File: README — Documentation for typhoeus_spec

typhoeus_spec_cache

Typhoeus is a great frontend to Curl/Curl::Multi. However, if you have a large spec suite that hits a number of external HTTP resources, you will quickly find yourself getting tired of caching all those resources by hand! Trust me. No seriously, I hate doing this by hand.

Thanks to Paul Dix for writing Typhoeus!

Usage

In any specs that hit HTTP via Typhoeus::Hydra, wrap them in a typhoeus_spec_cache block:

describe "making 30 expensive HTTP calls" do
  typhoeus_spec_cache("spec/cache/mycache") do |hydra|
    it "should get 10 URLs" do
      result = fetch_urls(10)
      result.should have(10).things
    end

    it "should get 20 URLs" do
      result = fetch_urls(20)
      result.should have(20).things
    end
  end
end

Create the cache directory:

mkdir -p spec/cache/mycache

On the first run of these two specs, Typhoeus will make 30 live HTTP requests. These will then be dumped out to .cache files in spec/cache/mycache, and used on subsequent runs, avoiding any future HTTP calls.

Auto-stubbing of Hydra

Calling typhoeus_spec_cache automatically stubs out calls to Typhoeus::Hydra#new with a single shared Hydra object.

Auto-managing of cache files

If the URLs your code hits ever changes, the typhoeus_spec_cache wrapper is smart about handling your cache. It will:

Only download new URLs it hasn't seen.
Remove any .cache files that aren't needed anymore.

This way, your cache directory will contain exactly the .cache files you need, and no more.

Installation

sudo gem install typhoeus_spec_cache

Then, add the following to your spec/spec_helper.rb

require 'typhoeus_spec_cache'

And add these two lines to your Spec::Runner.configure block in spec/spec_helper.rb:

Spec::Runner.configure do |config|
  config.include(Typhoeus::SpecCacheMacros::InstanceMethods)
  config.extend(Typhoeus::SpecCacheMacros::ClassMethods)
end

Note on Patches/Pull Requests