Back to Repositories

Testing Resque Web Interface Components in resque

This test suite validates the Resque web interface functionality, covering route handling, queue management, and status reporting features. It ensures proper behavior of the web dashboard across various endpoints and configurations.

Test Coverage Overview

The test suite provides comprehensive coverage of Resque’s web interface endpoints and functionality.

Key areas tested include:

Root path redirection
Overview page rendering
Queue management and display
Failed job handling
Working jobs status
Redis and Resque statistics

Edge cases covered include reverse proxy configuration and different failure backend scenarios.

Implementation Analysis

The testing approach uses RSpec with Rack::Test to simulate HTTP requests to the Resque web interface. It employs stub methods for failure backends and queue management, allowing isolated testing of UI components and response handling.

The implementation leverages RSpec’s describe blocks for logical grouping and before hooks for test setup, following BDD patterns.

Technical Details

Testing tools and configuration:

RSpec testing framework
Rack::Test for HTTP request simulation
Mock objects and stubs for failure handling
Redis backend integration
Custom URL prefix configuration support

Best Practices Demonstrated

The test suite exemplifies several testing best practices including proper separation of concerns, thorough endpoint coverage, and explicit error checking.

Notable practices include:

Consistent test structure and organization
Proper setup and teardown patterns
Explicit assertion messages
Comprehensive edge case handling
Modular test grouping

resque/resque

test/resque-web_test.rb

            
require 'test_helper'
require 'rack/test'
require 'resque/server'

describe "Resque web" do
  include Rack::Test::Methods

  def app
    Resque::Server.new
  end

  # Root path test
  describe "on GET to /" do
    before { get "/" }

    it "redirect to overview" do
      follow_redirect!
    end
  end

  # Global overview
  describe "on GET to /overview" do
    before { get "/overview" }

    it "should at least display 'queues'" do
      assert last_response.body.include?('Queues')
    end
  end

  describe "With append-prefix option on GET to /overview" do
    reverse_proxy_prefix = 'proxy_site/resque'
    Resque::Server.url_prefix = reverse_proxy_prefix
    before { get "/overview" }

    it "should contain reverse proxy prefix for asset urls and links" do
      assert last_response.body.include?(reverse_proxy_prefix)
    end
  end

  # Working jobs
  describe "on GET to /working" do
    before { get "/working" }

    it "should respond with success" do
      assert last_response.ok?, last_response.errors
    end
  end

  # Queues
  describe "on GET to /queues" do
    before { Resque::Failure.stubs(:count).returns(1) }

    describe "with Resque::Failure::RedisMultiQueue backend enabled" do
      it "should display failed queues" do
        with_failure_backend Resque::Failure::RedisMultiQueue do
          Resque::Failure.stubs(:queues).returns(
            [:queue1_failed, :queue2_failed]
          )

          get "/queues"
        end

        assert last_response.body.include?('queue1_failed')
        assert last_response.body.include?('queue2_failed')
      end

      it "should respond with success when no failed queues exists" do
        with_failure_backend Resque::Failure::RedisMultiQueue do
          Resque::Failure.stubs(:queues).returns([])

          get "/queues"
        end

        assert last_response.ok?, last_response.errors
      end
    end

    describe "Without Resque::Failure::RedisMultiQueue backend enabled" do
      it "should display queues when there are more than 1 failed queue" do
        Resque::Failure.stubs(:queues).returns(
          [:queue1_failed, :queue2_failed]
        )
        get "/queues"

        assert last_response.body.include?('queue1_failed')
        assert last_response.body.include?('queue2_failed')
      end

      it "should display 'failed' queue when there is 1 failed queue" do
        Resque::Failure.stubs(:queues).returns([:queue1])
        get "/queues"

        assert !last_response.body.include?('queue1')
        assert last_response.body.include?('failed')
      end

      it "should respond with success when no failed queues exists" do
        Resque::Failure.stubs(:queues).returns([])
        get "/queues"

        assert last_response.ok?, last_response.errors
      end
    end
  end

  # Failed
  describe "on GET to /failed" do
    before { get "/failed" }

    it "should respond with success" do
      assert last_response.ok?, last_response.errors
    end
  end

  # Stats
  describe "on GET to /stats/resque" do
    before { get "/stats/resque" }

    it "should respond with success" do
      assert last_response.ok?, last_response.errors
    end
  end

  describe "on GET to /stats/redis" do
    before { get "/stats/redis" }

    it "should respond with success" do
      assert last_response.ok?, last_response.errors
    end
  end

  describe "on GET to /stats/resque" do
    before { get "/stats/keys" }

    it "should respond with success" do
      assert last_response.ok?, last_response.errors
    end
  end

  describe "also works with slash at the end" do
    before { get "/working/" }

    it "should respond with success" do
      assert last_response.ok?, last_response.errors
    end
  end

end