How can I download a file from a URL and save it in Rails?

Question

I have a URL to an image which i want to save locally, so that I can use Paperclip to produce a thumbnail for my application. What's the best way to download and save the image? (I looked into ruby file handling but did not come across anything.)

Cutworm

Answer 1

Try this:

require 'open-uri'
open('image.png', 'wb') do |file|
  file << open('http://example.com/image.png').read
end

Georama

This works. Thanks. I tried this on an image. The image was not properly written to the file. As if it has been corrupted or something. Also could you guide me to do this same in a Rails application. Can i use this in a Rails application ? – Cutworm

Yes, you can use it in a rails application. The file not being written properly is probably because I forgot to open the destination in write-binary mode. I'll edit that now. – Georama

This reads the whole file in memory. The best way is IO.copy_stream(open('http://example.com/image.png'), 'destination.png') – Pia

this doesn't work for me. I've created a question about this: #37072431 – Titbit

rails 5, 2016, I didn't need to require 'open-uri', just used the code under it – Typical

this answer did not work for me. correct answer is here: https://mcmap.net/q/117388/-how-can-i-download-a-file-from-a-url-and-save-it-in-rails – Cardoso

Is it possible to provide a path of where the file should be stored? Currently, is being stored on root in Rails app – Eadwine

Extreme caution if the url can be provided by the user! sakurity.com/blog/2015/02/28/openuri.html – Imperialism

Answer 2

An even shorter version:

require 'open-uri'
download = open('http://example.com/image.png')
IO.copy_stream(download, '~/image.png')

To keep the same filename:

IO.copy_stream(download, "~/#{download.base_uri.to_s.split('/')[-1]}")

Vesta

And more efficient as the output is not stored as a string in memory. +1 – Pyle

Though it's actually useful having both answers displayed, so people like me can see both, think about the differences and then choose this answer consciously for the right reasons :) – Comity

I'm getting a security warning from Rubocop when using open. I replaced it with URI.parse(url).open instead. Read more about that cop here: rubydoc.info/gems/rubocop/RuboCop/Cop/Security/Open – Runoff

pure open does not work for me, had to use URI.open – Wholehearted

Answer 3

If you're using PaperClip, downloading from a URL is now handled automatically.

Assuming you've got something like:

class MyModel < ActiveRecord::Base
  has_attached_file :image, ...
end

On your model, just specify the image as a URL, something like this (written in deliberate longhand):

@my_model = MyModel.new
image_url = params[:image_url]
@my_model.image = URI.parse(image_url)

You'll probably want to put this in a method in your model. This will also work just fine on Heroku's temporary filesystem.

Paperclip will take it from there.

source: paperclip documentation

Herein

For future searchers, Paperclip no longer allows this by default: medium.com/in-the-weeds/… – Llama

Answer 4

I think this is the clearest way:

require 'open-uri'

File.write 'image.png', open('http://example.com/image.png').read

Mckellar

sage i am new to rails, where i can add this code in model or controller or in view – Beaverbrook

It depends, but typically you'd do something like this inside a model or controller method. In a Rails context, you're likely better off using the framework's tools like ActiveStorage, rather than writing directly to disk. – Mckellar

Answer 5

Possibly the simplest way:

require 'open-uri'
image_url = "https://i.imgur.com/ZWnhY9T.png"
IO.copy_stream(URI.open(image_url), 'destination.png')

Couchant

In my situation, I tried all the codes above. only this floor worked. The difference is copy_stream()'s first param URI.open(img_url), if i dont wrap the img_url by URI.open(), i got the following ERROR No such file or directory @ rb_sysopen. – Peevish

Is there a way to get the download progress? – Dianemarie

@Dianemarie sorry I don't know of a way, however it could make a great question to ask, as I suspect it would be valuable to others too. Please drop the link here if you ask it. – Couchant

I was able to do it in this gist: gist.github.com/pioz/… – Dianemarie

Answer 6

Check out Net::HTTP in the standard library. The documentation provides several examples on how to download documents using HTTP.

Waddell

This answer needs more visibility as Kernel#open enables not only file access but also process invocation by prefixing a pipe symbol (e.g., open("| ls")). So, it may lead to a serious security risk by using variable input to the argument of Kernel#open. – Campball

Answer 7

Using Ruby 3 and above, you'll get the following error using the accepted answer:

No such file or directory @ rb_sysopen - http://example.com/image.png (Errno::ENOENT)

The solution is to use URI.open in place of the Kernel.open. Example:

require "uri"

download = URI.open('http://example.com/image.png')
File.write('~/image.png', download)

Portingale

Answer 8

All above examples are great. In my case I just wanted to create a download link from the image from URL.

If you want to make it downloadable (to your downloads folder), you can use the following code in your controller:

require 'open-uri'
file_type = url.to_s.split(".")[-1]

send_data open(url).read, filename: "some_name.#{file_type}", type: "image/#{file_type}", disposition: "attachment"

Faultless