Release 2.0.1, merged in 1181 to HEAD.

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/branches/strict@1255 48356398-32a2-884e-a903-53898d9a118a

Doxyfile

@@ -4,7 +4,7 @@
 # Project related configuration options
 #---------------------------------------------------------------------------
 PROJECT_NAME           = HTML Purifier
-PROJECT_NUMBER         = 2.0.0
+PROJECT_NUMBER         = 2.0.1
 OUTPUT_DIRECTORY       = "C:/Documents and Settings/Edward/My Documents/My Webs/htmlpurifier/docs/doxygen"
 CREATE_SUBDIRS         = NO
 OUTPUT_LANGUAGE        = English

INSTALL

@@ -49,7 +49,7 @@ be standards compliant.  HTML Purifier can deal with these doctypes:
 * XHTML 1.0 Strict
 * HTML 4.01 Transitional
 * HTML 4.01 Strict
-* XHTML 1.1 sans Ruby
+* XHTML 1.1 (sans Ruby)
 
 ...and these character encodings:
 
@@ -117,8 +117,9 @@ websites):
 Note that HTML Purifier's support for non-Unicode encodings is crippled by the
 fact that any character not supported by that encoding will be silently
 dropped, EVEN if it is ampersand escaped.  If you want to work around
-this, you are welcome to read docs/enduser-utf8.html for a workaround,
+this, you are welcome to read docs/enduser-utf8.html for a fix,
-but please be cognizant of the issues the "solution" creates.
+but please be cognizant of the issues the "solution" creates (for this
+reason, I do not include the solution in this document).
 
 
 
@@ -148,7 +149,9 @@ Other supported doctypes include:
 There are more configuration directives which can be read about
 here: <http://htmlpurifier.org/live/configdoc/plain.html>  They're a bit boring,
 but they can help out for those of you who like to exert maximum control over
-your code.
+your code.  Some of the more interesting ones are configurable at the
+demo <http://htmlpurifier.org/demo.php> and are well worth looking into
+for your own system.
 
 
 
@@ -165,13 +168,15 @@ The interface is mind-numbingly simple:
     $clean_html = $purifier->purify( $dirty_html );
 
 That's it!  For more examples, check out docs/examples/ (they aren't very
-different though).  Also, SLOW gives advice on what to do if HTML Purifier
+different though).  Also, docs/enduser-slow.html gives advice on what to
-is slowing down your application.
+do if HTML Purifier is slowing down your application.
 
 
 
 6.   Quick install
 
+First, make sure library/HTMLPurifier/DefinitionCache/Serializer is
+writable by the webserver (see Section 7: Caching below for details).
 If your website is in UTF-8 and XHTML Transitional, use this code:
 
 <?php
@@ -198,6 +203,36 @@ If your website is in a different encoding or doctype, use this code:
 
 7. Caching
 
-HTML Purifier generates some cache files to speed up its execution. For
+HTML Purifier generates some cache files (generally one or two) to speed up
-maximum performance, make sure that library/HTMLPurifier/DefinitionCache/Serializer
+its execution. For maximum performance, make sure that
-is writeable by the webserver.
+library/HTMLPurifier/DefinitionCache/Serializer is writeable by the webserver.
+
+If you are in the library/ folder of HTML Purifier, you can set the
+appropriate permissions using:
+
+    chmod -R 0755 HTMLPurifier/DefinitionCache/Serializer
+
+If the above command doesn't work, you may need to assign write permissions
+to all. This may be necessary if your webserver runs as nobody, but is
+not recommended since it means any other user can write files in the
+directory. Use:
+
+    chmod -R 0777 HTMLPurifier/DefinitionCache/Serializer
+
+You can also chmod files via your FTP client; this option
+is usually accessible by right clicking the corresponding directory and
+then selecting "chmod" or "file permissions".
+
+Starting with 2.0.1, HTML Purifier will generate friendly error messages
+that will tell you exactly what you have to chmod the directory to, if in doubt,
+follow its advice.
+
+If you are unable or unwilling to give write permissions to the cache
+directory, you can either disable the cache (and suffer a performance
+hit):
+
+    $config->set('Core', 'DefinitionCache', null);
+
+Or move the cache directory somewhere else (no trailing slash):
+
+    $config->set('Cache', 'SerializerPath', '/home/user/absolute/path');

NEWS

@@ -9,6 +9,50 @@ NEWS ( CHANGELOG and HISTORY )                                     HTMLPurifier
     . Internal change
 ==========================
 
+2.0.1, released 2007-06-27
+! Tag auto-closing now based on a ChildDef heuristic rather than a
+  manually set auto_close array; some behavior may change
+! Experimental AutoFormat functionality added: auto-paragraph and
+  linkify your HTML input by setting %AutoFormat.AutoParagraph and
+  %AutoFormat.Linkify to true
+! Newlines normalized internally, and then converted back to the
+  value of PHP_EOL. If this is not desired, set your newline format
+  using %Output.Newline.
+! Beta error collection, messages are implemented for the most generic
+  cases involving Lexing or Strategies
+- Clean up special case code for <script> tags
+- Reorder includes for DefinitionCache decorators, fixes a possible
+  missing class error
+- Fixed bug where manually modified definitions were not saved via cache
+  (mostly harmless, except for the fact that it would be a little slower)
+- Configuration objects with different serials do not clobber each
+  others when revision numbers are unequal
+- Improve Serializer DefinitionCache directory permissions checks
+- DefinitionCache no longer throws errors when it encounters old
+  serial files that do not conform to the current style
+- Stray xmlns attributes removed from configuration documentation
+- configForm.php smoketest no longer has XSS vulnerability due to
+  unescaped print_r output
+- Printer adheres to configuration's directives on output format
+- Fix improperly named form field in ConfigForm printer
+. Rewire some test-cases to swallow errors rather than expect them
+. HTMLDefinition printer updated with some of the new attributes
+. DefinitionCache keys reordered to reflect precedence: version number,
+  hash, then revision number
+. %Core.DefinitionCache renamed to %Cache.DefinitionImpl
+. Interlinking in configuration documentation added using
+  Injector_PurifierLinkify
+. Directives now keep track of aliases to themselves
+. Error collector now requires a severity to be passed, use PHP's internal
+  error constants for this
+. HTMLPurifier_Config::getAllowedDirectivesForForm implemented, allows
+  much easier selective embedding of configuration values
+. Doctype objects now accept public and system DTD identifiers
+. %HTML.Doctype is now constrained by specific values, to specify a custom
+  doctype use new %HTML.CustomDoctype
+. ConfigForm truncates long directives to keep the form small, and does
+  not re-output namespaces
+
 2.0.0, released 2007-06-20
 # Completely refactored HTMLModuleManager, decentralizing safety
   information

TODO

@@ -7,31 +7,17 @@ TODO List
 ==========================
 
 2.1 release [Refactor, refactor!]
-
-
-
-
-
-
-
-
  # URI validation routines tighter (see docs/dev-code-quality.html) (COMPLEX)
  # Advanced URI filtering schemes (see docs/proposal-new-directives.txt)
+ # Ruby support
  - Configuration profiles: predefined directives set with one func call
  - Implement IDREF support (harder than it seems, since you cannot have
    IDREFs to non-existent IDs)
  - Allow non-ASCII characters in font names
- - Genericize special cases in RemoveForeignElements
 
 2.2 release [Error'ed]
  # Error logging for filtering/cleanup procedures
-    - Requires I18N facilities to be created first (COMPLEX)
  - XSS-attempt detection
- - More fine-grained control over escaping behavior
-
-
-
-
 
 2.3 release [Do What I Mean, Not What I Say]
  # Additional support for poorly written HTML
@@ -56,14 +42,9 @@ TODO List
    AttrDef class)
  # More control over allowed CSS properties (maybe modularize it in the
    same fashion!)
- # Formatters for plaintext (COMPLEX)
+ # Formatters for plaintext
-    - Auto-paragraphing (be sure to leverage fact that we know when things
-      shouldn't be paragraphed, such as lists and tables).
-    - Linkify URLs
     - Smileys
-    - Linkification for HTML Purifier docs: notably configuration and classes
+ - Standardize token armor for all areas of processing
- - Allow tags to be "armored", an internal flag that protects them
-   from validation and passes them out unharmed
  - Fixes for Firefox's inability to handle COL alignment props (Bug 915)
  - Automatically add non-breaking spaces to empty table cells when
    empty-cells:show is applied to have compatibility with Internet Explorer
@@ -75,12 +56,13 @@ TODO List
     - Hooks for adding custom processors to custom namespaced tags and
       attributes, offer default implementation
     - Lots of documentation and samples
- - XHTML 1.1 support
 
 Ongoing
  - Lots of profiling, make it faster!
  - Plugins for major CMSes (COMPLEX)
     - WordPress (mostly written, needs beta-testing)
+    - phpBB
+    - Phorum
     - eFiction
     - more! (look for ones that use WYSIWYGs)
  - Complete basic smoketests
@@ -89,7 +71,8 @@ Unknown release (on a scratch-an-itch basis)
  ? Semi-lossy dumb alternate character encoding transfor
  ? Have 'lang' attribute be checked against official lists, achieved by
    encoding all characters that have string entity equivalents
- - Explain how to use HTML Purifier in non-PHP languages
+ - Explain how to use HTML Purifier in non-PHP languages / create
+   a simple command line stub
  - Abstract ChildDef_BlockQuote to work with all elements that only
    allow blocks in them, required or optional
  - Reorganize Unit Tests
@@ -97,10 +80,11 @@ Unknown release (on a scratch-an-itch basis)
  - Reorganize configuration directives (Create more namespaces! Get messy!)
 
 Requested
- ? Native content compression, whitespace stripping (don't rely on Tidy, make
-   sure we don't remove from <pre> or related tags)
 
 Wontfix
  - Non-lossy smart alternate character encoding transformations (unless
    patch provided)
  - Pretty-printing HTML, users can use Tidy on the output on entire page
+ - Native content compression, whitespace stripping (don't rely on Tidy, make
+   sure we don't remove from <pre> or related tags): use gzip if this is
+   really important

VERSION

@@ -1 +1 @@
-2.0.0
+2.0.1

WHATSNEW

@@ -1,7 +1,12 @@
-HTML Purifier 2.0 is the culmination of two major architectural changes. 
+The 2.0.1 release introduces a number of stability and usability fixes,
-The first is Tidy, which enables HTML Purifier to both natively support 
+as well as a number of (disabled by default) experimental features.  The
-deprecated elements and also convert them to standards-compliant 
+security-minded should note that a reflected XSS vulnerability was patched
-alternatives. The second is the Advanced API, which enables users to 
+in smoketests/configForm.php; if you cannot upgrade immediately, please
-create new elements and attributes with ease. Keeping in line with a 
+delete that file (if that directory is not publically accessible, there
-commitment to high quality, there are also five esoteric bug-fixes and a 
+is no security risk).  The maintenance changes include more helpful file
-plethora of subtle improvements that enhance the library. 
+permissions errors, internal newline normalization, reordered includes
+to prevent a missing class definition in some setups, and better cache
+revision and id handling.  The two experimental features are auto-formatting
+(auto-paragraphing and linkification) and error collection, these can
+be enabled with %AutoFormat.AutoParagraph, %AutoFormat.Linkify and
+%Core.CollectErrors respectively.

benchmarks/ProfileDirectLex.php

@@ -15,5 +15,3 @@ $context = new HTMLPurifier_Context();
 for ($i = 0; $i < 10; $i++) {
     $tokens = $lexer->tokenizeHTML($input, $config, $context);
 }
-
-?>

configdoc/generate.php

@@ -22,6 +22,10 @@ error_reporting(E_ALL); // probably not possible to use E_STRICT
 require_once '../library/HTMLPurifier.auto.php';
 require_once 'library/ConfigDoc.auto.php';
 
+$purifier = HTMLPurifier::getInstance(array(
+    'AutoFormat.PurifierLinkify' => true
+));
+
 $schema = HTMLPurifier_ConfigSchema::instance();
 $style = 'plain'; // use $_GET in the future
 $configdoc = new ConfigDoc();
@@ -37,4 +41,3 @@ if (php_sapi_name() != 'cli') {
     echo 'Files generated successfully.';
 }
 
-?>

configdoc/library/ConfigDoc.auto.php

@@ -7,4 +7,3 @@
 set_include_path(dirname(__FILE__) . PATH_SEPARATOR . get_include_path() );
 require_once 'ConfigDoc.php';
 
-?>

configdoc/library/ConfigDoc.php

@@ -36,4 +36,3 @@ class ConfigDoc
     
 }
 
-?>

configdoc/library/ConfigDoc/HTMLXSLTProcessor.php

@@ -36,6 +36,7 @@ class ConfigDoc_HTMLXSLTProcessor
         // fudges for HTML backwards compatibility
         $out = str_replace('/>', ' />', $out); // <br /> not <br/>
         $out = str_replace(' xmlns=""', '', $out); // rm unnecessary xmlns
+        $out = str_replace(' xmlns="http://www.w3.org/1999/xhtml"', '', $out); // rm unnecessary xmlns
         if (class_exists('Tidy')) {
             // cleanup output
             $config = array(
@@ -59,4 +60,3 @@ class ConfigDoc_HTMLXSLTProcessor
     
 }
 
-?>

configdoc/library/ConfigDoc/XMLSerializer.php

@@ -23,4 +23,3 @@ class ConfigDoc_XMLSerializer
     
 }
 
-?>

configdoc/library/ConfigDoc/XMLSerializer/ConfigSchema.php

@@ -50,6 +50,12 @@ class ConfigDoc_XMLSerializer_ConfigSchema extends ConfigDoc_XMLSerializer
                     $dom_document->createElement('name', $name)
                 );
                 
+                $dom_aliases = $dom_document->createElement('aliases');
+                $dom_directive->appendChild($dom_aliases);
+                foreach ($info->directiveAliases as $alias) {
+                    $dom_aliases->appendChild($dom_document->createElement('alias', $alias));
+                }
+                
                 $dom_constraints = $dom_document->createElement('constraints');
                 $dom_directive->appendChild($dom_constraints);
                 
@@ -115,4 +121,3 @@ class ConfigDoc_XMLSerializer_ConfigSchema extends ConfigDoc_XMLSerializer
     
 }
 
-?>

configdoc/library/ConfigDoc/XMLSerializer/Types.php

@@ -24,4 +24,3 @@ class ConfigDoc_XMLSerializer_Types extends ConfigDoc_XMLSerializer
     
 }
 
-?>

configdoc/styles/plain.xsl

@@ -72,8 +72,16 @@
         <xsl:apply-templates />
     </xsl:template>
     <xsl:template match="directive/name">
+        <xsl:apply-templates select="../aliases/alias" mode="anchor" />
         <h3 id="{../@id}"><xsl:value-of select="../@id" /></h3>
     </xsl:template>
+    <xsl:template match="alias" mode="anchor">
+        <a id="{.}"></a>
+    </xsl:template>
+    
+    <!-- Do not pass through -->
+    <xsl:template match="alias"></xsl:template>
+    
     <xsl:template match="directive/constraints">
         <table class="constraints">
             <xsl:apply-templates />
@@ -89,8 +97,20 @@
                     </td>
                 </tr>
             </xsl:if>
+            <xsl:if test="../aliases/alias">
+                <xsl:apply-templates select="../aliases" mode="constraints" />
+            </xsl:if>
         </table>
     </xsl:template>
+    <xsl:template match="directive/aliases" mode="constraints">
+        <th>Aliases:</th>
+        <td>
+            <xsl:for-each select="alias">
+                <xsl:if test="position()&gt;1">, </xsl:if>
+                <xsl:value-of select="." />
+            </xsl:for-each>
+        </td>
+    </xsl:template>
     <xsl:template match="directive//description">
         <div class="description">
             <xsl:copy-of select="div/node()" />

docs/enduser-customize.html

@@ -17,11 +17,6 @@
 <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

docs/enduser-tidy.html

@@ -16,11 +16,6 @@
 <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>

docs/examples/basic.php

@@ -20,4 +20,3 @@ $pure_html = $purifier->purify($html);
 
 echo '<pre>' . htmlspecialchars($pure_html) . '</pre>';
 
-?>

library/HTMLPurifier.auto.php

@@ -7,4 +7,3 @@
 set_include_path(dirname(__FILE__) . PATH_SEPARATOR . get_include_path() );
 require_once 'HTMLPurifier.php';
 
-?>

library/HTMLPurifier.func.php

@@ -18,4 +18,3 @@ function HTMLPurifier($html, $config = null) {
     return $purifier->purify($html, $config);
 }
 
-?>

library/HTMLPurifier.php

@@ -22,7 +22,7 @@
  */
 
 /*
-    HTML Purifier 2.0.0 - Standards Compliant HTML Filtering
+    HTML Purifier 2.0.1 - Standards Compliant HTML Filtering
     Copyright (C) 2006 Edward Z. Yang
 
     This library is free software; you can redistribute it and/or
@@ -51,16 +51,9 @@ require_once 'HTMLPurifier/Generator.php';
 require_once 'HTMLPurifier/Strategy/Core.php';
 require_once 'HTMLPurifier/Encoder.php';
 
+require_once 'HTMLPurifier/ErrorCollector.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
@@ -81,7 +74,7 @@ This directive has been available since 2.0.0.
 class HTMLPurifier
 {
     
-    var $version = '2.0.0';
+    var $version = '2.0.1';
     
     var $config;
     var $filters;
@@ -139,15 +132,19 @@ class HTMLPurifier
         
         $context = new HTMLPurifier_Context();
         
+        // our friendly neighborhood generator, all primed with configuration too!
+        $this->generator->generateFromTokens(array(), $config, $context);
+        $context->register('Generator', $this->generator);
+        
         // set up global context variables
         if ($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'));
+            $language = $language_factory->create($config, $context);
             $context->register('Locale', $language);
             
-            $error_collector = new HTMLPurifier_ErrorCollector();
+            $error_collector = new HTMLPurifier_ErrorCollector($context);
-            $context->register('ErrorCollector', $language);
+            $context->register('ErrorCollector', $error_collector);
         }
         
         $html = HTMLPurifier_Encoder::convertToUTF8($html, $config, $context);
@@ -216,4 +213,3 @@ class HTMLPurifier
     
 }
 
-?>

library/HTMLPurifier/AttrCollections.php

@@ -127,4 +127,3 @@ class HTMLPurifier_AttrCollections
     
 }
 
-?>

library/HTMLPurifier/AttrDef.php

@@ -84,4 +84,3 @@ class HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS.php

@@ -66,4 +66,3 @@ class HTMLPurifier_AttrDef_CSS extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Background.php

@@ -84,4 +84,3 @@ class HTMLPurifier_AttrDef_CSS_Background extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php

@@ -127,4 +127,3 @@ class HTMLPurifier_AttrDef_CSS_BackgroundPosition extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Border.php

@@ -42,4 +42,3 @@ class HTMLPurifier_AttrDef_CSS_Border extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Color.php

@@ -98,4 +98,3 @@ class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Composite.php

@@ -35,4 +35,3 @@ class HTMLPurifier_AttrDef_CSS_Composite extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Font.php

@@ -148,4 +148,3 @@ class HTMLPurifier_AttrDef_CSS_Font extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/FontFamily.php

@@ -59,4 +59,3 @@ class HTMLPurifier_AttrDef_CSS_FontFamily extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Length.php

@@ -53,4 +53,3 @@ class HTMLPurifier_AttrDef_CSS_Length extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/ListStyle.php

@@ -77,4 +77,3 @@ class HTMLPurifier_AttrDef_CSS_ListStyle extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Multiple.php

@@ -55,4 +55,3 @@ class HTMLPurifier_AttrDef_CSS_Multiple extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Number.php

@@ -58,4 +58,3 @@ class HTMLPurifier_AttrDef_CSS_Number extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/Percentage.php

@@ -40,4 +40,3 @@ class HTMLPurifier_AttrDef_CSS_Percentage extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/TextDecoration.php

@@ -34,4 +34,3 @@ class HTMLPurifier_AttrDef_CSS_TextDecoration extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/CSS/URI.php

@@ -55,4 +55,3 @@ class HTMLPurifier_AttrDef_CSS_URI extends HTMLPurifier_AttrDef_URI
     
 }
 
-?>

library/HTMLPurifier/AttrDef/Enum.php

@@ -63,4 +63,3 @@ class HTMLPurifier_AttrDef_Enum extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/Bool.php

@@ -27,4 +27,3 @@ class HTMLPurifier_AttrDef_HTML_Bool extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/Color.php

@@ -32,4 +32,3 @@ class HTMLPurifier_AttrDef_HTML_Color extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/FrameTarget.php

@@ -31,4 +31,3 @@ class HTMLPurifier_AttrDef_HTML_FrameTarget extends HTMLPurifier_AttrDef_Enum
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/ID.php

@@ -118,4 +118,3 @@ class HTMLPurifier_AttrDef_HTML_ID extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/Length.php

@@ -41,4 +41,3 @@ class HTMLPurifier_AttrDef_HTML_Length extends HTMLPurifier_AttrDef_HTML_Pixels
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/LinkTypes.php

@@ -70,4 +70,3 @@ class HTMLPurifier_AttrDef_HTML_LinkTypes extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/MultiLength.php

@@ -41,4 +41,3 @@ class HTMLPurifier_AttrDef_HTML_MultiLength extends HTMLPurifier_AttrDef_HTML_Le
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/Nmtokens.php

@@ -48,4 +48,3 @@ class HTMLPurifier_AttrDef_HTML_Nmtokens extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/HTML/Pixels.php

@@ -34,4 +34,3 @@ class HTMLPurifier_AttrDef_HTML_Pixels extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/Integer.php

@@ -72,4 +72,3 @@ class HTMLPurifier_AttrDef_Integer extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/Lang.php

@@ -72,4 +72,3 @@ class HTMLPurifier_AttrDef_Lang extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/Text.php

@@ -14,4 +14,3 @@ class HTMLPurifier_AttrDef_Text extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/URI.php

@@ -302,4 +302,4 @@ class HTMLPurifier_AttrDef_URI extends HTMLPurifier_AttrDef
     
 }
 
-?>
+

library/HTMLPurifier/AttrDef/URI/Email.php

@@ -14,4 +14,3 @@ class HTMLPurifier_AttrDef_URI_Email extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/URI/Email/SimpleCheck.php

@@ -20,4 +20,3 @@ class HTMLPurifier_AttrDef_URI_Email_SimpleCheck extends HTMLPurifier_AttrDef_UR
     
 }
 
-?>

library/HTMLPurifier/AttrDef/URI/Host.php

@@ -51,4 +51,3 @@ class HTMLPurifier_AttrDef_URI_Host extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/URI/IPv4.php

@@ -39,4 +39,3 @@ class HTMLPurifier_AttrDef_URI_IPv4 extends HTMLPurifier_AttrDef
     
 }
 
-?>

library/HTMLPurifier/AttrDef/URI/IPv6.php

@@ -98,4 +98,3 @@ class HTMLPurifier_AttrDef_URI_IPv6 extends HTMLPurifier_AttrDef_URI_IPv4
     
 }
 
-?>

library/HTMLPurifier/AttrTransform.php

@@ -55,4 +55,3 @@ class HTMLPurifier_AttrTransform
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/BdoDir.php

@@ -28,4 +28,3 @@ class HTMLPurifier_AttrTransform_BdoDir extends HTMLPurifier_AttrTransform
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/BgColor.php

@@ -23,4 +23,3 @@ extends HTMLPurifier_AttrTransform {
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/BoolToCSS.php

@@ -36,4 +36,3 @@ extends HTMLPurifier_AttrTransform {
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/Border.php

@@ -17,4 +17,3 @@ class HTMLPurifier_AttrTransform_Border extends HTMLPurifier_AttrTransform {
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/EnumToCSS.php

@@ -57,4 +57,3 @@ class HTMLPurifier_AttrTransform_EnumToCSS extends HTMLPurifier_AttrTransform {
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/ImgRequired.php

@@ -51,4 +51,3 @@ class HTMLPurifier_AttrTransform_ImgRequired extends HTMLPurifier_AttrTransform
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/ImgSpace.php

@@ -44,4 +44,3 @@ extends HTMLPurifier_AttrTransform {
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/Lang.php

@@ -27,4 +27,3 @@ class HTMLPurifier_AttrTransform_Lang extends HTMLPurifier_AttrTransform
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/Length.php

@@ -26,4 +26,3 @@ class HTMLPurifier_AttrTransform_Length extends HTMLPurifier_AttrTransform
     
 }
 
-?>

library/HTMLPurifier/AttrTransform/Name.php

@@ -18,4 +18,3 @@ class HTMLPurifier_AttrTransform_Name extends HTMLPurifier_AttrTransform
     
 }
 
-?>

library/HTMLPurifier/AttrTypes.php

@@ -79,4 +79,4 @@ class HTMLPurifier_AttrTypes
     }
 }
 
-?>
+

library/HTMLPurifier/AttrValidator.php

@@ -1,32 +1,53 @@
 <?php
 
+/**
+ * Validates the attributes of a token. Doesn't manage required attributes
+ * very well. The only reason we factored this out was because RemoveForeignElements
+ * also needed it besides ValidateAttributes.
+ */
 class HTMLPurifier_AttrValidator
 {
     
-    
+    /**
-    function validateToken($token, &$config, &$context) {
+     * Validates the attributes of a token, returning a modified token
+     * that has valid tokens
+     * @param $token Reference to token to validate. We require a reference
+     *     because the operation this class performs on the token are
+     *     not atomic, so the context CurrentToken to be updated
+     *     throughout
+     * @param $config Instance of HTMLPurifier_Config
+     * @param $context Instance of HTMLPurifier_Context
+     */
+    function validateToken(&$token, &$config, &$context) {
             
         $definition = $config->getHTMLDefinition();
+        $e =& $context->get('ErrorCollector', true);
+        
+        // initialize CurrentToken if necessary
+        $current_token =& $context->get('CurrentToken', true);
+        if (!$current_token) $context->register('CurrentToken', $token);
+        
+        if ($token->type !== 'start' && $token->type !== 'empty') return $token;
         
         // create alias to global definition array, see also $defs
         // DEFINITION CALL
         $d_defs = $definition->info_global_attr;
         
-        // copy out attributes for easy manipulation
+        // reference attributes for easy manipulation
-        $attr = $token->attr;
+        $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);
+            $attr = $transform->transform($o = $attr, $config, $context);
+            if ($e && ($attr != $o)) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr);
         }
         
         // 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
+        foreach ($definition->info[$token->name]->attr_transform_pre as $transform) {
-            as $transform
+            $attr = $transform->transform($o = $attr, $config, $context);
-        ) {
+            if ($e && ($attr != $o)) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr);
-            $attr = $transform->transform($attr, $config, $context);
         }
         
         // create alias to this element's attribute definition array, see
@@ -34,6 +55,9 @@ class HTMLPurifier_AttrValidator
         // DEFINITION CALL
         $defs = $definition->info[$token->name]->attr;
         
+        $attr_key = false;
+        $context->register('CurrentAttr', $attr_key);
+        
         // iterate through all the attribute keypairs
         // Watch out for name collisions: $key has previously been used
         foreach ($attr as $attr_key => $value) {
@@ -67,9 +91,17 @@ class HTMLPurifier_AttrValidator
             
             // put the results into effect
             if ($result === false || $result === null) {
+                // this is a generic error message that should replaced
+                // with more specific ones when possible
+                if ($e) $e->send(E_ERROR, 'AttrValidator: Attribute removed');
+                
                 // remove the attribute
                 unset($attr[$attr_key]);
             } elseif (is_string($result)) {
+                // generally, if a substitution is happening, there
+                // was some sort of implicit correction going on. We'll
+                // delegate it to the attribute classes to say exactly what.
+                
                 // simple substitution
                 $attr[$attr_key] = $result;
             }
@@ -81,25 +113,27 @@ class HTMLPurifier_AttrValidator
             // others would prepend themselves).
         }
         
+        $context->destroy('CurrentAttr');
+        
         // post transforms
         
-        // ex. <x lang="fr"> to <x lang="fr" xml:lang="fr">
+        // global (error reporting untested)
         foreach ($definition->info_attr_transform_post as $transform) {
-            $attr = $transform->transform($attr, $config, $context);
+            $attr = $transform->transform($o = $attr, $config, $context);
+            if ($e && ($attr != $o)) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr);
         }
         
-        // ex. <bdo> to <bdo dir="ltr">
+        // local (error reporting untested)
         foreach ($definition->info[$token->name]->attr_transform_post as $transform) {
-            $attr = $transform->transform($attr, $config, $context);
+            $attr = $transform->transform($o = $attr, $config, $context);
+            if ($e && ($attr != $o)) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr);
         }
         
-        // commit changes
+        // destroy CurrentToken if we made it ourselves
-        $token->attr = $attr;
+        if (!$current_token) $context->destroy('CurrentToken');
-        return $token;
         
     }
     
     
 }
 
-?>

library/HTMLPurifier/CSSDefinition.php

@@ -226,4 +226,3 @@ class HTMLPurifier_CSSDefinition extends HTMLPurifier_Definition
     
 }
 
-?>

library/HTMLPurifier/ChildDef.php

@@ -36,6 +36,11 @@ class HTMLPurifier_ChildDef
      */
     var $allow_empty;
     
+    /**
+     * Lookup array of all elements that this definition could possibly allow
+     */
+    var $elements = array();
+    
     /**
      * Validates nodes according to definition and returns modification.
      * 
@@ -52,4 +57,4 @@ class HTMLPurifier_ChildDef
     }
 }
 
-?>
+

library/HTMLPurifier/ChildDef/Chameleon.php

@@ -35,6 +35,7 @@ class HTMLPurifier_ChildDef_Chameleon extends HTMLPurifier_ChildDef
     function HTMLPurifier_ChildDef_Chameleon($inline, $block) {
         $this->inline = new HTMLPurifier_ChildDef_Optional($inline);
         $this->block  = new HTMLPurifier_ChildDef_Optional($block);
+        $this->elements = $this->block->elements;
     }
     
     function validateChildren($tokens_of_children, $config, &$context) {
@@ -48,4 +49,3 @@ class HTMLPurifier_ChildDef_Chameleon extends HTMLPurifier_ChildDef
     }
 }
 
-?>

library/HTMLPurifier/ChildDef/Custom.php

@@ -44,6 +44,12 @@ class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef
         // COMPLICATED! AND MIGHT BE BUGGY! I HAVE NO CLUE WHAT I'M
         // DOING! Seriously: if there's problems, please report them.
         
+        // collect all elements into the $elements array
+        preg_match_all("/$el/", $reg, $matches);
+        foreach ($matches[0] as $match) {
+            $this->elements[$match] = true;
+        }
+        
         // setup all elements as parentheticals with leading commas
         $reg = preg_replace("/$el/", '(,\\0)', $reg);
         
@@ -85,4 +91,3 @@ class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef
     }
 }
 
-?>

library/HTMLPurifier/ChildDef/Empty.php

@@ -19,4 +19,3 @@ class HTMLPurifier_ChildDef_Empty extends HTMLPurifier_ChildDef
     }
 }
 
-?>

library/HTMLPurifier/ChildDef/Optional.php

@@ -20,4 +20,3 @@ class HTMLPurifier_ChildDef_Optional extends HTMLPurifier_ChildDef_Required
     }
 }
 
-?>

library/HTMLPurifier/ChildDef/Required.php

@@ -25,7 +25,7 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
             $elements = array_flip($elements);
             foreach ($elements as $i => $x) {
                 $elements[$i] = true;
-                if (empty($i)) unset($elements[$i]);
+                if (empty($i)) unset($elements[$i]); // remove blank
             }
         }
         $this->elements = $elements;
@@ -109,4 +109,3 @@ class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef
     }
 }
 
-?>

library/HTMLPurifier/ChildDef/StrictBlockquote.php

@@ -73,4 +73,3 @@ extends HTMLPurifier_ChildDef_Required
     }
 }
 
-?>

library/HTMLPurifier/ChildDef/Table.php

@@ -9,6 +9,8 @@ class HTMLPurifier_ChildDef_Table extends HTMLPurifier_ChildDef
 {
     var $allow_empty = false;
     var $type = 'table';
+    var $elements = array('tr' => true, 'tbody' => true, 'thead' => true,
+        'tfoot' => true, 'caption' => true, 'colgroup' => true, 'col' => true);
     function HTMLPurifier_ChildDef_Table() {}
     function validateChildren($tokens_of_children, $config, &$context) {
         if (empty($tokens_of_children)) return false;
@@ -139,4 +141,3 @@ class HTMLPurifier_ChildDef_Table extends HTMLPurifier_ChildDef
     }
 }
 
-?>

library/HTMLPurifier/Config.php

@@ -41,7 +41,7 @@ class HTMLPurifier_Config
     /**
      * HTML Purifier's version
      */
-    var $version = '2.0.0';
+    var $version = '2.0.1';
     
     /**
      * Two-level associative array of configuration directives
@@ -152,11 +152,15 @@ class HTMLPurifier_Config
     /**
      * Returns a md5 signature of a segment of the configuration object
      * that uniquely identifies that particular configuration
+     * @note Revision is handled specially and is removed from the batch
+     *       before processing!
      * @param $namespace Namespace to get serial for
      */
     function getBatchSerial($namespace) {
         if (empty($this->serials[$namespace])) {
-            $this->serials[$namespace] = md5(serialize($this->getBatch($namespace)));
+            $batch = $this->getBatch($namespace);
+            unset($batch['DefinitionRev']);
+            $this->serials[$namespace] = md5(serialize($batch));
         }
         return $this->serials[$namespace];
     }
@@ -243,14 +247,16 @@ class HTMLPurifier_Config
      *             called before it's been setup, otherwise won't work.
      */
     function &getHTMLDefinition($raw = false) {
-        return $this->getDefinition('HTML', $raw);
+        $def =& $this->getDefinition('HTML', $raw);
+        return $def; // prevent PHP 4.4.0 from complaining
     }
     
     /**
      * Retrieves reference to the CSS definition
      */
     function &getCSSDefinition($raw = false) {
-        return $this->getDefinition('CSS', $raw);
+        $def =& $this->getDefinition('CSS', $raw);
+        return $def;
     }
     
     /**
@@ -267,6 +273,7 @@ class HTMLPurifier_Config
             if (!empty($this->definitions[$type])) {
                 if (!$this->definitions[$type]->setup) {
                     $this->definitions[$type]->setup($this);
+                    $cache->set($this->definitions[$type], $this);
                 }
                 return $this->definitions[$type];
             }
@@ -298,7 +305,7 @@ class HTMLPurifier_Config
             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;
+                $false = new HTMLPurifier_Error();
                 return $false;
             }
             return $this->definitions[$type];
@@ -333,25 +340,78 @@ class HTMLPurifier_Config
         }
     }
     
+    /**
+     * Returns a list of array(namespace, directive) for all directives
+     * that are allowed in a web-form context as per an allowed
+     * namespaces/directives list.
+     * @param $allowed List of allowed namespaces/directives
+     * @static
+     */
+    static function getAllowedDirectivesForForm($allowed) {
+        $schema = HTMLPurifier_ConfigSchema::instance();
+        if ($allowed !== true) {
+             if (is_string($allowed)) $allowed = array($allowed);
+             $allowed_ns = array();
+             $allowed_directives = array();
+             $blacklisted_directives = array();
+             foreach ($allowed as $ns_or_directive) {
+                 if (strpos($ns_or_directive, '.') !== false) {
+                     // directive
+                     if ($ns_or_directive[0] == '-') {
+                         $blacklisted_directives[substr($ns_or_directive, 1)] = true;
+                     } else {
+                         $allowed_directives[$ns_or_directive] = true;
+                     }
+                 } else {
+                     // namespace
+                     $allowed_ns[$ns_or_directive] = true;
+                 }
+             }
+        }
+        $ret = array();
+        foreach ($schema->info as $ns => $keypairs) {
+            foreach ($keypairs as $directive => $def) {
+                if ($allowed !== true) {
+                    if (isset($blacklisted_directives["$ns.$directive"])) continue;
+                    if (!isset($allowed_directives["$ns.$directive"]) && !isset($allowed_ns[$ns])) continue;
+                }
+                if ($def->class == 'alias') continue;
+                if ($directive == 'DefinitionID' || $directive == 'DefinitionRev') continue;
+                $ret[] = array($ns, $directive);
+            }
+        }
+        return $ret;
+    }
+    
     /**
      * 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 $allowed List of allowed namespaces/directives 
      * @param $mq_fix Boolean whether or not to enable magic quotes fix
      * @static
      */
-    function loadArrayFromForm($array, $index, $mq_fix = true) {
+    static function loadArrayFromForm($array, $index, $allowed = true, $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)) {
+        $allowed = HTMLPurifier_Config::getAllowedDirectivesForForm($allowed);
-                unset($array[substr($key, 5)]);
+        $ret = array();
-                unset($array[$key]);
+        foreach ($allowed as $key) {
+            list($ns, $directive) = $key;
+            $skey = "$ns.$directive";
+            if (!empty($array["Null_$skey"])) {
+                $ret[$ns][$directive] = null;
+                continue;
             }
-            if ($mq) $array[$key] = stripslashes($value);
+            if (!isset($array[$skey])) continue;
+            $value = $mq ? stripslashes($array[$skey]) : $array[$skey];
+            $ret[$ns][$directive] = $value;
         }
-        return @HTMLPurifier_Config::create($array);
+        
+        $config = HTMLPurifier_Config::create($ret);
+        return $config;
     }
     
     /**
@@ -392,4 +452,4 @@ class HTMLPurifier_Config
     
 }
 
-?>
+

library/HTMLPurifier/ConfigDef.php

@@ -7,4 +7,3 @@ class HTMLPurifier_ConfigDef {
     var $class = false;
 }
 
-?>

library/HTMLPurifier/ConfigDef/Directive.php

@@ -61,6 +61,12 @@ class HTMLPurifier_ConfigDef_Directive extends HTMLPurifier_ConfigDef
      */
     var $aliases = array();
     
+    /**
+     * Advisory list of directive aliases, i.e. other directives that
+     * redirect here
+     */
+    var $directiveAliases = array();
+    
     /**
      * Adds a description to the array
      */
@@ -71,4 +77,3 @@ class HTMLPurifier_ConfigDef_Directive extends HTMLPurifier_ConfigDef
     
 }
 
-?>

library/HTMLPurifier/ConfigDef/DirectiveAlias.php

@@ -24,4 +24,3 @@ class HTMLPurifier_ConfigDef_DirectiveAlias extends HTMLPurifier_ConfigDef
     }
 }
 
-?>

library/HTMLPurifier/ConfigDef/Namespace.php

@@ -20,4 +20,3 @@ class HTMLPurifier_ConfigDef_Namespace extends HTMLPurifier_ConfigDef {
     
 }
 
-?>

library/HTMLPurifier/ConfigSchema.php

@@ -67,6 +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('AutoFormat', 'Configuration for activating auto-formatting functionality (also known as <code>Injector</code>s)');
+        $this->defineNamespace('AutoFormatParam', 'Configuration for customizing auto-formatting functionality');
         $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.');
@@ -293,6 +295,7 @@ class HTMLPurifier_ConfigSchema {
         $def->info[$namespace][$name] =
             new HTMLPurifier_ConfigDef_DirectiveAlias(
                 $new_namespace, $new_name);
+        $def->info[$new_namespace][$new_name]->directiveAliases[] = "$namespace.$name";
     }
     
     /**
@@ -401,4 +404,4 @@ class HTMLPurifier_ConfigSchema {
     }
 }
 
-?>
+

library/HTMLPurifier/ContentSets.php

@@ -147,4 +147,3 @@ class HTMLPurifier_ContentSets
     
 }
 
-?>

library/HTMLPurifier/Context.php

@@ -2,6 +2,8 @@
 
 /**
  * Registry object that contains information about the current context.
+ * @warning Is a bit buggy when variables are set to null: it thinks
+ *          they don't exist! So use false instead, please.
  */
 class HTMLPurifier_Context
 {
@@ -19,7 +21,7 @@ class HTMLPurifier_Context
      */
     function register($name, &$ref) {
         if (isset($this->_storage[$name])) {
-            trigger_error('Name collision, cannot re-register',
+            trigger_error("Name $name produces collision, cannot re-register",
                           E_USER_ERROR);
             return;
         }
@@ -29,11 +31,14 @@ class HTMLPurifier_Context
     /**
      * Retrieves a variable reference from the context.
      * @param $name String name
+     * @param $ignore_error Boolean whether or not to ignore error
      */
-    function &get($name) {
+    function &get($name, $ignore_error = false) {
         if (!isset($this->_storage[$name])) {
-            trigger_error('Attempted to retrieve non-existent variable',
+            if (!$ignore_error) {
+                trigger_error("Attempted to retrieve non-existent variable $name",
                               E_USER_ERROR);
+            }
             $var = null; // so we can return by reference
             return $var;
         }
@@ -46,7 +51,7 @@ class HTMLPurifier_Context
      */
     function destroy($name) {
         if (!isset($this->_storage[$name])) {
-            trigger_error('Attempted to destroy non-existent variable',
+            trigger_error("Attempted to destroy non-existent variable $name",
                           E_USER_ERROR);
             return;
         }
@@ -73,4 +78,3 @@ class HTMLPurifier_Context
     
 }
 
-?>

library/HTMLPurifier/Definition.php

@@ -38,4 +38,3 @@ class HTMLPurifier_Definition
     
 }
 
-?>

library/HTMLPurifier/DefinitionCache.php

@@ -4,6 +4,8 @@ require_once 'HTMLPurifier/DefinitionCache/Serializer.php';
 require_once 'HTMLPurifier/DefinitionCache/Null.php';
 
 require_once 'HTMLPurifier/DefinitionCache/Decorator.php';
+require_once 'HTMLPurifier/DefinitionCache/Decorator/Memory.php';
+require_once 'HTMLPurifier/DefinitionCache/Decorator/Cleanup.php';
 
 /**
  * Abstract class representing Definition cache managers that implements
@@ -36,8 +38,8 @@ class HTMLPurifier_DefinitionCache
      */
     function generateKey($config) {
         return $config->version . '-' . // possibly replace with function calls
-               $config->get($this->type, 'DefinitionRev') . '-' .
+               $config->getBatchSerial($this->type) . '-' .
-               $config->getBatchSerial($this->type);
+               $config->get($this->type, 'DefinitionRev');
     }
     
     /**
@@ -47,11 +49,17 @@ class HTMLPurifier_DefinitionCache
      * @param $config Instance of HTMLPurifier_Config to test against
      */
     function isOld($key, $config) {
-        list($version, $revision, $hash) = explode('-', $key, 3);
+        if (substr_count($key, '-') < 2) return true;
+        list($version, $hash, $revision) = explode('-', $key, 3);
         $compare = version_compare($version, $config->version);
-        if ($compare > 0) return false;
+        // version mismatch, is always old
-        if ($compare == 0 && $revision >= $config->get($this->type, 'DefinitionRev')) return false;
+        if ($compare != 0) return true;
-        return true;
+        // versions match, ids match, check revision number
+        if (
+            $hash == $config->getBatchSerial($this->type) &&
+            $revision < $config->get($this->type, 'DefinitionRev')
+        ) return true;
+        return false;
     }
     
     /**
@@ -118,4 +126,3 @@ class HTMLPurifier_DefinitionCache
     }
 }
 
-?>

library/HTMLPurifier/DefinitionCache/Decorator.php

@@ -2,9 +2,6 @@
 
 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
 {
     
@@ -60,4 +57,3 @@ class HTMLPurifier_DefinitionCache_Decorator extends HTMLPurifier_DefinitionCach
     
 }
 
-?>

library/HTMLPurifier/DefinitionCache/Decorator/Cleanup.php

@@ -42,4 +42,3 @@ class HTMLPurifier_DefinitionCache_Decorator_Cleanup extends
     
 }
 
-?>

library/HTMLPurifier/DefinitionCache/Decorator/Memory.php

@@ -45,4 +45,3 @@ class HTMLPurifier_DefinitionCache_Decorator_Memory extends
     
 }
 
-?>

library/HTMLPurifier/DefinitionCache/Decorator/Template.php.in

@@ -44,4 +44,3 @@ class HTMLPurifier_DefinitionCache_Decorator_Template extends
     
 }
 
-?>

library/HTMLPurifier/DefinitionCache/Null.php

@@ -34,4 +34,3 @@ class HTMLPurifier_DefinitionCache_Null extends HTMLPurifier_DefinitionCache
     
 }
 
-?>

library/HTMLPurifier/DefinitionCache/Serializer.php

@@ -21,14 +21,14 @@ class HTMLPurifier_DefinitionCache_Serializer extends
         if (!$this->checkDefType($def)) return;
         $file = $this->generateFilePath($config);
         if (file_exists($file)) return false;
-        $this->_prepareDir($config);
+        if (!$this->_prepareDir($config)) return false;
         return $this->_write($file, serialize($def));
     }
     
     function set($def, $config) {
         if (!$this->checkDefType($def)) return;
         $file = $this->generateFilePath($config);
-        $this->_prepareDir($config);
+        if (!$this->_prepareDir($config)) return false;
         return $this->_write($file, serialize($def));
     }
     
@@ -36,7 +36,7 @@ class HTMLPurifier_DefinitionCache_Serializer extends
         if (!$this->checkDefType($def)) return;
         $file = $this->generateFilePath($config);
         if (!file_exists($file)) return false;
-        $this->_prepareDir($config);
+        if (!$this->_prepareDir($config)) return false;
         return $this->_write($file, serialize($def));
     }
     
@@ -53,6 +53,7 @@ class HTMLPurifier_DefinitionCache_Serializer extends
     }
     
     function flush($config) {
+        if (!$this->_prepareDir($config)) return false;
         $dir = $this->generateDirectoryPath($config);
         $dh  = opendir($dir);
         while (false !== ($filename = readdir($dh))) {
@@ -63,7 +64,7 @@ class HTMLPurifier_DefinitionCache_Serializer extends
     }
     
     function cleanup($config) {
-        $this->_prepareDir($config);
+        if (!$this->_prepareDir($config)) return false;
         $dir = $this->generateDirectoryPath($config);
         $dh  = opendir($dir);
         while (false !== ($filename = readdir($dh))) {
@@ -88,9 +89,18 @@ class HTMLPurifier_DefinitionCache_Serializer extends
      * @note No trailing slash
      */
     function generateDirectoryPath($config) {
+        $base = $this->generateBaseDirectoryPath($config);
+        return $base . '/' . $this->type;
+    }
+    
+    /**
+     * Generates path to base directory that contains all definition type
+     * serials
+     */
+    function generateBaseDirectoryPath($config) {
         $base = $config->get('Cache', 'SerializerPath');
         $base = is_null($base) ? dirname(__FILE__) . '/Serializer' : $base;
-        return $base . '/' . $this->type;
+        return $base;
     }
     
     /**
@@ -116,14 +126,65 @@ class HTMLPurifier_DefinitionCache_Serializer extends
     
     /**
      * Prepares the directory that this type stores the serials in
+     * @return True if successful
      */
     function _prepareDir($config) {
         $directory = $this->generateDirectoryPath($config);
         if (!is_dir($directory)) {
-            mkdir($directory);
+            $base = $this->generateBaseDirectoryPath($config);
+            if (!is_dir($base)) {
+                trigger_error('Base directory '.$base.' does not exist,
+                    please create or change using %Cache.SerializerPath',
+                    E_USER_ERROR);
+                return false;
+            } elseif (!$this->_testPermissions($base)) {
+                return false;
             }
+            mkdir($directory);
+        } elseif (!$this->_testPermissions($directory)) {
+            return false;
+        }
+        return true;
+    }
+    
+    /**
+     * Tests permissions on a directory and throws out friendly
+     * error messages and attempts to chmod it itself if possible
+     */
+    function _testPermissions($dir) {
+        // early abort, if it is writable, everything is hunky-dory
+        if (is_writable($dir)) return true;
+        if (!is_dir($dir)) {
+            // generally, you'll want to handle this beforehand
+            // so a more specific error message can be given
+            trigger_error('Directory '.$dir.' does not exist',
+                E_USER_ERROR);
+            return false;
+        }
+        if (function_exists('posix_getuid')) {
+            // POSIX system, we can give more specific advice
+            if (fileowner($dir) === posix_getuid()) {
+                // we can chmod it ourselves
+                chmod($dir, 0755);
+                return true;
+            } elseif (filegroup($dir) === posix_getgid()) {
+                $chmod = '775';
+            } else {
+                // PHP's probably running as nobody, so we'll
+                // need to give global permissions
+                $chmod = '777';
+            }
+            trigger_error('Directory '.$dir.' not writable, '.
+                'please chmod to ' . $chmod,
+                E_USER_ERROR);
+        } else {
+            // generic error message
+            trigger_error('Directory '.$dir.' not writable, '.
+                'please alter file permissions',
+                E_USER_ERROR);
+        }
+        return false;
     }
     
 }
 
-?>

library/HTMLPurifier/DefinitionCacheFactory.php

@@ -3,7 +3,7 @@
 require_once 'HTMLPurifier/DefinitionCache.php';
 
 HTMLPurifier_ConfigSchema::define(
-    'Core', 'DefinitionCache', 'Serializer', 'string/null', '
+    'Cache', 'DefinitionImpl', '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
@@ -11,7 +11,12 @@ performance degradation). This directive has been available since 2.0.0.
 ');
 
 HTMLPurifier_ConfigSchema::defineAllowedValues(
-    'Core', 'DefinitionCache', array('Serializer')
+    'Cache', 'DefinitionImpl', array('Serializer')
+);
+
+HTMLPurifier_ConfigSchema::defineAlias(
+    'Core', 'DefinitionCache',
+    'Cache', 'DefinitionImpl'
 );
 
 
@@ -54,7 +59,7 @@ class HTMLPurifier_DefinitionCacheFactory
     function &create($type, $config) {
         // only one implementation as for right now, $config will
         // be used to determine implementation
-        $method = $config->get('Core', 'DefinitionCache');
+        $method = $config->get('Cache', 'DefinitionImpl');
         if ($method === null) {
             $null = new HTMLPurifier_DefinitionCache_Null($type);
             return $null;
@@ -87,4 +92,3 @@ class HTMLPurifier_DefinitionCacheFactory
     
 }
 
-?>

library/HTMLPurifier/Doctype.php

@@ -3,6 +3,8 @@
 /**
  * Represents a document type, contains information on which modules
  * need to be loaded.
+ * @note This class is inspected by Printer_HTMLDefinition->renderDoctype.
+ *       If structure changes, please update that function.
  */
 class HTMLPurifier_Doctype
 {
@@ -32,24 +34,33 @@ class HTMLPurifier_Doctype
      */
     var $aliases = array();
     
+    /**
+     * Public DTD identifier
+     */
+    var $dtdPublic;
+    
+    /**
+     * System DTD identifier
+     */
+    var $dtdSystem;
+    
     function HTMLPurifier_Doctype($name = null, $xml = true, $modules = array(),
-        $tidyModules = array(), $aliases = array()
+        $tidyModules = array(), $aliases = array(), $dtd_public = null, $dtd_system = null
     ) {
         $this->name         = $name;
         $this->xml          = $xml;
         $this->modules      = $modules;
         $this->tidyModules  = $tidyModules;
         $this->aliases      = $aliases;
+        $this->dtdPublic    = $dtd_public;
+        $this->dtdSystem    = $dtd_system;
     }
     
     /**
      * Clones the doctype, use before resolving modes and the like
      */
     function copy() {
-        return new HTMLPurifier_Doctype(
+        return unserialize(serialize($this));
-            $this->name, $this->xml, $this->modules, $this->tidyModules, $this->aliases
-        );
     }
 }
 
-?>

library/HTMLPurifier/DoctypeRegistry.php

@@ -44,14 +44,14 @@ class HTMLPurifier_DoctypeRegistry
      * @return Reference to registered doctype (usable for further editing)
      */
     function &register($doctype, $xml = true, $modules = array(),
-        $tidy_modules = array(), $aliases = array()
+        $tidy_modules = array(), $aliases = array(), $dtd_public = null, $dtd_system = null
     ) {
         if (!is_array($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
+                $doctype, $xml, $modules, $tidy_modules, $aliases, $dtd_public, $dtd_system
             );
         }
         $this->doctypes[$doctype->name] =& $doctype;
@@ -76,7 +76,7 @@ class HTMLPurifier_DoctypeRegistry
     function &get($doctype) {
         if (isset($this->aliases[$doctype])) $doctype = $this->aliases[$doctype];
         if (!isset($this->doctypes[$doctype])) {
-            trigger_error('Doctype ' . htmlspecialchars($doctype) . ' does not exist');
+            trigger_error('Doctype ' . htmlspecialchars($doctype) . ' does not exist', E_USER_ERROR);
             $anon = new HTMLPurifier_Doctype($doctype);
             return $anon;
         }
@@ -103,9 +103,9 @@ class HTMLPurifier_DoctypeRegistry
     function getDoctypeFromConfig($config) {
         // recommended test
         $doctype = $config->get('HTML', 'Doctype');
-        if ($doctype !== null) {
+        if (!empty($doctype)) return $doctype;
-            return $doctype;
+        $doctype = $config->get('HTML', 'CustomDoctype');
-        }
+        if (!empty($doctype)) return $doctype;
         // backwards-compatibility
         if ($config->get('HTML', 'XHTML')) {
             $doctype = 'XHTML 1.0';
@@ -122,4 +122,3 @@ class HTMLPurifier_DoctypeRegistry
     
 }
 
-?>

library/HTMLPurifier/ElementDef.php

@@ -3,6 +3,8 @@
 /**
  * Structure that stores an HTML element definition. Used by
  * HTMLPurifier_HTMLDefinition and HTMLPurifier_HTMLModule.
+ * @note This class is inspected by HTMLPurifier_Printer_HTMLDefinition.
+ *       Please update that class too.
  */
 class HTMLPurifier_ElementDef
 {
@@ -69,13 +71,6 @@ class HTMLPurifier_ElementDef
     
     
     
-    /**
-     * Lookup table of tags that close this tag. Used during parsing
-     * to make sure we don't attempt to nest unclosed tags.
-     * @public
-     */
-    var $auto_close = array();
-    
     /**
      * Does the element have a content model (#PCDATA | Inline)*? This
      * is important for chameleon ins and del processing in 
@@ -149,7 +144,6 @@ class HTMLPurifier_ElementDef
         }
         $this->_mergeAssocArray($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)) {
@@ -190,4 +184,4 @@ class HTMLPurifier_ElementDef
     
 }
 
-?>
+