[2.1.2] Implement experimental HTML5 parsing using PH5P

- Fix debugger so that tokens can be printed without an index
- Fix some broken PEAR unit tests

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@1383 48356398-32a2-884e-a903-53898d9a118a

maintenance/PH5P.patch

@@ -0,0 +1,45 @@
+--- old.php	2007-08-19 14:42:33.640625000 -0400
++++ new.php	2007-08-19 14:41:51.609375000 -0400
+@@ -211,7 +211,10 @@
+         // If nothing is returned, emit a U+0026 AMPERSAND character token.
+         // Otherwise, emit the character token that was returned.
+         $char = (!$entity) ? '&' : $entity;
+-        $this->emitToken($char);
++        $this->emitToken(array(
++            'type' => self::CHARACTR,
++            'data' => $char
++        ));
+ 
+         // Finally, switch to the data state.
+         $this->state = 'data';
+@@ -708,7 +711,7 @@
+         } elseif($char === '&') {
+             /* U+0026 AMPERSAND (&)
+             Switch to the entity in attribute value state. */
+-            $this->entityInAttributeValueState('non');
++            $this->entityInAttributeValueState();
+ 
+         } elseif($char === '>') {
+             /* U+003E GREATER-THAN SIGN (>)
+@@ -738,7 +741,8 @@
+             ? '&'
+             : $entity;
+ 
+-        $this->emitToken($char);
++        $last = count($this->token['attr']) - 1;
++        $this->token['attr'][$last]['value'] .= $char;
+     }
+ 
+     private function bogusCommentState() {
+@@ -1066,6 +1070,11 @@
+                     $this->char++;
+ 
+                     if(in_array($id, $this->entities)) {
++                        if ($e_name[$c-1] !== ';') {
++                            if ($c < $len && $e_name[$c] == ';') {
++                                $this->char++; // consume extra semicolon
++                            }
++                        }
+                         $entity = $id;
+                         break;
+                     }