0
0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2024-12-22 08:21:52 +00:00
htmlpurifier/docs/dev-advanced-api.html
2007-03-27 01:26:26 +00:00

207 lines
8.0 KiB
HTML

<?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="Functional specification for HTML Purifier's advanced API for defining custom filtering behavior." />
<link rel="stylesheet" type="text/css" href="style.css" />
<title>Advanced API - HTML Purifier</title>
</head><body>
<h1>Advanced API</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://hp.jpsband.org/">HTML Purifier</a> End-User Documentation</div>
<p>It makes no sense to adopt a <q>one-size-fits-all</q> approach to
filtersets: therefore, users must be able to define their own sets of
<q>allowed</q> elements, as well as switch in-between doctypes of HTML.</p>
<p>Our goals are to let the user:</p>
<dl>
<dt>Select</dt>
<dd><ul>
<li>Doctype</li>
<li>Filtersets: Rich / Plain / Full ...</li>
<li>Mode: Lenient / Correctional</li>
<li>Collections (?): Safe / Unsafe</li>
<li>Tags / Attributes / Modules</li>
</ul></dd>
<dt>Customize</dt>
<dd><ul>
<li>Tags / Attributes / Attribute Types</li>
<li>Filtersets</li>
<li>Root Node</li>
</ul></dd>
<dt>Create</dt>
<dd><ul>
<li>Modules / Tags / Attributes / Attribute Types</li>
<li>Filtersets</li>
<li>Doctype</li>
</ul></dd>
</dl>
<h2>Select</h2>
<h3>Selecting a Doctype</h3>
<p>By default, users will use a doctype-based, permissive but secure
whitelist. They must define a <strong>doctype</strong>, and this serves
as the first method of determining a filterset.</p>
<p class="technical">This identifier is based
on the name the W3C has given to the document type and <em>not</em>
the DTD identifier.</p>
<p>This parameter is set via the configuration object:</p>
<pre>$config->set('HTML', 'Doctype', 'XHTML 1.0 Transitional');</pre>
<p>Due to legacy, the default option is XHTML 1.0 Transitional, however, we
really shouldn't be guessing what the user's doctype is. Fortunantely,
people who can't be bothered to set this won't be bothered when their
pages stop validating.</p>
<h3>Selecting a Filterset</h3>
<p>However, selecting this doctype doesn't mean much, because if we
adhered exactly to the definition we would be letting XSS and other
nasties through. HTML Purifier must, in its filterset, allow a subset
of the doctype, which we shall call a <strong>filterset</strong>.</p>
<p>By default, HTML Purifier will use the <strong>Rich</strong>
filterset, which allows as many elements as possible with untrusted
sources. Other possible filtersets could be:</p>
<dl>
<dt>Full</dt>
<dd>Allows the full span of elements in the doctype, good if you want
HTML Purifier to work as a Tidy substitute but not to strip
anything out.</dd>
<dt>Plain</dt>
<dd>Provides a minimum set of tags for semantic markup of things
like blog comments.</dd>
</dl>
<p>Extension-authors would be able to define custom filtersets for
other users to use.</p>
<p>A possible call to select a filterset would be:</p>
<pre>$config->set('HTML', 'Filterset', 'Rich');</pre>
<h3>Selecting Mode</h3>
<p>Within filtersets, there are various <strong>modes</strong> of operation.
These indicate variant behaviors that, while not strictly changing the
allowed set of elements and attributes, will definitely affect the output.
Currently, we have two modes, which may be used together:</p>
<dl>
<dt>Lenient</dt>
<dd>Deprecated elements and attributes will be transformed into
standards-compliant alternatives when explicitly disallowed. For
example, in the XHTML 1.0 Strict doctype, a <code>center</code>
tag would be turned into a <code>div</code> with the CSS property
<code>text-align:center;</code>, but in XHTML 1.0 Transitional
the tag would be preserved. This mode is on by default.</dd>
<dt>Correctional</dt>
<dd>Deprecated elements and attributes will be transformed into
standards-compliant alternatives whenever possible. Referring
back to the previous example, the <code>center</code> tag would
be transformed in both cases. However, tags without a
reasonable standards-compliant alternative will be preserved
in their form. This mode is on by default. It may have
various levels of operation.</dd>
</dl>
<p>A possible call to select modes would be:</p>
<pre>$config->set('HTML', 'Mode', array('correctional', 'lenient'));</pre>
<p>If modes have extra parameters, a hash might work well:</p>
<pre>$config->set('HTML', 'Mode', array(
'correctional' => 9, // strongest level
'lenient' => true // this one's just boolean
));</pre>
<p>Modes may possibly be wrapped up with the filterset declaration:</p>
<pre>$config->set('HTML', 'Filterset', 'Rich: correctional, lenient');</pre>
<p>Further investigation in this field is necessary.</p>
<p>With regards to the various levels of operation conjectured in the
Correctional mode, this is prompted by the fact that a user may want to
correct certain problems but not others, for example, fix the <code>center</code>
tag but not the <code>u</code> tag, both of which are deprecated.
Having an integer <q>level</q> will not work very well for such fine
grained tweaking, but an array of specific settings might.</p>
<h3>Selecting Tags / Attributes / Modules</h3>
<p>If this cookie cutter approach doesn't appeal to a user, they may
decide to roll their own filterset by selecting modules, tags and
attributes to allow.</p>
<p class="technical">This would make use of the same facilities
as a filterset author would use, except that it would go under an
<q>anonymous</q> filterset that would be auto-selected if any of the
relevant module/tag/attribute selection configuration directives were
non-null.</p>
<p>In practice, this is the most commonly demanded feature. Most users are
perfectly happy defining a filterset that looks like:</p>
<pre>$config->setAllowedHTML('a[href,title],em,p,blockquote');</pre>
<p>We currently support a separated interface, which also must be preserved:</p>
<pre>$config->set('HTML', 'AllowedTags', 'a,em,p,blockquote');
$config->set('HTML', 'AllowedAttributes', 'a.href,a.title');</pre>
<p class="technical">The directive %HTML.Allowed is a convenience function
that may be fully expressed with the legacy interface, and thus is
given its own setter.</p>
<p>A user may also choose to allow modules:</p>
<pre>$config->set('HTML', 'AllowedModules', 'Hypertext,Text,Lists'); // or
$config->setAllowedHTML('Hypertext,Text,Lists');</pre>
<p>But it is not expected that this feature will be widely used.</p>
<p class="fixme">The granularity of these modules is too coarse for
the average user (for example, the core module loads everything from
the essential <code>p</code> tag to the not-so-safe <code>h1</code>
tag). How do we make this still a viable solution?</p>
<p class="technical">Modules are distinguished from regular tags by the
case of their first letter. While XML distinguishes between lower and uppercase
letters, in practice, most well-known XML languages use only lower-case
tag names for sake of consistency.</p>
<p class="technical">Considering that, internally speaking, as mandated by
the XHTML 1.1 Modularization specification, we have organized our
elements around modules, considerable gymnastics will be needed to
get this sort of functionality working.</p>
<h3>Unified selector</h3>
<p>Because selecting each and every one of these configuration options
is a chore, we may wish to offer a specialized configuration method
for selecting a filterset. Possibility:</p>
<pre>function selectFilter($doctype, $filterset, $mode)</pre>
<p>...which is simply a light wrapper over the individual configuration
calls. A custom config file format or text format could also be adopted.</p>
<div id="version">$Id$</div>
</body></html>