Helper to generate response headers

[itskingori]

Provides a handy helper to help generate the response headers either based off throttle data e.g. env['rack.attack.match_data'] or even env['rack.attack.throttle_data'][throttle_name].

To note, this is related to the discussion in #191 (and maybe #29) but since it's just a helper, it's all opt-in i.e. no opinions on whether you should generate the headers or not (or even where) ... you choose to use the helper if/when/where you want it.

Usage Examples

It can be used in the controller to set response headers (Rails example) ...

module API
  class BaseController < ApplicationController
    after_action :set_rate_limit_headers!

    def set_rate_limit_headers!
      throttle_data = request.env['rack.attack.throttle_data'][throttle_name]
      options = [include_limit: true, include_remaining: true, include_reset: true]
      rate_limit_headers = Rack::Attack::ResponseHeaders.new(throttle_data).generate(*options)
      rate_limit_headers.each { |name, value| response.set_header(name, value) }
    end
  end
end

Or it can be used in throttled_response ...

# Return rate limit data that's helpful for well-behaved clients such as how
# many seconds to wait until they can start sending requests again.
Rack::Attack.throttled_response_retry_after_header = true
Rack::Attack.throttled_response = lambda do |env|
  throttle_data = env['rack.attack.match_data']
  options = [include_limit: true, include_remaining: false, include_reset: true]
  headers = Rack::Attack::ResponseHeaders.new(throttle_data).generate(*options)
  headers['X-RateLimit-Remaining'] = '0'
  response = {
    errors: [
      {
        status: '429',
        title: 'Too Many Requests'
      }
    ]
  }.to_json

  [429, headers, [response]]
end