# Paperclip save to disk and s3

**URL:** <https://www.ruby-forum.com/t/paperclip-save-to-disk-and-s3/192908>\
**Category:** Rails\
**Created:** [August 8, 2010, 2:23am UTC](https://www.ruby-forum.com/t/paperclip-save-to-disk-and-s3/192908 "2010-08-08T02:23:40Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![df-sean](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/f07891/32.png) [@df-sean](https://www.ruby-forum.com/u/df-sean)\
**Post date:** [August 8, 2010, 2:23am UTC](https://www.ruby-forum.com/t/paperclip-save-to-disk-and-s3/192908/1 "2010-08-08T02:23:40Z")

</div>

I have a standard Paperclip setup that saves a file to my disk. In  
addition I would also like the file saved to my amazon s3 bucket.

```auto

after_save :copy_to_s3

  def copy_to_s3
   has_attached_file :photo,
   :storage => :s3,
   :s3_credentials => "#{RAILS_ROOT}/config/s3.yml",
   :styles => {
   :thumb => "100x100#",
   :small => "750x750>" }
  end

```

This is giving the error: undefined method `has\_attached\_file’

---

<div class="post-metadata">

**Author:** ![df-sean](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/f07891/32.png) [@df-sean](https://www.ruby-forum.com/u/df-sean)\
**Post date:** [August 8, 2010, 3:12am UTC](https://www.ruby-forum.com/t/paperclip-save-to-disk-and-s3/192908/2 "2010-08-08T03:12:48Z")

</div>

Because it’s in another scope of object  
And it’s nothing give to u, it’s one define some methods

U can open paperclip source and find out how call upload to s3, or u can  
use aws-s3 gem, which simple, s3 is http based storage, simple http  
query, and all a done ))

Sent from my iPad

---

<div class="post-metadata">

**Author:** ![df-sean](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/f07891/32.png) [@df-sean](https://www.ruby-forum.com/u/df-sean)\
**Post date:** [August 8, 2010, 8:16am UTC](https://www.ruby-forum.com/t/paperclip-save-to-disk-and-s3/192908/3 "2010-08-08T08:16:13Z")

</div>

Scratch the idea then.

So I found this gem called db2s3 which is used to backup the database  
with a simple rake command. I having trouble modifying it accordingly.

require ‘activesupport’  
require ‘aws/s3’  
require ‘tempfile’

class BLUECAT  
class Config  
end

def initialize  
end

def full\_backup

```
file_name = "system-#{Time.now.utc.strftime("%Y%m%d%H%M")}.gz"

store.store(file_name, open(backup_directory.path))
store.store(most_recent_dump_file_name, file_name)

```

end

private

def backup\_directory  
backup\_file = Tempfile.new(“dump”)

```
 cmd = "cp -r #{RAILS_ROOT}/public/system/ bucket"
 cmd += " | gzip > #{backup_file.path}"
 run(cmd)

 backup_file

```

end

def store  
@store ||= S3Store.new  
end

def most\_recent\_dump\_file\_name  
“most-recent-dump.txt”  
end

def run(command)  
result = system(command)  
raise(“error, process exited with status #{$?.exitstatus}”) unless  
result  
end

def db\_credentials  
ActiveRecord::Base.connection.instance\_eval { @config } # Dodgy!  
end

class S3Store  
def initialize  
@connected = false  
end

```
def ensure_connected
  return if @connected
  AWS::S3::Base.establish_connection!(BLUECAT::Config::S3.slice(:access_key_id, 

```

:secret\_access\_key).merge(:use\_ssl =\> true))  
AWS::S3::Bucket.create(bucket)  
@connected = true  
end

```
def store(file_name, file)
  ensure_connected
  AWS::S3::S3Object.store(file_name, file, bucket)
end

def fetch(file_name)
  ensure_connected
  AWS::S3::S3Object.find(file_name, bucket)

  file = Tempfile.new("dump")
  open(file.path, 'w') do |f|
    AWS::S3::S3Object.stream(file_name, bucket) do |chunk|
      f.write chunk
    end
  end
  file
end

def list
  ensure_connected
  AWS::S3::Bucket.find(bucket).objects.collect {|x| x.path }
end

def delete(file_name)
  if object = AWS::S3::S3Object.find(file_name, bucket)
    object.delete
  end
end

private

def bucket
  BLUECAT::Config::S3[:bucket]
end

```

end

end

In the backup\_directory method, I can’t figure out how to get it to copy  
the system directory into my s3 bucket.