Re: HTMLBBCodeParser patch

Wohaa, dude, you're almost putting in more work into the class than I ever did! :) Thanks alot! And yes, it's true, I've been busy working and studying for exams lately, and I don't read pear dev/general daily anymore (more or less twice a week nowadays *shame*). But, on the bright side, I'm still willing to put some work into the class, since I notice people are really using it (who'd ever thought!). So, Stan, we'll talk soon (aol: StijnDeReede, msn: stijndereede@hotmail.com, icq: 3386484, email: sjr@gmx.co.uk) on extending the class, making documentation and new features (strip_tags and nl2br are great, why didn't I think of such simple things!). Greets, Stijn de Reede Stan Lemon wrote:

Hey everyone, I've attached a patch for HTML_BBCodeParser. I've contacted Stij (the author) in regards to the patch, however, he seems to be rather busy so I figured I'd just send the code out. If the higher ups give a go I'd be more then happy to commit this code. None of the changes code in the patch is a major change, everything is very minor. Here's a brief summary of what I did: - In the current release you are not allowed to use numbers in tags, and this is both undocumented, and I believe unintended. Essentially if I created a tag called [h1] I would receive bad formatting problems because the current release does not parse the tag correctly at all. The output is rather ugly. I fixed the portion of the code which didn't account for numbers being included. Now it parses BBCode tags with numbers with no problems. - Currently the tag [b] and [B] are different, even though the average developer will want both to output <strong> Now I realized this may be by design so what I did was first created an option called "strict" and if toggled true it [b] and [B] are different, if toggled false [b] and [B} are the same. - I added two very small features, one called strip_tags and the other called nl2br (very original, I know) these features allow for the getText() method to return text which has had <br /> added, and/or remaining HTML tags stripped. This is beneficial for various security reasons and if you don't want output to run together the nl2br item is nice. (Note: I need to alter the methods in the list filter to clean up <br /> tags that have been added. This is not essential, but what it is, is a friendly "cleaner" utility. I had code started, just not finished.) They are not ridiculously major changes, but they are definitely beneficial. I will in the upcoming days also be releasing a patch to update and elaborate on the current phpdoc that exists in HTML_BBCodeParser. I would also like to receive some type of a go to develop a series of loadTag(), loadTags(), removeTag(), and removeTags() methods. Currently you have to develop a filter in order to add tags. This is nice for large sets of tags for a predefined filter, however, at the same time if I want to load a single tag I have to create a whole new filter in order to do so. There is a much easier way of handling this, and there is also an easy way to remove tags from the current set tags (e.g. Say I want to load the list filter but only allow for unordered lists instead of ordered list, I'd have to redo the entire class, rather then just running a removeTags("list"); type function. A patch for the above methods will be out shortly before my documentation patch. Anyhow, please take a look at my patch and try it out. Any input would be appreciated. Thanks. - Stan Lemon ------------------------------------------------------------------------ Index: HTML_BBCodeParser/BBCodeParser.php =================================================================== RCS file: /repository/pear/HTML_BBCodeParser/BBCodeParser.php,v retrieving revision 1.3 diff -u -r1.3 BBCodeParser.php --- HTML_BBCodeParser/BBCodeParser.php 24 Jul 2003 21:04:28 -0000 1.3 +++ HTML_BBCodeParser/BBCodeParser.php 6 Aug 2003 22:43:56 -0000 @@ -16,7 +16,7 @@

 // | Author: Stijn de Reede <sjr@gmx.co.uk>                               |

// +----------------------------------------------------------------------+ // -// $Id: BBCodeParser.php,v 1.3 2003/07/24 21:04:28 sjr Exp $ +// $Id$ // /** @@ -141,7 +141,10 @@

                             'open'          => '[',
                             'close'         => ']',
                             'xmlclose'      => true,
-                            'filters'       => 'Basic'
+                            'filters'       => 'Basic',
+                            'nl2br'			=> true,
+                            'strip_tags'	=> false,
+                            'strict'		=> false,
                          );
      /**

@@ -220,6 +223,16 @@

             $this->_filters[$filter] = new $class;
             $this->_definedTags = array_merge($this->_definedTags, $this->_filters[$filter]->_definedTags);
         }

+ + /* make sure all of the keys in definedTags are lowered for comparing */ + $lowerTags = array(); +

+        if ($this->_options['strict'] == false) {
+		    foreach ($this->_definedTags as $key => $value) {
+		        $lowerTags[strtolower($key)] = $value;
+		    }
+		    $this->_definedTags = $lowerTags;

+ }

     }
  @@ -313,7 +326,9 @@
                 } else {
                     /* possible valid tag */
                     $newPos = $closePos + 1;

+

                     $newTag = $this->_buildTag(substr($str, $strPos, $closePos - $strPos + 1));

+

                     if ( ($newTag !== false) ) {
                         $tag = $newTag;
                     } else {

@@ -362,9 +377,14 @@

     {
         $tag = array('text' => $str, 'attributes' => array());
 -        if (substr($str, 1, 1) == '/') {        /* closing tag */
+        /* Lower the current tag for comparison against the loaded tags */
+        if ($this->_options['strict'] == false) {
+            $str = strtolower($str);
+        }
 +        if (substr($str, 1, 1) == '/') {        /* closing tag */
             $tag['tag'] = substr($str, 2, strlen($str) - 3);

+

             if ( (in_array($tag['tag'], array_keys($this->_definedTags)) == false) ) {
                 return false;                   /* nope, it's not valid */
             } else {

@@ -372,7 +392,6 @@

                 return $tag;
             }
         } else {                                /* opening tag */

-

             $tag['type'] = 1;
             if ( (strpos($str, ' ') == true) && (strpos($str, '=') == false) ) {
                 return false;                   /* nope, it's not valid */

@@ -382,9 +401,12 @@

                split the tag with arguments and all */
             $oe = $this->_options['open_esc'];
             $ce = $this->_options['close_esc'];
-            preg_match("!$oe([a-z]+)[^$ce]*$ce!i", $str, $tagArray);

+

+            /* Retrieve the tag name */
+            preg_match("!$oe([a-z0-9]+)[^$ce]*$ce!i", $str, $tagArray);
             $tag['tag'] = $tagArray[1];
-            if ( (in_array($tag['tag'], array_keys($this->_definedTags)) == false) ) {

+

+            if ((in_array($tag['tag'], array_keys($this->_definedTags)) == false) ) {
                 return false;                   /* nope, it's not valid */
             }
 @@ -583,6 +605,15 @@
     */
     function setText($str)
     {
+    	/* If the user wants all html cleaned out, do so */
+        if ($this->_options['strip_tags'] == true) {
+            $str = strip_tags($str);
+        }
+    	/* If the user wants \n replaced with <br />, do so */
+        if ($this->_options['nl2br'] != true) {
+            $str = nl2br($str);
+        }

+

         $this->_text = $str;
     }