have_enqueued_job clashes with rspec-rails 3.4.0

[JonRowe]

In rspec-rails we implemented support for asserting on wether or not an ActiveJob has been queued, unfortunately we used have_enqueued_job as the name for this matcher as we were unaware of this gem. This means that rspec-sidekiq will now clash with rspec-rails.

Given that our matcher is more generic I propose that the have_enqueued_job matcher in rspec-sidekiq be renamed to have_enqueued_sidekiq_job and then aliased back to have_enqueued_job when it's not otherwise defined. (We use the alias trick a fair bit in rspec-expectations to provide extra grammar for matchers so it's a pretty accepted technique).

I'd be happy to work on a PR for this if it's acceptable to you :)

[mhenrixon]

👍

[aprescott]

I see @mrbrdo issued #84 for this fix. Until that gets reviewed, merged and included in a release, is there a temporary work-around I can use to unblock gem updates?

[aprescott]

Think I got one, actually:

# Temporarily work around a clash between rspec-rails 3.4.0 and
# rspec-sidekiq because of new ActiveJob matchers.
#
# See philostler/rspec-sidekiq#83
module RSpec
  module Sidekiq
    module Matchers
      def have_enqueued_sidekiq_job(*expected_arguments)
        HaveEnqueuedSidekiqJob.new expected_arguments
      end

      HaveEnqueuedSidekiqJob = HaveEnqueuedJob
    end
  end
end

[mrbrdo]

Or just use my branch in your gemfile.

[aprescott]

Git refs in a Gemfile lead to bulky vendor/cache directories, so I try to avoid them.

[fabn]

@philostler any ETA for this? I'd like to upgrade RSpec but this issue prevent me to do that.

[philostler]

Been busy with work etc recently, will review all PRs and Issues very soon.

[bkoltai]

+1 for PR!

[nickcoyne]

Was the PR not merged?

[philostler]

This should be fixed with #95 now