPurifier_HTMLModule_Hypertext() {
$this->info['a'] = new HTMLPurifier_ElementDef();
$this->info['a']->attr = array(
0 => array('Common'),
// 'accesskey' => 'Character',
// 'charset' => 'Charset',
'href' => 'URI',
//'hreflang' => 'LanguageCode',
'rel' => new HTMLPurifier_AttrDef_HTML_LinkTypes('rel'),
'rev' => new HTMLPurifier_AttrDef_HTML_LinkTypes('rev'),
//'tabindex' => 'Number',
//'type' => 'ContentType',
$a =& $this->addElement(
'a', true, 'Inline', 'Inline', 'Common',
array(
// 'accesskey' => 'Character',
// 'charset' => 'Charset',
'href' => 'URI',
// 'hreflang' => 'LanguageCode',
'rel' => new HTMLPurifier_AttrDef_HTML_LinkTypes('rel'),
'rev' => new HTMLPurifier_AttrDef_HTML_LinkTypes('rev'),
// 'tabindex' => 'Number',
// 'type' => 'ContentType',
)
);
$this->info['a']->content_model = '#PCDATA | Inline';
$this->info['a']->content_model_type = 'optional';
$this->info['a']->excludes = array('a' => true);
$a->excludes = array('a' => true);
}
}

24
library/HTMLPurifier/HTMLModule/Image.php
View File

@ -14,21 +14,21 @@ class HTMLPurifier_HTMLModule_Image extends HTMLPurifier_HTMLModule
{
var $name = 'Image';
var $elements = array('img');
var $content_sets = array('Inline' => 'img');
function HTMLPurifier_HTMLModule_Image() {
$this->info['img'] = new HTMLPurifier_ElementDef();
$this->info['img']->attr = array(
0 => array('Common'),
'alt' => 'Text',
'height' => 'Length',
'longdesc' => 'URI',
'src' => new HTMLPurifier_AttrDef_URI(true), // embedded
'width' => 'Length'
$img =& $this->addElement(
'img', true, 'Inline', 'Empty', 'Common',
array(
'alt*' => 'Text',
'height' => 'Length',
'longdesc' => 'URI',
'src*' => new HTMLPurifier_AttrDef_URI(true), // embedded
'width' => 'Length'
)
);
$this->info['img']->content_model_type = 'empty';
$this->info['img']->attr_transform_post[] =
// kind of strange, but splitting things up would be inefficient
$img->attr_transform_pre[] =
$img->attr_transform_post[] =
new HTMLPurifier_AttrTransform_ImgRequired();
}

125
library/HTMLPurifier/HTMLModule/Legacy.php
View File

@ -1,5 +1,7 @@
<?php
require_once 'HTMLPurifier/AttrDef/HTML/Bool.php';
/**
* XHTML 1.1 Legacy module defines elements that were previously
* deprecated.
@ -22,36 +24,115 @@ class HTMLPurifier_HTMLModule_Legacy extends HTMLPurifier_HTMLModule
// incomplete
var $name = 'Legacy';
var $elements = array('u', 's', 'strike');
var $non_standalone_elements = array('li', 'ol', 'address', 'blockquote');
function HTMLPurifier_HTMLModule_Legacy() {
// setup new elements
foreach ($this->elements as $name) {
$this->info[$name] = new HTMLPurifier_ElementDef();
// for u, s, strike, as more elements get added, add
// conditionals as necessary
$this->info[$name]->content_model = 'Inline | #PCDATA';
$this->info[$name]->content_model_type = 'optional';
$this->info[$name]->attr[0] = array('Common');
}
$this->addElement('basefont', true, 'Inline', 'Empty', false, array(
'color' => 'Color',
'face' => 'Text', // extremely broad, we should
'size' => 'Text', // tighten it
'id' => 'ID'
));
$this->addElement('center', true, 'Block', 'Flow', 'Common');
$this->addElement('dir', true, 'Block', 'Required: li', 'Common', array(
'compact' => 'Bool#compact'
));
$this->addElement('font', true, 'Inline', 'Inline', array('Core', 'I18N'), array(
'color' => 'Color',
'face' => 'Text', // extremely broad, we should
'size' => 'Text', // tighten it
));
$this->addElement('menu', true, 'Block', 'Required: li', 'Common', array(
'compact' => 'Bool#compact'
));
$this->addElement('s', true, 'Inline', 'Inline', 'Common');
$this->addElement('strike', true, 'Inline', 'Inline', 'Common');
$this->addElement('u', true, 'Inline', 'Inline', 'Common');
// setup modifications to old elements
foreach ($this->non_standalone_elements as $name) {
$this->info[$name] = new HTMLPurifier_ElementDef();
$this->info[$name]->standalone = false;
$align = 'Enum#left,right,center,justify';
$address =& $this->addBlankElement('address');
$address->content_model = 'Inline | #PCDATA | p';
$address->content_model_type = 'optional';
$address->child = false;
$blockquote =& $this->addBlankElement('blockquote');
$blockquote->content_model = 'Flow | #PCDATA';
$blockquote->content_model_type = 'optional';
$blockquote->child = false;
$br =& $this->addBlankElement('br');
$br->attr['clear'] = 'Enum#left,all,right,none';
$caption =& $this->addBlankElement('caption');
$caption->attr['align'] = 'Enum#top,bottom,left,right';
$div =& $this->addBlankElement('div');
$div->attr['align'] = $align;
$dl =& $this->addBlankElement('dl');
$dl->attr['compact'] = 'Bool#compact';
for ($i = 1; $i <= 6; $i++) {
$h =& $this->addBlankElement("h$i");
$h->attr['align'] = $align;
}
$this->info['li']->attr['value'] = new HTMLPurifier_AttrDef_Integer();
$this->info['ol']->attr['start'] = new HTMLPurifier_AttrDef_Integer();
$hr =& $this->addBlankElement('hr');
$hr->attr['align'] = $align;
$hr->attr['noshade'] = 'Bool#noshade';
$hr->attr['size'] = 'Pixels';
$hr->attr['width'] = 'Length';
$this->info['address']->content_model = 'Inline | #PCDATA | p';
$this->info['address']->content_model_type = 'optional';
$this->info['address']->child = false;
$img =& $this->addBlankElement('img');
$img->attr['align'] = 'Enum#top,middle,bottom,left,right';
$img->attr['border'] = 'Pixels';
$img->attr['hspace'] = 'Pixels';
$img->attr['vspace'] = 'Pixels';
$this->info['blockquote']->content_model = 'Flow | #PCDATA';
$this->info['blockquote']->content_model_type = 'optional';
$this->info['blockquote']->child = false;
// figure out this integer business
$li =& $this->addBlankElement('li');
$li->attr['value'] = new HTMLPurifier_AttrDef_Integer();
$li->attr['type'] = 'Enum#s:1,i,I,a,A,disc,square,circle';
$ol =& $this->addBlankElement('ol');
$ol->attr['compact'] = 'Bool#compact';
$ol->attr['start'] = new HTMLPurifier_AttrDef_Integer();
$ol->attr['type'] = 'Enum#s:1,i,I,a,A';
$p =& $this->addBlankElement('p');
$p->attr['align'] = $align;
$pre =& $this->addBlankElement('pre');
$pre->attr['width'] = 'Number';
// script omitted
$table =& $this->addBlankElement('table');
$table->attr['align'] = 'Enum#left,center,right';
$table->attr['bgcolor'] = 'Color';
$tr =& $this->addBlankElement('tr');
$tr->attr['bgcolor'] = 'Color';
$th =& $this->addBlankElement('th');
$th->attr['bgcolor'] = 'Color';
$th->attr['height'] = 'Length';
$th->attr['nowrap'] = 'Bool#nowrap';
$th->attr['width'] = 'Length';
$td =& $this->addBlankElement('td');
$td->attr['bgcolor'] = 'Color';
$td->attr['height'] = 'Length';
$td->attr['nowrap'] = 'Bool#nowrap';
$td->attr['width'] = 'Length';
$ul =& $this->addBlankElement('ul');
$ul->attr['compact'] = 'Bool#compact';
$ul->attr['type'] = 'Enum#square,disc,circle';
}

30
library/HTMLPurifier/HTMLModule/List.php
View File

@ -9,7 +9,6 @@ class HTMLPurifier_HTMLModule_List extends HTMLPurifier_HTMLModule
{
var $name = 'List';
var $elements = array('dl', 'dt', 'dd', 'ol', 'ul', 'li');
// According to the abstract schema, the List content set is a fully formed
// one or more expr, but it invariably occurs in an optional declaration
@ -19,26 +18,19 @@ class HTMLPurifier_HTMLModule_List extends HTMLPurifier_HTMLModule
// Furthermore, the actual XML Schema may disagree. Regardless,
// we don't have support for such nested expressions without using
// the incredibly inefficient and draconic Custom ChildDef.
var $content_sets = array('List' => 'dl | ol | ul', 'Flow' => 'List');
var $content_sets = array('Flow' => 'List');
function HTMLPurifier_HTMLModule_List() {
foreach ($this->elements as $element) {
$this->info[$element] = new HTMLPurifier_ElementDef();
$this->info[$element]->attr = array(0 => array('Common'));
if ($element == 'li' || $element == 'dd') {
$this->info[$element]->content_model = '#PCDATA | Flow';
$this->info[$element]->content_model_type = 'optional';
} elseif ($element == 'ol' || $element == 'ul') {
$this->info[$element]->content_model = 'li';
$this->info[$element]->content_model_type = 'required';
}
}
$this->info['dt']->content_model = '#PCDATA | Inline';
$this->info['dt']->content_model_type = 'optional';
$this->info['dl']->content_model = 'dt | dd';
$this->info['dl']->content_model_type = 'required';
// this could be a LOT more robust
$this->info['li']->auto_close = array('li' => true);
$this->addElement('ol', true, 'List', 'Required: li', 'Common');
$this->addElement('ul', true, 'List', 'Required: li', 'Common');
$this->addElement('dl', true, 'List', 'Required: dt | dd', 'Common');
$li =& $this->addElement('li', true, false, 'Flow', 'Common');
$li->auto_close = array('li' => true);
$this->addElement('dd', true, false, 'Flow', 'Common');
$this->addElement('dt', true, false, 'Inline', 'Common');
}
}

16
library/HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php Normal file
View File

@ -0,0 +1,16 @@
<?php
require_once 'HTMLPurifier/HTMLModule.php';
class HTMLPurifier_HTMLModule_NonXMLCommonAttributes extends HTMLPurifier_HTMLModule
{
var $name = 'NonXMLCommonAttributes';
var $attr_collections = array(
'Lang' => array(
'lang' => 'LanguageCode',
)
);
}
?>

23
library/HTMLPurifier/HTMLModule/Presentation.php
View File

@ -16,23 +16,16 @@ class HTMLPurifier_HTMLModule_Presentation extends HTMLPurifier_HTMLModule
{
var $name = 'Presentation';
var $elements = array('b', 'big', 'hr', 'i', 'small', 'sub', 'sup', 'tt');
var $content_sets = array(
'Block' => 'hr',
'Inline' => 'b | big | i | small | sub | sup | tt'
);
function HTMLPurifier_HTMLModule_Presentation() {
foreach ($this->elements as $element) {
$this->info[$element] = new HTMLPurifier_ElementDef();
$this->info[$element]->attr = array(0 => array('Common'));
if ($element == 'hr') {
$this->info[$element]->content_model_type = 'empty';
} else {
$this->info[$element]->content_model = '#PCDATA | Inline';
$this->info[$element]->content_model_type = 'optional';
}
}
$this->addElement('b', true, 'Inline', 'Inline', 'Common');
$this->addElement('big', true, 'Inline', 'Inline', 'Common');
$this->addElement('hr', true, 'Block', 'Empty', 'Common');
$this->addElement('i', true, 'Inline', 'Inline', 'Common');
$this->addElement('small', true, 'Inline', 'Inline', 'Common');
$this->addElement('sub', true, 'Inline', 'Inline', 'Common');
$this->addElement('sup', true, 'Inline', 'Inline', 'Common');
$this->addElement('tt', true, 'Inline', 'Inline', 'Common');
}
}

4
library/HTMLPurifier/HTMLModule/Scripting.php
View File

@ -46,8 +46,12 @@ class HTMLPurifier_HTMLModule_Scripting extends HTMLPurifier_HTMLModule
// blockquote's custom definition (we would use it but
// blockquote's contents are optional while noscript's contents
// are required)
// TODO: convert this to new syntax, main problem is getting
// both content sets working
foreach ($this->elements as $element) {
$this->info[$element] = new HTMLPurifier_ElementDef();
$this->info[$element]->safe = false;
}
$this->info['noscript']->attr = array( 0 => array('Common') );
$this->info['noscript']->content_model = 'Heading | List | Block';

107
library/HTMLPurifier/HTMLModule/Tables.php
View File

@ -10,75 +10,60 @@ class HTMLPurifier_HTMLModule_Tables extends HTMLPurifier_HTMLModule
{
var $name = 'Tables';
var $elements = array('caption', 'table', 'td', 'th', 'tr', 'col',
'colgroup', 'tbody', 'thead', 'tfoot');
var $content_sets = array('Block' => 'table');
function HTMLPurifier_HTMLModule_Tables() {
foreach ($this->elements as $e) {
$this->info[$e] = new HTMLPurifier_ElementDef();
$this->info[$e]->attr = array(0 => array('Common'));
$attr =& $this->info[$e]->attr;
if ($e == 'caption') continue;
if ($e == 'table'){
$attr['border'] = 'Pixels';
$attr['cellpadding'] = 'Length';
$attr['cellspacing'] = 'Length';
$attr['frame'] = new HTMLPurifier_AttrDef_Enum(array(
'void', 'above', 'below', 'hsides', 'lhs', 'rhs',
'vsides', 'box', 'border'
), false);
$attr['rules'] = new HTMLPurifier_AttrDef_Enum(array(
'none', 'groups', 'rows', 'cols', 'all'
), false);
$attr['summary'] = 'Text';
$attr['width'] = 'Length';
continue;
}
if ($e == 'col' || $e == 'colgroup') {
$attr['span'] = 'Number';
$attr['width'] = 'MultiLength';
}
if ($e == 'td' || $e == 'th') {
$attr['abbr'] = 'Text';
$attr['colspan'] = 'Number';
$attr['rowspan'] = 'Number';
}
$attr['align'] = new HTMLPurifier_AttrDef_Enum(array(
'left', 'center', 'right', 'justify', 'char'
), false);
$attr['valign'] = new HTMLPurifier_AttrDef_Enum(array(
'top', 'middle', 'bottom', 'baseline'
), false);
$attr['charoff'] = 'Length';
}
$this->info['caption']->content_model = '#PCDATA | Inline';
$this->info['caption']->content_model_type = 'optional';
// Is done directly because it doesn't leverage substitution
// mechanisms. True model is:
// 'caption?, ( col* | colgroup* ), (( thead?, tfoot?, tbody+ ) | ( tr+ ))'
$this->info['table']->child = new HTMLPurifier_ChildDef_Table();
$this->addElement('caption', true, false, 'Inline', 'Common');
$this->info['td']->content_model =
$this->info['th']->content_model = '#PCDATA | Flow';
$this->info['td']->content_model_type =
$this->info['th']->content_model_type = 'optional';
$this->addElement('table', true, 'Block',
new HTMLPurifier_ChildDef_Table(), 'Common',
array(
'border' => 'Pixels',
'cellpadding' => 'Length',
'cellspacing' => 'Length',
'frame' => 'Enum#void,above,below,hsides,lhs,rhs,vsides,box,border',
'rules' => 'Enum#none,groups,rows,cols,all',
'summary' => 'Text',
'width' => 'Length'
)
);
$this->info['tr']->content_model = 'td | th';
$this->info['tr']->content_model_type = 'required';
// common attributes
$cell_align = array(
'align' => 'Enum#left,center,right,justify,char',
'charoff' => 'Length',
'valign' => 'Enum#top,middle,bottom,baseline',
);
$this->info['col']->content_model_type = 'empty';
$cell_t = array_merge(
array(
'abbr' => 'Text',
'colspan' => 'Number',
'rowspan' => 'Number',
),
$cell_align
);
$this->addElement('td', true, false, 'Flow', 'Common', $cell_t);
$this->addElement('th', true, false, 'Flow', 'Common', $cell_t);
$this->info['colgroup']->content_model = 'col';
$this->info['colgroup']->content_model_type = 'optional';
$this->addElement('tr', true, false, 'Required: td | th', 'Common', $cell_align);
$this->info['tbody']->content_model =
$this->info['thead']->content_model =
$this->info['tfoot']->content_model = 'tr';
$this->info['tbody']->content_model_type =
$this->info['thead']->content_model_type =
$this->info['tfoot']->content_model_type = 'required';
$cell_col = array_merge(
array(
'span' => 'Number',
'width' => 'MultiLength',
),
$cell_align
);
$this->addElement('col', true, false, 'Empty', 'Common', $cell_col);
$colgroup =& $this->addElement('colgroup', true, false, 'Optional: col', 'Common', $cell_col);
$colgroup->auto_close = $this->makeLookup(
'thead', 'tbody', 'tfoot', 'tr'
);
$this->addElement('tbody', true, false, 'Required: tr', 'Common', $cell_align);
$this->addElement('thead', true, false, 'Required: tr', 'Common', $cell_align);
$this->addElement('tfoot', true, false, 'Required: tr', 'Common', $cell_align);
}

9
library/HTMLPurifier/HTMLModule/Target.php
View File

@ -9,13 +9,12 @@ class HTMLPurifier_HTMLModule_Target extends HTMLPurifier_HTMLModule
{
var $name = 'Target';
var $elements = array('a');
function HTMLPurifier_HTMLModule_Target() {
foreach ($this->elements as $e) {
$this->info[$e] = new HTMLPurifier_ElementDef();
$this->info[$e]->standalone = false;
$this->info[$e]->attr = array(
$elements = array('a');
foreach ($elements as $name) {
$e =& $this->addBlankElement($name);
$e->attr = array(
'target' => new HTMLPurifier_AttrDef_HTML_FrameTarget()
);
}

89
library/HTMLPurifier/HTMLModule/Text.php
View File

@ -10,65 +10,60 @@ require_once 'HTMLPurifier/HTMLModule.php';
* - Block Structural (div, p)
* - Inline Phrasal (abbr, acronym, cite, code, dfn, em, kbd, q, samp, strong, var)
* - Inline Structural (br, span)
* We have elected not to follow suite, but this may change.
* This module, functionally, does not distinguish between these
* sub-modules, but the code is internally structured to reflect
* these distinctions.
*/
class HTMLPurifier_HTMLModule_Text extends HTMLPurifier_HTMLModule
{
var $name = 'Text';
var $elements = array('abbr', 'acronym', 'address', 'blockquote',
'br', 'cite', 'code', 'dfn', 'div', 'em', 'h1', 'h2', 'h3',
'h4', 'h5', 'h6', 'kbd', 'p', 'pre', 'q', 'samp', 'span', 'strong',
'var');
var $content_sets = array(
'Heading' => 'h1 | h2 | h3 | h4 | h5 | h6',
'Block' => 'address | blockquote | div | p | pre',
'Inline' => 'abbr | acronym | br | cite | code | dfn | em | kbd | q | samp | span | strong | var',
'Flow' => 'Heading | Block | Inline'
);
function HTMLPurifier_HTMLModule_Text() {
foreach ($this->elements as $element) {
$this->info[$element] = new HTMLPurifier_ElementDef();
// attributes
if ($element == 'br') {
$this->info[$element]->attr = array(0 => array('Core'));
} elseif ($element == 'blockquote' || $element == 'q') {
$this->info[$element]->attr = array(0 => array('Common'), 'cite' => 'URI');
} else {
$this->info[$element]->attr = array(0 => array('Common'));
}
// content models
if ($element == 'br') {
$this->info[$element]->content_model_type = 'empty';
} elseif ($element == 'blockquote') {
$this->info[$element]->content_model = 'Heading | Block | List';
$this->info[$element]->content_model_type = 'optional';
} elseif ($element == 'div') {
$this->info[$element]->content_model = '#PCDATA | Flow';
$this->info[$element]->content_model_type = 'optional';
} else {
$this->info[$element]->content_model = '#PCDATA | Inline';
$this->info[$element]->content_model_type = 'optional';
}
}
// SGML permits exclusions for all descendants, but this is
// not possible with DTDs or XML Schemas. W3C has elected to
// use complicated compositions of content_models to simulate
// exclusion for children, but we go the simpler, SGML-style
// route of flat-out exclusions. Note that the Abstract Module
// is blithely unaware of such distinctions.
$this->info['pre']->excludes = array_flip(array(
'img', 'big', 'small',
'object', 'applet', 'font', 'basefont' // generally not allowed
));
$this->info['p']->auto_close = array_flip(array(
// Inline Phrasal -------------------------------------------------
$this->addElement('abbr', true, 'Inline', 'Inline', 'Common');
$this->addElement('acronym', true, 'Inline', 'Inline', 'Common');
$this->addElement('cite', true, 'Inline', 'Inline', 'Common');
$this->addElement('code', true, 'Inline', 'Inline', 'Common');
$this->addElement('dfn', true, 'Inline', 'Inline', 'Common');
$this->addElement('em', true, 'Inline', 'Inline', 'Common');
$this->addElement('kbd', true, 'Inline', 'Inline', 'Common');
$this->addElement('q', true, 'Inline', 'Inline', 'Common', array('cite' => 'URI'));
$this->addElement('samp', true, 'Inline', 'Inline', 'Common');
$this->addElement('strong', true, 'Inline', 'Inline', 'Common');
$this->addElement('var', true, 'Inline', 'Inline', 'Common');
// Inline Structural ----------------------------------------------
$this->addElement('span', true, 'Inline', 'Inline', 'Common');
$this->addElement('br', true, 'Inline', 'Empty', 'Core');
// Block Phrasal --------------------------------------------------
$this->addElement('address', true, 'Block', 'Inline', 'Common');
$this->addElement('blockquote', true, 'Block', 'Optional: Heading | Block | List', 'Common', array('cite' => 'URI') );
$pre =& $this->addElement('pre', true, 'Block', 'Inline', 'Common');
$pre->excludes = $this->makeLookup(
'img', 'big', 'small', 'object', 'applet', 'font', 'basefont' );
$this->addElement('h1', true, 'Heading', 'Inline', 'Common');
$this->addElement('h2', true, 'Heading', 'Inline', 'Common');
$this->addElement('h3', true, 'Heading', 'Inline', 'Common');
$this->addElement('h4', true, 'Heading', 'Inline', 'Common');
$this->addElement('h5', true, 'Heading', 'Inline', 'Common');
$this->addElement('h6', true, 'Heading', 'Inline', 'Common');
// Block Structural -----------------------------------------------
$p =& $this->addElement('p', true, 'Block', 'Inline', 'Common');
// this seems really ad hoc: implementing some general
// heuristics would probably be better
$p->auto_close = $this->makeLookup(
'address', 'blockquote', 'dd', 'dir', 'div', 'dl', 'dt',
'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'ol', 'p', 'pre',
'table', 'ul'
));
'table', 'ul' );
$this->addElement('div', true, 'Block', 'Flow', 'Common');
}
}

241
library/HTMLPurifier/HTMLModule/Tidy.php Normal file
View File

@ -0,0 +1,241 @@
<?php
require_once 'HTMLPurifier/HTMLModule.php';
HTMLPurifier_ConfigSchema::define(
'HTML', 'TidyLevel', 'medium', 'string', '
<p>General level of cleanliness the Tidy module should enforce.
There are four allowed values:</p>
<dl>
<dt>none</dt>
<dd>No extra tidying should be done</dd>
<dt>light</dt>
<dd>Only fix elements that would be discarded otherwise due to
lack of support in doctype</dd>
<dt>medium</dt>
<dd>Enforce best practices</dd>
<dt>heavy</dt>
<dd>Transform all deprecated elements and attributes to standards
compliant equivalents</dd>
</dl>
<p>This directive has been available since 2.0.0</p>
' );
HTMLPurifier_ConfigSchema::defineAllowedValues(
'HTML', 'TidyLevel', array('none', 'light', 'medium', 'heavy')
);
HTMLPurifier_ConfigSchema::define(
'HTML', 'TidyAdd', array(), 'lookup', '
Fixes to add to the default set of Tidy fixes as per your level. This
directive has been available since 2.0.0.
' );
HTMLPurifier_ConfigSchema::define(
'HTML', 'TidyRemove', array(), 'lookup', '
Fixes to remove from the default set of Tidy fixes as per your level. This
directive has been available since 2.0.0.
' );
/**
* Abstract class for a set of proprietary modules that clean up (tidy)
* poorly written HTML.
*/
class HTMLPurifier_HTMLModule_Tidy extends HTMLPurifier_HTMLModule
{
/**
* List of supported levels. Index zero is a special case "no fixes"
* level.
*/
var $levels = array(0 => 'none', 'light', 'medium', 'heavy');
/**
* Default level to place all fixes in. Disabled by default
*/
var $defaultLevel = null;
/**
* Lists of fixes used by getFixesForLevel(). Format is:
* HTMLModule_Tidy->fixesForLevel[$level] = array('fix-1', 'fix-2');
*/
var $fixesForLevel = array(
'light' => array(),
'medium' => array(),
'heavy' => array()
);
/**
* Lazy load constructs the module by determining the necessary
* fixes to create and then delegating to the populate() function.
* @todo Wildcard matching and error reporting when an added or
* subtracted fix has no effect.
*/
function construct($config) {
// create fixes, initialize fixesForLevel
$fixes = $this->makeFixes();
$this->makeFixesForLevel($fixes);
// figure out which fixes to use
$level = $config->get('HTML', 'TidyLevel');
$fixes_lookup = $this->getFixesForLevel($level);
// get custom fix declarations: these need namespace processing
$add_fixes = $config->get('HTML', 'TidyAdd');
$remove_fixes = $config->get('HTML', 'TidyRemove');
foreach ($fixes as $name => $fix) {
// needs to be refactored a little to implement globbing
if (
isset($remove_fixes[$name]) ||
(!isset($add_fixes[$name]) && !isset($fixes_lookup[$name]))
) {
unset($fixes[$name]);
}
}
// populate this module with necessary fixes
$this->populate($fixes);
}
/**
* Retrieves all fixes per a level, returning fixes for that specific
* level as well as all levels below it.
* @param $level String level identifier, see $levels for valid values
* @return Lookup up table of fixes
*/
function getFixesForLevel($level) {
if ($level == $this->levels[0]) {
return array();
}
$activated_levels = array();
for ($i = 1, $c = count($this->levels); $i < $c; $i++) {
$activated_levels[] = $this->levels[$i];
if ($this->levels[$i] == $level) break;
}
if ($i == $c) {
trigger_error(
'Tidy level ' . htmlspecialchars($level) . ' not recognized',
E_USER_WARNING
);
return array();
}
$ret = array();
foreach ($activated_levels as $level) {
foreach ($this->fixesForLevel[$level] as $fix) {
$ret[$fix] = true;
}
}
return $ret;
}
/**
* Dynamically populates the $fixesForLevel member variable using
* the fixes array. It may be custom overloaded, used in conjunction
* with $defaultLevel, or not used at all.
*/
function makeFixesForLevel($fixes) {
if (!isset($this->defaultLevel)) return;
if (!isset($this->fixesForLevel[$this->defaultLevel])) {
trigger_error(
'Default level ' . $this->defaultLevel . ' does not exist',
E_USER_ERROR
);
return;
}
$this->fixesForLevel[$this->defaultLevel] = array_keys($fixes);
}
/**
* Populates the module with transforms and other special-case code
* based on a list of fixes passed to it
* @param $lookup Lookup table of fixes to activate
*/
function populate($fixes) {
foreach ($fixes as $name => $fix) {
// determine what the fix is for
list($type, $params) = $this->getFixType($name);
switch ($type) {
case 'attr_transform_pre':
case 'attr_transform_post':
$attr = $params['attr'];
if (isset($params['element'])) {
$element = $params['element'];
if (empty($this->info[$element])) {
$e =& $this->addBlankElement($element);
} else {
$e =& $this->info[$element];
}
} else {
$type = "info_$type";
$e =& $this;
}
$f =& $e->$type;
$f[$attr] = $fix;
break;
case 'tag_transform':
$this->info_tag_transform[$params['element']] = $fix;
break;
case 'child':
case 'content_model_type':
$element = $params['element'];
if (empty($this->info[$element])) {
$e =& $this->addBlankElement($element);
} else {
$e =& $this->info[$element];
}
$e->$type = $fix;
break;
default:
trigger_error("Fix type $type not supported", E_USER_ERROR);
break;
}
}
}
/**
* Parses a fix name and determines what kind of fix it is, as well
* as other information defined by the fix
* @param $name String name of fix
* @return array(string $fix_type, array $fix_parameters)
* @note $fix_parameters is type dependant, see populate() for usage
* of these parameters
*/
function getFixType($name) {
// parse it
$property = $attr = null;
if (strpos($name, '#') !== false) list($name, $property) = explode('#', $name);
if (strpos($name, '@') !== false) list($name, $attr) = explode('@', $name);
// figure out the parameters
$params = array();
if ($name !== '') $params['element'] = $name;
if (!is_null($attr)) $params['attr'] = $attr;
// special case: attribute transform
if (!is_null($attr)) {
if (is_null($property)) $property = 'pre';
$type = 'attr_transform_' . $property;
return array($type, $params);
}
// special case: tag transform
if (is_null($property)) {
return array('tag_transform', $params);
}
return array($property, $params);
}
/**
* Defines all fixes the module will perform in a compact
* associative array of fix name to fix implementation.
* @abstract
*/
function makeFixes() {}
}
?>

18
library/HTMLPurifier/HTMLModule/Tidy/Proprietary.php Normal file
View File

@ -0,0 +1,18 @@
<?php
require_once 'HTMLPurifier/HTMLModule/Tidy.php';
class HTMLPurifier_HTMLModule_Tidy_Proprietary extends
HTMLPurifier_HTMLModule_Tidy
{
var $name = 'Tidy_Proprietary';
var $defaultLevel = 'light';
function makeFixes() {
return array();
}
}
?>

21
library/HTMLPurifier/HTMLModule/Tidy/XHTML.php Normal file
View File

@ -0,0 +1,21 @@
<?php
require_once 'HTMLPurifier/HTMLModule/Tidy.php';
require_once 'HTMLPurifier/AttrTransform/Lang.php';
class HTMLPurifier_HTMLModule_Tidy_XHTML extends
HTMLPurifier_HTMLModule_Tidy
{
var $name = 'Tidy_XHTML';
var $defaultLevel = 'medium';
function makeFixes() {
$r = array();
$r['@lang'] = new HTMLPurifier_AttrTransform_Lang();
return $r;
}
}
?>

193
library/HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php Normal file
View File

@ -0,0 +1,193 @@
<?php
require_once 'HTMLPurifier/HTMLModule/Tidy.php';
require_once 'HTMLPurifier/TagTransform/Simple.php';
require_once 'HTMLPurifier/TagTransform/Font.php';
require_once 'HTMLPurifier/AttrTransform/BgColor.php';
require_once 'HTMLPurifier/AttrTransform/BoolToCSS.php';
require_once 'HTMLPurifier/AttrTransform/Border.php';
require_once 'HTMLPurifier/AttrTransform/Name.php';
require_once 'HTMLPurifier/AttrTransform/Length.php';
require_once 'HTMLPurifier/AttrTransform/ImgSpace.php';
require_once 'HTMLPurifier/AttrTransform/EnumToCSS.php';
class HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4 extends
HTMLPurifier_HTMLModule_Tidy
{
function makeFixes() {
$r = array();
// == deprecated tag transforms ===================================
$r['font'] = new HTMLPurifier_TagTransform_Font();
$r['menu'] = new HTMLPurifier_TagTransform_Simple('ul');
$r['dir'] = new HTMLPurifier_TagTransform_Simple('ul');
$r['center'] = new HTMLPurifier_TagTransform_Simple('div', 'text-align:center;');
$r['u'] = new HTMLPurifier_TagTransform_Simple('span', 'text-decoration:underline;');
$r['s'] = new HTMLPurifier_TagTransform_Simple('span', 'text-decoration:line-through;');
$r['strike'] = new HTMLPurifier_TagTransform_Simple('span', 'text-decoration:line-through;');
// == deprecated attribute transforms =============================
$r['caption@align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
// we're following IE's behavior, not Firefox's, due
// to the fact that no one supports caption-side:right,
// W3C included (with CSS 2.1). This is a slightly
// unreasonable attribute!
'left' => 'text-align:left;',
'right' => 'text-align:right;',
'top' => 'caption-side:top;',
'bottom' => 'caption-side:bottom;' // not supported by IE
));
// @align for img -------------------------------------------------
$r['img@align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
'left' => 'float:left;',
'right' => 'float:right;',
'top' => 'vertical-align:top;',
'middle' => 'vertical-align:middle;',
'bottom' => 'vertical-align:baseline;',
));
// @align for table -----------------------------------------------
$r['table@align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
'left' => 'float:left;',
'center' => 'margin-left:auto;margin-right:auto;',
'right' => 'float:right;'
));
// @align for hr -----------------------------------------------
$r['hr@align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
// we use both text-align and margin because these work
// for different browsers (IE and Firefox, respectively)
// and the melange makes for a pretty cross-compatible
// solution
'left' => 'margin-left:0;margin-right:auto;text-align:left;',
'center' => 'margin-left:auto;margin-right:auto;text-align:center;',
'right' => 'margin-left:auto;margin-right:0;text-align:right;'
));
// @align for h1, h2, h3, h4, h5, h6, p, div ----------------------
// {{{
$align_lookup = array();
$align_values = array('left', 'right', 'center', 'justify');
foreach ($align_values as $v) $align_lookup[$v] = "text-align:$v;";
// }}}
$r['h1@align'] =
$r['h2@align'] =
$r['h3@align'] =
$r['h4@align'] =
$r['h5@align'] =
$r['h6@align'] =
$r['p@align'] =
$r['div@align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', $align_lookup);
// @bgcolor for table, tr, td, th ---------------------------------
$r['table@bgcolor'] =
$r['td@bgcolor'] =
$r['th@bgcolor'] =
new HTMLPurifier_AttrTransform_BgColor();
// @border for img ------------------------------------------------
$r['img@border'] = new HTMLPurifier_AttrTransform_Border();
// @clear for br --------------------------------------------------
$r['br@clear'] =
new HTMLPurifier_AttrTransform_EnumToCSS('clear', array(
'left' => 'clear:left;',
'right' => 'clear:right;',
'all' => 'clear:both;',
'none' => 'clear:none;',
));
// @height for td, th ---------------------------------------------
$r['td@height'] =
$r['th@height'] =
new HTMLPurifier_AttrTransform_Length('height');
// @hspace for img ------------------------------------------------
$r['img@hspace'] = new HTMLPurifier_AttrTransform_ImgSpace('hspace');
// @name for img, a -----------------------------------------------
$r['img@name'] =
$r['a@name'] = new HTMLPurifier_AttrTransform_Name();
// @noshade for hr ------------------------------------------------
// this transformation is not precise but often good enough.
// different browsers use different styles to designate noshade
$r['hr@noshade'] =
new HTMLPurifier_AttrTransform_BoolToCSS(
'noshade',
'color:#808080;background-color:#808080;border:0;'
);
// @nowrap for td, th ---------------------------------------------
$r['td@nowrap'] =
$r['th@nowrap'] =
new HTMLPurifier_AttrTransform_BoolToCSS(
'nowrap',
'white-space:nowrap;'
);
// @size for hr --------------------------------------------------
$r['hr@size'] = new HTMLPurifier_AttrTransform_Length('size', 'height');
// @type for li, ol, ul -------------------------------------------
// {{{
$ul_types = array(
'disc' => 'list-style-type:disc;',
'square' => 'list-style-type:square;',
'circle' => 'list-style-type:circle;'
);
$ol_types = array(
'1' => 'list-style-type:decimal;',
'i' => 'list-style-type:lower-roman;',
'I' => 'list-style-type:upper-roman;',
'a' => 'list-style-type:lower-alpha;',
'A' => 'list-style-type:upper-alpha;'
);
$li_types = $ul_types + $ol_types;
// }}}
$r['ul@type'] = new HTMLPurifier_AttrTransform_EnumToCSS('type', $ul_types);
$r['ol@type'] = new HTMLPurifier_AttrTransform_EnumToCSS('type', $ol_types, true);
$r['li@type'] = new HTMLPurifier_AttrTransform_EnumToCSS('type', $li_types, true);
// @vspace for img ------------------------------------------------
$r['img@vspace'] = new HTMLPurifier_AttrTransform_ImgSpace('vspace');
// @width for hr, td, th ------------------------------------------
$r['td@width'] =
$r['th@width'] =
$r['hr@width'] = new HTMLPurifier_AttrTransform_Length('width');
return $r;
}
}
class HTMLPurifier_HTMLModule_Tidy_Transitional extends
HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4
{
var $name = 'Tidy_Transitional';
var $defaultLevel = 'heavy';
}
class HTMLPurifier_HTMLModule_Tidy_Strict extends
HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4
{
var $name = 'Tidy_Strict';
var $defaultLevel = 'light';
}
?>

27
library/HTMLPurifier/HTMLModule/Tidy/XHTMLStrict.php Normal file
View File

@ -0,0 +1,27 @@
<?php
require_once 'HTMLPurifier/HTMLModule/Tidy.php';
require_once 'HTMLPurifier/ChildDef/StrictBlockquote.php';
class HTMLPurifier_HTMLModule_Tidy_XHTMLStrict extends
HTMLPurifier_HTMLModule_Tidy
{
var $name = 'Tidy_XHTMLStrict';
var $defaultLevel = 'light';
function makeFixes() {
$r = array();
$r['blockquote#content_model_type'] = 'strictblockquote';
return $r;
}
var $defines_child_def = true;
function getChildDef($def) {
if ($def->content_model_type != 'strictblockquote') return false;
return new HTMLPurifier_ChildDef_StrictBlockquote($def->content_model);
}
}
?>

200
library/HTMLPurifier/HTMLModule/TransformToStrict.php
View File

@ -1,200 +0,0 @@
<?php
require_once 'HTMLPurifier/ChildDef/StrictBlockquote.php';
require_once 'HTMLPurifier/TagTransform/Simple.php';
require_once 'HTMLPurifier/TagTransform/Center.php';
require_once 'HTMLPurifier/TagTransform/Font.php';
require_once 'HTMLPurifier/AttrTransform/Lang.php';
require_once 'HTMLPurifier/AttrTransform/BgColor.php';
require_once 'HTMLPurifier/AttrTransform/BoolToCSS.php';
require_once 'HTMLPurifier/AttrTransform/Border.php';
require_once 'HTMLPurifier/AttrTransform/Name.php';
require_once 'HTMLPurifier/AttrTransform/Length.php';
require_once 'HTMLPurifier/AttrTransform/ImgSpace.php';
require_once 'HTMLPurifier/AttrTransform/EnumToCSS.php';
/**
* Proprietary module that transforms deprecated elements into Strict
* HTML (see HTML 4.01 and XHTML 1.0) when possible.
*/
class HTMLPurifier_HTMLModule_TransformToStrict extends HTMLPurifier_HTMLModule
{
var $name = 'TransformToStrict';
// we're actually modifying these elements, not defining them
var $elements = array('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p',
'blockquote', 'table', 'td', 'th', 'tr', 'img', 'a', 'hr', 'br',
'caption', 'ul', 'ol', 'li');
var $info_tag_transform = array(
// placeholders, see constructor for definitions
'font' => false,
'menu' => false,
'dir' => false,
'center'=> false
);
var $attr_collections = array(
'Lang' => array(
'lang' => false // placeholder
)
);
var $info_attr_transform_post = array(
'lang' => false // placeholder
);
function HTMLPurifier_HTMLModule_TransformToStrict() {
// behavior with transformations when there's another CSS property
// working on it is interesting: the CSS will *always* override
// the deprecated attribute, whereas an inline CSS declaration will
// override the corresponding declaration in, say, an external
// stylesheet. This behavior won't affect most people, but it
// does represent an operational difference we CANNOT fix.
// deprecated tag transforms
$this->info_tag_transform['font'] = new HTMLPurifier_TagTransform_Font();
$this->info_tag_transform['menu'] = new HTMLPurifier_TagTransform_Simple('ul');
$this->info_tag_transform['dir'] = new HTMLPurifier_TagTransform_Simple('ul');
$this->info_tag_transform['center'] = new HTMLPurifier_TagTransform_Center();
foreach ($this->elements as $name) {
$this->info[$name] = new HTMLPurifier_ElementDef();
$this->info[$name]->standalone = false;
}
// deprecated attribute transforms
// align battery
$align_lookup = array();
$align_values = array('left', 'right', 'center', 'justify');
foreach ($align_values as $v) $align_lookup[$v] = "text-align:$v;";
$this->info['h1']->attr_transform_pre['align'] =
$this->info['h2']->attr_transform_pre['align'] =
$this->info['h3']->attr_transform_pre['align'] =
$this->info['h4']->attr_transform_pre['align'] =
$this->info['h5']->attr_transform_pre['align'] =
$this->info['h6']->attr_transform_pre['align'] =
$this->info['p'] ->attr_transform_pre['align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', $align_lookup);
// xml:lang <=> lang mirroring, implement in TransformToStrict,
// this is overridden in TransformToXHTML11
$this->info_attr_transform_post['lang'] = new HTMLPurifier_AttrTransform_Lang();
$this->attr_collections['Lang']['lang'] = new HTMLPurifier_AttrDef_Lang();
// this should not be applied to XHTML 1.0 Transitional, ONLY
// XHTML 1.0 Strict. We may need three classes
$this->info['blockquote']->content_model_type = 'strictblockquote';
$this->info['blockquote']->child = false; // recalculate please!
$this->info['table']->attr_transform_pre['bgcolor'] =
$this->info['tr']->attr_transform_pre['bgcolor'] =
$this->info['td']->attr_transform_pre['bgcolor'] =
$this->info['th']->attr_transform_pre['bgcolor'] = new HTMLPurifier_AttrTransform_BgColor();
$this->info['img']->attr_transform_pre['border'] = new HTMLPurifier_AttrTransform_Border();
$this->info['img']->attr_transform_pre['name'] =
$this->info['a']->attr_transform_pre['name'] = new HTMLPurifier_AttrTransform_Name();
$this->info['td']->attr_transform_pre['width'] =
$this->info['th']->attr_transform_pre['width'] =
$this->info['hr']->attr_transform_pre['width'] = new HTMLPurifier_AttrTransform_Length('width');
$this->info['td']->attr_transform_pre['nowrap'] =
$this->info['th']->attr_transform_pre['nowrap'] = new HTMLPurifier_AttrTransform_BoolToCSS('nowrap', 'white-space:nowrap;');
$this->info['td']->attr_transform_pre['height'] =
$this->info['th']->attr_transform_pre['height'] = new HTMLPurifier_AttrTransform_Length('height');
$this->info['img']->attr_transform_pre['hspace'] = new HTMLPurifier_AttrTransform_ImgSpace('hspace');
$this->info['img']->attr_transform_pre['vspace'] = new HTMLPurifier_AttrTransform_ImgSpace('vspace');
$this->info['hr']->attr_transform_pre['size'] = new HTMLPurifier_AttrTransform_Length('size', 'height');
// this transformation is not precise but often good enough.
// different browsers use different styles to designate noshade
$this->info['hr']->attr_transform_pre['noshade'] = new HTMLPurifier_AttrTransform_BoolToCSS('noshade', 'color:#808080;background-color:#808080;border: 0;');
$this->info['br']->attr_transform_pre['clear'] =
new HTMLPurifier_AttrTransform_EnumToCSS('clear', array(
'left' => 'clear:left;',
'right' => 'clear:right;',
'all' => 'clear:both;',
'none' => 'clear:none;',
));
// this is a slightly unreasonable attribute
$this->info['caption']->attr_transform_pre['align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
// we're following IE's behavior, not Firefox's, due
// to the fact that no one supports caption-side:right,
// W3C included (with CSS 2.1)
'left' => 'text-align:left;',
'right' => 'text-align:right;',
'top' => 'caption-side:top;',
'bottom' => 'caption-side:bottom;' // not supported by IE
));
$this->info['table']->attr_transform_pre['align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
'left' => 'float:left;',
'center' => 'margin-left:auto;margin-right:auto;',
'right' => 'float:right;'
));
$this->info['img']->attr_transform_pre['align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
'left' => 'float:left;',
'right' => 'float:right;',
'top' => 'vertical-align:top;',
'middle' => 'vertical-align:middle;',
'bottom' => 'vertical-align:baseline;',
));
$this->info['hr']->attr_transform_pre['align'] =
new HTMLPurifier_AttrTransform_EnumToCSS('align', array(
'left' => 'margin-left:0;margin-right:auto;text-align:left;',
'center' => 'margin-left:auto;margin-right:auto;text-align:center;',
'right' => 'margin-left:auto;margin-right:0;text-align:right;'
));
$ul_types = array(
'disc' => 'list-style-type:disc;',
'square' => 'list-style-type:square;',
'circle' => 'list-style-type:circle;'
);
$ol_types = array(
'1' => 'list-style-type:decimal;',
'i' => 'list-style-type:lower-roman;',
'I' => 'list-style-type:upper-roman;',
'a' => 'list-style-type:lower-alpha;',
'A' => 'list-style-type:upper-alpha;'
);
$li_types = $ul_types + $ol_types;
$this->info['ul']->attr_transform_pre['type'] =
new HTMLPurifier_AttrTransform_EnumToCSS('type', $ul_types);
$this->info['ol']->attr_transform_pre['type'] =
new HTMLPurifier_AttrTransform_EnumToCSS('type', $ol_types, true);
$this->info['li']->attr_transform_pre['type'] =
new HTMLPurifier_AttrTransform_EnumToCSS('type', $li_types, true);
}
var $defines_child_def = true;
function getChildDef($def) {
if ($def->content_model_type != 'strictblockquote') return false;
return new HTMLPurifier_ChildDef_StrictBlockquote($def->content_model);
}
}
?>

36
library/HTMLPurifier/HTMLModule/TransformToXHTML11.php
View File

@ -1,36 +0,0 @@
<?php
require_once 'HTMLPurifier/AttrTransform/Lang.php';
/**
* Proprietary module that transforms XHTML 1.0 deprecated aspects into
* XHTML 1.1 compliant ones, when possible. For maximum effectiveness,
* HTMLPurifier_HTMLModule_TransformToStrict must also be loaded
* (otherwise, elements that were deprecated from Transitional to Strict
* will not be transformed).
*
* XHTML 1.1 compliant document are automatically XHTML 1.0 compliant too,
* although they may not be as friendly to legacy browsers.
*/
class HTMLPurifier_HTMLModule_TransformToXHTML11 extends HTMLPurifier_HTMLModule
{
var $name = 'TransformToXHTML11';
var $attr_collections = array(
'Lang' => array(
'lang' => false // remove it
)
);
var $info_attr_transform_post = array(
'lang' => false // remove it
);
function HTMLPurifier_HTMLModule_TransformToXHTML11() {
$this->info_attr_transform_pre['lang'] = new HTMLPurifier_AttrTransform_Lang();
}
}
?>

16
library/HTMLPurifier/HTMLModule/XMLCommonAttributes.php Normal file
View File

@ -0,0 +1,16 @@
<?php
require_once 'HTMLPurifier/HTMLModule.php';
class HTMLPurifier_HTMLModule_XMLCommonAttributes extends HTMLPurifier_HTMLModule
{
var $name = 'XMLCommonAttributes';
var $attr_collections = array(
'Lang' => array(
'xml:lang' => 'LanguageCode',
)
);
}
?>

595
library/HTMLPurifier/HTMLModuleManager.php
View File

@ -2,6 +2,8 @@
require_once 'HTMLPurifier/HTMLModule.php';
require_once 'HTMLPurifier/ElementDef.php';
require_once 'HTMLPurifier/Doctype.php';
require_once 'HTMLPurifier/DoctypeRegistry.php';
require_once 'HTMLPurifier/ContentSets.php';
require_once 'HTMLPurifier/AttrTypes.php';
@ -23,14 +25,20 @@ require_once 'HTMLPurifier/HTMLModule/Image.php';
require_once 'HTMLPurifier/HTMLModule/StyleAttribute.php';
require_once 'HTMLPurifier/HTMLModule/Legacy.php';
require_once 'HTMLPurifier/HTMLModule/Target.php';
require_once 'HTMLPurifier/HTMLModule/Scripting.php';
require_once 'HTMLPurifier/HTMLModule/XMLCommonAttributes.php';
require_once 'HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php';
// proprietary modules
require_once 'HTMLPurifier/HTMLModule/TransformToStrict.php';
require_once 'HTMLPurifier/HTMLModule/TransformToXHTML11.php';
// tidy modules
require_once 'HTMLPurifier/HTMLModule/Tidy.php';
require_once 'HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php';
require_once 'HTMLPurifier/HTMLModule/Tidy/XHTML.php';
require_once 'HTMLPurifier/HTMLModule/Tidy/XHTMLStrict.php';
require_once 'HTMLPurifier/HTMLModule/Tidy/Proprietary.php';
HTMLPurifier_ConfigSchema::define(
'HTML', 'Doctype', null, 'string/null',
'Doctype to use, valid values are HTML 4.01 Transitional, HTML 4.01 '.
'Doctype to use, pre-defined values are HTML 4.01 Transitional, HTML 4.01 '.
'Strict, XHTML 1.0 Transitional, XHTML 1.0 Strict, XHTML 1.1. '.
'Technically speaking this is not actually a doctype (as it does '.
'not identify a corresponding DTD), but we are using this name '.
@ -38,173 +46,159 @@ HTMLPurifier_ConfigSchema::define(
'like %Core.XHTML or %HTML.Strict.'
);
HTMLPurifier_ConfigSchema::define(
'HTML', 'Trusted', false, 'bool',
'Indicates whether or not the user input is trusted or not. If the '.
'input is trusted, a more expansive set of allowed tags and attributes '.
'will be used. This directive has been available since 2.0.0.'
);
HTMLPurifier_ConfigSchema::define(
'HTML', 'AllowedModules', null, 'lookup/null', '
<p>
A doctype comes with a set of usual modules to use. Without having
to mucking about with the doctypes, you can quickly activate or
disable these modules by specifying which modules you wish to allow
with this directive. This is most useful for unit testing specific
modules, although end users may find it useful for their own ends.
</p>
<p>
If you specify a module that does not exist, the manager will silently
fail to use it, so be careful! User-defined modules are not affected
by this directive. Modules defined in %HTML.CoreModules are not
affected by this directive. This directive has been available since 2.0.0.
</p>
');
HTMLPurifier_ConfigSchema::define(
'HTML', 'CoreModules', array(
'Structure' => true,
'Text' => true,
'Hypertext' => true,
'List' => true,
'NonXMLCommonAttributes' => true,
'XMLCommonAttributes' => true,
'CommonAttributes' => true
), 'lookup', '
<p>
Certain modularized doctypes (XHTML, namely), have certain modules
that must be included for the doctype to be an conforming document
type: put those modules here. By default, XHTML\'s core modules
are used. You can set this to a blank array to disable core module
protection, but this is not recommended. This directive has been
available since 2.0.0.
</p>
');
class HTMLPurifier_HTMLModuleManager
{
/**
* Array of HTMLPurifier_Module instances, indexed by module's class name.
* All known modules, regardless of use, are in this array.
* Instance of HTMLPurifier_DoctypeRegistry
* @public
*/
var $doctypes;
/**
* Instance of current doctype
* @public
*/
var $doctype;
/**
* Instance of HTMLPurifier_AttrTypes
* @public
*/
var $attrTypes;
/**
* Active instances of modules for the specified doctype are
* indexed, by name, in this array.
*/
var $modules = array();
/**
* String doctype we will validate against. See $validModules for use.
*
* @note
* There is a special doctype '*' that acts both as the "default"
* doctype if a customized system only defines one doctype and
* also a catch-all doctype that gets merged into all the other
* module collections. When possible, use a private collection to
* share modules between doctypes: this special doctype is to
* make life more convenient for users.
* Array of recognized HTMLPurifier_Module instances, indexed by
* module's class name. This array is usually lazy loaded, but a
* user can overload a module by pre-emptively registering it.
*/
var $doctype;
var $doctypeAliases = array(); /**< Lookup array of strings to real doctypes */
var $registeredModules = array();
/**
* Associative array: $collections[$type][$doctype] = list of modules.
* This is used to logically separate types of functionality so that
* based on the doctype and other configuration settings they may
* be easily switched and on and off. Custom setups may not need
* to use this abstraction, opting to have only one big collection
* with one valid doctype.
* List of extra modules that were added by the user using addModule().
* These get unconditionally merged into the current doctype, whatever
* it may be.
*/
var $collections = array();
var $userModules = array();
/**
* Modules that may be used in a valid doctype of this kind.
* Correctional and leniency modules should not be placed in this
* array unless the user said so: don't stuff every possible lenient
* module for this doctype in here.
* Associative array of element name to list of modules that have
* definitions for the element; this array is dynamically filled.
*/
var $validModules = array();
var $validCollections = array(); /**< Collections to merge into $validModules */
/**
* Modules that we will allow in input, subset of $validModules. Single
* element definitions may result in us consulting validModules.
*/
var $activeModules = array();
var $activeCollections = array(); /**< Collections to merge into $activeModules */
var $counter = 0; /**< Designates next available integer order for modules. */
var $initialized = false; /**< Says whether initialize() was called */
/**
* Specifies what doctype to siphon new modules from addModule() to,
* or false to disable the functionality. Must be used in conjunction
* with $autoCollection.
*/
var $autoDoctype = false;
/**
* Specifies what collection to siphon new modules from addModule() to,
* or false to disable the functionality. Must be used in conjunction
* with $autoCollection.
*/
var $autoCollection = false;
/** Associative array of element name to defining modules (always array) */
var $elementLookup = array();
/** List of prefixes we should use for resolving small names */
/** List of prefixes we should use for registering small names */
var $prefixes = array('HTMLPurifier_HTMLModule_');
var $contentSets; /**< Instance of HTMLPurifier_ContentSets */
var $attrTypes; /**< Instance of HTMLPurifier_AttrTypes */
var $contentSets; /**< Instance of HTMLPurifier_ContentSets */
var $attrCollections; /**< Instance of HTMLPurifier_AttrCollections */
/**
* @param $blank If true, don't do any initializing
*/
function HTMLPurifier_HTMLModuleManager($blank = false) {
/** If set to true, unsafe elements and attributes will be allowed */
var $trusted = false;
function HTMLPurifier_HTMLModuleManager() {
// the only editable internal object. The rest need to
// be manipulated through modules
// editable internal objects
$this->attrTypes = new HTMLPurifier_AttrTypes();
$this->doctypes = new HTMLPurifier_DoctypeRegistry();
if (!$blank) $this->initialize();
// setup default HTML doctypes
}
function initialize() {
$this->initialized = true;
// load default modules to the recognized modules list (not active)
$modules = array(
// define
'CommonAttributes',
'Text', 'Hypertext', 'List', 'Presentation',
'Edit', 'Bdo', 'Tables', 'Image', 'StyleAttribute',
'Target',
// define-redefine
'Legacy',
// redefine
'TransformToStrict', 'TransformToXHTML11'
// module reuse
$common = array(
'CommonAttributes', 'Text', 'Hypertext', 'List',
'Presentation', 'Edit', 'Bdo', 'Tables', 'Image',
'StyleAttribute', 'Scripting'
);
foreach ($modules as $module) {
$this->addModule($module);
}
$transitional = array('Legacy', 'Target');
$xml = array('XMLCommonAttributes');
$non_xml = array('NonXMLCommonAttributes');
// Safe modules for supported doctypes. These are included
// in the valid and active module lists by default
$this->collections['Safe'] = array(
'_Common' => array( // leading _ indicates private
'CommonAttributes', 'Text', 'Hypertext', 'List',
'Presentation', 'Edit', 'Bdo', 'Tables', 'Image',
'StyleAttribute'
),
// HTML definitions, defer to XHTML definitions
'HTML 4.01 Transitional' => array(array('XHTML 1.0 Transitional')),
'HTML 4.01 Strict' => array(array('XHTML 1.0 Strict')),
// XHTML definitions
'XHTML 1.0 Transitional' => array( array('XHTML 1.0 Strict'), 'Legacy', 'Target' ),
'XHTML 1.0 Strict' => array(array('_Common')),
'XHTML 1.1' => array(array('_Common')),
$this->doctypes->register(
'HTML 4.01 Transitional', false,
array_merge($common, $transitional, $non_xml),
array('Tidy_Transitional', 'Tidy_Proprietary')
);
// Modules that specify elements that are unsafe from untrusted
// third-parties. These should be registered in $validModules but
// almost never $activeModules unless you really know what you're
// doing.
$this->collections['Unsafe'] = array();
// Modules to import if lenient mode (attempt to convert everything
// to a valid representation) is on. These must not be in $validModules
// unless specified so.
$this->collections['Lenient'] = array(
'HTML 4.01 Strict' => array(array('XHTML 1.0 Strict')),
'XHTML 1.0 Strict' => array('TransformToStrict'),
'XHTML 1.1' => array(array('XHTML 1.0 Strict'), 'TransformToXHTML11')
$this->doctypes->register(
'HTML 4.01 Strict', false,
array_merge($common, $non_xml),
array('Tidy_Strict', 'Tidy_Proprietary')
);
// Modules to import if correctional mode (correct everything that
// is feasible to strict mode) is on. These must not be in $validModules
// unless specified so.
$this->collections['Correctional'] = array(
'HTML 4.01 Transitional' => array(array('XHTML 1.0 Transitional')),
'XHTML 1.0 Transitional' => array('TransformToStrict'), // probably want a different one
$this->doctypes->register(
'XHTML 1.0 Transitional', true,
array_merge($common, $transitional, $xml, $non_xml),
array('Tidy_Transitional', 'Tidy_XHTML', 'Tidy_Proprietary')
);
// User-space modules, custom code or whatever
$this->collections['Extension'] = array();
$this->doctypes->register(
'XHTML 1.0 Strict', true,
array_merge($common, $xml, $non_xml),
array('Tidy_Strict', 'Tidy_XHTML', 'Tidy_XHTMLStrict', 'Tidy_Proprietary')
);
// setup active versus valid modules. ORDER IS IMPORTANT!
// definition modules
$this->makeCollectionActive('Safe');
$this->makeCollectionValid('Unsafe');
// redefinition modules
$this->makeCollectionActive('Lenient');
$this->makeCollectionActive('Correctional');
$this->autoDoctype = '*';
$this->autoCollection = 'Extension';
$this->doctypes->register(
'XHTML 1.1', true,
array_merge($common, $xml),
array('Tidy_Strict', 'Tidy_XHTML', 'Tidy_Proprietary') // Tidy_XHTML1_1
);
}
/**
* Adds a module to the recognized module list. This does not
* do anything else: the module must be added to a corresponding
* collection to be "activated".
* Registers a module to the recognized module list, useful for
* overloading pre-existing modules.
* @param $module Mixed: string module name, with or without
* HTMLPurifier_HTMLModule prefix, or instance of
* subclass of HTMLPurifier_HTMLModule.
@ -217,10 +211,15 @@ class HTMLPurifier_HTMLModuleManager
* - Check for literal object name
* - Throw fatal error
* If your object name collides with an internal class, specify
* your module manually.
* your module manually. All modules must have been included
* externally: registerModule will not perform inclusions for you!
* @warning If your module has the same name as an already loaded
* module, your module will overload the old one WITHOUT
* warning.
*/
function addModule($module) {
function registerModule($module) {
if (is_string($module)) {
// attempt to load the module
$original_module = $module;
$ok = false;
foreach ($this->prefixes as $prefix) {
@ -240,16 +239,19 @@ class HTMLPurifier_HTMLModuleManager
}
$module = new $module();
}
$module->order = $this->counter++; // assign then increment
$this->modules[$module->name] = $module;
if ($this->autoDoctype !== false && $this->autoCollection !== false) {
$this->collections[$this->autoCollection][$this->autoDoctype][] = $module->name;
if (empty($module->name)) {
trigger_error('Module instance of ' . get_class($module) . ' must have name');
return;
}
$this->registeredModules[$module->name] = $module;
}
/**
* Safely tests for class existence without invoking __autoload in PHP5
* or greater.
* @param $name String class name to test
* @note If any other class needs it, we'll need to stash in a
* conjectured "compatibility" class
* @private
*/
function _classExists($name) {
@ -265,55 +267,63 @@ class HTMLPurifier_HTMLModuleManager
}
/**
* Makes a collection active, while also making it valid if not
* already done so. See $activeModules for the semantics of "active".
* @param $collection_name Name of collection to activate
* Adds a module to the current doctype by first registering it,
* and then tacking it on to the active doctype
*/
function makeCollectionActive($collection_name) {
if (!in_array($collection_name, $this->validCollections)) {
$this->makeCollectionValid($collection_name);
}
$this->activeCollections[] = $collection_name;
function addModule($module) {
$this->registerModule($module);
if (is_object($module)) $module = $module->name;
$this->userModules[] = $module;
}
/**
* Makes a collection valid. See $validModules for the semantics of "valid"
*/
function makeCollectionValid($collection_name) {
$this->validCollections[] = $collection_name;
}
/**
* Adds a class prefix that addModule() will use to resolve a
* Adds a class prefix that registerModule() will use to resolve a
* string name to a concrete class
*/
function addPrefix($prefix) {
$this->prefixes[] = (string) $prefix;
$this->prefixes[] = $prefix;
}
/**
* Performs processing on modules, after being called you may
* use getElement() and getElements()
* @param $config Instance of HTMLPurifier_Config
*/
function setup($config) {
// load up the autocollection
if ($this->autoCollection !== false) {
$this->makeCollectionActive($this->autoCollection);
$this->trusted = $config->get('HTML', 'Trusted');
// generate
$this->doctype = $this->doctypes->make($config);
$modules = $this->doctype->modules;
// take out the default modules that aren't allowed
$lookup = $config->get('HTML', 'AllowedModules');
$special_cases = $config->get('HTML', 'CoreModules');
if (is_array($lookup)) {
foreach ($modules as $k => $m) {
if (isset($special_cases[$m])) continue;
if (!isset($lookup[$m])) unset($modules[$k]);
}
}
// retrieve the doctype
$this->doctype = $this->getDoctype($config);
if (isset($this->doctypeAliases[$this->doctype])) {
$this->doctype = $this->doctypeAliases[$this->doctype];
// merge in custom modules
$modules = array_merge($modules, $this->userModules);
foreach ($modules as $module) {
$this->processModule($module);
}
// process module collections to module name => module instance form
foreach ($this->collections as $col_i => $x) {
$this->processCollections($this->collections[$col_i]);
foreach ($this->doctype->tidyModules as $module) {
$this->processModule($module);
if (method_exists($this->modules[$module], 'construct')) {
$this->modules[$module]->construct($config);
}
}
$this->validModules = $this->assembleModules($this->validCollections);
$this->activeModules = $this->assembleModules($this->activeCollections);
// setup lookup table based on all valid modules
foreach ($this->validModules as $module) {
foreach ($this->modules as $module) {
foreach ($module->info as $name => $def) {
if (!isset($this->elementLookup[$name])) {
$this->elementLookup[$name] = array();
@ -324,214 +334,51 @@ class HTMLPurifier_HTMLModuleManager
// note the different choice
$this->contentSets = new HTMLPurifier_ContentSets(
// content models that contain non-allowed elements are
// harmless because RemoveForeignElements will ensure
// they never get in anyway, and there is usually no
// reason why you should want to restrict a content
// model beyond what is mandated by the doctype.
// Note, however, that this means redefinitions of
// content models can't be tossed in validModels willy-nilly:
// that stuff still is regulated by configuration.
$this->validModules
// content set assembly deals with all possible modules,
// not just ones deemed to be "safe"
$this->modules
);
$this->attrCollections = new HTMLPurifier_AttrCollections(
$this->attrTypes,
// only explicitly allowed modules are allowed to affect
// the global attribute collections. This mean's there's
// a distinction between loading the Bdo module, and the
// bdo element: Bdo will enable the dir attribute on all
// elements, while bdo will only define the bdo element,
// which will not have an editable directionality. This might
// catch people who are loading only elements by surprise, so
// we should consider loading an entire module if all the
// elements it defines are requested by the user, especially
// if it affects the global attribute collections.
$this->activeModules
// there is no way to directly disable a global attribute,
// but using AllowedAttributes or simply not including
// the module in your custom doctype should be sufficient
$this->modules
);
}
/**
* Takes a list of collections and merges together all the defined
* modules for the current doctype from those collections.
* @param $collections List of collection suffixes we should grab
* modules from (like 'Safe' or 'Lenient')
* Takes a module and adds it to the active module collection,
* registering it if necessary.
*/
function assembleModules($collections) {
$modules = array();
$numOfCollectionsUsed = 0;
foreach ($collections as $name) {
$disable_global = false;
if (!isset($this->collections[$name])) {
trigger_error("$name collection is undefined", E_USER_ERROR);
continue;
}
$cols = $this->collections[$name];
if (isset($cols[$this->doctype])) {
if (isset($cols[$this->doctype]['*'])) {
unset($cols[$this->doctype]['*']);
$disable_global = true;
}
$modules += $cols[$this->doctype];
$numOfCollectionsUsed++;
}
// accept catch-all doctype
if (
$this->doctype !== '*' &&
isset($cols['*']) &&
!$disable_global
) {
$modules += $cols['*'];
}
function processModule($module) {
if (!isset($this->registeredModules[$module]) || is_object($module)) {
$this->registerModule($module);
}
if ($numOfCollectionsUsed < 1) {
// possible XSS injection if user-specified doctypes
// are allowed
trigger_error("Doctype {$this->doctype} does not exist, ".
"check for typos (if you desire a doctype that allows ".
"no elements, use an empty array collection)", E_USER_ERROR);
}
return $modules;
$this->modules[$module] = $this->registeredModules[$module];
}
/**
* Takes a collection and performs inclusions and substitutions for it.
* @param $cols Reference to collections class member variable
* Retrieves merged element definitions.
* @return Array of HTMLPurifier_ElementDef
*/
function processCollections(&$cols) {
// $cols is the set of collections
// $col_i is the name (index) of a collection
// $col is a collection/list of modules
// perform inclusions
foreach ($cols as $col_i => $col) {
$seen = array();
if (!empty($col[0]) && is_array($col[0])) {
$seen[$col_i] = true; // recursion reporting
$includes = $col[0];
unset($cols[$col_i][0]); // remove inclusions value, recursion guard
} else {
$includes = array();
}
if (empty($includes)) continue;
for ($i = 0; isset($includes[$i]); $i++) {
$inc = $includes[$i];
if (isset($seen[$inc])) {
trigger_error(
"Circular inclusion detected in $col_i collection",
E_USER_ERROR
);
continue;
} else {
$seen[$inc] = true;
}
if (!isset($cols[$inc])) {
trigger_error(
"Collection $col_i tried to include undefined ".
"collection $inc", E_USER_ERROR);
continue;
}
foreach ($cols[$inc] as $module) {
if (is_array($module)) { // another inclusion!
foreach ($module as $inc2) $includes[] = $inc2;
continue;
}
$cols[$col_i][] = $module; // merge in the other modules
}
}
}
// replace with real modules, invert module from list to
// assoc array of module name to module instance
foreach ($cols as $col_i => $col) {
$ignore_global = false;
$order = array();
foreach ($col as $module_i => $module) {
unset($cols[$col_i][$module_i]);
if (is_array($module)) {
trigger_error("Illegal inclusion array at index".
" $module_i found collection $col_i, inclusion".
" arrays must be at start of collection (index 0)",
E_USER_ERROR);
continue;
}
if ($module_i === '*' && $module === false) {
$ignore_global = true;
continue;
}
if (!isset($this->modules[$module])) {
trigger_error(
"Collection $col_i references undefined ".
"module $module",
E_USER_ERROR
);
continue;
}
$module = $this->modules[$module];
$cols[$col_i][$module->name] = $module;
$order[$module->name] = $module->order;
}
array_multisort(
$order, SORT_ASC, SORT_NUMERIC, $cols[$col_i]
);
if ($ignore_global) $cols[$col_i]['*'] = false;
}
// delete pseudo-collections
foreach ($cols as $col_i => $col) {
if ($col_i[0] == '_') unset($cols[$col_i]);
}
}
/**
* Retrieves the doctype from the configuration object
*/
function getDoctype($config) {
$doctype = $config->get('HTML', 'Doctype');
if ($doctype !== null) {
return $doctype;
}
if (!$this->initialized) {
// don't do HTML-oriented backwards compatibility stuff
// use either the auto-doctype, or the catch-all doctype
return $this->autoDoctype ? $this->autoDoctype : '*';
}
// this is backwards-compatibility stuff
if ($config->get('Core', 'XHTML')) {
$doctype = 'XHTML 1.0';
} else {
$doctype = 'HTML 4.01';
}
if ($config->get('HTML', 'Strict')) {
$doctype .= ' Strict';
} else {
$doctype .= ' Transitional';
}
return $doctype;
}
/**
* Retrieves merged element definitions for all active elements.
* @note We may want to generate an elements array during setup
* and pass that on, because a specific combination of
* elements may trigger the loading of a module.
* @param $config Instance of HTMLPurifier_Config, for determining
* stray elements.
*/
function getElements($config) {
function getElements() {
$elements = array();
foreach ($this->activeModules as $module) {
foreach ($this->modules as $module) {
foreach ($module->info as $name => $v) {
if (isset($elements[$name])) continue;
$elements[$name] = $this->getElement($name, $config);
// if element is not safe, don't use it
if (!$this->trusted && ($v->safe === false)) continue;
$elements[$name] = $this->getElement($name);
}
}
// standalone elements now loaded
// remove dud elements, this happens when an element that
// appeared to be safe actually wasn't
foreach ($elements as $n => $v) {
if ($v === false) unset($elements[$n]);
}
return $elements;
@ -540,13 +387,16 @@ class HTMLPurifier_HTMLModuleManager
/**
* Retrieves a single merged element definition
* @param $name Name of element
* @param $config Instance of HTMLPurifier_Config, may not be necessary.
* @param $trusted Boolean trusted overriding parameter: set to true
* if you want the full version of an element
* @return Merged HTMLPurifier_ElementDef
*/
function getElement($name, $config) {
function getElement($name, $trusted = null) {
$def = false;
if ($trusted === null) $trusted = $this->trusted;
$modules = $this->validModules;
$modules = $this->modules;
if (!isset($this->elementLookup[$name])) {
return false;
@ -555,9 +405,23 @@ class HTMLPurifier_HTMLModuleManager
foreach($this->elementLookup[$name] as $module_name) {
$module = $modules[$module_name];
$new_def = $module->info[$name];
// copy is used because, ideally speaking, the original
// definition should not be modified. Usually, this will
// make no difference, but for consistency's sake
$new_def = $module->info[$name]->copy();
// refuse to create/merge in a definition that is deemed unsafe
if (!$trusted && ($new_def->safe === false)) {
$def = false;
continue;
}
if (!$def && $new_def->standalone) {
// element with unknown safety is not to be trusted.
// however, a merge-in definition with undefined safety
// is fine
if (!$trusted && !$new_def->safe) continue;
$def = $new_def;
} elseif ($def) {
$def->mergeIn($new_def);
@ -583,6 +447,13 @@ class HTMLPurifier_HTMLModuleManager
$this->contentSets->generateChildDef($def, $module);
}
// add information on required attributes
foreach ($def->attr as $attr_name => $attr_def) {
if ($attr_def->required) {
$def->required_attr[] = $attr_name;
}
}
return $def;

22
library/HTMLPurifier/Language.php
View File

@ -41,16 +41,34 @@ class HTMLPurifier_Language
}
/**
* Retrieves a localised message. Does not perform any operations.
* Retrieves a localised message.
* @param $key string identifier of message
* @return string localised message
*/
function getMessage($key) {
if (!$this->_loaded) $this->load();
if (!isset($this->messages[$key])) return '';
if (!isset($this->messages[$key])) return "[$key]";
return $this->messages[$key];
}
/**
* Formats a localised message with passed parameters
* @param $key string identifier of message
* @param $param Parameter to substitute in (arbitrary number)
* @return string localised message
*/
function formatMessage($key) {
if (!$this->_loaded) $this->load();
if (!isset($this->messages[$key])) return "[$key]";
$raw = $this->messages[$key];
$args = func_get_args();
$substitutions = array();
for ($i = 1; $i < count($args); $i++) {
$substitutions['$' . $i] = $args[$i];
}
return strtr($raw, $substitutions);
}
}
?>

2
library/HTMLPurifier/Language/messages/en.php
View File

@ -7,6 +7,8 @@ $messages = array(
'htmlpurifier' => 'HTML Purifier',
'pizza' => 'Pizza', // for unit testing purposes
);
?>

16
library/HTMLPurifier/LanguageFactory.php
View File

@ -100,15 +100,15 @@ class HTMLPurifier_LanguageFactory
// you can bypass the conditional include by loading the
// file yourself
if (file_exists($file) && !class_exists($class)) {
include_once $file;
}
include_once $file;
}
}
if (!class_exists($class)) {
// go fallback
$fallback = HTMLPurifier_Language::getFallbackFor($code);
$fallback = HTMLPurifier_LanguageFactory::getFallbackFor($code);
$depth++;
$lang = Language::factory( $fallback );
$lang = HTMLPurifier_LanguageFactory::factory( $fallback );
$depth--;
} else {
$lang = new $class;
@ -172,15 +172,15 @@ class HTMLPurifier_LanguageFactory
// merge fallback with current language
foreach ( $this->keys as $key ) {
if (isset($cache[$key]) && isset($fallback_cache[$key])) {
if (isset($cache[$key]) && isset($fallback_cache[$key])) {
if (isset($this->mergeable_keys_map[$key])) {
$cache[$key] = $cache[$key] + $fallback_cache[$key];
} elseif (isset($this->mergeable_keys_list[$key])) {
$cache[$key] = array_merge( $fallback_cache[$key], $cache[$key] );
}
} else {
$cache[$key] = $fallback_cache[$key];
}
} else {
$cache[$key] = $fallback_cache[$key];
}
}
}

168
library/HTMLPurifier/Lexer.php
View File

@ -4,6 +4,14 @@ require_once 'HTMLPurifier/Token.php';
require_once 'HTMLPurifier/Encoder.php';
require_once 'HTMLPurifier/EntityParser.php';
// implementations
require_once 'HTMLPurifier/Lexer/DirectLex.php';
if (version_compare(PHP_VERSION, "5", ">=")) {
// You can remove the if statement if you are running PHP 5 only.
// We ought to get the strict version to follow those rules.
require_once 'HTMLPurifier/Lexer/DOMLex.php';
}
HTMLPurifier_ConfigSchema::define(
'Core', 'AcceptFullDocuments', true, 'bool',
'This parameter determines whether or not the filter should accept full '.
@ -11,6 +19,52 @@ HTMLPurifier_ConfigSchema::define(
'drop all sections except the content between body.'
);
HTMLPurifier_ConfigSchema::define(
'Core', 'LexerImpl', null, 'mixed/null', '
<p>
This parameter determines what lexer implementation can be used. The
valid values are:
</p>
<dl>
<dt><em>null</em></dt>
<dd>
Recommended, the lexer implementation will be auto-detected based on
your PHP-version and configuration.
</dd>
<dt><em>string</em> lexer identifier</dt>
<dd>
This is a slim way of manually overridding the implementation.
Currently recognized values are: DOMLex (the default PHP5 implementation)
and DirectLex (the default PHP4 implementation). Only use this if
you know what you are doing: usually, the auto-detection will
manage things for cases you aren\'t even aware of.
</dd>
<dt><em>object</em> lexer instance</dt>
<dd>
Super-advanced: you can specify your own, custom, implementation that
implements the interface defined by <code>HTMLPurifier_Lexer</code>.
I may remove this option simply because I don\'t expect anyone
to use it.
</dd>
</dl>
<p>
This directive has been available since 2.0.0.
</p>
'
);
HTMLPurifier_ConfigSchema::define(
'Core', 'MaintainLineNumbers', false, 'bool', '
<p>
If true, HTML Purifier will add line number information to all tokens.
This is useful when error reporting is turned on, but can result in
significant performance degradation and should not be used when
unnecessary. This directive must be used with the DirectLex lexer,
as the DOMLex lexer does not (yet) support this functionality. This directive
has been available since 2.0.0.
</p>
');
/**
* Forgivingly lexes HTML (SGML-style) markup into tokens.
*
@ -55,11 +109,83 @@ HTMLPurifier_ConfigSchema::define(
class HTMLPurifier_Lexer
{
// -- STATIC ----------------------------------------------------------
/**
* Retrieves or sets the default Lexer as a Prototype Factory.
*
* Depending on what PHP version you are running, the abstract base
* Lexer class will determine which concrete Lexer is best for you:
* HTMLPurifier_Lexer_DirectLex for PHP 4, and HTMLPurifier_Lexer_DOMLex
* for PHP 5 and beyond. This general rule has a few exceptions to it
* involving special features that only DirectLex implements.
*
* @static
*
* @note The behavior of this class has changed, rather than accepting
* a prototype object, it now accepts a configuration object.
* To specify your own prototype, set %Core.LexerImpl to it.
* This change in behavior de-singletonizes the lexer object.
*
* @note In PHP4, it is possible to call this factory method from
* subclasses, such usage is not recommended and not
* forwards-compatible.
*
* @param $prototype Optional prototype lexer or configuration object
* @return Concrete lexer.
*/
static function create($config) {
if (!($config instanceof HTMLPurifier_Config)) {
$lexer = $config;
trigger_error("Passing a prototype to
HTMLPurifier_Lexer::create() is deprecated, please instead
use %Core.LexerImpl", E_USER_WARNING);
} else {
$lexer = $config->get('Core', 'LexerImpl');
}
if (is_object($lexer)) {
return $lexer;
}
if (is_null($lexer)) { do {
// auto-detection algorithm
// once PHP DOM implements native line numbers, or we
// hack out something using XSLT, remove this stipulation
if ($config->get('Core', 'MaintainLineNumbers')) {
$lexer = 'DirectLex';
break;
}
if (version_compare(PHP_VERSION, "5", ">=") && // check for PHP5
class_exists('DOMDocument')) { // check for DOM support
$lexer = 'DOMLex';
} else {
$lexer = 'DirectLex';
}
} while(0); } // do..while so we can break
// instantiate recognized string names
switch ($lexer) {
case 'DOMLex':
return new HTMLPurifier_Lexer_DOMLex();
case 'DirectLex':
return new HTMLPurifier_Lexer_DirectLex();
default:
trigger_error("Cannot instantiate unrecognized Lexer type " . htmlspecialchars($lexer), E_USER_ERROR);
}
}
// -- CONVENIENCE MEMBERS ---------------------------------------------
function HTMLPurifier_Lexer() {
$this->_entity_parser = new HTMLPurifier_EntityParser();
}
/**
* Most common entity to raw value conversion table for special entities.
* @protected
@ -123,46 +249,6 @@ class HTMLPurifier_Lexer
trigger_error('Call to abstract class', E_USER_ERROR);
}
/**
* Retrieves or sets the default Lexer as a Prototype Factory.
*
* Depending on what PHP version you are running, the abstract base
* Lexer class will determine which concrete Lexer is best for you:
* HTMLPurifier_Lexer_DirectLex for PHP 4, and HTMLPurifier_Lexer_DOMLex
* for PHP 5 and beyond.
*
* Passing the optional prototype lexer parameter will override the
* default with your own implementation. A copy/reference of the prototype
* lexer will now be returned when you request a new lexer.
*
* @static
*
* @note
* Though it is possible to call this factory method from subclasses,
* such usage is not recommended.
*
* @param $prototype Optional prototype lexer.
* @return Concrete lexer.
*/
static function create($prototype = null) {
// we don't really care if it's a reference or a copy
static $lexer = null;
if ($prototype) {
$lexer = $prototype;
}
if (empty($lexer)) {
if (version_compare(PHP_VERSION, "5", ">=") && // check for PHP5
class_exists('DOMDocument')) { // check for DOM support
require_once 'HTMLPurifier/Lexer/DOMLex.php';
$lexer = new HTMLPurifier_Lexer_DOMLex();
} else {
require_once 'HTMLPurifier/Lexer/DirectLex.php';
$lexer = new HTMLPurifier_Lexer_DirectLex();
}
}
return $lexer;
}
/**
* Translates CDATA sections into regular sections (through escaping).
*

23
library/HTMLPurifier/Lexer/DOMLex.php
View File

@ -53,20 +53,17 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
'</head><body><div>'.$string.'</div></body></html>';
$doc = new DOMDocument();
$doc->encoding = 'UTF-8'; // technically does nothing, but whatever
$doc->encoding = 'UTF-8'; // theoretically, the above has this covered
// DOM will toss errors if the HTML its parsing has really big
// problems, so we're going to mute them. This can cause problems
// if a custom error handler that doesn't implement error_reporting
// is set, as noted by a Drupal plugin of HTML Purifier. Consider
// making our own error reporter to temporarily load in
@$doc->loadHTML($string);
set_error_handler(array($this, 'muteErrorHandler'));
$doc->loadHTML($string);
restore_error_handler();
$tokens = array();
$this->tokenizeDOM(
$doc->getElementsByTagName('html')->item(0)-> // html
getElementsByTagName('body')->item(0)-> // body
getElementsByTagName('div')->item(0) // div
$doc->getElementsByTagName('html')->item(0)-> // <html>
getElementsByTagName('body')->item(0)-> // <body>
getElementsByTagName('div')->item(0) // <div>
, $tokens);
return $tokens;
}
@ -82,7 +79,6 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
* @returns Tokens of node appended to previously passed tokens.
*/
protected function tokenizeDOM($node, &$tokens, $collect = false) {
// recursive goodness!
// intercept non element nodes. WE MUST catch all of them,
// but we're not getting the character reference nodes because
@ -147,6 +143,11 @@ class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer
return $array;
}
/**
* An error handler that mutes all errors
*/
public function muteErrorHandler($errno, $errstr) {}
}
?>

104
library/HTMLPurifier/Lexer/DirectLex.php
View File

@ -2,6 +2,20 @@
require_once 'HTMLPurifier/Lexer.php';
HTMLPurifier_ConfigSchema::define(
'Core', 'DirectLexLineNumberSyncInterval', 0, 'int', '
<p>
Specifies the number of tokens the DirectLex line number tracking
implementations should process before attempting to resyncronize the
current line count by manually counting all previous new-lines. When
at 0, this functionality is disabled. Lower values will decrease
performance, and this is only strictly necessary if the counting
algorithm is buggy (in which case you should report it as a bug).
This has no effect when %Core.MaintainLineNumbers is disabled or DirectLex is
not being used. This directive has been available since 2.0.0.
</p>
');
/**
* Our in-house implementation of a parser.
*
@ -32,9 +46,17 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
$inside_tag = false; // whether or not we're parsing the inside of a tag
$array = array(); // result array
$maintain_line_numbers = $config->get('Core', 'MaintainLineNumbers');
$current_line = 1;
$nl = PHP_EOL;
// how often to manually recalculate. This will ALWAYS be right,
// but it's pretty wasteful. Set to 0 to turn off
$synchronize_interval = $config->get('Core', 'DirectLexLineNumberSyncInterval');
// infinite loop protection
// has to be pretty big, since html docs can be big
// we're allow two hundred thousand tags... more than enough?
// NOTE: this is also used for synchronization, so watch out
$loops = 0;
while(true) {
@ -42,10 +64,21 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
// infinite loop protection
if (++$loops > 200000) return array();
// recalculate lines
if (
$maintain_line_numbers && // line number tracking is on
$synchronize_interval && // synchronization is on
$cursor > 0 && // cursor is further than zero
$loops % $synchronize_interval === 0 // time to synchronize!
) {
$current_line = 1 + $this->substrCount($html, $nl, 0, $cursor);
}
$position_next_lt = strpos($html, '<', $cursor);
$position_next_gt = strpos($html, '>', $cursor);
// triggers on "<b>asdf</b>" but not "asdf <b></b>"
// special case to set up context
if ($position_next_lt === $cursor) {
$inside_tag = true;
$cursor++;
@ -53,7 +86,7 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
if (!$inside_tag && $position_next_lt !== false) {
// We are not inside tag and there still is another tag to parse
$array[] = new
$token = new
HTMLPurifier_Token_Text(
$this->parseData(
substr(
@ -61,6 +94,11 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
)
)
);
if ($maintain_line_numbers) {
$token->line = $current_line;
$current_line += $this->substrCount($html, $nl, $cursor, $position_next_lt - $cursor);
}
$array[] = $token;
$cursor = $position_next_lt + 1;
$inside_tag = true;
continue;
@ -69,7 +107,7 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
// If we're already at the end, break
if ($cursor === strlen($html)) break;
// Create Text of rest of string
$array[] = new
$token = new
HTMLPurifier_Token_Text(
$this->parseData(
substr(
@ -77,6 +115,8 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
)
)
);
if ($maintain_line_numbers) $token->line = $current_line;
$array[] = $token;
break;
} elseif ($inside_tag && $position_next_gt !== false) {
// We are in tag and it is well formed
@ -89,12 +129,17 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
substr($segment, 0, 3) == '!--' &&
substr($segment, $strlen_segment-2, 2) == '--'
) {
$array[] = new
$token = new
HTMLPurifier_Token_Comment(
substr(
$segment, 3, $strlen_segment - 5
)
);
if ($maintain_line_numbers) {
$token->line = $current_line;
$current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
}
$array[] = $token;
$inside_tag = false;
$cursor = $position_next_gt + 1;
continue;
@ -104,7 +149,12 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
$is_end_tag = (strpos($segment,'/') === 0);
if ($is_end_tag) {
$type = substr($segment, 1);
$array[] = new HTMLPurifier_Token_End($type);
$token = new HTMLPurifier_Token_End($type);
if ($maintain_line_numbers) {
$token->line = $current_line;
$current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
}
$array[] = $token;
$inside_tag = false;
$cursor = $position_next_gt + 1;
continue;
@ -114,7 +164,7 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
// have accidently grabbed an emoticon. Translate into
// text and go our merry way
if (!ctype_alnum($segment[0])) {
$array[] = new
$token = new
HTMLPurifier_Token_Text(
'<' .
$this->parseData(
@ -122,6 +172,11 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
) .
'>'
);
if ($maintain_line_numbers) {
$token->line = $current_line;
$current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
}
$array[] = $token;
$cursor = $position_next_gt + 1;
$inside_tag = false;
continue;
@ -142,10 +197,15 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
if ($position_first_space >= $strlen_segment) {
if ($is_self_closing) {
$array[] = new HTMLPurifier_Token_Empty($segment);
$token = new HTMLPurifier_Token_Empty($segment);
} else {
$array[] = new HTMLPurifier_Token_Start($segment);
$token = new HTMLPurifier_Token_Start($segment);
}
if ($maintain_line_numbers) {
$token->line = $current_line;
$current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
}
$array[] = $token;
$inside_tag = false;
$cursor = $position_next_gt + 1;
continue;
@ -169,21 +229,29 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
}
if ($is_self_closing) {
$array[] = new HTMLPurifier_Token_Empty($type, $attr);
$token = new HTMLPurifier_Token_Empty($type, $attr);
} else {
$array[] = new HTMLPurifier_Token_Start($type, $attr);
$token = new HTMLPurifier_Token_Start($type, $attr);
}
if ($maintain_line_numbers) {
$token->line = $current_line;
$current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
}
$array[] = $token;
$cursor = $position_next_gt + 1;
$inside_tag = false;
continue;
} else {
$array[] = new
$token = new
HTMLPurifier_Token_Text(
'<' .
$this->parseData(
substr($html, $cursor)
)
);
if ($maintain_line_numbers) $token->line = $current_line;
// no cursor scroll? Hmm...
$array[] = $token;
break;
}
break;
@ -191,6 +259,22 @@ class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
return $array;
}
/**
* PHP 4 compatible substr_count that implements offset and length
*/
function substrCount($haystack, $needle, $offset, $length) {
static $oldVersion;
if ($oldVersion === null) {
$oldVersion = version_compare(PHP_VERSION, '5.1', '<');
}
if ($oldVersion) {
$haystack = substr($haystack, $offset, $length);
return substr_count($haystack, $needle);
} else {
return substr_count($haystack, $needle, $offset, $length);
}
}
/**
* Takes the inside of an HTML tag and makes an assoc array of attributes.
*

18
library/HTMLPurifier/Printer.php
View File

@ -4,6 +4,8 @@ require_once 'HTMLPurifier/Generator.php';
require_once 'HTMLPurifier/Token.php';
require_once 'HTMLPurifier/Encoder.php';
// OUT OF DATE, NEEDS UPDATING!
class HTMLPurifier_Printer
{
@ -26,9 +28,9 @@ class HTMLPurifier_Printer
/**
* Main function that renders object or aspect of that object
* @param $config Configuration object
* @note Parameters vary depending on printer
*/
function render($config) {}
// function render() {}
/**
* Returns a start tag
@ -64,6 +66,18 @@ class HTMLPurifier_Printer
$this->end($tag);
}
function elementEmpty($tag, $attr = array()) {
return $this->generator->generateFromToken(
new HTMLPurifier_Token_Empty($tag, $attr)
);
}
function text($text) {
return $this->generator->generateFromToken(
new HTMLPurifier_Token_Text($text)
);
}
/**
* Prints a simple key/value row in a table.
* @param $name Key

Some files were not shown because too many files have changed in this diff Show More