gitlab-org--gitlab-foss/spec/spec_helper.rb

341 lines
10 KiB
Ruby
Raw Normal View History

require './spec/simplecov_env'
SimpleCovEnv.start!
ENV["RAILS_ENV"] = 'test'
ENV["IN_MEMORY_APPLICATION_SETTINGS"] = 'true'
2018-07-02 06:43:06 -04:00
require File.expand_path('../config/environment', __dir__)
2014-03-13 18:34:12 -04:00
require 'rspec/rails'
2015-06-10 01:38:36 -04:00
require 'shoulda/matchers'
2015-11-12 04:52:20 -05:00
require 'rspec/retry'
require 'rspec-parameterized'
2016-05-21 21:17:15 -04:00
rspec_profiling_is_configured =
ENV['RSPEC_PROFILING_POSTGRES_URL'].present? ||
ENV['RSPEC_PROFILING']
branch_can_be_profiled =
ENV['GITLAB_DATABASE'] == 'postgresql' &&
(ENV['CI_COMMIT_REF_NAME'] == 'master' ||
ENV['CI_COMMIT_REF_NAME'] =~ /rspec-profile/)
if rspec_profiling_is_configured && (!ENV.key?('CI') || branch_can_be_profiled)
require 'rspec_profiling/rspec'
end
if ENV['CI'] && ENV['KNAPSACK_GENERATE_REPORT'] && !ENV['NO_KNAPSACK']
2016-06-08 05:43:13 -04:00
require 'knapsack'
Knapsack::Adapters::RSpecAdapter.bind
end
2014-03-13 18:34:12 -04:00
# require rainbow gem String monkeypatch, so we can test SystemChecks
require 'rainbow/ext/string'
Rainbow.enabled = false
2014-03-13 18:34:12 -04:00
# Requires supporting ruby files with custom matchers and macros, etc,
# in spec/support/ and its subdirectories.
# Requires helpers, and shared contexts/examples first since they're used in other support files
# Load these first since they may be required by other helpers
require Rails.root.join("spec/support/helpers/git_helpers.rb")
# Then the rest
Dir[Rails.root.join("spec/support/helpers/*.rb")].each { |f| require f }
Dir[Rails.root.join("spec/support/shared_contexts/*.rb")].each { |f| require f }
Dir[Rails.root.join("spec/support/shared_examples/*.rb")].each { |f| require f }
Dir[Rails.root.join("spec/support/**/*.rb")].each { |f| require f }
2014-03-13 18:34:12 -04:00
quality_level = Quality::TestLevel.new
2014-03-13 18:34:12 -04:00
RSpec.configure do |config|
config.use_transactional_fixtures = true
config.use_instantiated_fixtures = false
config.fixture_path = Rails.root
2014-03-13 18:34:12 -04:00
2015-11-12 04:52:20 -05:00
config.verbose_retry = true
config.display_try_failure_messages = true
config.infer_spec_type_from_file_location!
2019-05-03 02:49:45 -04:00
config.full_backtrace = !!ENV['CI']
config.define_derived_metadata(file_path: %r{(ee)?/spec/.+_spec\.rb\z}) do |metadata|
location = metadata[:location]
metadata[:level] = quality_level.level_for(location)
metadata[:api] = true if location =~ %r{/spec/requests/api/}
# do not overwrite type if it's already set
next if metadata.key?(:type)
match = location.match(%r{/spec/([^/]+)/})
metadata[:type] = match[1].singularize.to_sym if match
end
config.include LicenseHelpers
config.include ActiveJob::TestHelper
config.include ActiveSupport::Testing::TimeHelpers
config.include CycleAnalyticsHelpers
config.include ExpectOffense
config.include FactoryBot::Syntax::Methods
config.include FixtureHelpers
config.include GitlabRoutingHelper
config.include StubFeatureFlags
config.include StubGitlabCalls
config.include StubGitlabData
config.include ExpectNextInstanceOf
config.include TestEnv
config.include Devise::Test::ControllerHelpers, type: :controller
config.include Devise::Test::IntegrationHelpers, type: :feature
config.include LoginHelpers, type: :feature
config.include SearchHelpers, type: :feature
config.include WaitHelpers, type: :feature
config.include EmailHelpers, :mailer, type: :mailer
config.include Warden::Test::Helpers, type: :request
config.include Gitlab::Routing, type: :routing
config.include Devise::Test::ControllerHelpers, type: :view
config.include ApiHelpers, :api
config.include CookieHelper, :js
config.include InputHelper, :js
config.include SelectionHelper, :js
config.include InspectRequests, :js
config.include WaitForRequests, :js
config.include LiveDebugger, :js
config.include MigrationsHelpers, :migration
config.include RedisHelpers
2018-06-14 09:33:35 -04:00
config.include Rails.application.routes.url_helpers, type: :routing
config.include PolicyHelpers, type: :policy
config.include MemoryUsageHelper
config.include ExpectRequestWithStatus, type: :request
2014-03-13 18:34:12 -04:00
if ENV['CI']
# This includes the first try, i.e. tests will be run 4 times before failing.
config.default_retry_count = 4
config.reporter.register_listener(
RspecFlaky::Listener.new,
:example_passed,
:dump_summary)
end
2014-03-13 18:34:12 -04:00
config.before(:suite) do
Timecop.safe_mode = true
2014-07-31 12:24:53 -04:00
TestEnv.init
2014-03-13 18:34:12 -04:00
end
config.after(:all) do
TestEnv.clean_test_path
end
config.before do |example|
# Enable all features by default for testing
allow(Feature).to receive(:enabled?) { true }
enabled = example.metadata[:enable_rugged].present?
# Disable Rugged features by default
Gitlab::Git::RuggedImpl::Repository::FEATURE_FLAGS.each do |flag|
allow(Feature).to receive(:enabled?).with(flag).and_return(enabled)
end
allow(Gitlab::GitalyClient).to receive(:can_use_disk?).and_return(enabled)
# The following can be removed when we remove the staged rollout strategy
# and we can just enable it using instance wide settings
# (ie. ApplicationSetting#auto_devops_enabled)
allow(Feature).to receive(:enabled?)
.with(:force_autodevops_on_by_default, anything)
.and_return(false)
Gitlab::ThreadMemoryCache.cache_backend.clear
end
config.around(:example, :quarantine) do
# Skip tests in quarantine unless we explicitly focus on them.
skip('In quarantine') unless config.inclusion_filter[:quarantine]
end
2017-06-09 13:12:51 -04:00
config.before(:example, :request_store) do
RequestStore.begin!
end
config.after(:example, :request_store) do
RequestStore.end!
RequestStore.clear!
end
2018-12-11 01:29:58 -05:00
config.after do
2018-06-05 10:17:04 -04:00
Fog.unmock! if Fog.mock?
end
2018-12-11 01:29:58 -05:00
config.after do
Gitlab::CurrentSettings.clear_in_memory_application_settings!
end
config.before(:example, :mailer) do
reset_delivered_emails!
end
config.before(:example, :prometheus) do
matching_files = File.join(::Prometheus::Client.configuration.multiprocess_files_dir, "*.db")
Dir[matching_files].map { |filename| File.delete(filename) if File.file?(filename) }
Gitlab::Metrics.reset_registry!
end
config.around(:each, :use_clean_rails_memory_store_caching) do |example|
caching_store = Rails.cache
Rails.cache = ActiveSupport::Cache::MemoryStore.new
example.run
Rails.cache = caching_store
end
config.around(:each, :clean_gitlab_redis_cache) do |example|
redis_cache_cleanup!
example.run
redis_cache_cleanup!
end
config.around(:each, :clean_gitlab_redis_shared_state) do |example|
redis_shared_state_cleanup!
example.run
redis_shared_state_cleanup!
end
config.around(:each, :clean_gitlab_redis_queues) do |example|
redis_queues_cleanup!
example.run
redis_queues_cleanup!
end
Use CTEs for nested groups and authorizations This commit introduces the usage of Common Table Expressions (CTEs) to efficiently retrieve nested group hierarchies, without having to rely on the "routes" table (which is an _incredibly_ inefficient way of getting the data). This requires a patch to ActiveRecord (found in the added initializer) to work properly as ActiveRecord doesn't support WITH statements properly out of the box. Unfortunately MySQL provides no efficient way of getting nested groups. For example, the old routes setup could easily take 5-10 seconds depending on the amount of "routes" in a database. Providing vastly different logic for both MySQL and PostgreSQL will negatively impact the development process. Because of this the various nested groups related methods return empty relations when used in combination with MySQL. For project authorizations the logic is split up into two classes: * Gitlab::ProjectAuthorizations::WithNestedGroups * Gitlab::ProjectAuthorizations::WithoutNestedGroups Both classes get the fresh project authorizations (= as they should be in the "project_authorizations" table), including nested groups if PostgreSQL is used. The logic of these two classes is quite different apart from their public interface. This complicates development a bit, but unfortunately there is no way around this. This commit also introduces Gitlab::GroupHierarchy. This class can be used to get the ancestors and descendants of a base relation, or both by using a UNION. This in turn is used by methods such as: * Namespace#ancestors * Namespace#descendants * User#all_expanded_groups Again this class relies on CTEs and thus only works on PostgreSQL. The Namespace methods will return an empty relation when MySQL is used, while User#all_expanded_groups will return only the groups a user is a direct member of. Performance wise the impact is quite large. For example, on GitLab.com Namespace#descendants used to take around 580 ms to retrieve data for a particular user. Using CTEs we are able to reduce this down to roughly 1 millisecond, returning the exact same data. == On The Fly Refreshing Refreshing of authorizations on the fly (= when users.authorized_projects_populated was not set) is removed with this commit. This simplifies the code, and ensures any queries used for authorizations are not mutated because they are executed in a Rails scope (e.g. Project.visible_to_user). This commit includes a migration to schedule refreshing authorizations for all users, ensuring all of them have their authorizations in place. Said migration schedules users in batches of 5000, with 5 minutes between every batch to smear the load around a bit. == Spec Changes This commit also introduces some changes to various specs. For example, some specs for ProjectTeam assumed that creating a personal project would _not_ lead to the owner having access, which is incorrect. Because we also no longer refresh authorizations on the fly for new users some code had to be added to the "empty_project" factory. This chunk of code ensures that the owner's permissions are refreshed after creating the project, something that is normally done in Projects::CreateService.
2017-04-24 11:19:22 -04:00
config.around(:each, :use_clean_rails_memory_store_fragment_caching) do |example|
caching_store = ActionController::Base.cache_store
ActionController::Base.cache_store = ActiveSupport::Cache::MemoryStore.new
ActionController::Base.perform_caching = true
example.run
ActionController::Base.perform_caching = false
ActionController::Base.cache_store = caching_store
end
config.around(:each, :use_sql_query_cache) do |example|
ActiveRecord::Base.cache do
example.run
end
end
# The :each scope runs "inside" the example, so this hook ensures the DB is in the
# correct state before any examples' before hooks are called. This prevents a
# problem where `ScheduleIssuesClosedAtTypeChange` (or any migration that depends
# on background migrations being run inline during test setup) can be broken by
# altering Sidekiq behavior in an unrelated spec like so:
#
# around do |example|
# Sidekiq::Testing.fake! do
# example.run
# end
# end
config.before(:context, :migration) do
schema_migrate_down!
end
# Each example may call `migrate!`, so we must ensure we are migrated down every time
config.before(:each, :migration) do
use_fake_application_settings
schema_migrate_down!
end
config.after(:context, :migration) do
schema_migrate_up!
Gitlab::CurrentSettings.clear_in_memory_application_settings!
end
Use CTEs for nested groups and authorizations This commit introduces the usage of Common Table Expressions (CTEs) to efficiently retrieve nested group hierarchies, without having to rely on the "routes" table (which is an _incredibly_ inefficient way of getting the data). This requires a patch to ActiveRecord (found in the added initializer) to work properly as ActiveRecord doesn't support WITH statements properly out of the box. Unfortunately MySQL provides no efficient way of getting nested groups. For example, the old routes setup could easily take 5-10 seconds depending on the amount of "routes" in a database. Providing vastly different logic for both MySQL and PostgreSQL will negatively impact the development process. Because of this the various nested groups related methods return empty relations when used in combination with MySQL. For project authorizations the logic is split up into two classes: * Gitlab::ProjectAuthorizations::WithNestedGroups * Gitlab::ProjectAuthorizations::WithoutNestedGroups Both classes get the fresh project authorizations (= as they should be in the "project_authorizations" table), including nested groups if PostgreSQL is used. The logic of these two classes is quite different apart from their public interface. This complicates development a bit, but unfortunately there is no way around this. This commit also introduces Gitlab::GroupHierarchy. This class can be used to get the ancestors and descendants of a base relation, or both by using a UNION. This in turn is used by methods such as: * Namespace#ancestors * Namespace#descendants * User#all_expanded_groups Again this class relies on CTEs and thus only works on PostgreSQL. The Namespace methods will return an empty relation when MySQL is used, while User#all_expanded_groups will return only the groups a user is a direct member of. Performance wise the impact is quite large. For example, on GitLab.com Namespace#descendants used to take around 580 ms to retrieve data for a particular user. Using CTEs we are able to reduce this down to roughly 1 millisecond, returning the exact same data. == On The Fly Refreshing Refreshing of authorizations on the fly (= when users.authorized_projects_populated was not set) is removed with this commit. This simplifies the code, and ensures any queries used for authorizations are not mutated because they are executed in a Rails scope (e.g. Project.visible_to_user). This commit includes a migration to schedule refreshing authorizations for all users, ensuring all of them have their authorizations in place. Said migration schedules users in batches of 5000, with 5 minutes between every batch to smear the load around a bit. == Spec Changes This commit also introduces some changes to various specs. For example, some specs for ProjectTeam assumed that creating a personal project would _not_ lead to the owner having access, which is incorrect. Because we also no longer refresh authorizations on the fly for new users some code had to be added to the "empty_project" factory. This chunk of code ensures that the owner's permissions are refreshed after creating the project, something that is normally done in Projects::CreateService.
2017-04-24 11:19:22 -04:00
config.around(:each, :nested_groups) do |example|
example.run if Group.supports_nested_objects?
Use CTEs for nested groups and authorizations This commit introduces the usage of Common Table Expressions (CTEs) to efficiently retrieve nested group hierarchies, without having to rely on the "routes" table (which is an _incredibly_ inefficient way of getting the data). This requires a patch to ActiveRecord (found in the added initializer) to work properly as ActiveRecord doesn't support WITH statements properly out of the box. Unfortunately MySQL provides no efficient way of getting nested groups. For example, the old routes setup could easily take 5-10 seconds depending on the amount of "routes" in a database. Providing vastly different logic for both MySQL and PostgreSQL will negatively impact the development process. Because of this the various nested groups related methods return empty relations when used in combination with MySQL. For project authorizations the logic is split up into two classes: * Gitlab::ProjectAuthorizations::WithNestedGroups * Gitlab::ProjectAuthorizations::WithoutNestedGroups Both classes get the fresh project authorizations (= as they should be in the "project_authorizations" table), including nested groups if PostgreSQL is used. The logic of these two classes is quite different apart from their public interface. This complicates development a bit, but unfortunately there is no way around this. This commit also introduces Gitlab::GroupHierarchy. This class can be used to get the ancestors and descendants of a base relation, or both by using a UNION. This in turn is used by methods such as: * Namespace#ancestors * Namespace#descendants * User#all_expanded_groups Again this class relies on CTEs and thus only works on PostgreSQL. The Namespace methods will return an empty relation when MySQL is used, while User#all_expanded_groups will return only the groups a user is a direct member of. Performance wise the impact is quite large. For example, on GitLab.com Namespace#descendants used to take around 580 ms to retrieve data for a particular user. Using CTEs we are able to reduce this down to roughly 1 millisecond, returning the exact same data. == On The Fly Refreshing Refreshing of authorizations on the fly (= when users.authorized_projects_populated was not set) is removed with this commit. This simplifies the code, and ensures any queries used for authorizations are not mutated because they are executed in a Rails scope (e.g. Project.visible_to_user). This commit includes a migration to schedule refreshing authorizations for all users, ensuring all of them have their authorizations in place. Said migration schedules users in batches of 5000, with 5 minutes between every batch to smear the load around a bit. == Spec Changes This commit also introduces some changes to various specs. For example, some specs for ProjectTeam assumed that creating a personal project would _not_ lead to the owner having access, which is incorrect. Because we also no longer refresh authorizations on the fly for new users some code had to be added to the "empty_project" factory. This chunk of code ensures that the owner's permissions are refreshed after creating the project, something that is normally done in Projects::CreateService.
2017-04-24 11:19:22 -04:00
end
config.around(:each, :postgresql) do |example|
example.run if Gitlab::Database.postgresql?
end
config.around(:each, :mysql) do |example|
example.run if Gitlab::Database.mysql?
end
# This makes sure the `ApplicationController#can?` method is stubbed with the
# original implementation for all view specs.
config.before(:each, type: :view) do
allow(view).to receive(:can?) do |*args|
Ability.allowed?(*args)
end
end
2018-01-03 03:07:03 -05:00
config.before(:each, :http_pages_enabled) do |_|
allow(Gitlab.config.pages).to receive(:external_http).and_return(['1.1.1.1:80'])
end
config.before(:each, :https_pages_enabled) do |_|
allow(Gitlab.config.pages).to receive(:external_https).and_return(['1.1.1.1:443'])
end
config.before(:each, :http_pages_disabled) do |_|
allow(Gitlab.config.pages).to receive(:external_http).and_return(false)
end
config.before(:each, :https_pages_disabled) do |_|
allow(Gitlab.config.pages).to receive(:external_https).and_return(false)
end
# We can't use an `around` hook here because the wrapping transaction
# is not yet opened at the time that is triggered
config.prepend_before do
Gitlab::Database.set_open_transactions_baseline
end
config.append_after do
Gitlab::Database.reset_open_transactions_baseline
end
2013-03-26 05:13:30 -04:00
end
# add simpler way to match asset paths containing digest strings
RSpec::Matchers.define :match_asset_path do |expected|
match do |actual|
path = Regexp.escape(expected)
extname = Regexp.escape(File.extname(expected))
digest_regex = Regexp.new(path.sub(extname, "(?:-\\h+)?#{extname}") << '$')
digest_regex =~ actual
end
failure_message do |actual|
"expected that #{actual} would include an asset path for #{expected}"
end
failure_message_when_negated do |actual|
"expected that #{actual} would not include an asset path for #{expected}"
end
end
FactoryBot::SyntaxRunner.class_eval do
2015-10-05 06:02:26 -04:00
include RSpec::Mocks::ExampleMethods
end
ActiveRecord::Migration.maintain_test_schema!
Shoulda::Matchers.configure do |config|
config.integrate do |with|
with.test_framework :rspec
with.library :rails
end
end
# Prevent Rugged from picking up local developer gitconfig.
Rugged::Settings['search_path_global'] = Rails.root.join('tmp/tests').to_s