Add full documentation. Implement deferred ChildDef to HTMLModule. Add missing attributes for table, switched some to Number. Add necessary includes to module files. Add pre exclusions. Printer now ksorts arrays before output. Exclude ins/del from descendants_are_inline flagging.

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@717 48356398-32a2-884e-a903-53898d9a118a

library/HTMLPurifier/HTMLModule/Bdo.php

@@ -1,6 +1,7 @@
 <?php
 
 require_once 'HTMLPurifier/HTMLModule.php';
+require_once 'HTMLPurifier/AttrTransform/BdoDir.php';
 
 /**
  * XHTML 1.1 Bi-directional Text Module, defines elements that
@@ -23,13 +24,19 @@ class HTMLPurifier_HTMLModule_Bdo extends HTMLPurifier_HTMLModule
         $this->info['bdo']->attr = array(
             0 => array('Core'),
             '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)
             'lang' => 'Lang',
             'xml:lang' => 'Lang'
         );
         $this->info['bdo']->content_model = '#PCDATA | Inline';
         $this->info['bdo']->content_model_type = 'optional';
-        $this->info['bdo']->content_model_type = 'optional';
-        $this->info['bdo']->attr_transform_post[] = new HTMLPurifier_AttrTransform_BdoDir();
+        // provides fallback behavior if dir's missing (dir is required)
+        $this->info['bdo']->attr_transform_post[] =
+            new HTMLPurifier_AttrTransform_BdoDir();
     }
     
 }

library/HTMLPurifier/HTMLModule/Edit.php

@@ -1,9 +1,11 @@
 <?php
 
 require_once 'HTMLPurifier/HTMLModule.php';
+require_once 'HTMLPurifier/ChildDef/Chameleon.php';
 
 /**
- * XHTML 1.1 Edit Module, defines editing-related elements. Text Extension Module.
+ * XHTML 1.1 Edit Module, defines editing-related elements. Text Extension
+ * Module.
  */
 class HTMLPurifier_HTMLModule_Edit extends HTMLPurifier_HTMLModule
 {
@@ -20,11 +22,24 @@ class HTMLPurifier_HTMLModule_Edit extends HTMLPurifier_HTMLModule
                 'cite' => 'URI',
                 // 'datetime' => 'Datetime' // Datetime not implemented
             );
-            $this->info[$element]->content_model = '#PCDATA | Inline ! #PCDATA | Flow';
+            // 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';
         }
     }
     
+    var $defines_child_def = true;
+    function getChildDef($def) {
+        if ($def->content_model_type != 'chameleon') return false;
+        $value = explode('!', $def->content_model);
+        return new HTMLPurifier_ChildDef_Chameleon($value[0], $value[1]);
+    }
+    
 }
 
 ?>

library/HTMLPurifier/HTMLModule/Image.php

@@ -2,8 +2,13 @@
 
 require_once 'HTMLPurifier/HTMLModule.php';
 
+require_once 'HTMLPurifier/AttrDef/URI.php';
+require_once 'HTMLPurifier/AttrTransform/ImgRequired.php';
+
 /**
  * XHTML 1.1 Image Module provides basic image embedding.
+ * @note There is specialized code for removing empty images in
+ *       HTMLPurifier_Strategy_RemoveForeignElements
  */
 class HTMLPurifier_HTMLModule_Image extends HTMLPurifier_HTMLModule
 {

library/HTMLPurifier/HTMLModule/List.php

@@ -10,11 +10,14 @@ class HTMLPurifier_HTMLModule_List extends HTMLPurifier_HTMLModule
     
     var $elements = array('dl', 'dt', 'dd', 'ol', 'ul', 'li');
     var $info = array();
-    // technically speaking, the List content set is a fully formed
+    // 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
     // so we're not going to do that subtlety. It might cause trouble
     // if a user defines "List" and expects that multiple lists are
     // allowed to be specified, but then again, that's not very intuitive.
+    // 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');
     
     function HTMLPurifier_HTMLModule_List() {
@@ -33,6 +36,7 @@ class HTMLPurifier_HTMLModule_List extends HTMLPurifier_HTMLModule
         $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);
     }
     

library/HTMLPurifier/HTMLModule/Presentation.php

@@ -3,7 +3,14 @@
 require_once 'HTMLPurifier/HTMLModule.php';
 
 /**
- * XHTML 1.1 Presentation Module, defines hypertext links. Text Extension Module.
+ * XHTML 1.1 Presentation Module, defines simple presentation-related
+ * markup. Text Extension Module.
+ * @note The official XML Schema and DTD specs further divide this into
+ *       two modules:
+ *          - Block Presentation (hr)
+ *          - Inline Presentation (b, big, i, small, sub, sup, tt)
+ *       We have chosen not to heed this distinction, as content_sets
+ *       provides satisfactory disambiguation.
  */
 class HTMLPurifier_HTMLModule_Presentation extends HTMLPurifier_HTMLModule
 {

library/HTMLPurifier/HTMLModule/StyleAttribute.php

@@ -1,14 +1,18 @@
 <?php
 
 require_once 'HTMLPurifier/HTMLModule.php';
+require_once 'HTMLPurifier/AttrDef/CSS.php';
 
 /**
- * XHTML 1.1 Edit Module, defines editing-related elements. Text Extension Module.
+ * XHTML 1.1 Edit Module, defines editing-related elements. Text Extension
+ * Module.
  */
 class HTMLPurifier_HTMLModule_StyleAttribute extends HTMLPurifier_HTMLModule
 {
     var $attr_collection = array(
-        'Style' => array('style' => false),
+        // The inclusion routine differs from the Abstract Modules but
+        // is in line with the DTD and XML Schemas.
+        'Style' => array('style' => false), // see constructor
         'Core' => array(0 => array('Style'))
     );
     

library/HTMLPurifier/HTMLModule/Tables.php

@@ -1,6 +1,7 @@
 <?php
 
 require_once 'HTMLPurifier/HTMLModule.php';
+require_once 'HTMLPurifier/ChildDef/Table.php';
 
 /**
  * XHTML 1.1 Tables Module, fully defines accessible table elements.
@@ -34,7 +35,15 @@ class HTMLPurifier_HTMLModule_Tables extends HTMLPurifier_HTMLModule
                 $attr['width'] = 'Length';
                 continue;
             }
-            if ($e == 'td' || $e == 'th') $attr['abbr'] = 'Text';
+            if ($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);
@@ -46,7 +55,10 @@ class HTMLPurifier_HTMLModule_Tables extends HTMLPurifier_HTMLModule
         $this->info['caption']->content_model = '#PCDATA | Inline';
         $this->info['caption']->content_model_type = 'optional';
         
-        $this->info['table']->content_model = 'caption?, ( col* | colgroup* ), (( thead?, tfoot?, tbody+ ) | ( tr+ ))';
+        // Is done directly because it doesn't leverage substitution
+        // mechanisms. True model is:
+        // 'caption?, ( col* | colgroup* ), (( thead?, tfoot?, tbody+ ) | ( tr+ ))'
+        $this->info['table']->content_model = new HTMLPurifier_ChildDef_Table();
         $this->info['table']->content_model_type = 'table';
         
         $this->info['td']->content_model = 

library/HTMLPurifier/HTMLModule/Text.php

@@ -54,11 +54,21 @@ class HTMLPurifier_HTMLModule_Text extends HTMLPurifier_HTMLModule
                 $this->info[$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(
-                'address', 'blockquote', 'dd', 'dir', 'div', 'dl', 'dt',
-                'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'ol', 'p', 'pre',
-                'table', 'ul'
-            ));
+            'address', 'blockquote', 'dd', 'dir', 'div', 'dl', 'dt',
+            'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'ol', 'p', 'pre',
+            'table', 'ul'
+        ));
     }
     
 }