使用Paperclip / Rails为AWS S3正确设置URL / PATH

Question

I have two models: Collection and Letter . Collection has many Letter s and Letter obviously belongs_to Collection .

Below is my Letter.rb file:

class Letter < ActiveRecord::Base
    belongs_to :collection

    has_attached_file :pdf,
                    :url => "/pdf/:id/:style/:basename.:extension",
                    :path => "/pdf/:id/:style/:basename.:extension",
                    :s3_host_name => host_name,
                    :storage => :s3,
                    :bucket  => ENV['S3_BUCKET_NAME'],
                    :s3_credentials => {
                                        :access_key_id =>#ENV['AWS_ACCESS_KEY_ID'],
                                        :secret_access_key =>ENV['AWS_SECRET_ACCESS_KEY']
                                    }

    validates_attachment :pdf,
                     :content_type => {
                         :content_type =>
                             ["application/pdf", "text/plain", /\Aimage\/.*\Z/, "application/msword"]
                     }

end

I have a question about url and path attribute in has_attached_file .

Instead of setting letter model's id in path, I would like to set collection's id. Additionally, I also wanna put title which is Letter's attribute. Let's say @collection 's id is 1. @collection has @letter1 and @letter2. When I save the files to AWS S3, I want to save it under /pdf/1(which is collection_id)/:title . How can I write this in url and path ?

Answer 1

In Paperclip, you can use interpolations for that.

Your has_attached_file method would look somewhat like this;

has_attached_file :image, :default_url => "/pdf/:collection_id/:title/:basename.:extension"

Create an interpolation file, called paperclip.rb or interpolations.rb in the config/initializers directory (rails picks up any script in that folder on startup), which contains code that looks somewhat like this;

Paperclip.interpolates :collection_id do |attachment, style|
    attachment.instance.collection_id
end

Add the :title interpolation in the same way; add it to your has_attached_file urls, and create a second interpolation for that.

You can read more about this at https://github.com/thoughtbot/paperclip/wiki/Interpolations

In your case, I would suggest to also include the :id of the Letter in the URL, as it might be possible the user uploads two documents with the same title which might conflict.

has_attached_file :image, :default_url => "/pdf/:collection_id/:id/:title"

Paperclip uses the interpolations :basename , :extension and :style by default to create a unique path for the file.

• :basename is the base file name of the uploaded file
• :extension is the extension of that uploaded file
• :style is the "style" or size of that uploaded file

You can specify multiple styles (like thumnails in various versions). The default style is "original", which will contain the original uploaded file.

Read more about styles here; https://github.com/thoughtbot/paperclip/wiki/Thumbnail-Generation

Always try to keep the original file as it can be handy in the future; when your site/application layout changes and new thumbnail sizes are required. You can rebuild/regenerate your whole thumbnail library from the original version.

Read more about generating/regenerating thumbnails here; https://github.com/thoughtbot/paperclip/wiki/Thumbnail-Generation#generatingregenerating-your-thumbnails

Answer 2

Using gem aws-sdk for a ror application for uploading images to s3 Uploading images to a fixed bucket with different folders for each object or application. The s3 keeps a limitation on the number of buckets creattion whereas there is no limitation for content inside a bucket.

This code will upload image for a user to s3 using aws-sdk gem. The bucket and the image uploaded are made public so that the images uploaded are directly accessible. The input it takes is the image complete path where it is present, folder in which it should be uploaded and user_id for whom it should be uploaded.

  def save_screenshot_to_s3(image_location, folder_name,user_id)
    service = AWS::S3.new(:access_key_id => ACCESS_KEY_ID,
                          :secret_access_key => SECRET_ACCESS_KEY)
    bucket_name = "app-images"
    if(service.buckets.include?(bucket_name))
      bucket = service.buckets[bucket_name]
    else
      bucket = service.buckets.create(bucket_name)
    end
    bucket.acl = :public_read
    key = folder_name.to_s + "/" + File.basename(image_location)
    s3_file = service.buckets[bucket_name].objects[key].write(:file => image_location)
    s3_file.acl = :public_read
    user = User.where(id: user_id).first
    user.image = s3_file.public_url.to_s
    user.save
  end   

Use key = folder_name.to_s + "/" + File.basename(image_location) to customize the path you want to have.