Rails How to use aws-sdk for native s3 communication

With out using paperclip

Source: https://github.com/amazonwebservices/aws-sdk-for-ruby/blob/master/lib/aws/s3.rb

  # Provides an expressive, object-oriented interface to Amazon S3.

  #

  # To use Amazon S3 you must first 

  # {sign up here}[http://aws.amazon.com/s3/].

  #

  # For more information about Amazon S3, see:

  #

  # * {Amazon S3}[http://aws.amazon.com/s3/]

  # * {Amazon S3 Documentation}[http://aws.amazon.com/documentation/s3/]

  #

  # = Credentials

  #

  # You can setup default credentials for all AWS services via 

  # AWS.config:

  #

  #   AWS.config(

  #     :access_key_id => 'YOUR_ACCESS_KEY_ID',

  #     :secret_access_key => 'YOUR_SECRET_ACCESS_KEY')

  #

  # Or you can set them directly on the S3 interface:

  #

  #   s3 = AWS::S3.new(

  #     :access_key_id => 'YOUR_ACCESS_KEY_ID',

  #     :secret_access_key => 'YOUR_SECRET_ACCESS_KEY')

  #

  # = Buckets

  #

  # Before you can upload files to S3, you need to create a bucket.

  #

  #   s3 = AWS::S3.new

  #   bucket = s3.buckets.create('my-bucket')

  #

  # If a bucket already exists, you can get a reference to the bucket.

  #

  #   bucket = s3.buckets['my-bucket'] # no request made

  #

  # You can also enumerate all buckets in your account.

  #

  #   s3.buckets.each do |bucket|

  #     puts bucket.name

  #   end

  #

  # See {BucketCollection} and {Bucket} for more information on working 

  # with buckets.

  #

  # = Objects

  #

  # Buckets contain objects.  Each object in a bucket has a unique key.

  #

  # == Getting an Object

  #

  # If the object already exists, you can get a reference to the object.

  #

  #   # makes no request, returns an AWS::S3::S3Object

  #   obj = bucket.objects['key']

  #

  # == Reading and Writing an Object

  #

  # The example above returns an {S3Object}.  You call {S3Object#write} and 

  # {S3Object#read} to upload to and download from S3 respectively.

  #

  #   # streaming upload a file to S3

  #   obj.write(Pathname.new('/path/to/file.txt'))

  #

  #   # streaming download from S3 to a file on disk

  #   File.open('file.txt', 'w') do |file|

  #     obj.read do |chunk|

  #        file.write(chunk)

  #     end

  #   end

  #

  # == Enumerating Objects

  #

  # You can enumerate objects in your buckets.

  #

  #   # enumerate ALL objects in the bucket (even if the bucket contains 

  #   # more than 1k objects)

  #   bucket.objects.each do |obj|

  #     puts obj.key

  #   end

  #

  #   # enumerate at most 20 objects with the given prefix

  #   bucket.objects.with_prefix('photos/').each(:limit => 20).each do |photo|

  #     puts photo.key

  #   end

  #

  # See {ObjectCollection} and {S3Object} for more information on working

  # with objects.

  #