optimize url detection

src/Context/UrlAnalyzer.php

@@ -4,7 +4,9 @@ declare(strict_types=1);
 
 namespace App\Context;
 
-use RuntimeException;
+use fivefilters\Readability\Configuration;
+use fivefilters\Readability\ParseException;
+use fivefilters\Readability\Readability;
 
 /**
  * UrlAnalyzer
@@ -65,30 +67,29 @@ final class UrlAnalyzer
             ],
         ]);
 
-        $handle = @fopen($url, 'rb', false, $context);
-        if ($handle === false) {
+        $html = @file_get_contents($url, false, $context);
+        if ($html === false || $html === '') {
             return '';
         }
 
+        $config = new Configuration();
+        $config->setFixRelativeURLs(true);
+        $config->setOriginalURL($url);
+
+        $readability = new Readability($config);
+
         try {
-            $html = '';
-            while (!feof($handle) && strlen($html) < $this->maxChars * 2) {
-                $html .= fread($handle, 1024);
-            }
-        } finally {
-            fclose($handle);
-        }
-
-        if ($html === '') {
+            $readability->parse($html);
+            $content = $readability->getContent() ?? '';
+        } catch (ParseException) {
             return '';
         }
 
-        // Remove script and style blocks
-        $html = preg_replace('~<script[^>]*>.*?</script>~is', '', $html) ?? $html;
-        $html = preg_replace('~<style[^>]*>.*?</style>~is', '', $html) ?? $html;
+        if ($content === '') {
+            return '';
+        }
 
-        // Strip remaining HTML and normalize whitespace
-        $text = strip_tags($html);
+        $text = strip_tags($content);
         $text = preg_replace('/\s+/u', ' ', $text) ?? $text;
 
         return mb_substr(trim($text), 0, $this->maxChars);