gitlab-org--gitlab-foss/app/models/namespace.rb

244 lines
6.9 KiB
Ruby
Raw Normal View History

class Namespace < ActiveRecord::Base
acts_as_paranoid without_default_scope: true
include CacheMarkdownField
include Sortable
include Gitlab::ShellAdapter
include Gitlab::CurrentSettings
include Gitlab::VisibilityLevel
include Routable
include AfterCommitQueue
include Storage::LegacyNamespace
# Prevent users from creating unreasonably deep level of nesting.
# The number 20 was taken based on maximum nesting level of
# Android repo (15) + some extra backup.
NUMBER_OF_ANCESTORS_ALLOWED = 20
cache_markdown_field :description, pipeline: :description
has_many :projects, dependent: :destroy # rubocop:disable Cop/ActiveRecordDependent
has_many :project_statistics
belongs_to :owner, class_name: "User"
belongs_to :parent, class_name: "Namespace"
has_many :children, class_name: "Namespace", foreign_key: :parent_id
has_one :chat_team, dependent: :destroy # rubocop:disable Cop/ActiveRecordDependent
validates :owner, presence: true, unless: ->(n) { n.type == "Group" }
2015-02-03 00:15:44 -05:00
validates :name,
2015-12-07 16:17:24 -05:00
presence: true,
uniqueness: { scope: :parent_id },
length: { maximum: 255 },
namespace_name: true
2015-02-03 00:15:44 -05:00
validates :description, length: { maximum: 255 }
2015-02-03 00:15:44 -05:00
validates :path,
2015-12-07 16:17:12 -05:00
presence: true,
length: { maximum: 255 },
dynamic_path: true
validate :nesting_level_allowed
delegate :name, to: :owner, allow_nil: true, prefix: true
after_commit :refresh_access_of_projects_invited_groups, on: :update, if: -> { previous_changes.key?('share_with_group_lock') }
before_create :sync_share_with_group_lock_with_parent
before_update :sync_share_with_group_lock_with_parent, if: :parent_changed?
2017-09-05 20:10:30 -04:00
after_update :force_share_with_group_lock_on_descendants, if: -> { share_with_group_lock_changed? && share_with_group_lock? }
# Legacy Storage specific hooks
after_update :move_dir, if: :path_changed?
before_destroy(prepend: true) { prepare_for_destroy }
after_destroy :rm_dir
scope :for_user, -> { where('type IS NULL') }
2012-11-22 23:24:09 -05:00
scope :with_statistics, -> do
joins('LEFT JOIN project_statistics ps ON ps.namespace_id = namespaces.id')
.group('namespaces.id')
.select(
'namespaces.*',
'COALESCE(SUM(ps.storage_size), 0) AS storage_size',
'COALESCE(SUM(ps.repository_size), 0) AS repository_size',
'COALESCE(SUM(ps.lfs_objects_size), 0) AS lfs_objects_size',
'COALESCE(SUM(ps.build_artifacts_size), 0) AS build_artifacts_size'
)
end
class << self
def by_path(path)
2015-12-14 21:53:52 -05:00
find_by('lower(path) = :value', value: path.downcase)
end
# Case insensetive search for namespace by path or name
def find_by_path_or_name(path)
find_by("lower(path) = :path OR lower(name) = :path", path: path.downcase)
end
# Searches for namespaces matching the given query.
#
# This method uses ILIKE on PostgreSQL and LIKE on MySQL.
#
# query - The search query as a String
#
# Returns an ActiveRecord::Relation
def search(query)
t = arel_table
pattern = "%#{query}%"
where(t[:name].matches(pattern).or(t[:path].matches(pattern)))
end
def clean_path(path)
path = path.dup
2015-04-21 09:49:16 -04:00
# Get the email username by removing everything after an `@` sign.
path.gsub!(/@.*\z/, "")
2015-04-21 09:49:16 -04:00
# Remove everything that's not in the list of allowed characters.
path.gsub!(/[^a-zA-Z0-9_\-\.]/, "")
# Remove trailing violations ('.atom', '.git', or '.')
path.gsub!(/(\.atom|\.git|\.)*\z/, "")
# Remove leading violations ('-')
path.gsub!(/\A\-+/, "")
# Users with the great usernames of "." or ".." would end up with a blank username.
# Work around that by setting their username to "blank", followed by a counter.
path = "blank" if path.blank?
uniquify = Uniquify.new
uniquify.string(path) { |s| Namespace.find_by_path_or_name(s) }
end
end
def visibility_level_field
:visibility_level
end
def to_param
full_path
end
def human_name
owner_name
end
def any_project_has_container_registry_tags?
all_projects.any?(&:has_container_registry_tags?)
end
def send_update_instructions
projects.each do |project|
project.send_move_instructions("#{full_path_was}/#{project.path}")
end
end
def kind
type == 'Group' ? 'group' : 'user'
end
def find_fork_of(project)
2015-12-14 21:53:52 -05:00
projects.joins(:forked_project_link).find_by('forked_project_links.forked_from_project_id = ?', project.id)
end
def lfs_enabled?
# User namespace will always default to the global setting
Gitlab.config.lfs.enabled
end
def shared_runners_enabled?
projects.with_shared_runners.any?
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
# Returns all the ancestors of the current namespaces.
def ancestors
return self.class.none unless parent_id
2017-06-21 09:48:12 -04:00
Gitlab::GroupHierarchy
.new(self.class.where(id: parent_id))
.base_and_ancestors
end
def self_and_ancestors
return self.class.where(id: id) unless parent_id
Gitlab::GroupHierarchy
.new(self.class.where(id: id))
.base_and_ancestors
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
# Returns all the descendants of the current namespace.
def descendants
2017-06-21 09:48:12 -04:00
Gitlab::GroupHierarchy
.new(self.class.where(parent_id: id))
.base_and_descendants
end
def self_and_descendants
Gitlab::GroupHierarchy
.new(self.class.where(id: id))
.base_and_descendants
end
def user_ids_for_project_authorizations
[owner_id]
end
def parent_changed?
parent_id_changed?
end
# Includes projects from this namespace and projects from all subgroups
# that belongs to this namespace
def all_projects
Project.inside_path(full_path)
end
def has_parent?
parent.present?
end
def subgroup?
has_parent?
end
def soft_delete_without_removing_associations
# We can't use paranoia's `#destroy` since this will hard-delete projects.
# Project uses `pending_delete` instead of the acts_as_paranoia gem.
self.deleted_at = Time.now
end
private
def refresh_access_of_projects_invited_groups
2017-06-21 09:48:12 -04:00
Group
.joins(project_group_links: :project)
.where(projects: { namespace_id: id })
.find_each(&:refresh_members_authorized_projects)
end
def nesting_level_allowed
if ancestors.count > Group::NUMBER_OF_ANCESTORS_ALLOWED
errors.add(:parent_id, "has too deep level of nesting")
end
end
def sync_share_with_group_lock_with_parent
2017-09-05 20:10:30 -04:00
if parent&.share_with_group_lock?
self.share_with_group_lock = true
end
end
def force_share_with_group_lock_on_descendants
return unless Group.supports_nested_groups?
# We can't use `descendants.update_all` since Rails will throw away the WITH
# RECURSIVE statement. We also can't use WHERE EXISTS since we can't use
# different table aliases, hence we're just using WHERE IN. Since we have a
# maximum of 20 nested groups this should be fine.
Namespace.where(id: descendants.select(:id))
.update_all(share_with_group_lock: true)
end
end