When making the new command insfrastructure I had missed that
`bin/rails runner some_file.rb some args` would pass the extra
args onto the file in `ARGV`.
Now fixed by allowing the command to take extra args again, and
make sure to remove the file name from `ARGV`.
Rails routes (even rake routes in previous versions) output showed incorrect routes when an application use resource :controller, implying that edit_controller_path match with controller#show.
The order of the output has changed to correct this. View #26606 for more information.
Added a test case, change unit test in rake to expect the new output.
Since the output of resource :controller is changing, the string spected of the railties/test/application/rake_test.rb test_rails_routes_with_controller_environment had to be modified.
Reset a new session directly after its creation in
`ActionDispatch::IntegrationTest#open_session`. Reset the session to a clean
state before making it available to the client's test code.
Issue #22742 reports unexpected behavior of integration tests that run multiple
sessions. For example an `ActionDispatch::Flash` instance is shared across
multiple sessions, though a client code will rightfully assume that each new
session has its own flash hash.
The following test failed due to this behavior:
class Issue22742Test < ActionDispatch::IntegrationTest
test 'issue #22742' do
integration_session # initialize first session
a = open_session
b = open_session
refute_same(a.integration_session, b.integration_session)
end
end
Instead of creating a new `ActionDispatch::Integration::Session` instance,
the same instance is shared across all newly opened test sessions. This is
due to the way how new test sessions are created in
`ActionDispatch::IntegrationTest#open_session`. The already existing
`ActionDispatch::IntegrationTest` instance is duplicated with `Object#dup`,
This approach was introduced in commit 15c31c7639b. `Object#dup` copies the
instance variables, but not the objects they reference. Therefore this issue
only occurred when the current test instance had been tapped in such a way that
the instance variable `@integration_session` was initialized before creating the
new test session.
Close#22742
[Tawan Sierek + Sina Sadeghian]
- New apps generated on master and latest bundler give warning about
"github" source being insecure.
- Use the same solution used for Rails master in the generated app's
Gemfile to fix this issue.
`rails plugin new` with `--full` and `--skip-sprockets` options generates a dummy application that throws `NoMethodError`.
```
% rails plugin new my_engine -S --full --skip-gemspec
% cd my_engine
% bin/rails test
rails aborted!
NoMethodError: undefined method `assets' for #<Rails::Application::Configuration:0x007f83aa1e6100>
```
This removes the following warnings.
```ruby
rails/railties/lib/rails/railtie.rb:186: warning: instance variable @rake_tasks not initialized
rails/railties/lib/rails/railtie.rb:186: warning: instance variable @rake_tasks not initialized
rails/railties/lib/rails/railtie.rb:186: warning: instance variable @load_console not initialized
rails/railties/lib/rails/railtie.rb:186: warning: instance variable @rake_tasks not initialized
```
This reverts commit aaf561d26cf9a879ff40190b625155015c6225da, reversing
changes made to 83776676e54b387fc52081ff31cfd7710bae9e03.
Reason: It is not a good idea to connect to the database on generators
I'm sorry for causing #24168. I wasn't aware of --no-helper and --no-assets.
So I'm adding them to the help message.
I'm still not sure how to show `--no-test-framework` though.
The test framework should not be autoloaded in production mode. Before
this commit, the testing railtie would extend AS::TestCase. This caused
AS::TestCase to be preloaded regardless of the environment in which we
were running.
This commit just moves the code that adds line filtering support in to
the test command where we actually execute the test runner. That allows
us to maintain the line runner feature but only load the minimal amount
of code we need.
When the initial evented monitor feature was written, the latest version of listen
was the 3.0.x series. Since then the listen project has moved on to the 3.1.x series.
This patch allows the use of the new versions.
29f0fbd5db643b885665e4a14c7de3bf6c0d9c96 has changed that always call
`Minitest.autorun` in `active_support/testing/autorun`.
Therefore, when used directly `Minitest.run`, tests are performed twice.
Makes sure we get support for `Minitest.after_run` hooks and don't
accidentally run tests twice.
Requires overriding minitest `run` to respect the TESTOPTS variable.
The Rails test runner supports three ways to run tests: directly, via rake, or ruby.
When Running with Ruby ala `ruby -Itest test/models/post_test.rb` our test file would
be evaluated first, requiring `test_helper` and then `active_support/testing/autorun`
that would then require the test file (which it hadn't been before) thus reevaluating
it. This caused exceptions if using Active Support's declarative syntax.
Fix this by shifting around when we set the how we're run to closer mimick the require
order.
If we're running with `bin/rails test` the test command file is run first and we then
set `run_with_rails_extension`, later we hit `active_support/testing/autorun` and do
nothing — because we've been run elsewhere.
If we at this point haven't set `run_with_rails_extension` we've been running with
`ruby` this whole time and thus we set that.
We should always trigger `Minitest.autorun` as it doesn't hurt to call it twice.
Consolidate the two methods into a single one that better brings out the intent of
why they're there.
