Adds detection for BBC bots

matomo-org · Dec 25, 2023 · a6fc878 · a6fc878
1 parent e4b14f6
commit a6fc878
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 0 deletions.
diff --git a/Tests/fixtures/bots.yml b/Tests/fixtures/bots.yml
@@ -5899,3 +5899,21 @@
     producer:
       name: Open Search Foundation e.V.
       url: https://openwebsearch.eu/
+-
+  user_agent: Page Monitor (https://confluence.dev.bbc.co.uk/display/men/Page+Monitor)
+  bot:
+    name: BBC Page Monitor
+    category: Site Monitor
+    url: https://confluence.dev.bbc.co.uk/display/men/Page+Monitor
+    producer:
+      name: BBC
+      url: https://www.bbc.com/
+-
+  user_agent: BBC-Forge-URL-Monitor-Twisted
+  bot:
+    name: BBC Forge URL Monitor
+    category: Site Monitor
+    url: https://www.bbc.com/
+    producer:
+      name: BBC
+      url: https://www.bbc.com/
diff --git a/regexes/bots.yml b/regexes/bots.yml
@@ -3516,6 +3516,22 @@
     name: 'Open Search Foundation e.V.'
     url: 'https://openwebsearch.eu/'
 
+- regex: 'bbc.co.uk/display/men/Page\+Monitor'
+  name: 'BBC Page Monitor'
+  category: 'Site Monitor'
+  url: 'https://confluence.dev.bbc.co.uk/display/men/Page+Monitor'
+  producer:
+    name: 'BBC'
+    url: 'https://www.bbc.com/'
+
+- regex: 'BBC-Forge-URL-Monitor-Twisted'
+  name: 'BBC Forge URL Monitor'
+  category: 'Site Monitor'
+  url: 'https://www.bbc.com/'
+  producer:
+    name: 'BBC'
+    url: 'https://www.bbc.com/'
+
 # Generic detections
 - regex: '[a-z0-9\-_]*((?<!cu|power[ _]|m[ _])bot(?![ _]TAB|[ _]?5[0-9]|[ _]Senior|[ _]Junior)|crawler|crawl|checker|archiver|transcoder|spider|^firefox$|^chrome$)([^a-z]|$)'
   name: 'Generic Bot'