Currently, we skip the Migration/EnsureFactoryForTable cop when
running in CE. This results in a Lint/RedundantCopDisableDirective
offense when there's a
"rubocop: disable Migration/EnsureFactoryForTable" comment.

To avoid this, we now run the Migration/EnsureFactoryForTable cop
when a disable comment is present. This prevents the redundant
disable directive offense.

Changelog: other

Moves the `Gitlab/DocUrl` cop to the
`Gitlab/DocumentationLinks/HardcodedUrl` namespace.

When `feature_category` is missing RuboCop set the whole
`RSpec.describe` as offended node which made some IDEs like RubyMine
highlight the complete file making it hard to read.

This MR ensures that only `RSpec.describe ...` (without `do`) is marked
making the offenses easier on the eyes.

Implement RuboCop rule to prevent misuse of allow_access_with_scope
method. This ensures the method is only used at the class level, not
nested in namespaces. Because, if this method is called multiple times
on the same class, the scopes are all aggregated

Changelog: other

As we want to be able to round trip this YAML, but unfortunate the Ruby
Psych library does not preserve comments.

To point to the correct example

This commit updates our existing cop to prevent adding columns
on large tables.

In the GitLab monolith codebase, we should use the `Gitlab::HTTP`
wrapper instead of `Gitlab::HTTP_V2` because we need to pass options
that are related to the GitLab instance setting.

We go through sizes in ascending order, so the largest partition sets
the final `table_size` for the table dictionary

We've stopped support for Ruby 3.1 some time ago.

As we want to enable this cop after fixing all the offenses.

This table is excessively large because it is a many-to-many
relationship between vulnerability findings and ci pipelines. We've
removed all usages of this table from the application, so it can now be
dropped and all code associated with it can be deleted.

Changelog: removed
EE: true

This commit includes IgnoreColumns by default in ApplicationRecord
in order to reduce the number of line we include while removing a column.
We seem to use this helper quite a lot, it might makes sense to include
it by default in all models.

This creates a cop that ensures that `help_page_path` usages link to
existing documentation.
As the documentation gets updated, it's easy to forget that we link to
it from many parts of the product. Those links can get outdated as
documented are removed, section moved around, titles changed, etc.
This cop should ensure the product links stay in sync with the
documentation.

We should use .md in all links to markdown documentation
files for better maintainability.

Changelog: changed
EE: true

Changelog: other

This enforces some of the guidelines added around use of `ID` type
added in https://gitlab.com/gitlab-org/gitlab/-/merge_requests/152002.

- placing an ee only class under an ee/**/ee subdirectory
  is not the right placement and we should alert on it.
- see https://docs.gitlab.com/ee/development/ee_features.html#extend-ce-features-with-ee-backend-code
- see https://gitlab.com/gitlab-org/gitlab/-/issues/378167

MR: https://gitlab.com/gitlab-org/gitlab/-/merge_requests/165146

MR: https://gitlab.com/gitlab-org/gitlab/-/merge_requests/165640

Changelog: removed

This new batched migration will add partition_id in the where
clause in order to leverage partition pruning and not
scanning all partitions.

`...` matches on zero arguments which causes it to flag things like
`build.id`. `_` matches on at least one argument, so it should only flag
things like `build(:model)`.

Co-authored-by: Vincy Wilson <vwilson@gitlab.com>

Like for `Types` and `Resolvers` we ignore the namespace
`Types::PermissionTypes` for now, considering anything under
that as required to be organized per bounded context.

This commit adds a new RuboCop rule `Gitlab/NoFindInWorkers` to prevent
the use of `.find` in workers. It also enabled "grace period" and generated TODOs.

Co-authored-by: Peter Leitzen <pleitzen@gitlab.com>

Now that 'audit events' is sentence-cased in the UI, let's make the
docs and other ancillary assets match.

Let's clean up the rest of these and bring them to sentence case.

Add a graphql API to return the multimatches in a single file.
So that frontend can display the multimatches in a single file
at one place. API will raie an error if performed other than for Zoekt.

Changelog: added
MR: https://gitlab.com/gitlab-org/gitlab/-/merge_requests/155984
EE: true

This reverts merge request !158822

Changelog: changed