gitlab-org--gitlab-foss/app/services/event_create_service.rb

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

272 lines
8.9 KiB
Ruby
Raw Normal View History

# frozen_string_literal: true
# EventCreateService class
#
# Used for creating events feed on dashboard after certain user action
#
# Ex.
# EventCreateService.new.new_issue(issue, current_user)
#
class EventCreateService
IllegalActionError = Class.new(StandardError)
def open_issue(issue, current_user)
create_record_event(issue, current_user, :created)
end
def close_issue(issue, current_user)
create_record_event(issue, current_user, :closed)
end
def reopen_issue(issue, current_user)
create_record_event(issue, current_user, :reopened)
end
def open_mr(merge_request, current_user)
create_record_event(merge_request, current_user, :created).tap do
track_event(event_action: :created, event_target: MergeRequest, author_id: current_user.id)
track_snowplow_event(merge_request, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::MERGE_REQUEST_ACTION,
:create, 'merge_requests_users')
end
end
def close_mr(merge_request, current_user)
create_record_event(merge_request, current_user, :closed).tap do
track_event(event_action: :closed, event_target: MergeRequest, author_id: current_user.id)
track_snowplow_event(merge_request, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::MERGE_REQUEST_ACTION,
:close, 'merge_requests_users')
end
end
def reopen_mr(merge_request, current_user)
create_record_event(merge_request, current_user, :reopened)
end
def merge_mr(merge_request, current_user)
create_record_event(merge_request, current_user, :merged).tap do
track_event(event_action: :merged, event_target: MergeRequest, author_id: current_user.id)
track_snowplow_event(merge_request, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::MERGE_REQUEST_ACTION,
:merge, 'merge_requests_users')
end
end
def open_milestone(milestone, current_user)
create_record_event(milestone, current_user, :created)
end
def close_milestone(milestone, current_user)
create_record_event(milestone, current_user, :closed)
end
def reopen_milestone(milestone, current_user)
create_record_event(milestone, current_user, :reopened)
end
def destroy_milestone(milestone, current_user)
create_record_event(milestone, current_user, :destroyed)
end
def leave_note(note, current_user)
create_record_event(note, current_user, :commented).tap do
if note.is_a?(DiffNote) && note.for_merge_request?
track_event(event_action: :commented, event_target: MergeRequest, author_id: current_user.id)
track_snowplow_event(note, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::MERGE_REQUEST_ACTION,
:comment, 'merge_requests_users')
end
end
end
def join_project(project, current_user)
create_event(project, current_user, :joined)
end
def leave_project(project, current_user)
create_event(project, current_user, :left)
end
def expired_leave_project(project, current_user)
create_event(project, current_user, :expired)
end
def create_project(project, current_user)
create_event(project, current_user, :created)
end
def push(project, current_user, push_data)
create_push_event(PushEventPayloadService, project, current_user, push_data)
end
def bulk_push(project, current_user, push_data)
create_push_event(BulkPushEventPayloadService, project, current_user, push_data)
end
def save_designs(current_user, create: [], update: [])
records = create.zip([:created].cycle) + update.zip([:updated].cycle)
return [] if records.empty?
if create.any?
track_snowplow_event(create.first, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::DESIGN_ACTION,
:create, 'design_users')
end
if update.any?
track_snowplow_event(update.first, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::DESIGN_ACTION,
:update, 'design_users')
end
create_record_events(records, current_user)
end
def destroy_designs(designs, current_user)
return [] unless designs.present?
track_snowplow_event(designs.first, current_user,
Gitlab::UsageDataCounters::TrackUniqueEvents::DESIGN_ACTION,
:destroy, 'design_users')
create_record_events(designs.zip([:destroyed].cycle), current_user)
end
# Create a new wiki page event
#
# @param [WikiPage::Meta] wiki_page_meta The event target
# @param [User] author The event author
# @param [Symbol] action One of the Event::WIKI_ACTIONS
# @param [String] fingerprint The de-duplication fingerprint
#
# The fingerprint, if provided, should be sufficient to find duplicate events.
# Suitable values would be, for example, the current page SHA.
#
# @return [Event] the event
def wiki_event(wiki_page_meta, author, action, fingerprint)
raise IllegalActionError, action unless Event::WIKI_ACTIONS.include?(action)
track_event(event_action: action, event_target: wiki_page_meta.class, author_id: author.id)
duplicate = Event.for_wiki_meta(wiki_page_meta).for_fingerprint(fingerprint).first
return duplicate if duplicate.present?
create_record_event(wiki_page_meta, author, action, fingerprint.presence)
end
def approve_mr(merge_request, current_user)
create_record_event(merge_request, current_user, :approved)
end
private
def create_record_event(record, current_user, status, fingerprint = nil)
create_event(record.resource_parent, current_user, status,
fingerprint: fingerprint,
target_id: record.id,
target_type: record.class.name)
end
# If creating several events, this method will insert them all in a single
# statement
#
# @param [[Eventable, Symbol, String]] a list of tuples of records, a valid status, and fingerprint
# @param [User] the author of the event
def create_record_events(tuples, current_user)
base_attrs = {
created_at: Time.now.utc,
updated_at: Time.now.utc,
author_id: current_user.id
}
attribute_sets = tuples.map do |record, status, fingerprint|
action = Event.actions[status]
raise IllegalActionError, "#{status} is not a valid status" if action.nil?
parent_attrs(record.resource_parent)
.merge(base_attrs)
.merge(action: action, fingerprint: fingerprint, target_id: record.id, target_type: record.class.name)
end
result = Event.insert_all(attribute_sets, returning: %w[id])
tuples.each do |record, status, _|
track_event(event_action: status, event_target: record.class, author_id: current_user.id)
end
result
end
def create_push_event(service_class, project, current_user, push_data)
Migrate events into a new format This commit migrates events data in such a way that push events are stored much more efficiently. This is done by creating a shadow table called "events_for_migration", and a table called "push_event_payloads" which is used for storing push data of push events. The background migration in this commit will copy events from the "events" table into the "events_for_migration" table, push events in will also have a row created in "push_event_payloads". This approach allows us to reclaim space in the next release by simply swapping the "events" and "events_for_migration" tables, then dropping the old events (now "events_for_migration") table. The new table structure is also optimised for storage space, and does not include the unused "title" column nor the "data" column (since this data is moved to "push_event_payloads"). == Newly Created Events Newly created events are inserted into both "events" and "events_for_migration", both using the exact same primary key value. The table "push_event_payloads" in turn has a foreign key to the _shadow_ table. This removes the need for recreating and validating the foreign key after swapping the tables. Since the shadow table also has a foreign key to "projects.id" we also don't have to worry about orphaned rows. This approach however does require some additional storage as we're duplicating a portion of the events data for at least 1 release. The exact amount is hard to estimate, but for GitLab.com this is expected to be between 10 and 20 GB at most. The background migration in this commit deliberately does _not_ update the "events" table as doing so would put a lot of pressure on PostgreSQL's auto vacuuming system. == Supporting Both Old And New Events Application code has also been adjusted to support push events using both the old and new data formats. This is done by creating a PushEvent class which extends the regular Event class. Using Rails' Single Table Inheritance system we can ensure the right class is used for the right data, which in this case is based on the value of `events.action`. To support displaying old and new data at the same time the PushEvent class re-defines a few methods of the Event class, falling back to their original implementations for push events in the old format. Once all existing events have been migrated the various push event related methods can be removed from the Event model, and the calls to `super` can be removed from the methods in the PushEvent model. The UI and event atom feed have also been slightly changed to better handle this new setup, fortunately only a few changes were necessary to make this work. == API Changes The API only displays push data of events in the new format. Supporting both formats in the API is a bit more difficult compared to the UI. Since the old push data was not really well documented (apart from one example that used an incorrect "action" nmae) I decided that supporting both was not worth the effort, especially since events will be migrated in a few days _and_ new events are created in the correct format.
2017-07-10 11:43:57 -04:00
# We're using an explicit transaction here so that any errors that may occur
# when creating push payload data will result in the event creation being
# rolled back as well.
Rework how recent push events are retrieved Whenever you push to a branch GitLab will show a button to create a merge request (should one not exist already). The underlying code to display this data was quite inefficient. For example, it involved multiple slow queries just to figure out what the most recent push event was. This commit changes the way this data is retrieved so it's much faster. This is achieved by caching the ID of the last push event on every push, which is then retrieved when loading certain pages. Database queries are only executed if necessary and the cached data is removed automatically once a merge request has been created, or 2 hours after being stored. A trade-off of this approach is that we _only_ track the last event. Previously if you were to push to branch A and B then create a merge request for branch B we'd still show the widget for branch A. As of this commit this is no longer the case, instead we will only show the widget for the branch you pushed to most recently. Once a merge request exists the widget is no longer displayed. Alternative solutions are either too complex and/or too slow, hence the decision was made to settle for this trade-off. Performance Impact ------------------ In the best case scenario (= a user didn't push anything for more than 2 hours) we perform a single Redis GET per page. Should there be cached data we will run a single (and lightweight) SQL query to get the event data from the database. If a merge request already exists we will run an additional DEL to remove the cache key. The difference in response timings can vary a bit per project. On GitLab.com the 99th percentile of time spent in User#recent_push hovers between 100 milliseconds and 1 second, while the mean hovers around 50 milliseconds. With the changes in this MR the expected time spent in User#recent_push is expected to be reduced down to just a few milliseconds. Fixes https://gitlab.com/gitlab-org/gitlab-ce/issues/35990
2017-09-01 06:50:14 -04:00
event = Event.transaction do
new_event = create_event(project, current_user, :pushed)
Migrate events into a new format This commit migrates events data in such a way that push events are stored much more efficiently. This is done by creating a shadow table called "events_for_migration", and a table called "push_event_payloads" which is used for storing push data of push events. The background migration in this commit will copy events from the "events" table into the "events_for_migration" table, push events in will also have a row created in "push_event_payloads". This approach allows us to reclaim space in the next release by simply swapping the "events" and "events_for_migration" tables, then dropping the old events (now "events_for_migration") table. The new table structure is also optimised for storage space, and does not include the unused "title" column nor the "data" column (since this data is moved to "push_event_payloads"). == Newly Created Events Newly created events are inserted into both "events" and "events_for_migration", both using the exact same primary key value. The table "push_event_payloads" in turn has a foreign key to the _shadow_ table. This removes the need for recreating and validating the foreign key after swapping the tables. Since the shadow table also has a foreign key to "projects.id" we also don't have to worry about orphaned rows. This approach however does require some additional storage as we're duplicating a portion of the events data for at least 1 release. The exact amount is hard to estimate, but for GitLab.com this is expected to be between 10 and 20 GB at most. The background migration in this commit deliberately does _not_ update the "events" table as doing so would put a lot of pressure on PostgreSQL's auto vacuuming system. == Supporting Both Old And New Events Application code has also been adjusted to support push events using both the old and new data formats. This is done by creating a PushEvent class which extends the regular Event class. Using Rails' Single Table Inheritance system we can ensure the right class is used for the right data, which in this case is based on the value of `events.action`. To support displaying old and new data at the same time the PushEvent class re-defines a few methods of the Event class, falling back to their original implementations for push events in the old format. Once all existing events have been migrated the various push event related methods can be removed from the Event model, and the calls to `super` can be removed from the methods in the PushEvent model. The UI and event atom feed have also been slightly changed to better handle this new setup, fortunately only a few changes were necessary to make this work. == API Changes The API only displays push data of events in the new format. Supporting both formats in the API is a bit more difficult compared to the UI. Since the old push data was not really well documented (apart from one example that used an incorrect "action" nmae) I decided that supporting both was not worth the effort, especially since events will be migrated in a few days _and_ new events are created in the correct format.
2017-07-10 11:43:57 -04:00
service_class.new(new_event, push_data).execute
Rework how recent push events are retrieved Whenever you push to a branch GitLab will show a button to create a merge request (should one not exist already). The underlying code to display this data was quite inefficient. For example, it involved multiple slow queries just to figure out what the most recent push event was. This commit changes the way this data is retrieved so it's much faster. This is achieved by caching the ID of the last push event on every push, which is then retrieved when loading certain pages. Database queries are only executed if necessary and the cached data is removed automatically once a merge request has been created, or 2 hours after being stored. A trade-off of this approach is that we _only_ track the last event. Previously if you were to push to branch A and B then create a merge request for branch B we'd still show the widget for branch A. As of this commit this is no longer the case, instead we will only show the widget for the branch you pushed to most recently. Once a merge request exists the widget is no longer displayed. Alternative solutions are either too complex and/or too slow, hence the decision was made to settle for this trade-off. Performance Impact ------------------ In the best case scenario (= a user didn't push anything for more than 2 hours) we perform a single Redis GET per page. Should there be cached data we will run a single (and lightweight) SQL query to get the event data from the database. If a merge request already exists we will run an additional DEL to remove the cache key. The difference in response timings can vary a bit per project. On GitLab.com the 99th percentile of time spent in User#recent_push hovers between 100 milliseconds and 1 second, while the mean hovers around 50 milliseconds. With the changes in this MR the expected time spent in User#recent_push is expected to be reduced down to just a few milliseconds. Fixes https://gitlab.com/gitlab-org/gitlab-ce/issues/35990
2017-09-01 06:50:14 -04:00
new_event
Migrate events into a new format This commit migrates events data in such a way that push events are stored much more efficiently. This is done by creating a shadow table called "events_for_migration", and a table called "push_event_payloads" which is used for storing push data of push events. The background migration in this commit will copy events from the "events" table into the "events_for_migration" table, push events in will also have a row created in "push_event_payloads". This approach allows us to reclaim space in the next release by simply swapping the "events" and "events_for_migration" tables, then dropping the old events (now "events_for_migration") table. The new table structure is also optimised for storage space, and does not include the unused "title" column nor the "data" column (since this data is moved to "push_event_payloads"). == Newly Created Events Newly created events are inserted into both "events" and "events_for_migration", both using the exact same primary key value. The table "push_event_payloads" in turn has a foreign key to the _shadow_ table. This removes the need for recreating and validating the foreign key after swapping the tables. Since the shadow table also has a foreign key to "projects.id" we also don't have to worry about orphaned rows. This approach however does require some additional storage as we're duplicating a portion of the events data for at least 1 release. The exact amount is hard to estimate, but for GitLab.com this is expected to be between 10 and 20 GB at most. The background migration in this commit deliberately does _not_ update the "events" table as doing so would put a lot of pressure on PostgreSQL's auto vacuuming system. == Supporting Both Old And New Events Application code has also been adjusted to support push events using both the old and new data formats. This is done by creating a PushEvent class which extends the regular Event class. Using Rails' Single Table Inheritance system we can ensure the right class is used for the right data, which in this case is based on the value of `events.action`. To support displaying old and new data at the same time the PushEvent class re-defines a few methods of the Event class, falling back to their original implementations for push events in the old format. Once all existing events have been migrated the various push event related methods can be removed from the Event model, and the calls to `super` can be removed from the methods in the PushEvent model. The UI and event atom feed have also been slightly changed to better handle this new setup, fortunately only a few changes were necessary to make this work. == API Changes The API only displays push data of events in the new format. Supporting both formats in the API is a bit more difficult compared to the UI. Since the old push data was not really well documented (apart from one example that used an incorrect "action" nmae) I decided that supporting both was not worth the effort, especially since events will be migrated in a few days _and_ new events are created in the correct format.
2017-07-10 11:43:57 -04:00
end
track_event(event_action: :pushed, event_target: Project, author_id: current_user.id)
namespace = project.namespace
if Feature.enabled?(:route_hll_to_snowplow, namespace)
Gitlab::Tracking.event(self.class.to_s, 'action_active_users_project_repo', namespace: namespace, user: current_user, project: project)
end
Rework how recent push events are retrieved Whenever you push to a branch GitLab will show a button to create a merge request (should one not exist already). The underlying code to display this data was quite inefficient. For example, it involved multiple slow queries just to figure out what the most recent push event was. This commit changes the way this data is retrieved so it's much faster. This is achieved by caching the ID of the last push event on every push, which is then retrieved when loading certain pages. Database queries are only executed if necessary and the cached data is removed automatically once a merge request has been created, or 2 hours after being stored. A trade-off of this approach is that we _only_ track the last event. Previously if you were to push to branch A and B then create a merge request for branch B we'd still show the widget for branch A. As of this commit this is no longer the case, instead we will only show the widget for the branch you pushed to most recently. Once a merge request exists the widget is no longer displayed. Alternative solutions are either too complex and/or too slow, hence the decision was made to settle for this trade-off. Performance Impact ------------------ In the best case scenario (= a user didn't push anything for more than 2 hours) we perform a single Redis GET per page. Should there be cached data we will run a single (and lightweight) SQL query to get the event data from the database. If a merge request already exists we will run an additional DEL to remove the cache key. The difference in response timings can vary a bit per project. On GitLab.com the 99th percentile of time spent in User#recent_push hovers between 100 milliseconds and 1 second, while the mean hovers around 50 milliseconds. With the changes in this MR the expected time spent in User#recent_push is expected to be reduced down to just a few milliseconds. Fixes https://gitlab.com/gitlab-org/gitlab-ce/issues/35990
2017-09-01 06:50:14 -04:00
Users::LastPushEventService.new(current_user)
.cache_last_push_event(event)
Users::ActivityService.new(current_user).execute
end
def create_event(resource_parent, current_user, status, attributes = {})
attributes.reverse_merge!(
action: status,
author_id: current_user.id
)
attributes.merge!(parent_attrs(resource_parent))
if attributes[:fingerprint].present?
Event.safe_find_or_create_by!(attributes)
else
Event.create!(attributes)
end
end
def parent_attrs(resource_parent)
resource_parent_attr = case resource_parent
when Project
:project_id
when Group
:group_id
end
return {} unless resource_parent_attr
{ resource_parent_attr => resource_parent.id }
end
def track_event(**params)
Gitlab::UsageDataCounters::TrackUniqueEvents.track_event(**params)
end
def track_snowplow_event(record, current_user, category, action, label)
return unless Feature.enabled?(:route_hll_to_snowplow_phase2)
project = record.project
Gitlab::Tracking.event(
category.to_s,
action.to_s,
label: label,
project: project,
namespace: project.namespace,
user: current_user
)
end
end
EventCreateService.prepend_mod_with('EventCreateService')