Improved handling of ignore_user_agents (OpenMage#3238)

Co-authored-by: Fabrizio Balliano <fabrizio.balliano@gmail.com>
empiricompany · May 13, 2023 · c0d136a · c0d136a
1 parent 0e8319c
commit c0d136a
Show file tree

Hide file tree

Showing 2 changed files with 33 additions and 9 deletions.
diff --git a/app/code/core/Mage/Log/Model/Visitor.php b/app/code/core/Mage/Log/Model/Visitor.php
@@ -101,17 +101,22 @@ public function __construct(array $data = [])
     protected function _construct()
     {
         $this->_init('log/visitor');
-        $userAgent = $this->_httpHelper->getHttpUserAgent();
+        if ($this->_logCondition->isLogDisabled()) {
+            $this->_skipRequestLogging = true;
+            return;
+        }
+
         $ignoreAgents = $this->_config->getNode('global/ignore_user_agents');
         if ($ignoreAgents) {
             $ignoreAgents = $ignoreAgents->asArray();
-            if (in_array($userAgent, $ignoreAgents)) {
-                $this->_skipRequestLogging = true;
+            $userAgent = $this->_httpHelper->getHttpUserAgent();
+            foreach ($ignoreAgents as $ignoreAgent) {
+                if (stripos($userAgent, $ignoreAgent) !== false) {
+                    $this->_skipRequestLogging = true;
+                    break;
+                }
             }
         }
-        if ($this->_logCondition->isLogDisabled()) {
-            $this->_skipRequestLogging = true;
-        }
     }
 
     /**

diff --git a/app/code/core/Mage/Log/etc/config.xml b/app/code/core/Mage/Log/etc/config.xml
@@ -29,9 +29,28 @@
             </entities>
         </ignoredModules>
         <ignore_user_agents>
-            <google1>Googlebot/1.0 (googlebot@googlebot.com http://googlebot.com/)</google1>
-            <google2>Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)</google2>
-            <google3>Googlebot/2.1 (+http://www.googlebot.com/bot.html)</google3>
+            <adsbot>AdsBot</adsbot>
+            <alphabot>AlphaBot</alphabot>
+            <amazonbot>Amazonbot</amazonbot>
+            <bingbot>bingbot</bingbot>
+            <blexbot>BLEXBot</blexbot>
+            <dotbot>DotBot</dotbot>
+            <facebookexternalhit>facebookexternalhit</facebookexternalhit>
+            <google>Googlebot</google>
+            <google-site-verification>Google-Site-Verification</google-site-verification>
+            <hrefsbot>AhrefsBot</hrefsbot>
+            <istellabot>istellabot</istellabot>
+            <mauibot>MauiBot</mauibot>
+            <mj12bot>MJ12bot</mj12bot>
+            <mod_pagespeed>mod_pagespeed</mod_pagespeed>
+            <opensiteexplorer>spbot</opensiteexplorer>
+            <pinterestbot>Pinterestbot</pinterestbot>
+            <semrushbot>SemrushBot</semrushbot>
+            <seotester>SEOTesterBot</seotester>
+            <yandexbot>YandexBot</yandexbot>
+            <yandeximage>YandexImages</yandeximage>
+            <yandexmobilebot>YandexMobileBot</yandexmobilebot>
+            <zoominfobot>ZoominfoBot</zoominfobot>
         </ignore_user_agents>
         <helpers>
             <log>