s3copy.rb

ruby_block "copy S3 objects from production to staging" do
  block do
    require 'rubygems'
    require 'right_aws'

    aws_access_key_id = credentials['amazon_access_key_id']
    aws_secret_access_key = credentials['amazon_secret_access_key']
    target_bucket = 'source'
    destination_bucket = 'destination'

    s3 = RightAws::S3Interface.new(aws_access_key_id, aws_secret_access_key)

    copied_keys = Array.new
    s3.incrementally_list_bucket(destination_bucket) do |key_set| 
      copied_keys << key_set[:contents].map{|k| k[:key]}.flatten 
    end
    copied_keys.flatten!

    s3.incrementally_list_bucket(target_bucket) do |key_set|
      key_set[:contents].each do |key|
        key = key[:key]
        if copied_keys.include?(key)
          Chef::Log.info "#{destination_bucket} #{key} already exists. Skipping..."
        else
          Chef::Log.info "Copying #{target_bucket} #{key}"

          retries = 0
          begin
            s3.copy(target_bucket, key, destination_bucket)
          rescue Exception => e
            Chef::Log.error "cannot copy key, #{e.inspect}\nretrying #{retries} out of 10 times..."
            retries += 1
            retry if retries <= 10
          end

          Chef::Log.info "Copying ACL for #{target_bucket} #{key}"

          retries = 0
          begin
            acl = s3.get_acl(target_bucket, key)
            s3.put_acl(destination_bucket, key, acl[:object])
          rescue Exception => e
            Chef::Log.error "cannot copy ACL, #{e.inspect}\nretrying #{retries} out of 10 times..."
            retries += 1
            retry if retries <= 10
          end
        end
      end
    end
  end
end