[myexperiment-hackers] [1887] trunk: Viewing and download counts for all contributables (packs, files, workflows) are now only updated when a validation of user-agent against a list of patterns that some major bots follow is done .

[noreply]

Revision

1887

Author

alekses6

Date

2008-10-27 09:50:15 -0400 (Mon, 27 Oct 2008)

Log Message

Viewing and download counts for all contributables (packs, files, workflows) are now only updated when a validation of user-agent against a list of patterns that some major bots follow is done.

The list of patterns can easily be managed from environment_private.rb.

Modified Paths

• trunk/app/controllers/application.rb
• trunk/app/controllers/blobs_controller.rb
• trunk/app/controllers/packs_controller.rb
• trunk/app/controllers/workflows_controller.rb
• trunk/config/environment_private.rb.pre

Diff

Modified: trunk/app/controllers/application.rb (1886 => 1887)

--- trunk/app/controllers/application.rb	2008-10-24 17:18:34 UTC (rev 1886)
+++ trunk/app/controllers/application.rb	2008-10-27 13:50:15 UTC (rev 1887)
@@ -22,6 +22,22 @@
     request.host_with_port
   end
   
+  
+  def allow_statistics_logging
+    # check if the current viewing/download is to be logged
+    # (i.e. request is sent not by a bot and is legitimate)
+    allow_logging = true
+    BOT_IGNORE_LIST.each do |pattern|
+      if request.env['HTTP_USER_AGENT'].match(pattern)
+        allow_logging = false
+        break
+      end
+    end
+    
+    return allow_logging
+  end
+  
+  
   def can_manage_pages?
     return admin?  # from authenticated_system
   end

Modified: trunk/app/controllers/blobs_controller.rb (1886 => 1887)

--- trunk/app/controllers/blobs_controller.rb	2008-10-24 17:18:34 UTC (rev 1886)
+++ trunk/app/controllers/blobs_controller.rb	2008-10-27 13:50:15 UTC (rev 1887)
@@ -37,7 +37,9 @@
   
   # GET /files/1;download
   def download
-    @download = Download.create(:contribution => @blob.contribution, :user => (logged_in? ? current_user : nil))
+    if allow_statistics_logging
+      @download = Download.create(:contribution => @blob.contribution, :user => (logged_in? ? current_user : nil))
+    end
     
     send_data(@blob.content_blob.data, :filename => @blob.local_name, :type => @blob.content_type)
     
@@ -71,7 +73,9 @@
   
   # GET /files/1
   def show
-    @viewing = Viewing.create(:contribution => @blob.contribution, :user => (logged_in? ? current_user : nil))
+    if allow_statistics_logging
+      @viewing = Viewing.create(:contribution => @blob.contribution, :user => (logged_in? ? current_user : nil))
+    end
     
     respond_to do |format|
       format.html # show.rhtml

Modified: trunk/app/controllers/packs_controller.rb (1886 => 1887)

--- trunk/app/controllers/packs_controller.rb	2008-10-24 17:18:34 UTC (rev 1886)
+++ trunk/app/controllers/packs_controller.rb	2008-10-27 13:50:15 UTC (rev 1887)
@@ -48,7 +48,9 @@
   
   # GET /packs/1
   def show
-    @viewing = Viewing.create(:contribution => @pack.contribution, :user => (logged_in? ? current_user : nil))
+    if allow_statistics_logging
+      @viewing = Viewing.create(:contribution => @pack.contribution, :user => (logged_in? ? current_user : nil))
+    end
     
     respond_to do |format|
       format.html # show.rhtml
@@ -73,6 +75,10 @@
     
     @pack.create_zip(current_user, image_hash)
     
+    if allow_statistics_logging
+      @download = Download.create(:contribution => @pack.contribution, :user => (logged_in? ? current_user : nil))
+    end
+    
     send_file @pack.archive_file_path, :disposition => 'attachment'
   end
   

Modified: trunk/app/controllers/workflows_controller.rb (1886 => 1887)

--- trunk/app/controllers/workflows_controller.rb	2008-10-24 17:18:34 UTC (rev 1886)
+++ trunk/app/controllers/workflows_controller.rb	2008-10-27 13:50:15 UTC (rev 1887)
@@ -157,7 +157,9 @@
   
   # GET /workflows/1;download
   def download
-    @download = Download.create(:contribution => @workflow.contribution, :user => (logged_in? ? current_user : nil))
+    if allow_statistics_logging
+      @download = Download.create(:contribution => @workflow.contribution, :user => (logged_in? ? current_user : nil))
+    end
     
     send_data(@viewing_version.content_blob.data, :filename => @workflow.filename(@viewing_version_number), :type => @workflow.content_type)
   end
@@ -225,7 +227,9 @@
 
   # GET /workflows/1
   def show
-    @viewing = Viewing.create(:contribution => @workflow.contribution, :user => (logged_in? ? current_user : nil))
+    if allow_statistics_logging
+      @viewing = Viewing.create(:contribution => @workflow.contribution, :user => (logged_in? ? current_user : nil))
+    end
 
     respond_to do |format|
       format.html # show.rhtml

Modified: trunk/config/environment_private.rb.pre (1886 => 1887)

--- trunk/config/environment_private.rb.pre	2008-10-24 17:18:34 UTC (rev 1886)
+++ trunk/config/environment_private.rb.pre	2008-10-27 13:50:15 UTC (rev 1887)
@@ -71,4 +71,14 @@
 INVITATION_EMAIL_LIMIT = 10 
 
 # The maximum file size allowed for workflows
-WORKFLOW_UPLOAD_MAX_BYTES = 20971520
\ No newline at end of file
+WORKFLOW_UPLOAD_MAX_BYTES = 20971520
+
+
+# =========== Settings for Download and Viewing Logging ===========
+
+# Patterns in this list will be used to be checked against "user-agent" in http request headers
+# to filter out bots from usage statistics.
+#
+# It is essential to put patterns into the ignore list in a single quotes - this will enable the
+# patterns to be treated as regular expressions, not just strings
+BOT_IGNORE_LIST = ['Googlebot', 'Slurp', 'msnbot', 'crawler', 'bot', 'heritrix']
\ No newline at end of file