Added fingerprinting support

Leverage browser caching and proxy caching by setting far future Expires
headers and changing filenames when file contents change. This can make
your web app faster for users and also reduce your bandwidth costs.

By adding the column :avatar_fingerprint to our db table and including
:fingerprint in the attachment filename, we ensure the filename will
change whenever the file contents do.

  has_attached_file :avatar,
    :styles => { :medium => "300x300>", :thumb => "100x100>" },
    :path => "users/:id/:attachment/:fingerprint-:style.:extension",
    :storage => :s3,
    :s3_headers => {'Expires' => 1.year.from_now.httpdate},
    :s3_credentials => "#{RAILS_ROOT}/config/s3.yml",
    :include_updated_timestamp => false

This enables us to set far future expire headers so that browsers
don't need to check for a newer version. If a change does occur,
say because a user uploads a new avatar, the new filename will
be rendered in your html and the cached version will be ignored.

The example above will set Expires headers in S3. If you're using
local storage you can configure your webserver to do something similar.

We disable the timestamped query string because some proxies refuse
to cache items with query strings.

For more info on optimizing for caching:

http://code.google.com/speed/page-speed/docs/caching.html

lib/paperclip.rb

@@ -26,6 +26,7 @@
 # See the +has_attached_file+ documentation for more details.
 
 require 'erb'
+require 'digest'
 require 'tempfile'
 require 'paperclip/version'
 require 'paperclip/upfile'

lib/paperclip/attachment.rb

@@ -15,6 +15,7 @@ module Paperclip
         :default_url       => "/:attachment/:style/missing.png",
         :default_style     => :original,
         :storage           => :filesystem,
+        :include_updated_timestamp => true,
         :whiny             => Paperclip.options[:whiny] || Paperclip.options[:whiny_thumbnails]
       }
     end
@@ -39,6 +40,7 @@ module Paperclip
       @default_url       = options[:default_url]
       @default_style     = options[:default_style]
       @storage           = options[:storage]
+      @include_updated_timestamp = options[:include_updated_timestamp]
       @whiny             = options[:whiny_thumbnails] || options[:whiny]
       @convert_options   = options[:convert_options]
       @processors        = options[:processors]
@@ -90,6 +92,7 @@ module Paperclip
       instance_write(:file_name,       uploaded_file.original_filename.strip)
       instance_write(:content_type,    uploaded_file.content_type.to_s.strip)
       instance_write(:file_size,       uploaded_file.size.to_i)
+      instance_write(:fingerprint,     uploaded_file.fingerprint)
       instance_write(:updated_at,      Time.now)
 
       @dirty = true
@@ -98,6 +101,7 @@ module Paperclip
 
       # Reset the file size if the original file was reprocessed.
       instance_write(:file_size, @queued_for_write[:original].size.to_i)
+      instance_write(:fingerprint, @queued_for_write[:original].fingerprint)
     ensure
       uploaded_file.close if close_uploaded_file
     end
@@ -109,7 +113,7 @@ module Paperclip
     # security, however, for performance reasons.  set
     # include_updated_timestamp to false if you want to stop the attachment
     # update time appended to the url
-    def url style_name = default_style, include_updated_timestamp = true
+    def url(style_name = default_style, include_updated_timestamp = @include_updated_timestamp)
       url = original_filename.nil? ? interpolate(@default_url, style_name) : interpolate(@url, style_name)
       include_updated_timestamp && updated_at ? [url, updated_at].compact.join(url.include?("?") ? "&" : "?") : url
     end
@@ -174,6 +178,12 @@ module Paperclip
       instance_read(:file_size) || (@queued_for_write[:original] && @queued_for_write[:original].size)
     end
 
+    # Returns the hash of the file as originally assigned, and lives in the
+    # <attachment>_fingerprint attribute of the model.
+    def fingerprint
+      instance_read(:fingerprint) || (@queued_for_write[:original] && @queued_for_write[:original].fingerprint)
+    end
+
     # Returns the content_type of the file as originally assigned, and lives
     # in the <attachment>_content_type attribute of the model.
     def content_type

lib/paperclip/interpolations.rb

@@ -88,6 +88,11 @@ module Paperclip
       attachment.instance.id
     end
 
+    # Returns the fingerprint of the instance.
+    def fingerprint attachment, style_name
+      attachment.fingerprint
+    end
+
     # Returns the id of the instance in a split path form. e.g. returns
     # 000/001/234 for an id of 1234.
     def id_partition attachment, style_name

lib/paperclip/upfile.rb

@@ -32,18 +32,26 @@ module Paperclip
     def size
       File.size(self)
     end
+
+    # Returns the hash of the file.
+    def fingerprint
+      Digest::MD5.hexdigest(self.read)
+    end
   end
 end
 
 if defined? StringIO
   class StringIO
-    attr_accessor :original_filename, :content_type
+    attr_accessor :original_filename, :content_type, :fingerprint
     def original_filename
       @original_filename ||= "stringio.txt"
     end
     def content_type
       @content_type ||= "text/plain"
     end
+    def fingerprint
+      @fingerprint ||= Digest::MD5.hexdigest(self.string)
+    end
   end
 end
 

test/attachment_test.rb

@@ -446,6 +446,8 @@ class AttachmentTest < Test::Unit::TestCase
       @not_file = mock
       @tempfile = mock
       @not_file.stubs(:nil?).returns(false)
+      @not_file.stubs(:fingerprint).returns('bd94545193321376b70136f8b223abf8')
+      @tempfile.stubs(:fingerprint).returns('bd94545193321376b70136f8b223abf8')
       @not_file.expects(:size).returns(10)
       @tempfile.expects(:size).returns(10)
       @not_file.expects(:to_tempfile).returns(@tempfile)
@@ -754,5 +756,29 @@ class AttachmentTest < Test::Unit::TestCase
         assert_equal @file.size, @dummy.avatar.size
       end
     end
+
+    context "and avatar_fingerprint column" do
+      setup do
+        ActiveRecord::Base.connection.add_column :dummies, :avatar_fingerprint, :string
+        rebuild_class
+        @dummy = Dummy.new
+      end
+
+      should "not error when assigned an attachment" do
+        assert_nothing_raised { @dummy.avatar = @file }
+      end
+
+      should "return the right value when sent #avatar_fingerprint" do
+        @dummy.avatar = @file
+        assert_equal 'aec488126c3b33c08a10c3fa303acf27', @dummy.avatar_fingerprint
+      end
+
+      should "return the right value when saved, reloaded, and sent #avatar_fingerprint" do
+        @dummy.avatar = @file
+        @dummy.save
+        @dummy = Dummy.find(@dummy.id)
+        assert_equal 'aec488126c3b33c08a10c3fa303acf27', @dummy.avatar_fingerprint
+      end
+    end
   end
 end

test/helper.rb

@@ -84,6 +84,7 @@ def rebuild_model options = {}
     table.column :avatar_content_type, :string
     table.column :avatar_file_size, :integer
     table.column :avatar_updated_at, :datetime
+    table.column :avatar_fingerprint, :string
   end
   rebuild_class options
 end
@@ -103,6 +104,7 @@ class FakeModel
                 :avatar_file_size,
                 :avatar_last_updated,
                 :avatar_content_type,
+                :avatar_fingerprint,
                 :id
 
   def errors