Skip to content

Latest commit

 

History

History
83 lines (67 loc) · 2.9 KB

README.md

File metadata and controls

83 lines (67 loc) · 2.9 KB

Cloudinary Storage

In this setup, we'll be implementing a system when the messages are passed as follows :

  1. User browses to the 'add new image' form (GET: /images/new)
  2. The choose a file, and click submit, which sends a POST to /images
  3. The server receives that form with attached image, and send the file to Cloudinary (or some other cloud storage location)
  • this needs authentication
  1. Cloudinary stores the file and sends a response to our server, with details about where it's saved the image (url)
  2. On our server, we save the image location (db), then send some response to the user (e.g. show the list of all the uploaded images, or status OK)

Setting it up

  1. setup up a cloudinary account : https://cloudinary.com/users/register/free
  2. check out their Node guide : http://cloudinary.com/documentation/node_integration#getting_started_guide
  1. get an image you have uploaded
const cloudinary = require('cloudinary') 

cloudinary.config({ 
  cloud_name: 'fakeName', 
  api_key: 'fakeKey', 
  api_secret: 'fakeSecret' 
})

cloudinary.uploader.upload('radiolariands.jpeg', function(result) { 
  console.log(result) 
})

Cool, got this response :

// response
{ public_id: 'geojastuh5g0vibhm8o8',
  version: 1490220497,
  signature: 'dd7425570d05f9892fdd30044938dc3697c84e4e',
  width: 400,
  height: 300,
  format: 'jpg',
  resource_type: 'image',
  created_at: '2017-03-22T22:08:17Z',
  tags: [],
  bytes: 24633,
  type: 'upload',
  etag: '1208302ed19bbc0e3f6a52f299ddfe4a',
  url: 'http://res.cloudinary.com/db7jm5eee/image/upload/v1490220497/geojastuh5g0vibhm8o8.jpg',
  secure_url: 'https://res.cloudinary.com/db7jm5eee/image/upload/v1490220497/geojastuh5g0vibhm8o8.jpg',
  original_filename: 'radiolariands' }

  1. get dotenv set up to handle your keys (hiding them in the gitignored .env file)

  2. get file a file upload set up with express and plug that into cloudinary

  • set up formidable
    • prove formidable can parse files
  • get file from formidable to be given to cloudinary

///

Building this into an app

Things to consider :

  • what do you want to display to the user?
  • where are you going to record the url cloundinary give you?
    • probably a database table
  • happy + unhappy paths:
    • what happens when cloudinary returns an error
    • what sort of files do you accept? (zip/ pdf/ doc/ svg/ etc.)
    • what if a user uploads 2 images?
    • if these env vars are connected to my credit card, should I share them with my team? (spoiler: NO)
  • get your environment variables onto a server
  • how would you change this setup for a client-side + API setup
    • the form can be a particular view
    • the POST is probably to something like /api/v1/images

Diagram

alt tag