Adeed decoding of rss HTML descriptions

Author: ssddanbrown

app/Rss/RssParser.php

@@ -37,7 +37,7 @@ public function parseRssDataToPosts(string $rssData): array
             $date = DateTime::createFromFormat('D, d M Y H:i:s T', $item->pubDate ?? '');
             $postData = [
                 'title' => substr(strval($item->title ?? ''), 0, 250),
-                'description' => substr(strval($item->description ?? ''), 0, 1000),
+                'description' => $this->formatDescription(strval($item->description) ?: ''),
                 'url' => strval($item->link ?? ''),
                 'published_at' => $date ? $date->getTimestamp() : 0,
             ];
@@ -52,6 +52,17 @@ public function parseRssDataToPosts(string $rssData): array
         return $posts;
     }
 
+    protected function formatDescription(string $description): string
+    {
+        $decoded = html_entity_decode(strip_tags($description));
+        
+        if (strlen($decoded) > 200) {
+            return substr($decoded, 0, 200) . '...';
+        }
+
+        return $decoded;
+    }
+
     /**
      * @param array{title: string, description: string, url: string, published_at: int} $item
      */

tests/Unit/RssParserTest.php

@@ -133,4 +133,27 @@ public function test_invalid_posts_are_not_returned()
         $this->assertCount(0, $posts);
     }
 
+    public function test_descriptions_in_html_are_parsed()
+    {
+        $parser = new RssParser();
+
+        $posts = $parser->rssDataToPosts(<<<END
+<?xml version="1.0" encoding="utf-8" standalone="yes"?>
+<rss version="2.0">
+  <channel>
+    <item>
+      <title>BookStack Release v22.06</title>
+      <link>https://www.bookstackapp.com/blog/bookstack-release-v22-06/</link>
+      <pubDate>Fri, 24 Jun 2022 11:00:00 +0000</pubDate>
+      <guid>https://www.bookstackapp.com/blog/bookstack-release-v22-06/</guid>
+      <description>&lt;span a=&quot;b&quot;&gt;Some really cool text&lt;/span&gt; &amp;amp; with &amp;pound; entities within</description>
+    </item>
+  </channel>
+</rss>
+END
+        );
+
+        $this->assertEquals('Some really cool text & with £ entities within', $posts[0]->description);
+    }
+
 }