Gem Version Build Status Code Climate Coverage Status


http.rb is an easy-to-use client library for making requests from Ruby. It uses a simple method chaining system for building requests, similar to Python's Requests.

Under the hood, http.rb uses http_parser.rb, a fast HTTP parsing native extension based on the Node.js parser and a Java port thereof. This library isn't just yet another wrapper around Net::HTTP. It implements the HTTP protocol natively and outsources the parsing to native extensions.

Another Ruby HTTP library? Why should I care?

There are a lot of HTTP libraries to choose from in the Ruby ecosystem. So why would you choose this one?

Top three reasons:

  1. Clean API: http.rb offers an easy-to-use API that should be a breath of fresh air after using something like Net::HTTP.

  2. Maturity: http.rb is one of the most mature Ruby HTTP clients, supporting features like persistent connections and fine-grained timeouts.

  3. Performance: using native parsers and a clean, lightweight implementation, http.rb achieves the best performance of any Ruby HTTP library which implements the HTTP protocol in Ruby instead of C:

    HTTP client time
    curb (persistent) 2.519088
    em-http-request 2.731645
    Typhoeus 2.851911
    StreamlyFFI (persistent) 2.853786
    http.rb (persistent) 2.970702
    http.rb 3.588964
    HTTParty 3.931913
    Net::HTTP 3.959342
    Net::HTTP (persistent) 4.043674
    open-uri 4.479817
    Excon (persistent) 4.618361
    Excon 4.701262
    RestClient 26.832668

Benchmarks performed using excon's benchmarking tool

Help and Discussion

If you need help or just want to talk about the http.rb, visit the http.rb Google Group:!forum/httprb

You can join by email by sending a message to:

If you believe you've found a bug, please report it at:


Add this line to your application's Gemfile:

gem "http"

And then execute:

$ bundle

Or install it yourself as:

$ gem install http

Inside of your Ruby program do:

require "http" pull it in as a dependency.


Please see the http.rb wiki for more detailed documentation and usage notes.

Basic Usage

Here's some simple examples to get you started:

GET requests

>> HTTP.get("").to_s
=> "<html><head><meta http-equiv=\"content-type\" content=..."

That's all it takes! To obtain an HTTP::Response object instead of the response body, all we have to do is omit the #to_s on the end:

>> HTTP.get("")
=> #<HTTP/1.0 200 OK @headers={"Content-Type"=>"text/html; charset=UTF-8", "Date"=>"Fri, ...>
 => #<HTTP::Response/1.1 200 OK @headers={"Content-Type"=>"text/html; ...>

We can also obtain an HTTP::Response::Body object for this response:

>> HTTP.get("").body
 => #<HTTP::Response::Body:814d7aac @streaming=false>

The response body can be streamed with HTTP::Response::Body#readpartial:

>> HTTP.get("").body.readpartial
 => "<!doctype html><html "

In practice you'll want to bind the HTTP::Response::Body to a local variable (e.g. "body") and call readpartial on it repeatedly until it returns nil.

POST requests

Making POST requests is simple too. Want to POST a form?"", :form => {:foo => "42"})

Making GET requests with query string parameters is as simple.

HTTP.get("", :params => {:foo => "bar"})

Want to POST with a specific body, JSON for instance?"", :json => { :foo => "42" })

Or just a plain body?"", :body => "foo=42&bar=baz")

Posting a file?"", :form => {
  :username => "ixti",
  :avatar   =>"/home/ixit/avatar.png")

It's easy!

Proxy Support

Making request behind proxy is as simple as making them directly. Just specify hostname (or IP address) of your proxy server and its port, and here you go:

HTTP.via("proxy-hostname.local", 8080)

Proxy needs authentication? No problem:

HTTP.via("proxy-hostname.local", 8080, "username", "password")

Adding Headers

The HTTP gem uses the concept of chaining to simplify requests. Let's say you want to get the latest commit of this library from GitHub in JSON format. One way we could do this is by tacking a filename on the end of the URL:


The GitHub API happens to support this approach, but really this is a bit of a hack that makes it easy for people typing URLs into the address bars of browsers to perform the act of content negotiation. Since we have access to the full, raw power of HTTP, we can perform content negotiation the way HTTP intends us to, by using the Accept header:

HTTP.headers(:accept => "application/json")

This requests JSON from GitHub. GitHub is smart enough to understand our request and returns a response with Content-Type: application/json.

Shorter alias exists for HTTP.headers:

HTTP[:accept => "application/json"]

Authorization Header

With HTTP Basic Authentication using a username and password:

HTTP.basic_auth(:user => "user", :pass => "pass")
# <HTTP::Headers {"Authorization"=>"Basic dXNlcjpwYXNz"}>

Or with plain as-is value:

HTTP.auth("Bearer VGhlIEhUVFAgR2VtLCBST0NLUw")
# <HTTP::Headers {"Authorization"=>"Bearer VGhlIEhUVFAgR2VtLCBST0NLUw"}>

And Chain all together!

HTTP.basic_auth(:user => "user", :pass => "pass")
  .headers("Cookie" => "9wq3w")

Content Negotiation

As important a concept as content negotiation is to HTTP, it sure should be easy, right? But usually it's not, and so we end up adding ".json" onto the ends of our URLs because the existing mechanisms make it too hard. It should be easy:


This adds the appropriate Accept header for retrieving a JSON response for the given resource.

Reuse HTTP connection: HTTP Keep-Alive

If you have many successive requests against the same host, you better want to reuse the same connection again and again:

contents = []
targets = %w(Hypertext_Transfer_Protocol Git GitHub Linux Hurd)
HTTP.persistent('') do |http|
  targets.each { |target| contents << http.get("/wiki/#{target}") }

Celluloid::IO Support

http.rb makes it simple to make multiple concurrent HTTP requests from a Celluloid::IO actor. Here's a parallel HTTP fetcher combining http.rb with Celluloid::IO:

require "celluloid/io"
require "http"

class HttpFetcher
  include Celluloid::IO

  def fetch(url)
    HTTP.get(url, socket_class: Celluloid::IO::TCPSocket)

There's a little more to it, but that's the core idea!


By default, HTTP does not timeout on a request. You can enable per operation (each read/write/connect call) or global (sum of all read/write/connect calls).

Per operation timeouts are what Net::HTTP and the majority of HTTP clients do:

HTTP.timeout(:per_operation, :write => 2, :connect => 5, :read => 10)
  .get ""

# For convinience, you can omit timeout type in this case. So following has
# same result as the above:

HTTP.timeout(:write => 2, :connect => 5, :read => 10).get ""

Global timeouts let you set an upper bound of how long a request can take, without having to rely on Timeout.timeout:

HTTP.timeout(:global, :write => 1, :connect => 1, :read => 1)
  .get ""

Uses a timeout of 3 seconds, for the entire get call.

Warning! You cannot use Celluloid::IO with timeouts currently.

Supported Ruby Versions

This library aims to support and is tested against the following Ruby versions:

  • Ruby 1.9.3
  • Ruby 2.0.0
  • Ruby 2.1.x
  • Ruby 2.2.x
  • JRuby 1.7.x
  • JRuby 9000

If something doesn't work on one of these versions, it's a bug.

This library may inadvertently work (or seem to work) on other Ruby versions, however support will only be provided for the versions listed above.

If you would like this library to support another Ruby version or implementation, you may volunteer to be a maintainer. Being a maintainer entails making sure all tests run and pass on that implementation. When something breaks on your implementation, you will be responsible for providing patches in a timely fashion. If critical issues for a particular implementation exist at the time of a major release, support for that Ruby version may be dropped.

Contributing to http.rb

  • Fork http.rb on GitHub
  • Make your changes
  • Ensure all tests pass (bundle exec rake)
  • Send a pull request
  • If we like them we'll merge them
  • If we've accepted a patch, feel free to ask for commit access!


Copyright (c) 2011-2015 Tony Arcieri, Erik Michaels-Ober, Alexey V. Zapparov, Zachary Anker. See LICENSE.txt for further details.