`...` generates an anonymous block, it's basically a shortcut
for `*, **, &`. So to look more similar to tools that introspect
method signatures, it's best to continue to use an anonymous block.
Previously, only the PrismRenderParser or RipperRenderParser would be
tested depending on if the Prism gem is available. This meant that
PrismRenderParser was being tested on Ruby 3.3 and RipperRenderParser
was tested on Ruby < 3.3. Additionally, if someone were to add prism to
the rails/rails Gemfile because they wrote a tool that uses it then the
RipperRenderParser would end up completely untested.
This commit is a small refactor to enable testing both RenderParsers in
all Ruby versions so that the prism gem can be added to the Gemfile.
Generally the idea is:
- use Prism to parse the file into AST + Comments
- transform each comment block into plain RDoc (instead of RDoc in a
comment)
- use RDoc's ToMarkdown class to get a Markdown representation of the
comment
- transform the Markdown representation back into a comment
- write a new file, skipping the lines that were previously RDoc
comments and instead inserting the new Markdown comments
A little extra work has to be down for metaprogrammed documentation
because ToMarkdown turns RDoc directives into H1s. So for these cases,
the directive is first split off the top before doing the ToMarkdown
transformation and then added back afterwards.
In SQLite3-Ruby version 2.0, I would like to make row arrays frozen. I
tested the change, and it only breaks this test, so I'm changing the
test. I don't think we should be mutating the objects that the database
adapter returns
At GitHub we get a fair number of Trilogy `ETIMEDOUT` errors (for known
reasons that we might be able to improve somewhat, but I doubt we'll
make them go away entirely). These are very much retryable network
errors, so it'd be handy if these `ETIMEDOUT` errors were translated to
`ConnectionFailed` instead of `StatementInvalid`, making them
`retryable_connection_error`s.
ed2bc92b82/activerecord/lib/active_record/connection_adapters/abstract_adapter.rb (L1077)
We're already translating `ECONNRESET` (via matching on the error
message) and `EPIPE` to `ConnectionFailed`. Rather than adding another
case, this commit treats all of the Trilogy `SystemCallError` subclasses
as `ConnectionFailed`.
This requires bumping trilogy 2.7 so we can get
https://github.com/trilogy-libraries/trilogy/pull/143
Fixes https://github.com/rails/rails/issues/50853
The video analyzer was relying on the positional reference of the Display
Matrix side_data to fetch the rotation value. However, the side_data is
not guaranteed to be in the same position. For instance, HDR videos shot
on iOS have "DOVI configuration record" side_data in the first position,
followed by the "Display Matrix" side data containing the rotation value.
This fix removes the positional reference and explicitely searches for
the "Display Matrix" side_data to retrieve the rotation angle.
The code: Book.joins(:author, :reviews)
This code will return books that have an author and at least 1
review.
English text implies it returns the book and the author:
"return all books with their author .."
Changing the text to imply it returns books with an author.
Closes [#49574][]
Issue #49574 outlines how an Active Record join model accessed through a
`has_many ..., through:` association is unable to infer an appropriate
`:inverse_of` association by pluralizing a predictably pluralizable
class name.
This commit resolves that issue by also checking a model's reflections
for a pluralized inverse name in addition to whatever's provided through
the `:as` option or inferred in the singular.
The issue shares a code sample:
```ruby
ActiveRecord::Schema.define do
create_table "listings", force: :cascade do |t|
t.bigint "list_id", null: false
t.bigint "pin_id", null: false
end
create_table "lists", force: :cascade do |t|
end
create_table "pins", force: :cascade do |t|
end
end
class Pin < ActiveRecord::Base
has_many :listings
end
class List < ActiveRecord::Base
has_many :listings
has_many :pins, through: :listings
end
class Listing < ActiveRecord::Base
belongs_to :list
belongs_to :pin
end
class BugTest < Minitest::Test
def test_association_stuff
list = List.create!
pin = list.pins.new
pin.save!
assert_equal [pin], list.reload.pins
assert_equal 1, list.reload.pins.count
end
end
```
Unfortunately, there isn't a one-to-one mapping in the test suite's
`test/model` directory for this type of structure. The most similar
associations were between the `Book`, `Subscription`, and `Subscriber`.
For the sake of ease, this commit wraps the test block in a new
`skipping_validations_for` helper method to ignore validations declared
on the `Subscription` join table model.
[#49574]: https://github.com/rails/rails/issues/49574
Following the discussion in #50770, the new format will be:
`[dasherized-app-name]([colorized-env])>`
For example, if the app's module name is `MyApp`, the prompt will be:
`my-app(dev)>`, where the `dev` part will be colored.
Update railties/lib/rails/commands/console/console_command.rb
Co-authored-by: Jean Boussier <jean.boussier@gmail.com>
Update changelog
Setting `TERM=dumb` is enough to disable colorization in those tests.
Mocked app should also load console helpers
Move console method extension to IRBConsole's constructor
Test IRB autocompletion control without mutating ENV
Control IRB autocompletion without mutating ENV
Correct local variable name
Avoid corrupting the cache by mutating the return value, and also
sligthly reduce memory usage when the quoting format often return
an unmodified string.
- This is a fix needed to unblock
https://github.com/rails/rails/pull/50284,
because Active Storage relies on a Active Record bug.
The problem is best understood with a small snippet:
```
blob = ActiveStorage::Blob.new
ActiveStorage::Attachment.new(blob: blob)
ActiveStorage::Attachment.new(blob: blob)
# Currently:
p blob.attachments #=> #<ActiveRecord::Associations::CollectionProxy []>
# Once the Active Record bug is fixed:
p blob.attachments #=> #<ActiveRecord::Associations::CollectionProxy [#<ActiveStorage::Attachment id: nil, name: nil, record_type: nil, record_id: nil, blob_id: nil, created_at: nil>, #<ActiveStorage::Attachment id: nil, name: nil, record_type: nil, record_id: nil, blob_id: nil, created_at: nil>]>
# Trying to save the blob would result in trying to create 2 attachments which
# fails because of unique constrainsts.
```
### Code path
The real code path that does what the snippet above does is located here:
9c3ffab47c/activestorage/lib/active_storage/attached/many.rb (L52)
It's basically doing this:
```
user.images.attach "photo1.png"
# Initialize a Blob record and an Attachment
user.images.attach "photo2.png"
# Get the Blob from above, create another Attachment
# Initialize a new Blob record and an new Attachment
# rinse and repeat every time `attach` is called
```
Basically each time we call `attach`, we grab the previous blobs that were attached
(and that already have an Attachment record), and
### Solution
- Explicitly set the `inverse_of`, so that it behaves as if #50284 is shipped
- Don't build a new attachment for blob already having one.
### Tests
I didn't add tests, the test suite is already well covered, adding the `inverse_of`
without any changes breaks the test suite already. You can try by running
for instance the `activestorage/test/models/attached/many_test.rb`.
When Rake parses an argument string with no tasks, it sets the top-level
task as "default". Prior to this commit, if no default task was defined
(for example, if an app was generated with `--skip-test` and didn't
define its own default task), `Rails::Command::RakeCommand` would raise
`UnrecognizedCommandError`, preventing Rake from displaying the task
list.
This commit changes `Rails::Command::RakeCommand` to let Rake handle the
"default" task.
Fixes#50700.
Jean Boussier