Adds get all medians to Cycle Analytics model

app/models/cycle_analytics.rb

@@ -6,6 +6,12 @@ class CycleAnalytics
     @options = options
   end
+  def self.all_medians_per_stage(projects, options)
+    STAGES.each_with_object({}) do |stage_name, hsh|
+      hsh[stage_name] = Gitlab::CycleAnalytics::Stage[stage_name].new(projects: projects, options: options).medians&.values || []
+    end
+  end
+
   def summary
     @summary ||= ::Gitlab::CycleAnalytics::StageSummary.new(@project,
                                                             from: @options[:from],

lib/gitlab/cycle_analytics/base_query.rb

@@ -14,7 +14,7 @@ module Gitlab
       def stage_query
         query = mr_closing_issues_table.join(issue_table).on(issue_table[:id].eq(mr_closing_issues_table[:issue_id]))
           .join(issue_metrics_table).on(issue_table[:id].eq(issue_metrics_table[:issue_id]))
-          .where(issue_table[:project_id].eq(@project.id)) # rubocop:disable Gitlab/ModuleWithInstanceVariables
+          .where(issue_table[:project_id].in(project_ids)) # rubocop:disable Gitlab/ModuleWithInstanceVariables
           .where(issue_table[:created_at].gteq(@options[:from])) # rubocop:disable Gitlab/ModuleWithInstanceVariables
         # Load merge_requests
@@ -22,9 +22,14 @@ module Gitlab
           .on(mr_table[:id].eq(mr_closing_issues_table[:merge_request_id]))
           .join(mr_metrics_table)
           .on(mr_table[:id].eq(mr_metrics_table[:merge_request_id]))
+          .project(issue_table[:project_id].as("project_id"))
         query
       end
+
+      def project_ids
+        @projects.map(&:id)
+      end
     end
   end
 end

lib/gitlab/cycle_analytics/base_stage.rb

@@ -3,8 +3,8 @@ module Gitlab
     class BaseStage
       include BaseQuery
-      def initialize(project:, options:)
-        @project = project
+      def initialize(projects:, options:)
+        @projects = projects
         @options = options
       end
@@ -20,7 +20,7 @@ module Gitlab
         raise NotImplementedError.new("Expected #{self.name} to implement title")
       end
-      def median
+      def medians
         cte_table = Arel::Table.new("cte_table_for_#{name}")
         # Build a `SELECT` query. We find the first of the `end_time_attrs` that isn't `NULL` (call this end_time).
@@ -31,7 +31,7 @@ module Gitlab
           cte_table,
           subtract_datetimes(base_query.dup, start_time_attrs, end_time_attrs, name.to_s))
-        median_datetime(cte_table, interval_query, name)
+        median_datetimes(cte_table, interval_query, name)
       end
       def name

lib/gitlab/cycle_analytics/usage_data.rb

+module Gitlab
+  module CycleAnalytics
+    class UsageData
+      PROJECTS_LIMIT = 10
+
+      attr_reader :projects, :options
+
+      def initialize(projects, options)
+        @projects = projects
+        @options = options
+      end
+
+      def to_json
+        total = 0
+        values = {}
+
+        medians_per_stage.each do |stage_name, medians|
+          medians = medians.compact
+
+          stage_values = {
+            average: calc_average(medians),
+            sd: standard_deviation(medians),
+            missing: projects.length - medians.length
+          }
+
+          total += stage_values.values.compact.sum
+          values[stage_name] = stage_values
+        end
+
+        values[:total] = total
+
+        { avg_cycle_analytics: values }
+      end
+
+      private
+
+      def medians_per_stage
+        @medians_per_stage ||= ::CycleAnalytics.all_medians_per_stage(projects, options)
+      end
+
+      def calc_average(values)
+        return if values.empty?
+
+        (values.sum / values.length).to_i
+      end
+
+      def sample_variance(values)
+        return 0 if values.length <= 1
+
+        avg = calc_average(values)
+        sum = values.inject(0) do |acc, val|
+          acc + (val - avg)**2
+        end
+
+        sum / (values.length - 1)
+      end
+
+      def standard_deviation(values)
+        Math.sqrt(sample_variance(values)).to_i
+      end
+    end
+  end
+end
+

lib/gitlab/database/median.rb

@@ -2,7 +2,7 @@
 module Gitlab
   module Database
     module Median
-      def median_datetime(arel_table, query_so_far, column_sym)
+      def median_datetimes(arel_table, query_so_far, column_sym)
         median_queries =
           if Gitlab::Database.postgresql?
             pg_median_datetime_sql(arel_table, query_so_far, column_sym)
@@ -13,16 +13,16 @@ module Gitlab
         results = Array.wrap(median_queries).map do |query|
           ActiveRecord::Base.connection.execute(query)
         end
-        extract_median(results).presence
+        extract_medians(results).presence
       end
-      def extract_median(results)
+      def extract_medians(results)
         result = results.compact.first
         if Gitlab::Database.postgresql?
-          result = result.first.presence
-          median = result['median'] if result
-          median.to_f if median
+          result.values.map do |id, median|
+            [id, median&.to_f]
+          end.to_h
         elsif Gitlab::Database.mysql?
           result.to_a.flatten.first
         end
@@ -69,17 +69,20 @@ module Gitlab
           cte_table,
           arel_table
             .project(
+              arel_table[:project_id],
               arel_table[column_sym].as(column_sym.to_s),
-              Arel::Nodes::Over.new(Arel::Nodes::NamedFunction.new("row_number", []),
-                                    Arel::Nodes::Window.new.order(arel_table[column_sym])).as('row_id'),
-              arel_table.project("COUNT(1)").as('ct')).
+              Arel::Nodes::Over.new(Arel::Nodes::NamedFunction.new("rank", []),
+                                    Arel::Nodes::Window.new.partition(arel_table[:project_id])
+                                      .order(arel_table[column_sym])).as('row_id'),
+              arel_table.from(arel_table.alias).project("COUNT(*)").where(arel_table[:project_id].eq(arel_table.alias[:project_id])).as('ct')).
             # Disallow negative values
             where(arel_table[column_sym].gteq(zero_interval)))
         # From the CTE, select either the middle row or the middle two rows (this is accomplished
         # by 'where cte.row_id between cte.ct / 2.0 AND cte.ct / 2.0 + 1'). Find the average of the
         # selected rows, and this is the median value.
-        cte_table.project(average([extract_epoch(cte_table[column_sym])], "median"))
+        cte_table.project(cte_table[:project_id])
+          .project(average([extract_epoch(cte_table[column_sym])], "median"))
           .where(
             Arel::Nodes::Between.new(
               cte_table[:row_id],
@@ -90,6 +93,8 @@ module Gitlab
             )
           )
           .with(query_so_far, cte)
+          .group(cte_table[:project_id])
+          .order(cte_table[:project_id])
           .to_sql
       end

lib/gitlab/usage_data.rb

@@ -9,6 +9,7 @@ module Gitlab
         license_usage_data.merge(system_usage_data)
                           .merge(features_usage_data)
                           .merge(components_usage_data)
+                          .merge(cycle_analytics_usage_data)
       end
       def to_json(force_refresh: false)
@@ -71,6 +72,12 @@ module Gitlab
         }
       end
+      def cycle_analytics_usage_data
+        projects = Project.sorted_by_activity.limit(Gitlab::CycleAnalytics::UsageData::PROJECTS_LIMIT)
+
+        Gitlab::CycleAnalytics::UsageData.new(projects, { from: 7.days.ago }).to_json
+      end
+
       def features_usage_data
         features_usage_data_ce
       end