kotovalexarian-likes-gitlab
/
gitlab-org--gitlab-foss
1
0
Fork 0
Code Releases Activity
gitlab-org--gitlab-foss/app/models/cycle_analytics.rb

65 lines
2.3 KiB
Ruby
Raw Normal View History

Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
class CycleAnalytics
def issue
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
calculate_metric(Queries::issues,
-> (data_point) { data_point[:issue].created_at },
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
[Queries::issue_first_associated_with_milestone_at, Queries::issue_first_added_to_list_label_at])
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
end
def plan
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
calculate_metric(Queries::issues,
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
[Queries::issue_first_associated_with_milestone_at, Queries::issue_first_added_to_list_label_at],
Queries::issue_closing_merge_request_opened_at)
Add the "Code" Cycle Analytics section. 1. Record the `wip_flag_first_removed_at` and `first_assigned_to_user_other_than_author` metrics for a merge request. Use a `merge_request_metrics` table, similar to the one for `issues`. Metrics are recorded `after_save`. 2. Move larger queries to a `CycleAnalytics::Queries` module.
2016-08-25 05:54:41 +00:00
end
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
Add the "Code" Cycle Analytics section. 1. Record the `wip_flag_first_removed_at` and `first_assigned_to_user_other_than_author` metrics for a merge request. Use a `merge_request_metrics` table, similar to the one for `issues`. Metrics are recorded `after_save`. 2. Move larger queries to a `CycleAnalytics::Queries` module.
2016-08-25 05:54:41 +00:00
def code
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
calculate_metric(Queries::merge_requests_closing_issues,
-> (data_point) { data_point[:merge_request].created_at },
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
[Queries::merge_request_first_assigned_to_user_other_than_author_at, Queries::merge_request_wip_flag_first_removed_at])
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
end
Add the "Test" cycle analytics section.
2016-08-25 09:19:18 +00:00
def test
calculate_metric(Queries::merge_requests_closing_issues,
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
Queries::merge_request_build_started_at,
Queries::merge_request_build_finished_at)
Add the "Test" cycle analytics section.
2016-08-25 09:19:18 +00:00
end
Add the "Review" cycle analytics section.
2016-08-25 06:25:07 +00:00
def review
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
calculate_metric(Queries::merge_requests_closing_issues,
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
[Queries::merge_request_first_assigned_to_user_other_than_author_at, Queries::merge_request_wip_flag_first_removed_at],
[Queries::merge_request_first_closed_at, Queries::merge_request_merged_at])
Add the "Review" cycle analytics section.
2016-08-25 06:25:07 +00:00
end
Add the "Staging" cycle analytics section.
2016-08-25 08:39:48 +00:00
def staging
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
calculate_metric(Queries::merge_requests_closing_issues,
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
Queries::merge_request_merged_at,
Queries::merge_request_deployed_to_any_environment_at)
Add the "Staging" cycle analytics section.
2016-08-25 08:39:48 +00:00
end
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
def production
calculate_metric(Queries::merge_requests_closing_issues,
-> (data_point) { data_point[:issue].created_at },
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
Queries::merge_request_deployed_to_production_at)
Add the "Production" cycle analytics section. 1. Rewrite the `Queries` module to work off a `data_point` hash, with `issue` and `merge_request` as keys. The "production" query needs both an issue and a merge request to make it's calculation, so it makes sense to keep things consistent and provide the same data (issue + merge request) for all queries.
2016-08-25 09:06:01 +00:00
end
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
private
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
def calculate_metric(data, start_time_fns, end_time_fns)
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
times = data.map do |data_point|
Allow multiple queries for each cycle analytics section. 1. Pass in an array of queries - the first to return a value will be used. This makes it easier to add more heuristics later. 2. Convert all queries with 'or' in the title to two separate queries. 3. Rename all `mr_` methods to `merge_request_`
2016-08-25 09:34:12 +00:00
start_time = Array.wrap(start_time_fns).map { |fn| fn[data_point] }.compact.first
end_time = Array.wrap(end_time_fns).map { |fn| fn[data_point] }.compact.first
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
if start_time.present? && end_time.present?
end_time - start_time
end
end
median(times.compact)
end
def median(coll)
Add the "Code" Cycle Analytics section. 1. Record the `wip_flag_first_removed_at` and `first_assigned_to_user_other_than_author` metrics for a merge request. Use a `merge_request_metrics` table, similar to the one for `issues`. Metrics are recorded `after_save`. 2. Move larger queries to a `CycleAnalytics::Queries` module.
2016-08-25 05:54:41 +00:00
return if coll.empty?
Add the "Plan" Cycle Analytics query. 1. Move from raw SQL to ActiveRecord. 2. Add a non-persisted `CycleAnalytics` model to store all the queries.
2016-08-25 05:02:27 +00:00
size = coll.length
(coll[size / 2] + coll[(size - 1) / 2]) / 2.0
end
end