Inspired by https://gist.github.com/bantic/4080793. A ruby class to copy files from one AWS S3 bucket to another one supporting aws-sdk v.2

s3_sync_service.rb

require 'aws-sdk'

class S3SyncService

  attr_reader :from_bucket, :to_bucket, :logger
  attr_accessor :debug

  # from_credentials and to_credentials are both hashes with these keys:
  #  * :access_key_id
  #  * :secret_access_key
  #  * :bucket
  #  * :region
  def initialize(from_credentials, to_credentials)
    @from_bucket = init_bucket(from_credentials.symbolize_keys)
    @to_bucket = init_bucket(to_credentials.symbolize_keys)
    @object_counts = { sync: 0, skip: 0 }
  end

  def perform(output = STDOUT)
    create_logger(output)

    @logger.info 'Synchronization started'

    total = @from_bucket.objects.count
    progress = 0

    @from_bucket.objects.each do |object|
      if object_requires_sync?(object)
        sync object
      else
        skip object
      end

      progress += 1
      yield total, progress
    end

    @logger.info 'Synchronization finished.'
    @logger.info "Synced #{@object_counts[:sync]}, skipped #{@object_counts[:skip]}."
  end

  private

  def init_bucket(creds)
    credentials = Aws::Credentials.new(creds[:access_key_id], creds[:secret_access_key])
    s3 = Aws::S3::Resource.new(region: creds[:region], credentials: credentials)
    bucket = s3.bucket(creds[:bucket])
    bucket = bucket.create_bucket(creds[:bucket]) unless bucket.exists?
    bucket
  end

  def sync(object)
    logger.debug "Syncing #{stats object}"
    object.copy_to(bucket: @to_bucket.name, key: object.key)
    @object_counts[:sync] += 1
  end

  def skip(object)
    @logger.debug "Skipped #{stats object}"
    @object_counts[:skip] += 1
  end

  def object_requires_sync?(object)
    to_object = @to_bucket.object(object.key)
    !to_object.exists? || to_object.etag != object.etag
  rescue => e
    @logger.debug "#{e.message || e.inspect}"
    true
  end

  def create_logger(output)
    @logger = Logger.new(output).tap do |l|
      l.level = debug ? Logger::DEBUG : Logger::INFO
    end
  end

  def stats(object)
    content_length_in_kb = object.content_length / 1024
    "#{object.key} #{content_length_in_kb}k " +
      "#{object.last_modified.strftime("%b %d %Y %H:%M")}"
  end

end

=begin
Example usage:
 from_creds = {access_key_id:"AAA", secret_access_key:"BBB", bucket:"first-bucket", region: "us-east-1"}
 from_creds = {access_key_id:"CCC", secret_access_key:"DDD", bucket:"second-bucket", region: "us-east-1"}

 syncer = S3SyncService.new(from_creds, to_creds)
 syncer.debug = true # log each object
 syncer.perform do |total_number_of_objects, progress|
  # process synchronization progress
 end
=end