[1.7.0] ChildDef_Custom's regex generation has been improved, removing several false positives

git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@1173 48356398-32a2-884e-a903-53898d9a118a

NEWS

@@ -46,6 +46,8 @@ NEWS ( CHANGELOG and HISTORY )                                     HTMLPurifier
 - StrictBlockquote child definition refrains from wrapping whitespace
   in tags now.
 - Bug resulting from tag transforms to non-allowed elements fixed
+- ChildDef_Custom's regex generation has been improved, removing several
+  false positives
 . Unit test for ElementDef created, ElementDef behavior modified to
   be more flexible
 . Added convenience functions for HTMLModule constructors

WHATSNEW

@@ -3,5 +3,5 @@ The first is Tidy, which enables HTML Purifier to both natively support
 deprecated elements and also convert them to standards-compliant 
 alternatives. The second is the Advanced API, which enables users to 
 create new elements and attributes with ease. Keeping in line with a 
-commitment to high quality, there are also four esoteric bug-fixes and a 
+commitment to high quality, there are also five esoteric bug-fixes and a 
 plethora of subtle improvements that enhance the library. 

library/HTMLPurifier/ChildDef/Custom.php

@@ -38,8 +38,21 @@ class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef
         if ($raw{0} != '(') {
             $raw = "($raw)";
         }
-        $reg = str_replace(',', ',?', $raw);
+        $el = '[#a-zA-Z0-9_.-]+';
-        $reg = preg_replace('/([#a-zA-Z0-9_.-]+)/', '(,?\\0)', $reg);
+        $reg = $raw;
+        
+        // COMPLICATED! AND MIGHT BE BUGGY! I HAVE NO CLUE WHAT I'M
+        // DOING! Seriously: if there's problems, please report them.
+        
+        // setup all elements as parentheticals with leading commas
+        $reg = preg_replace("/$el/", '(,\\0)', $reg);
+        
+        // remove commas when they were not solicited
+        $reg = preg_replace("/([^,(|]\(+),/", '\\1', $reg);
+        
+        // remove all non-paranthetical commas: they are handled by first regex
+        $reg = preg_replace("/,\(/", '(', $reg);
+        
         $this->_pcre_regex = $reg;
     }
     function validateChildren($tokens_of_children, $config, &$context) {
@@ -60,11 +73,11 @@ class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef
                 $list_of_children .= $token->name . ',';
             }
         }
-        $list_of_children = rtrim($list_of_children, ',');
+        // add leading comma to deal with stray comma declarations
-        
+        $list_of_children = ',' . rtrim($list_of_children, ',');
         $okay =
             preg_match(
-                '/^'.$this->_pcre_regex.'$/',
+                '/^,?'.$this->_pcre_regex.'$/',
                 $list_of_children
             );
         

tests/HTMLPurifier/ChildDef/CustomTest.php

@@ -19,6 +19,49 @@ class HTMLPurifier_ChildDef_CustomTest extends HTMLPurifier_ChildDefHarness
         
     }
     
+    function testNesting() {
+        $this->obj = new HTMLPurifier_ChildDef_Custom('(a,b,(c|d))+');
+        $this->assertResult('', false);
+        $this->assertResult('<a /><b /><c /><a /><b /><d />');
+        $this->assertResult('<a /><b /><c /><d />', false);
+    }
+    
+    function testNestedEitherOr() {
+        $this->obj = new HTMLPurifier_ChildDef_Custom('b,(a|(c|d))+');
+        $this->assertResult('', false);
+        $this->assertResult('<b /><a /><c /><d />');
+        $this->assertResult('<b /><d /><a /><a />');
+        $this->assertResult('<b /><a />');
+        $this->assertResult('<acd />', false);
+    }
+    
+    function testNestedQuantifier() {
+        $this->obj = new HTMLPurifier_ChildDef_Custom('(b,c+)*');
+        $this->assertResult('');
+        $this->assertResult('<b /><c />');
+        $this->assertResult('<b /><c /><c /><c />');
+        $this->assertResult('<b /><c /><b /><c />');
+        $this->assertResult('<b /><c /><b />', false);
+    }
+    
+    function testEitherOr() {
+        
+        $this->obj = new HTMLPurifier_ChildDef_Custom('a|b');
+        $this->assertResult('', false);
+        $this->assertResult('<a />');
+        $this->assertResult('<b />');
+        $this->assertResult('<a /><b />', false);
+        
+    }
+    
+    function testCommafication() {
+        
+        $this->obj = new HTMLPurifier_ChildDef_Custom('a,b');
+        $this->assertResult('<a /><b />');
+        $this->assertResult('<ab />', false);
+        
+    }
+    
 }
 
 ?>