gistfile1.txt

urls = ['http://www.google.com', 'http://www.github.com', 'http://www.twitter.com']
 
concurrent_url_grabber = Enumerator.new do |y|
  urls.map do |url|
    Thread.new do
      result = http_request(url)
      # Need a mutex here for multithreaded VMs
      y << result
    end
  end.each(&:join)
end
 
concurrent_url_grabbler.map(&:code) # => ['200', '200', '200']

FYI if you use each(&:value), instead of each(&:join), you don't need to worry about the mutex (and it's still thread-safe).

@jstorimer I had a different take on the difference between Thread#value and Thread#join. I thought #value calls #join internally and returns a value, while #join returns the instance of the new Thread? Is that not correct? If it is I do not see how there is no need for mutex at first glance.