htmlpurifier update - compatibility issue with language library autoloader

7 files changed, 267 insertions, 98 deletions

diff --git a/library/HTMLPurifier/Injector/AutoParagraph.php b/library/HTMLPurifier/Injector/AutoParagraph.php
index afa760892..4afdd128d 100644
--- a/library/HTMLPurifier/Injector/AutoParagraph.php
+++ b/library/HTMLPurifier/Injector/AutoParagraph.php
@@ -8,17 +8,31 @@
  */
 class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
 {
-
+    /**
+     * @type string
+     */
     public $name = 'AutoParagraph';
+
+    /**
+     * @type array
+     */
     public $needed = array('p');
 
-    private function _pStart() {
+    /**
+     * @return HTMLPurifier_Token_Start
+     */
+    private function _pStart()
+    {
         $par = new HTMLPurifier_Token_Start('p');
         $par->armor['MakeWellFormed_TagClosedError'] = true;
         return $par;
     }
 
-    public function handleText(&$token) {
+    /**
+     * @param HTMLPurifier_Token_Text $token
+     */
+    public function handleText(&$token)
+    {
         $text = $token->data;
         // Does the current parent allow <p> tags?
         if ($this->allowsElement('p')) {
@@ -72,11 +86,9 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
                     //                   ----
                 }
             }
-        // Is the current parent a <p> tag?
-        } elseif (
-            !empty($this->currentNesting) &&
-            $this->currentNesting[count($this->currentNesting)-1]->name == 'p'
-        ) {
+            // Is the current parent a <p> tag?
+        } elseif (!empty($this->currentNesting) &&
+            $this->currentNesting[count($this->currentNesting) - 1]->name == 'p') {
             // State 3.1: ...<p>PAR1
             //                  ----
 
@@ -84,7 +96,7 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
             //                  ------------
             $token = array();
             $this->_splitText($text, $token);
-        // Abort!
+            // Abort!
         } else {
             // State 4.1: ...<b>PAR1
             //                  ----
@@ -94,7 +106,11 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
         }
     }
 
-    public function handleElement(&$token) {
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleElement(&$token)
+    {
         // We don't have to check if we're already in a <p> tag for block
         // tokens, because the tag would have been autoclosed by MakeWellFormed.
         if ($this->allowsElement('p')) {
@@ -102,7 +118,6 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
                 if ($this->_isInline($token)) {
                     // State 1: <div>...<b>
                     //                  ---
-
                     // Check if this token is adjacent to the parent token
                     // (seek backwards until token isn't whitespace)
                     $i = null;
@@ -110,31 +125,24 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
 
                     if (!$prev instanceof HTMLPurifier_Token_Start) {
                         // Token wasn't adjacent
-
-                        if (
-                            $prev instanceof HTMLPurifier_Token_Text &&
+                        if ($prev instanceof HTMLPurifier_Token_Text &&
                             substr($prev->data, -2) === "\n\n"
                         ) {
                             // State 1.1.4: <div><p>PAR1</p>\n\n<b>
                             //                                  ---
-
                             // Quite frankly, this should be handled by splitText
                             $token = array($this->_pStart(), $token);
                         } else {
                             // State 1.1.1: <div><p>PAR1</p><b>
                             //                              ---
-
                             // State 1.1.2: <div><br /><b>
                             //                         ---
-
                             // State 1.1.3: <div>PAR<b>
                             //                      ---
                         }
-
                     } else {
                         // State 1.2.1: <div><b>
                         //                   ---
-
                         // Lookahead to see if <p> is needed.
                         if ($this->_pLookAhead()) {
                             // State 1.3.1: <div><b>PAR1\n\nPAR2
@@ -166,24 +174,20 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
 
                 $i = null;
                 if ($this->backward($i, $prev)) {
-                    if (
-                        !$prev instanceof HTMLPurifier_Token_Text
-                    ) {
+                    if (!$prev instanceof HTMLPurifier_Token_Text) {
                         // State 3.1.1: ...</p>{p}<b>
                         //                        ---
-
                         // State 3.2.1: ...</p><div>
                         //                     -----
-
-                        if (!is_array($token)) $token = array($token);
+                        if (!is_array($token)) {
+                            $token = array($token);
+                        }
                         array_unshift($token, new HTMLPurifier_Token_Text("\n\n"));
                     } else {
                         // State 3.1.2: ...</p>\n\n{p}<b>
                         //                            ---
-
                         // State 3.2.2: ...</p>\n\n<div>
                         //                         -----
-
                         // Note: PAR<ELEM> cannot occur because PAR would have been
                         // wrapped in <p> tags.
                     }
@@ -192,7 +196,6 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
         } else {
             // State 2.2: <ul><li>
             //                ----
-
             // State 2.4: <p><b>
             //               ---
         }
@@ -201,18 +204,17 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
     /**
      * Splits up a text in paragraph tokens and appends them
      * to the result stream that will replace the original
-     * @param $data String text data that will be processed
+     * @param string $data String text data that will be processed
      *    into paragraphs
-     * @param $result Reference to array of tokens that the
+     * @param HTMLPurifier_Token[] $result Reference to array of tokens that the
      *    tags will be appended onto
-     * @param $config Instance of HTMLPurifier_Config
-     * @param $context Instance of HTMLPurifier_Context
      */
-    private function _splitText($data, &$result) {
+    private function _splitText($data, &$result)
+    {
         $raw_paragraphs = explode("\n\n", $data);
-        $paragraphs  = array(); // without empty paragraphs
+        $paragraphs = array(); // without empty paragraphs
         $needs_start = false;
-        $needs_end   = false;
+        $needs_end = false;
 
         $c = count($raw_paragraphs);
         if ($c == 1) {
@@ -285,26 +287,33 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
             array_pop($result); // removes \n\n
             array_pop($result); // removes </p>
         }
-
     }
 
     /**
      * Returns true if passed token is inline (and, ergo, allowed in
      * paragraph tags)
+     * @param HTMLPurifier_Token $token
+     * @return bool
      */
-    private function _isInline($token) {
+    private function _isInline($token)
+    {
         return isset($this->htmlDefinition->info['p']->child->elements[$token->name]);
     }
 
     /**
      * Looks ahead in the token list and determines whether or not we need
      * to insert a <p> tag.
+     * @return bool
      */
-    private function _pLookAhead() {
-        $this->current($i, $current);
-        if ($current instanceof HTMLPurifier_Token_Start) $nesting = 1;
-        else $nesting = 0;
+    private function _pLookAhead()
+    {
+        if ($this->currentToken instanceof HTMLPurifier_Token_Start) {
+            $nesting = 1;
+        } else {
+            $nesting = 0;
+        }
         $ok = false;
+        $i = null;
         while ($this->forwardUntilEndToken($i, $current, $nesting)) {
             $result = $this->_checkNeedsP($current);
             if ($result !== null) {
@@ -318,9 +327,12 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
     /**
      * Determines if a particular token requires an earlier inline token
      * to get a paragraph. This should be used with _forwardUntilEndToken
+     * @param HTMLPurifier_Token $current
+     * @return bool
      */
-    private function _checkNeedsP($current) {
-        if ($current instanceof HTMLPurifier_Token_Start){
+    private function _checkNeedsP($current)
+    {
+        if ($current instanceof HTMLPurifier_Token_Start) {
             if (!$this->_isInline($current)) {
                 // <div>PAR1<div>
                 //      ----
@@ -339,7 +351,6 @@ class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector
         }
         return null;
     }
-
 }
 
 // vim: et sw=4 sts=4
diff --git a/library/HTMLPurifier/Injector/DisplayLinkURI.php b/library/HTMLPurifier/Injector/DisplayLinkURI.php
index 9dce9bd08..c19b1bc27 100644
--- a/library/HTMLPurifier/Injector/DisplayLinkURI.php
+++ b/library/HTMLPurifier/Injector/DisplayLinkURI.php
@@ -5,15 +5,29 @@
  */
 class HTMLPurifier_Injector_DisplayLinkURI extends HTMLPurifier_Injector
 {
-
+    /**
+     * @type string
+     */
     public $name = 'DisplayLinkURI';
+
+    /**
+     * @type array
+     */
     public $needed = array('a');
 
-    public function handleElement(&$token) {
+    /**
+     * @param $token
+     */
+    public function handleElement(&$token)
+    {
     }
 
-    public function handleEnd(&$token) {
-        if (isset($token->start->attr['href'])){
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleEnd(&$token)
+    {
+        if (isset($token->start->attr['href'])) {
             $url = $token->start->attr['href'];
             unset($token->start->attr['href']);
             $token = array($token, new HTMLPurifier_Token_Text(" ($url)"));
diff --git a/library/HTMLPurifier/Injector/Linkify.php b/library/HTMLPurifier/Injector/Linkify.php
index 296dac282..069708c25 100644
--- a/library/HTMLPurifier/Injector/Linkify.php
+++ b/library/HTMLPurifier/Injector/Linkify.php
@@ -5,12 +5,24 @@
  */
 class HTMLPurifier_Injector_Linkify extends HTMLPurifier_Injector
 {
-
+    /**
+     * @type string
+     */
     public $name = 'Linkify';
+
+    /**
+     * @type array
+     */
     public $needed = array('a' => array('href'));
 
-    public function handleText(&$token) {
-        if (!$this->allowsElement('a')) return;
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleText(&$token)
+    {
+        if (!$this->allowsElement('a')) {
+            return;
+        }
 
         if (strpos($token->data, '://') === false) {
             // our really quick heuristic failed, abort
@@ -21,7 +33,8 @@ class HTMLPurifier_Injector_Linkify extends HTMLPurifier_Injector
 
         // there is/are URL(s). Let's split the string:
         // Note: this regex is extremely permissive
-        $bits = preg_split('#((?:https?|ftp)://[^\s\'"<>()]+)#S', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE);
+        $bits = preg_split('#((?:https?|ftp)://[^\s\'",<>()]+)#Su', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE);
+
 
         $token = array();
 
@@ -30,7 +43,9 @@ class HTMLPurifier_Injector_Linkify extends HTMLPurifier_Injector
         // $l = is link
         for ($i = 0, $c = count($bits), $l = false; $i < $c; $i++, $l = !$l) {
             if (!$l) {
-                if ($bits[$i] === '') continue;
+                if ($bits[$i] === '') {
+                    continue;
+                }
                 $token[] = new HTMLPurifier_Token_Text($bits[$i]);
             } else {
                 $token[] = new HTMLPurifier_Token_Start('a', array('href' => $bits[$i]));
@@ -38,9 +53,7 @@ class HTMLPurifier_Injector_Linkify extends HTMLPurifier_Injector
                 $token[] = new HTMLPurifier_Token_End('a');
             }
         }
-
     }
-
 }
 
 // vim: et sw=4 sts=4
diff --git a/library/HTMLPurifier/Injector/PurifierLinkify.php b/library/HTMLPurifier/Injector/PurifierLinkify.php
index ad2455a91..cb9046f33 100644
--- a/library/HTMLPurifier/Injector/PurifierLinkify.php
+++ b/library/HTMLPurifier/Injector/PurifierLinkify.php
@@ -6,19 +6,43 @@
  */
 class HTMLPurifier_Injector_PurifierLinkify extends HTMLPurifier_Injector
 {
-
+    /**
+     * @type string
+     */
     public $name = 'PurifierLinkify';
+
+    /**
+     * @type string
+     */
     public $docURL;
+
+    /**
+     * @type array
+     */
     public $needed = array('a' => array('href'));
 
-    public function prepare($config, $context) {
+    /**
+     * @param HTMLPurifier_Config $config
+     * @param HTMLPurifier_Context $context
+     * @return string
+     */
+    public function prepare($config, $context)
+    {
         $this->docURL = $config->get('AutoFormat.PurifierLinkify.DocURL');
         return parent::prepare($config, $context);
     }
 
-    public function handleText(&$token) {
-        if (!$this->allowsElement('a')) return;
-        if (strpos($token->data, '%') === false) return;
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleText(&$token)
+    {
+        if (!$this->allowsElement('a')) {
+            return;
+        }
+        if (strpos($token->data, '%') === false) {
+            return;
+        }
 
         $bits = preg_split('#%([a-z0-9]+\.[a-z0-9]+)#Si', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE);
         $token = array();
@@ -28,18 +52,20 @@ class HTMLPurifier_Injector_PurifierLinkify extends HTMLPurifier_Injector
         // $l = is link
         for ($i = 0, $c = count($bits), $l = false; $i < $c; $i++, $l = !$l) {
             if (!$l) {
-                if ($bits[$i] === '') continue;
+                if ($bits[$i] === '') {
+                    continue;
+                }
                 $token[] = new HTMLPurifier_Token_Text($bits[$i]);
             } else {
-                $token[] = new HTMLPurifier_Token_Start('a',
-                    array('href' => str_replace('%s', $bits[$i], $this->docURL)));
+                $token[] = new HTMLPurifier_Token_Start(
+                    'a',
+                    array('href' => str_replace('%s', $bits[$i], $this->docURL))
+                );
                 $token[] = new HTMLPurifier_Token_Text('%' . $bits[$i]);
                 $token[] = new HTMLPurifier_Token_End('a');
             }
         }
-
     }
-
 }
 
 // vim: et sw=4 sts=4
diff --git a/library/HTMLPurifier/Injector/RemoveEmpty.php b/library/HTMLPurifier/Injector/RemoveEmpty.php
index 638bfca03..cd885722e 100644
--- a/library/HTMLPurifier/Injector/RemoveEmpty.php
+++ b/library/HTMLPurifier/Injector/RemoveEmpty.php
@@ -2,10 +2,44 @@
 
 class HTMLPurifier_Injector_RemoveEmpty extends HTMLPurifier_Injector
 {
+    /**
+     * @type HTMLPurifier_Context
+     */
+    private $context;
 
-    private $context, $config, $attrValidator, $removeNbsp, $removeNbspExceptions;
+    /**
+     * @type HTMLPurifier_Config
+     */
+    private $config;
 
-    public function prepare($config, $context) {
+    /**
+     * @type HTMLPurifier_AttrValidator
+     */
+    private $attrValidator;
+
+    /**
+     * @type bool
+     */
+    private $removeNbsp;
+
+    /**
+     * @type bool
+     */
+    private $removeNbspExceptions;
+
+    /**
+     * @type array
+     * TODO: make me configurable
+     */
+    private $_exclude = array('colgroup' => 1, 'th' => 1, 'td' => 1, 'iframe' => 1);
+
+    /**
+     * @param HTMLPurifier_Config $config
+     * @param HTMLPurifier_Context $context
+     * @return void
+     */
+    public function prepare($config, $context)
+    {
         parent::prepare($config, $context);
         $this->config = $config;
         $this->context = $context;
@@ -14,38 +48,54 @@ class HTMLPurifier_Injector_RemoveEmpty extends HTMLPurifier_Injector
         $this->attrValidator = new HTMLPurifier_AttrValidator();
     }
 
-    public function handleElement(&$token) {
-        if (!$token instanceof HTMLPurifier_Token_Start) return;
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleElement(&$token)
+    {
+        if (!$token instanceof HTMLPurifier_Token_Start) {
+            return;
+        }
         $next = false;
-        for ($i = $this->inputIndex + 1, $c = count($this->inputTokens); $i < $c; $i++) {
-            $next = $this->inputTokens[$i];
+        $deleted = 1; // the current tag
+        for ($i = count($this->inputZipper->back) - 1; $i >= 0; $i--, $deleted++) {
+            $next = $this->inputZipper->back[$i];
             if ($next instanceof HTMLPurifier_Token_Text) {
-                if ($next->is_whitespace) continue;
+                if ($next->is_whitespace) {
+                    continue;
+                }
                 if ($this->removeNbsp && !isset($this->removeNbspExceptions[$token->name])) {
                     $plain = str_replace("\xC2\xA0", "", $next->data);
                     $isWsOrNbsp = $plain === '' || ctype_space($plain);
-                    if ($isWsOrNbsp) continue;
+                    if ($isWsOrNbsp) {
+                        continue;
+                    }
                 }
             }
             break;
         }
         if (!$next || ($next instanceof HTMLPurifier_Token_End && $next->name == $token->name)) {
-            if ($token->name == 'colgroup') return;
+            if (isset($this->_exclude[$token->name])) {
+                return;
+            }
             $this->attrValidator->validateToken($token, $this->config, $this->context);
             $token->armor['ValidateAttributes'] = true;
-            if (isset($token->attr['id']) || isset($token->attr['name'])) return;
-            $token = $i - $this->inputIndex + 1;
-            for ($b = $this->inputIndex - 1; $b > 0; $b--) {
-                $prev = $this->inputTokens[$b];
-                if ($prev instanceof HTMLPurifier_Token_Text && $prev->is_whitespace) continue;
+            if (isset($token->attr['id']) || isset($token->attr['name'])) {
+                return;
+            }
+            $token = $deleted + 1;
+            for ($b = 0, $c = count($this->inputZipper->front); $b < $c; $b++) {
+                $prev = $this->inputZipper->front[$b];
+                if ($prev instanceof HTMLPurifier_Token_Text && $prev->is_whitespace) {
+                    continue;
+                }
                 break;
             }
             // This is safe because we removed the token that triggered this.
-            $this->rewind($b - 1);
+            $this->rewindOffset($b+$deleted);
             return;
         }
     }
-
 }
 
 // vim: et sw=4 sts=4
diff --git a/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php b/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php
index b21313470..9ee7aa84d 100644
--- a/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php
+++ b/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php
@@ -5,25 +5,45 @@
  */
 class HTMLPurifier_Injector_RemoveSpansWithoutAttributes extends HTMLPurifier_Injector
 {
+    /**
+     * @type string
+     */
     public $name = 'RemoveSpansWithoutAttributes';
+
+    /**
+     * @type array
+     */
     public $needed = array('span');
 
+    /**
+     * @type HTMLPurifier_AttrValidator
+     */
     private $attrValidator;
 
     /**
-     * Used by AttrValidator
+     * Used by AttrValidator.
+     * @type HTMLPurifier_Config
      */
     private $config;
+
+    /**
+     * @type HTMLPurifier_Context
+     */
     private $context;
 
-    public function prepare($config, $context) {
+    public function prepare($config, $context)
+    {
         $this->attrValidator = new HTMLPurifier_AttrValidator();
         $this->config = $config;
         $this->context = $context;
         return parent::prepare($config, $context);
     }
 
-    public function handleElement(&$token) {
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleElement(&$token)
+    {
         if ($token->name !== 'span' || !$token instanceof HTMLPurifier_Token_Start) {
             return;
         }
@@ -39,8 +59,8 @@ class HTMLPurifier_Injector_RemoveSpansWithoutAttributes extends HTMLPurifier_In
         }
 
         $nesting = 0;
-        $spanContentTokens = array();
-        while ($this->forwardUntilEndToken($i, $current, $nesting)) {}
+        while ($this->forwardUntilEndToken($i, $current, $nesting)) {
+        }
 
         if ($current instanceof HTMLPurifier_Token_End && $current->name === 'span') {
             // Mark closing span tag for deletion
@@ -50,7 +70,11 @@ class HTMLPurifier_Injector_RemoveSpansWithoutAttributes extends HTMLPurifier_In
         }
     }
 
-    public function handleEnd(&$token) {
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleEnd(&$token)
+    {
         if ($token->markForDeletion) {
             $token = false;
         }
diff --git a/library/HTMLPurifier/Injector/SafeObject.php b/library/HTMLPurifier/Injector/SafeObject.php
index 9e178ce01..3d17e07af 100644
--- a/library/HTMLPurifier/Injector/SafeObject.php
+++ b/library/HTMLPurifier/Injector/SafeObject.php
@@ -6,29 +6,61 @@
  */
 class HTMLPurifier_Injector_SafeObject extends HTMLPurifier_Injector
 {
+    /**
+     * @type string
+     */
     public $name = 'SafeObject';
+
+    /**
+     * @type array
+     */
     public $needed = array('object', 'param');
 
+    /**
+     * @type array
+     */
     protected $objectStack = array();
-    protected $paramStack  = array();
 
-    // Keep this synchronized with AttrTransform/SafeParam.php
+    /**
+     * @type array
+     */
+    protected $paramStack = array();
+
+    /**
+     * Keep this synchronized with AttrTransform/SafeParam.php.
+     * @type array
+     */
     protected $addParam = array(
         'allowScriptAccess' => 'never',
         'allowNetworking' => 'internal',
     );
+
+    /**
+     * @type array
+     */
     protected $allowedParam = array(
         'wmode' => true,
         'movie' => true,
         'flashvars' => true,
         'src' => true,
+        'allowFullScreen' => true, // if omitted, assume to be 'false'
     );
 
-    public function prepare($config, $context) {
+    /**
+     * @param HTMLPurifier_Config $config
+     * @param HTMLPurifier_Context $context
+     * @return void
+     */
+    public function prepare($config, $context)
+    {
         parent::prepare($config, $context);
     }
 
-    public function handleElement(&$token) {
+    /**
+     * @param HTMLPurifier_Token $token
+     */
+    public function handleElement(&$token)
+    {
         if ($token->name == 'object') {
             $this->objectStack[] = $token;
             $this->paramStack[] = array();
@@ -50,16 +82,15 @@ class HTMLPurifier_Injector_SafeObject extends HTMLPurifier_Injector
                 // attribute, which we need if a type is specified. This is
                 // *very* Flash specific.
                 if (!isset($this->objectStack[$i]->attr['data']) &&
-                    ($token->attr['name'] == 'movie' || $token->attr['name'] == 'src')) {
+                    ($token->attr['name'] == 'movie' || $token->attr['name'] == 'src')
+                ) {
                     $this->objectStack[$i]->attr['data'] = $token->attr['value'];
                 }
                 // Check if the parameter is the correct value but has not
                 // already been added
-                if (
-                    !isset($this->paramStack[$i][$n]) &&
+                if (!isset($this->paramStack[$i][$n]) &&
                     isset($this->addParam[$n]) &&
-                    $token->attr['name'] === $this->addParam[$n]
-                ) {
+                    $token->attr['name'] === $this->addParam[$n]) {
                     // keep token, and add to param stack
                     $this->paramStack[$i][$n] = true;
                 } elseif (isset($this->allowedParam[$n])) {
@@ -75,7 +106,8 @@ class HTMLPurifier_Injector_SafeObject extends HTMLPurifier_Injector
         }
     }
 
-    public function handleEnd(&$token) {
+    public function handleEnd(&$token)
+    {
         // This is the WRONG way of handling the object and param stacks;
         // we should be inserting them directly on the relevant object tokens
         // so that the global stack handling handles it.
@@ -84,7 +116,6 @@ class HTMLPurifier_Injector_SafeObject extends HTMLPurifier_Injector
             array_pop($this->paramStack);
         }
     }
-
 }
 
 // vim: et sw=4 sts=4