[1.7.0] Finish implementing legacy elements, begin implementing legacy attributes

- Migrated most unit tests over to XHTML 1.0 Strict to preserve transformation behavior
- Created %Core.ColorKeywords to be shared between CSS_Color and HTML_Color
- Added AttrDef_HTML_Color as AttrType Color
- HTMLPurifier_Config::create(HTMLPurifier_Config $config) now clones the object
- Attribute minimization for HTML implemented in Generator
- Move div@align fix from proprietary to regular set
- Color keywords now map to full six digit hexadecimal codes
- Harness will now tack on per-use-case configuration

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@1084 48356398-32a2-884e-a903-53898d9a118a

TODO

@@ -13,9 +13,10 @@ TODO List
     - Reorganize configuration directives
     - Set up doctype object inside configuration object
     - Set up anonymous module management by HTMLDefinition
-    - Revamp/create unit tests for modules, esp. the Tidy ones
  # Implement HTMLDefinition caching using serialize
  # Implement all deprecated tags and attributes
+ # Create parsing/standards compliance smoketest
+ # Reorganize Unit Tests
  - Parse TinyMCE-style whitelist into our %HTML.Allow* whitelists (possibly
    do this earlier)
  ? HTML interface for tweaking configuration to see changes

library/HTMLPurifier/AttrDef/CSS/Color.php

@@ -2,6 +2,31 @@
 
 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 1.7.0.
+');
+
 /**
  * Validates Color as defined by CSS.
  */
@@ -10,25 +35,8 @@ class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef
     
     function validate($color, $config, &$context) {
         
-        static $colors = array(
+        static $colors = null;
-            'maroon'    => '#800000',
+        if ($colors === null) $colors = $config->get('Core', 'ColorKeywords');
-            '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'
-        );
         
         $color = trim($color);
         if (!$color) return false;

library/HTMLPurifier/AttrDef/HTML/Bool.php

@@ -0,0 +1,23 @@
+<?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) {$this->name = $name;}
+    
+    function validate($string, $config, &$context) {
+        if (empty($string)) return false;
+        return $name;
+    }
+    
+}
+
+?>

library/HTMLPurifier/AttrDef/HTML/Color.php

@@ -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";
+        
+    }
+    
+}
+
+?>

library/HTMLPurifier/AttrTypes.php

@@ -34,6 +34,7 @@ class HTMLPurifier_AttrTypes
         $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)
         $this->info['Number']   = new HTMLPurifier_AttrDef_Integer(false, false, true);

library/HTMLPurifier/Config.php

@@ -70,7 +70,9 @@ class HTMLPurifier_Config
      * @return Configured HTMLPurifier_Config object
      */
     function create($config) {
-        if (is_a($config, 'HTMLPurifier_Config')) return $config;
+        if (is_a($config, 'HTMLPurifier_Config')) {
+            $config = $config->conf; // create a clone
+        }
         $ret = HTMLPurifier_Config::createDefault();
         if (is_string($config)) $ret->loadIni($config);
         elseif (is_array($config)) $ret->loadArray($config);

library/HTMLPurifier/Generator.php

@@ -74,6 +74,12 @@ class HTMLPurifier_Generator
      */
     var $_scriptFix = false;
     
+    /**
+     * Cache of HTMLDefinition
+     * @private
+     */
+    var $_def;
+    
     /**
      * Generates HTML from an array of tokens.
      * @param $tokens Array of HTMLPurifier_Token
@@ -89,6 +95,8 @@ class HTMLPurifier_Generator
         $doctype = $config->getDoctype();
         $this->_xhtml = $doctype->xml;
         
+        $this->_def = $config->getHTMLDefinition();
+        
         if (!$tokens) return '';
         for ($i = 0, $size = count($tokens); $i < $size; $i++) {
             if ($this->_scriptFix && $tokens[$i]->name === 'script') {
@@ -136,14 +144,14 @@ 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 ? ' /': '' )
                 . '>';
@@ -174,13 +182,16 @@ class HTMLPurifier_Generator
      * @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).'" ';
         }

library/HTMLPurifier/HTMLModule/Legacy.php

@@ -1,5 +1,7 @@
 <?php
 
+require_once 'HTMLPurifier/AttrDef/HTML/Bool.php';
+
 /**
  * XHTML 1.1 Legacy module defines elements that were previously 
  * deprecated.
@@ -24,9 +26,28 @@ class HTMLPurifier_HTMLModule_Legacy extends HTMLPurifier_HTMLModule
     var $name = 'Legacy';
     
     function HTMLPurifier_HTMLModule_Legacy() {
-        $this->addElement('u', true, 'Inline', 'Inline', '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' => new HTMLPurifier_AttrDef_HTML_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' => new HTMLPurifier_AttrDef_HTML_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
         
@@ -36,6 +57,8 @@ class HTMLPurifier_HTMLModule_Legacy extends HTMLPurifier_HTMLModule
         $ol =& $this->addBlankElement('ol');
         $ol->attr['start'] = new HTMLPurifier_AttrDef_Integer();
         
+        $align = new HTMLPurifier_AttrDef_Enum(array('left', 'right', 'center', 'justify'));
+        
         $address =& $this->addBlankElement('address');
         $address->content_model = 'Inline | #PCDATA | p';
         $address->content_model_type = 'optional';
@@ -46,6 +69,24 @@ class HTMLPurifier_HTMLModule_Legacy extends HTMLPurifier_HTMLModule
         $blockquote->content_model_type = 'optional';
         $blockquote->child = false;
         
+        $br =& $this->addBlankElement('br');
+        $br->attr['clear'] = new HTMLPurifier_AttrDef_Enum(array('left', 'all', 'right', 'none'));
+        
+        $caption =& $this->addBlankElement('caption');
+        $caption->attr['align'] = new HTMLPurifier_AttrDef_Enum(array('top', 'bottom', 'left', 'right'));
+        
+        $div =& $this->addBlankElement('div');
+        $div->attr['align'] = $align;
+        
+        // dl.compact omitted
+        
+        for ($i = 1; $i <= 6; $i++) {
+            $h =& $this->addBlankElement("h$i");
+            $h->attr['align'] = $align;
+        }
+        
+        // to be continued...
+        
     }
     
 }

library/HTMLPurifier/HTMLModule/Tidy/Proprietary.php

@@ -10,16 +10,7 @@ class HTMLPurifier_HTMLModule_Tidy_Proprietary extends
     var $defaultLevel = 'light';
     
     function makeFixes() {
-        $r = array();
+        return array();
-        
-        // {{{ // duplicated from XHTMLAndHTML4: not sure how to factor out
-            $align_lookup = array();
-            $align_values = array('left', 'right', 'center', 'justify');
-            foreach ($align_values as $v) $align_lookup[$v] = "text-align:$v;";
-        // }}}
-        $r['div@align'] = new HTMLPurifier_AttrTransform_EnumToCSS('align', $align_lookup);
-        
-        return $r;
     }
     
 }

library/HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php

@@ -88,6 +88,7 @@ class HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4 extends
         $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 ---------------------------------
@@ -179,7 +180,7 @@ class HTMLPurifier_HTMLModule_Tidy_Transitional extends
       HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4
 {
     var $name = 'Tidy_Transitional';
-    var $defaultLevel = 'light'; // switch this to heavy once we implement legacy fully
+    var $defaultLevel = 'heavy';
 }
 
 class HTMLPurifier_HTMLModule_Tidy_Strict extends

tests/HTMLPurifier/AttrDef/CSS/BorderTest.php

@@ -10,9 +10,9 @@ class HTMLPurifier_AttrDef_CSS_BorderTest extends HTMLPurifier_AttrDefHarness
         $config = HTMLPurifier_Config::createDefault();
         $this->def = new HTMLPurifier_AttrDef_CSS_Border($config);
         
-        $this->assertDef('thick solid red', 'thick solid #F00');
+        $this->assertDef('thick solid red', 'thick solid #FF0000');
         $this->assertDef('thick solid');
-        $this->assertDef('solid red', 'solid #F00');
+        $this->assertDef('solid red', 'solid #FF0000');
         $this->assertDef('1px solid #000');
         
     }

tests/HTMLPurifier/AttrDef/CSS/ColorTest.php

@@ -25,7 +25,7 @@ class HTMLPurifier_AttrDef_CSS_ColorTest extends HTMLPurifier_AttrDefHarness
         $this->assertDef('rgb(256,-23,34)', 'rgb(255,0,34)');
         
         // color keywords, of course
-        $this->assertDef('red', '#F00');
+        $this->assertDef('red', '#FF0000');
         
         // maybe hex transformations would be another nice feature
         // at the very least transform rgb percent to rgb integer

tests/HTMLPurifier/AttrDef/HTML/ColorTest.php

@@ -0,0 +1,23 @@
+<?php
+
+require_once 'HTMLPurifier/AttrDefHarness.php';
+require_once 'HTMLPurifier/AttrDef/HTML/Color.php';
+
+class HTMLPurifier_AttrDef_HTML_ColorTest extends HTMLPurifier_AttrDefHarness
+{
+    
+    function test() {
+        $this->def = new HTMLPurifier_AttrDef_HTML_Color();
+        $this->assertDef('', false);
+        $this->assertDef('foo', false);
+        $this->assertDef('43', false);
+        $this->assertDef('red', '#FF0000');
+        $this->assertDef('#FF0000');
+        $this->assertDef('#453443');
+        $this->assertDef('453443', '#453443');
+        $this->assertDef('#345', '#334455');
+        $this->assertDef('120', '#112200');
+    }
+}
+
+?>

tests/HTMLPurifier/GeneratorTest.php

@@ -89,13 +89,21 @@ class HTMLPurifier_GeneratorTest extends HTMLPurifier_Harness
         $expect[4] = 'title="Theta is ' . $theta_char . '"';
         
         foreach ($inputs as $i => $input) {
-            $result = $this->obj->generateAttributes($input);
+            $result = $this->obj->generateAttributes($input, 'irrelevant');
             $this->assertIdentical($result, $expect[$i]);
             paintIf($result, $result != $expect[$i]);
         }
         
     }
     
+    function test_generateAttributes_minimized() {
+        $gen = new HTMLPurifier_Generator();
+        $context = new HTMLPurifier_Context();
+        $gen->generateFromTokens(array(), HTMLPurifier_Config::create(array('HTML.Doctype' => 'HTML 4.01 Transitional')), $context);
+        $result = $gen->generateAttributes(array('compact' => 'compact'), 'menu');
+        $this->assertIdentical($result, 'compact');
+    }
+    
     function test_generateFromTokens() {
         
         $this->func = 'generateFromTokens';

tests/HTMLPurifier/Harness.php

@@ -63,12 +63,13 @@ class HTMLPurifier_Harness extends UnitTestCase
      *                       context object.
      */
     function assertResult($input, $expect = true,
-        $config_array = false, $context_array = array()
+        $config_array = array(), $context_array = array()
     ) {
         
         // setup config 
         if ($this->config) {
             $config = HTMLPurifier_Config::create($this->config);
+            $config->loadArray($config_array);
         } else {
             $config = HTMLPurifier_Config::create($config_array);
         }

tests/HTMLPurifier/Strategy/FixNestingTest.php

@@ -13,6 +13,8 @@ class HTMLPurifier_Strategy_FixNestingTest extends HTMLPurifier_StrategyHarness
     
     function test() {
         
+        $this->config = array('HTML.Doctype' => 'XHTML 1.0 Strict');
+        
         // legal inline
         $this->assertResult('<b>Bold text</b>');
         

tests/HTMLPurifier/Strategy/MakeWellFormedTest.php

@@ -13,6 +13,8 @@ class HTMLPurifier_Strategy_MakeWellFormedTest extends HTMLPurifier_StrategyHarn
     
     function test() {
         
+        $this->config = array('HTML.Doctype' => 'XHTML 1.0 Strict');
+        
         $this->assertResult('');
         $this->assertResult('This is <b>bold text</b>.');
         

tests/HTMLPurifier/Strategy/RemoveForeignElementsTest.php

@@ -14,6 +14,7 @@ class HTMLPurifier_Strategy_RemoveForeignElementsTest
     
     function test() {
         
+        $this->config = array('HTML.Doctype' => 'XHTML 1.0 Strict');
         
         $this->assertResult('');
         

tests/HTMLPurifier/Strategy/ValidateAttributesTest.php

@@ -11,6 +11,7 @@ class HTMLPurifier_Strategy_ValidateAttributesTest extends
     function setUp() {
         parent::setUp();
         $this->obj = new HTMLPurifier_Strategy_ValidateAttributes();
+        $this->config = array('HTML.Doctype' => 'XHTML 1.0 Strict');
     }
     
     function testEmpty() {
@@ -297,7 +298,8 @@ class HTMLPurifier_Strategy_ValidateAttributesTest extends
         $this->assertResult(
             '<a href="foo" target="_top" />',
             true,
-            array('Attr.AllowedFrameTargets' => '_top')
+            array('Attr.AllowedFrameTargets' => '_top',
+                'HTML.Doctype' => 'XHTML 1.0 Transitional')
         );
         $this->assertResult(
             '<a href="foo" target="_top" />',

tests/test_files.php

@@ -20,6 +20,7 @@ $test_files[] = 'AttrDef/CSS/TextDecorationTest.php';
 $test_files[] = 'AttrDef/CSS/URITest.php';
 $test_files[] = 'AttrDef/CSSTest.php';
 $test_files[] = 'AttrDef/EnumTest.php';
+$test_files[] = 'AttrDef/HTML/ColorTest.php';
 $test_files[] = 'AttrDef/HTML/IDTest.php';
 $test_files[] = 'AttrDef/HTML/LengthTest.php';
 $test_files[] = 'AttrDef/HTML/FrameTargetTest.php';