Delegate child session assertions to parent session

[sambostock]

Summary

This fixes #32142, in which assertions made in sub-sessions in ActionDispatch::IntegrationTest are not counted in the test run report.

By overriding the assertions accessor methods on the child session to use the parent's assertions, the count is accurately tracked.

Other Information

• Is this the appropriate fix, or is there a better way to delegate to the parent session?
• Is this the appropriate test? Should test report output be asserted on instead?

---

cc. @aud

[rails-bot]

Thanks for the pull request, and welcome! The Rails team is excited to review your changes, and you should hear from @rafaelfranca (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. Due to the way GitHub handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

This repository is being automatically checked for code quality issues using Code Climate. You can see results for this analysis in the PR status below. Newly introduced issues should be fixed before a Pull Request is considered ready to review.

Please see the contribution instructions for more information.

[sambostock]

@rafaelfranca I've made the changes you requested. Let me know what you think, and I can squash the commits down.

[sambostock]

I found myself working on some integration tests and realized this never shipped 😅

• I've rebased on latest master, and squashed my previous commits.
• I realized that in that a grandchild session is opened by opening a session inside a child session, the delegation is indirect. Therefore, I pushed another commit which switches to storing the root session, not the parent one.
• I noticed another relevant accessor that could result in miscounts would be failures. Should it also be proxied to the root_session? What about the various other related variables?

[rafaelfranca]

• I noticed another relevant accessor that could result in miscounts would be failures. Should it also be proxied to the root_session? What about the various other related variables?

I think we should make sure all of them work.

[sambostock]

I did some experimenting, and it turns out I was wrong about all those variables.

As far as I can tell, they're defined on the report, not the test itself. The only relevant instance variables are @assertions and @failures.

@assertions is already handled, and the following test (adapted from #32142's example) shows that @failures already works on Rails 5.2.3.

@failures test

#! /usr/bin/env ruby
# frozen_string_literal: true

begin
  require "bundler/inline"
rescue LoadError => e
  $stderr.puts "Bundler version 1.10 or later is required. Please update your Bundler"
  raise e
end

gemfile(true) { gem 'rails', source: 'https://rubygems.org' }
require 'rails/all'
require "minitest/autorun"

class FailuresTest < ActionDispatch::IntegrationTest
  test('flunk outside session') { assert(true) } # control
  test('flunk in session') { open_session.assert(true) }
end

class BugTest < Minitest::Test
  def test_control
    test_instance = FailuresTest.new('flunk_outside_session').run
    refute_predicate test_instance.failures, :empty?
  end

  def test_actual
    test_instance = FailuresTest.new('flunk_in_session').run
    refute_predicate test_instance.failures, :empty?
  end
end

I suspect this is because failures already bubble up due to their use of exceptions.

---

I have once again squashed and rebased, so unless there is other feedback, this should be good to go!

[rails-bot]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.
Thank you for your contributions.