Corrections

* Corrected GoogleBridge (URI extraction was incorrect)
* Corrected ATOM format:
* mime-type was incorrect
* Hyperlinks were not clickable.
* non-UTF8 characters are now properly filtered.
* Corrected HTML format output:
* Hyperlinks were not clickable.
* Corrected error message when SimpleHtmlDom library is not installed.
* Added changelog.

CHANGELOG.md

@@ -0,0 +1,21 @@
+rss-bridge Changelog
+===
+
+Alpha 0.1
+===
+ * Firt tagged version.
+ * Includes refactoring.
+ * Unstable.
+
+Current development version
+===
+ * Corrected GoogleBridge (URI extraction was incorrect)
+ * Corrected ATOM format:
+   * mime-type was incorrect
+   * Hyperlinks were not clickable.
+   * non-UTF8 characters are now properly filtered.
+ * Corrected HTML format output:
+   * Hyperlinks were not clickable.
+ * Corrected error message when SimpleHtmlDom library is not installed.
+ * Added changelog.
+ 

README.md

@@ -1,8 +1,6 @@
 rss-bridge
 ===
 
-Version alpha 0.1
-
 rss-bridge is a collection of independant php scripts capable of generating ATOM feed for specific pages which don't have one.
 
 Supported sites/pages
@@ -10,19 +8,15 @@ Supported sites/pages
 
  * `FlickrExplore` : [Latest interesting images](http://www.flickr.com/explore) from Flickr.
  * `GoogleSearch` : Most recent results from Google Search. Parameters:
-   * q=keyword : Keyword search.
+ * `Twitter` : Twitter. Can return keyword/hashtag search or user timline.
- * `Twitter` : Twitter. Parameters:
-   * q=keyword : Keyword search.
-   * u=username : Get user timeline.
 
-Easy new bridge system (detail below) !
 
 Output format
 ===
-Output format can be used in any rss-bridge:
+Output format can take several forms:
 
- * `Atom` : ATOM Feed.
+ * `Atom` : ATOM Feed, for use in RSS/Feed readers
- * `Json` : Json
+ * `Json` : Json, for consumption by other application.
  * `Html` : html page
  * `Plaintext` : raw text (php object, as returned by print_r)
    
@@ -35,7 +29,7 @@ Requirements
 ===
 
  * php 5.3
- * [PHP Simple HTML DOM Parser](http://simplehtmldom.sourceforge.net). (Put `simple_html_dom.php` in `vendor/simplehtmldom`).
+ * [PHP Simple HTML DOM Parser](http://simplehtmldom.sourceforge.net). (Put `simple_html_dom.php` in `vendor/simplehtmldom/`).
  * Ssl lib activated in PHP config
 
  
@@ -46,7 +40,8 @@ I'm sebsauvage, webmaster of [sebsauvage.net](http://sebsauvage.net), author of
 Thanks to [Mitsukarenai](https://github.com/Mitsukarenai) for the inspiration.
 
 Patch :
-- Yves ASTIER (Draeli) : PHP optimizations, fixes, dynamic brigde/format list with all stuff behind and extend cache system. Mail : contact@yves-astier.com
+
+ * Yves ASTIER ([Draeli](https://github.com/Draeli)) : PHP optimizations, fixes, dynamic brigde/format list with all stuff behind and extend cache system. Mail : contact@yves-astier.com
 
 Licence
 ===
@@ -56,7 +51,7 @@ Code is public domain.
 Technical notes
 ===
   * There is a cache so that source services won't ban you even if you hammer the rss-bridge with requests. Each bridge has a different duration for the cache. The `cache` subdirectory will be automatically created. You can purge it whenever you want.
-  * To implement a new rss-bridge, create a new class in `bridges` directory and extends with `BridgeAbstract`. Look at existing bridges for examples. For items you generate in `$this->items`, only `uri` and `title` are mandatory in each item. `timestamp` and `content` are optional but recommended. Any additional key will be ignored by ATOM feed (but outputed to jSon). If you want your new bridge appear in `index.php`, don't forget add annotation.
+  * To implement a new rss-bridge, create a new class in `bridges` subdirectory. Look at existing bridges for examples. For items you generate in `$this->items`, only `uri` and `title` are mandatory in each item. `timestamp` and `content` are optional but recommended. Any additional key will be ignored by ATOM feed (but outputed to jSon).
 
 Rant
 ===

bridges/GoogleSearchBridge.php

@@ -28,8 +28,13 @@ class GoogleSearchBridge extends BridgeAbstract{
         $emIsRes = $html->find('div[id=ires]',0);
         if( !is_null($emIsRes) ){
             foreach($emIsRes->find('li[class=g]') as $element) {
-                $item = new \Item();
+                $item = new Item();
-                $item->uri = $element->find('a[href]',0)->href;
+                
+                // Extract direct URL from google href (eg. /url?q=...)
+                $t = $element->find('a[href]',0)->href;
+                $item->uri = 'http://google.com'.$t;
+                parse_str(parse_url($t, PHP_URL_QUERY),$parameters);
+                if (isset($parameters['q'])) { $item->uri = $parameters['q']; }
                 $item->title = $element->find('h3',0)->plaintext;
                 $item->content = $element->find('span[class=st]',0)->plaintext;
                 $this->items[] = $item;

formats/AtomFormat.php

@@ -26,7 +26,8 @@ class AtomFormat extends FormatAbstract{
             $entryTitle = is_null($data->title) ? '' : $data->title;
             $entryUri = is_null($data->uri) ? '' : $data->uri;
             $entryTimestamp = is_null($data->timestamp) ? '' : date(DATE_ATOM, $data->timestamp);
-            $entryContent = is_null($data->content) ? '' : '<![CDATA[' . htmlentities($data->content) . ']]>';
+            // We prevent content from closing the CDATA too early.
+            $entryContent = is_null($data->content) ? '' : '<![CDATA[' . $this->sanitizeHtml(str_replace(']]>','',$data->content)) . ']]>';
 
             $entries .= <<<EOD
 
@@ -65,14 +66,22 @@ EOD;
 {$entries}
 </feed>
 EOD;
+        
+        // Remove invalid non-UTF8 characters
 
+        // We cannot use iconv because of a bug in some versions of iconv.
+        // See http://www.php.net/manual/fr/function.iconv.php#108643
+        //$toReturn = iconv("UTF-8", "UTF-8//IGNORE", $toReturn);  
+        // So we use mb_convert_encoding instead:
+        ini_set('mbstring.substitute_character', 'none');
+        $toReturn= mb_convert_encoding($toReturn, 'UTF-8', 'UTF-8'); 
         return $toReturn;
     }
 
     public function display(){
-        // $this
+        $this
-            // ->setContentType('application/atom+xml; charset=' . $this->getCharset())
+            ->setContentType('application/atom+xml; charset=utf8')  // We force UTF-8 in ATOM output.
-            // ->callContentType();
+            ->callContentType();
 
         return parent::display();
     }

formats/HtmlFormat.php

@@ -16,10 +16,9 @@ class HtmlFormat extends FormatAbstract{
         $entries = '';
         foreach($this->getDatas() as $data){
             $entryUri = is_null($data->uri) ? $uri : $data->uri;
-            $entryTitle = is_null($data->title) ? '' : htmlspecialchars(strip_tags($data->title));
+            $entryTitle = is_null($data->title) ? '' : $this->sanitizeHtml(strip_tags($data->title));
             $entryTimestamp = is_null($data->timestamp) ? '' : '<small>' . date(DATE_ATOM, $data->timestamp) . '</small>';
-            $entryContent = is_null($data->content) ? '' : '<p>' . $data->content . '</p>';
+            $entryContent = is_null($data->content) ? '' : '<p>' . $this->sanitizeHtml($data->content). '</p>';
-
             $entries .= <<<EOD
 
         <div class="rssitem">
@@ -52,7 +51,7 @@ EOD;
         return $toReturn;
     }
 
-    public function display(){
+    public function display() {
         $this
             ->setContentType('text/html; charset=' . $this->getCharset())
             ->callContentType();

lib/Format.php

@@ -90,6 +90,23 @@ abstract class FormatAbstract implements FormatInterface{
 
         return $this->extraInfos;
     }
+    
+    /**
+     * Sanitized html while leaving it functionnal.
+     * The aim is to keep html as-is (with clickable hyperlinks)
+     * while reducing annoying and potentially dangerous things.
+     * Yes, I know sanitizing HTML 100% is an impossible task.
+     * Maybe we'll switch to http://htmlpurifier.org/
+     * or http://www.bioinformatics.org/phplabware/internal_utilities/htmLawed/index.php
+     */
+    public function sanitizeHtml($html)
+    {
+        $html = str_replace('<script','<&zwnj;script',$html); // Disable scripts, but leave them visible.
+        $html = str_replace('<iframe','<&zwnj;iframe',$html);
+        $html = str_replace('<link','<&zwnj;link',$html);
+        // We leave alone object and embed so that videos can play in RSS readers.
+        return $html;
+    }
 }
 
 class Format{

lib/RssBridge.php

@@ -15,7 +15,7 @@ require __DIR__ . '/Cache.php';
 
 $vendorLibSimpleHtmlDom = __DIR__ . PATH_VENDOR . '/simplehtmldom/simple_html_dom.php';
 if( !file_exists($vendorLibSimpleHtmlDom) ){
-    throw new \HttpException('"PHP Simple HTML DOM Parser" is missing. Get it from http://simplehtmldom.sourceforge.net and place the script "simple_html_dom.php" in the same folder to allow me to work.', 500);
+    throw new \HttpException('"PHP Simple HTML DOM Parser" library is missing. Get it from http://simplehtmldom.sourceforge.net and place the script "simple_html_dom.php" in '.substr(PATH_VENDOR,4) . '/simplehtmldom/', 500);
 }
 require_once $vendorLibSimpleHtmlDom;