Managed to reproduce CI failure at https://buildkite.com/rails/rails-nightly/builds/133#018d7bb8-8a32-4978-8e36-d7cb9b067813/1196-1204 . It would also reproduce against the released versions of Ruby because this is triggered by minitest v5.22.0 change. ebb468c81c
To avoid all of railties CI failures, pin minitest version to 5.21 tentatively.
* Steps to reproduce
```ruby
git clone https://github.com/rails/rails
cd rails
rm Gemfile.lock
bundle install
cd railties
bin/test test/application/test_runner_test.rb -n test_system_tests_are_not_run_with_the_default_test_command
```
* Expected behavior
It should pass.
* Actual behavior
```ruby
$ bin/test test/application/test_runner_test.rb -n test_system_tests_are_not_run_with_the_default_test_command
Run options: -n test_system_tests_are_not_run_with_the_default_test_command --seed 14574
F
Failure:
ApplicationTests::TestRunnerTest#test_system_tests_are_not_run_with_the_default_test_command [test/application/test_runner_test.rb:1191]:
Expected /0\ runs,\ 0\ assertions,\ 0\ failures,\ 0\ errors,\ 0\ skips/ to match "Nothing ran for filter: \nRunning 0 tests in a single process (parallelization threshold is 50)\nRun options: --seed 45713\n\n# Running:\n\n".
bin/test test/application/test_runner_test.rb:1179
Finished in 9.982314s, 0.1002 runs/s, 0.2004 assertions/s.
1 runs, 2 assertions, 1 failures, 0 errors, 0 skips
$
```
Followup: https://github.com/rails/rails/pull/50923
Instead of stopping on the first frame that isn't in
Action Dispatch, we should return the first frame that
isn't filtered by the backtrace cleaner.
The following was added to the `ActiveJob::Callbacks`,
`ActiveModel::Callbacks` and `AbstractController:Callbacks` docs
in #29072:
> NOTE: Calling the same callback multiple times will overwrite
> previous callback definitions.
The comment refers to "calling" callbacks but seems to be about defining
callbacks, as mentioned in the PR description.
In the ActiveJob and AbstractController docs, I believe this will be
misinterpreted as referring to setting callbacks, which, as far as I can
tell, does not have such a restriction.
In the ActiveModel docs, I believe it would be slightly clearer to
replace "calling" with "defining".
Minitest::Test does not support all of the same options as
ActiveSupport::TestCase, such as running bin/test <filename>:<lineno>
and -n /regex/. Trying to use these options on this file would just run
all of the Minitest::Tests no matter what options were passed.
This commit fixes the ability to use those options by using
ActiveSupport::TestCase (like every other test in the repo).
Before:
```
$ bin/test test/template/dependency_tracker_test.rb:217
Running 59 tests in parallel using 8 processes
Run options: --seed 42725
.........................................................
Finished in 0.322759s, 176.6024 runs/s, 176.6024 assertions/s.
57 runs, 57 assertions, 0 failures, 0 errors, 0 skips
```
After:
```
$ bin/test test/template/dependency_tracker_test.rb:217
Running 59 tests in parallel using 8 processes
Run options: --seed 15213
...
Finished in 0.359162s, 8.3528 runs/s, 8.3528 assertions/s.
3 runs, 3 assertions, 0 failures, 0 errors, 0 skips
```
The context of this example suggests what it really wants to
point to is `query_constraints` instead of `query_by`.
Also apply fixed-width font to the reference to the methods correctly.
The VideoAnalyzer uses ffprobe (part of ffmpeg) to determine the
rotation of a video. The VideoAnalyzer#angle returns tags["rotate"] if
it is available, else it attempts to get "rotation" from the Display
Matrix side_data.
However, ffprobe exhibits different behavior with different versions:
ffprobe version 4.4.2-0ubuntu on Ubuntu 22.04 returns a "rotate" : "90"
tag.
In contrast, ffprobe version 6.0-6ubuntu1 on Ubuntu 23.10 does not
return the "rotate" tag; instead, it determines the angle from the
display matrix side_data, which returns "rotation": -90.
To account for this difference, we now check for both values in the test.
The puma documentation and code seem to disagree about
whether `preload_app!` is the default depending on the number
of workers.
Worst case it's the default and making it explicit is no big deal.
* Switch ActiveSupport::TestCase teardown and setup callbacks to run in setup and teardown minitest lifecycle hooks.
Minitest provides `setup` and `teardown` lifecycle hooks to run code in. In general it is best practice to when defining your own test case class, to use `Minitest::TestCase.setup` and `Minitest::TestCase.teardown` instead of `before_setup` and `after_teardown`.
Per Minitest's Documentation on Lifecycle Hooks: https://docs.ruby-lang.org/en/2.1.0/MiniTest/Unit/LifecycleHooks.html
> before_setup()
> Runs before every test, before setup. This hook is meant for libraries to extend minitest. It is not meant to be used by test developers.
> after_teardown()
> Runs after every test, after teardown. This hook is meant for libraries to extend minitest. It is not meant to be used by test developers.
Since the `setup` and `teardown` ActiveSupport::TestCase callbacks are in essence user code, it makes sense to run during their corresponding Minitest Lifecycle hooks.
* Ensure test fixutres are torndown on errors in superclass after_teardown code.
By not adding wrapping the `teardown_fixtures` code, its possible that super raises an error and prevents the existing database transaction from rolling back.
`super` in general should only be calling `Minitest::Testcase.after_teardown` however, if another library were to override `Minitest::Testcase.after_teardown`, like the popular gem [rspec-mocks](https://github.com/rspec/rspec-mocks/blob/main/lib/rspec/mocks/minitest_integration.rb#L23) does, it causes all subsequent tests to retain any changes that were made to the database in the original test that errors.
* Remove unnecessary setup and teardown methods in tests
* update activesupport Changelog
* Fix linter issues in CHANGELOG
* fix tests with improper setup and teardown method definitions
* Fix final CHANGELOG lint
* Revert "Fix final CHANGELOG lint"
This reverts commit f30682eb62.
* Revert "fix tests with improper setup and teardown method definitions"
This reverts commit 1d5b88c873.
* Revert "Fix linter issues in CHANGELOG"
This reverts commit 60e89bd189.
* Revert "update activesupport Changelog"
This reverts commit 0f19bc324f.
* Revert "Remove unnecessary setup and teardown methods in tests"
This reverts commit e5673f179a.
* Revert "Switch ActiveSupport::TestCase teardown and setup callbacks to run in setup and teardown minitest lifecycle hooks."
This reverts commit d08d92d861.
* Rescue Minitest::Assertion errors in ActiveSupport::TestCase.teardown callback code to ensure all other after_teardown methods are called.
* Fix name of test class
* remove unused MyError class
* Fix module to not be in global namespace
Co-authored-by: Rafael Mendonça França <rafael@rubyonrails.org>
In recent Ruby versions some pure C functions have been moved into
some semi-ruby code and now have an `<internal:something>` location.
They should be silenced like stdlib etc.
This commit addresses CI failure
at https://buildkite.com/rails/rails-nightly/builds/108#018d57ac-4f2a-45f1-86b9-9015a7b0a463/1165-2002
* Error addressed by this commit
```
$ ruby -v
ruby 3.4.0dev (2024-01-30T10:19:23Z master 86547fd69d) [x86_64-linux]
$ rm Gemfile.lock
$ bundle install
$ cd activesupport
$ bin/test test/core_ext/object/to_query_test.rb -n test_hash_not_sorted_lexicographically_for_nested_structure
... snip ...
E
Error:
ToQueryTest#test_hash_not_sorted_lexicographically_for_nested_structure:
NameError: uninitialized constant ToQueryTest::URI
test/core_ext/object/to_query_test.rb:90:in `test_hash_not_sorted_lexicographically_for_nested_structure'
bin/test test/core_ext/object/to_query_test.rb:79
Finished in 0.000385s, 2597.3082 runs/s, 0.0000 assertions/s.
1 runs, 0 assertions, 0 failures, 1 errors, 0 skips
$
```
This behavior has been introduced since Ruby 3.4.0dev vendors URI from top level to `Gem::URI`
via d64d0b5423 .
Here is the RubyGems one.
https://github.com/rubygems/rubygems/pull/7386
Not that `inspect` is anywhere close to be a hostspot, but just
to not allow a less defficient pattern.
Instead of allocating an UnboundMethod and bind_call it, we can
directly replace the method.
* Add ".[key]" to the abstract namespaces [ci skip]
The section "Error Message Scopes" describes how the Rails look up the
namespaces for a validation error message with code snippets.
The first code snippet shows abstract namespaces like
`errors.attributes.[attribute_name]` while the second one demonstrates
the materialized one like `errors.attributes.name.blank`.
The first code snippet lacks the validation error "key." My team member
was confused and created a translation without "key", so I think it
would be helpful to add ".[key]" to the abstract namespaces.
* Move abstract key list after :blank examples
Co-authored-by: Rafael Mendonça França <rafael@rubyonrails.org>
Add a `compress` option to ActiveRecord::Encryption::Encryptor, which
defaults to `true`. When set to `false`, the encryptor will never
compress the data.
This is useful for cases where the data is already compressed.
This can be used with the `encryptor` option in the model:
```ruby
class Record < ApplicationRecord
encrypts :field, encryptor: ActiveRecord::Encryption::Encryptor.new(compress: false)
end
```
When the application has more than one database configuration, only
the primary was being loaded by default on boot time. All the other
connection pools were loading the schema cache lazily.
This logic can be found in:
351a8d9bc9/activerecord/lib/active_record/connection_adapters/pool_config.rb (L13)351a8d9bc9/activerecord/lib/active_record/railtie.rb (L149-L178)
The lazy code path wasn't using the same logic to determine the name
of the default schema cache file that the Railties uses, so it
was loading the schema cache of the primary config to all the other
configs. If no table name coincided nothing bad would happen, but if
table names coincided, the schema would be completly wrong in all
non-primary connections.
Yasuo Honda