Refactors median code to work with both single and multiple projects

app/models/cycle_analytics.rb

@@ -7,8 +7,8 @@ class CycleAnalytics
   end
 
   def all_medians_per_stage
-    STAGES.each_with_object({}) do |stage_name, hsh|
-      hsh[stage_name] = self[stage_name].median
+    STAGES.each_with_object({}) do |stage_name, medians_per_stage|
+      medians_per_stage[stage_name] = self[stage_name].median
     end
   end
 

app/serializers/analytics_stage_entity.rb

@@ -7,6 +7,8 @@ class AnalyticsStageEntity < Grape::Entity
   expose :description
 
   expose :median, as: :value do |stage|
-    distance_of_time_in_words(stage.median) if stage.median && !(stage.median.blank? || stage.median.zero?)
+    if stage.median && !(stage.median.nil? || stage.median.zero?)
+      distance_of_time_in_words(stage.median)
+    end
   end
 end

lib/gitlab/cycle_analytics/base_query.rb

@@ -8,14 +8,14 @@ module Gitlab
       private
 
       def base_query
-        @base_query ||= stage_query([@project.id]) # rubocop:disable Gitlab/ModuleWithInstanceVariables
+        @base_query ||= stage_query(@project.id) # rubocop:disable Gitlab/ModuleWithInstanceVariables
       end
 
       def stage_query(project_ids)
         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]))
           .project(issue_table[:project_id].as("project_id"))
-          .where(issue_table[:project_id].in(project_ids))
+          .where(issue_table[:project_id].in(Array(project_ids)))
           .where(issue_table[:created_at].gteq(@options[:from])) # rubocop:disable Gitlab/ModuleWithInstanceVariables
 
         # Load merge_requests

lib/gitlab/cycle_analytics/base_stage.rb

@@ -31,8 +31,12 @@ module Gitlab
           interval_query = Arel::Nodes::As.new(cte_table,
             subtract_datetimes(stage_query(project_ids), start_time_attrs, end_time_attrs, name.to_s))
 
-          median_datetimes(cte_table, interval_query, name, :project_id)&.each do |project_id, median|
-            loader.call(project_id, median)
+          if project_ids.size == 1
+            loader.call(@project.id, median_datetime(cte_table, interval_query, name))
+          else
+            median_datetimes(cte_table, interval_query, name, :project_id)&.each do |project_id, median|
+              loader.call(project_id, median)
+            end
           end
         end
       end

lib/gitlab/cycle_analytics/usage_data.rb

@@ -5,27 +5,21 @@ module Gitlab
 
       attr_reader :projects, :options
 
-      def initialize(projects, options)
-        @projects = projects
-        @options = options
+      def initialize
+        @projects = Project.sorted_by_activity.limit(PROJECTS_LIMIT)
+        @options = { from: 7.days.ago }
       end
 
       def to_json
         total = 0
-        values = {}
 
-        medians_per_stage.each do |stage_name, medians|
-          medians = medians.map(&:presence).compact
+        values =
+          medians_per_stage.each_with_object({}) do |(stage_name, medians), hsh|
+            calculations = stage_values(medians)
 
-          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
+            total += calculations.values.compact.sum
+            hsh[stage_name] = calculations
+          end
 
         values[:total] = total
 
@@ -43,26 +37,36 @@ module Gitlab
         end
       end
 
+      def stage_values(medians)
+        medians = medians.map(&:presence).compact
+        average = calc_average(medians)
+
+        {
+          average: average,
+          sd: standard_deviation(medians, average),
+          missing: projects.length - medians.length
+        }
+      end
+
       def calc_average(values)
         return if values.empty?
 
         (values.sum / values.length).to_i
       end
 
-      def sample_variance(values)
+      def standard_deviation(values, average)
+        Math.sqrt(sample_variance(values, average)).to_i
+      end
+
+      def sample_variance(values, average)
         return 0 if values.length <= 1
 
-        avg = calc_average(values)
         sum = values.inject(0) do |acc, val|
-          acc + (val - avg)**2
+          acc + (val - average)**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

@@ -2,32 +2,34 @@
 module Gitlab
   module Database
     module Median
+      def median_datetime(arel_table, query_so_far, column_sym)
+        extract_median(execute_queries(arel_table, query_so_far, column_sym)).presence
+      end
+
       def median_datetimes(arel_table, query_so_far, column_sym, partition_column)
-        median_queries =
-          if Gitlab::Database.postgresql?
-            pg_median_datetime_sql(arel_table, query_so_far, column_sym, partition_column)
-          elsif Gitlab::Database.mysql?
-            mysql_median_datetime_sql(arel_table, query_so_far, column_sym)
-          end
-
-        results = Array.wrap(median_queries).map do |query|
-          ActiveRecord::Base.connection.execute(query)
-        end
-        extract_medians(results).presence
+        extract_medians(execute_queries(arel_table, query_so_far, column_sym, partition_column)).presence
       end
 
-      def extract_medians(results)
+      def extract_median(results)
         result = results.compact.first
 
         if Gitlab::Database.postgresql?
-          result.values.map do |id, median|
-            [id.to_i, median&.to_f]
-          end.to_h
+          result = result.first.presence
+
+          result['median']&.to_f if result
         elsif Gitlab::Database.mysql?
           result.to_a.flatten.first
         end
       end
 
+      def extract_medians(results)
+        return {} if Gitlab::Database.mysql?
+
+        results.compact.first.values.map do |id, median|
+          [id.to_i, median&.to_f]
+        end.to_h
+      end
+
       def mysql_median_datetime_sql(arel_table, query_so_far, column_sym)
         query = arel_table
                 .from(arel_table.project(Arel.sql('*')).order(arel_table[column_sym]).as(arel_table.table_name))
@@ -53,7 +55,7 @@ module Gitlab
         ]
       end
 
-      def pg_median_datetime_sql(arel_table, query_so_far, column_sym, partition_column)
+      def pg_median_datetime_sql(arel_table, query_so_far, column_sym, partition_column = nil)
         # Create a CTE with the column we're operating on, row number (after sorting by the column
         # we're operating on), and count of the table we're operating on (duplicated across) all rows
         # of the CTE. For example, if we're looking to find the median of the `projects.star_count`
@@ -64,49 +66,107 @@ module Gitlab
         #           5 |      1 |  3
         #           9 |      2 |  3
         #          15 |      3 |  3
+        #
+        #  If a partition column is used we will do the same operation but for separate partitions,
+        #  when that happens the CTE might look like this:
+        #
+        #  project_id | star_count | row_id | ct
+        # ------------+------------+--------+----
+        #           1 |          5 |     1 |  2
+        #           1 |          9 |     2 |  2
+        #           2 |         10 |     1 |  3
+        #           2 |         15 |     2 |  3
+        #           2 |         20 |     3 |  3
         cte_table = Arel::Table.new("ordered_records")
+
         cte = Arel::Nodes::As.new(
           cte_table,
-          arel_table
-            .project(
-              arel_table[partition_column],
-              arel_table[column_sym].as(column_sym.to_s),
-              Arel::Nodes::Over.new(Arel::Nodes::NamedFunction.new("rank", []),
-                                    Arel::Nodes::Window.new.partition(arel_table[partition_column])
-                                      .order(arel_table[column_sym])).as('row_id'),
-              arel_table.from(arel_table.alias)
-                .project("COUNT(*)")
-                .where(arel_table[partition_column].eq(arel_table.alias[partition_column])).as('ct')).
+          arel_table.project(*rank_rows(arel_table, column_sym, partition_column)).
             # 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(cte_table[partition_column])
-          .project(average([extract_epoch(cte_table[column_sym])], "median"))
-          .where(
-            Arel::Nodes::Between.new(
-              cte_table[:row_id],
-              Arel::Nodes::And.new(
-                [(cte_table[:ct] / Arel.sql('2.0')),
-                 (cte_table[:ct] / Arel.sql('2.0') + 1)]
+        result =
+          cte_table
+            .project(*median_projections(cte_table, column_sym, partition_column))
+            .where(
+              Arel::Nodes::Between.new(
+                cte_table[:row_id],
+                Arel::Nodes::And.new(
+                  [(cte_table[:ct] / Arel.sql('2.0')),
+                   (cte_table[:ct] / Arel.sql('2.0') + 1)]
+                )
               )
             )
-          )
-          .with(query_so_far, cte)
-          .group(cte_table[partition_column])
-          .order(cte_table[partition_column])
-          .to_sql
+            .with(query_so_far, cte)
+
+        result.group(cte_table[partition_column]).order(cte_table[partition_column]) if partition_column
+
+        result.to_sql
       end
 
       private
 
+      def median_queries(arel_table, query_so_far, column_sym, partition_column = nil)
+        if Gitlab::Database.postgresql?
+          pg_median_datetime_sql(arel_table, query_so_far, column_sym, partition_column)
+        elsif Gitlab::Database.mysql?
+          mysql_median_datetime_sql(arel_table, query_so_far, column_sym)
+        end
+      end
+
+      def execute_queries(arel_table, query_so_far, column_sym, partition_column = nil)
+        queries = median_queries(arel_table, query_so_far, column_sym, partition_column)
+
+        Array.wrap(queries).map { |query| ActiveRecord::Base.connection.execute(query) }
+      end
+
       def average(args, as)
         Arel::Nodes::NamedFunction.new("AVG", args, as)
       end
 
+      def rank_rows(arel_table, column_sym, partition_column)
+        column_row = arel_table[column_sym].as(column_sym.to_s)
+
+        if partition_column
+          partition_row = arel_table[partition_column]
+          row_id =
+            Arel::Nodes::Over.new(
+              Arel::Nodes::NamedFunction.new('rank', []),
+              Arel::Nodes::Window.new.partition(arel_table[partition_column])
+                .order(arel_table[column_sym])
+            ).as('row_id')
+
+          count = arel_table.from(arel_table.alias)
+                    .project('COUNT(*)')
+                    .where(arel_table[partition_column].eq(arel_table.alias[partition_column]))
+                    .as('ct')
+
+          [partition_row, column_row, row_id, count]
+        else
+          row_id =
+            Arel::Nodes::Over.new(
+              Arel::Nodes::NamedFunction.new('row_number', []),
+              Arel::Nodes::Window.new.order(arel_table[column_sym])
+            ).as('row_id')
+
+          count = arel_table.project("COUNT(1)").as('ct')
+
+          [column_row, row_id, count]
+        end
+      end
+
+      def median_projections(table, column_sym, partition_column)
+        if partition_column
+          [table[partition_column],
+           average([extract_epoch(table[column_sym])], "median")]
+        else
+          [average([extract_epoch(table[column_sym])], "median")]
+        end
+      end
+
       def extract_epoch(arel_attribute)
         Arel.sql(%Q{EXTRACT(EPOCH FROM "#{arel_attribute.relation.name}"."#{arel_attribute.name}")})
       end

lib/gitlab/usage_data.rb

@@ -73,12 +73,7 @@ module Gitlab
       end
 
       def cycle_analytics_usage_data
-        # We only want to generate this data for instances that use PostgreSQL
-        return {} if Gitlab::Database.mysql?
-
-        projects = Project.sorted_by_activity.limit(Gitlab::CycleAnalytics::UsageData::PROJECTS_LIMIT)
-
-        Gitlab::CycleAnalytics::UsageData.new(projects, { from: 7.days.ago }).to_json
+        Gitlab::CycleAnalytics::UsageData.new.to_json
       end
 
       def features_usage_data

spec/features/cycle_analytics_spec.rb

@@ -41,7 +41,7 @@ feature 'Cycle Analytics', :js do
         allow_any_instance_of(Gitlab::ReferenceExtractor).to receive(:issues).and_return([issue])
         project.add_master(user)
 
-        create_cycle
+        @build = create_cycle(user, project, issue, mr, milestone, pipeline)
         deploy_master
 
         sign_in(user)
@@ -117,7 +117,7 @@ feature 'Cycle Analytics', :js do
       project.add_guest(guest)
 
       allow_any_instance_of(Gitlab::ReferenceExtractor).to receive(:issues).and_return([issue])
-      create_cycle
+      create_cycle(user, project, issue, mr, milestone, pipeline)
       deploy_master
 
       sign_in(guest)
@@ -166,16 +166,6 @@ feature 'Cycle Analytics', :js do
     expect(find('.stage-events')).to have_content("!#{mr.iid}")
   end
 
-  def create_cycle
-    issue.update(milestone: milestone)
-    pipeline.run
-
-    @build = create(:ci_build, pipeline: pipeline, status: :success, author: user)
-
-    merge_merge_requests_closing_issue(issue)
-    ProcessCommitWorker.new.perform(project.id, user.id, mr.commits.last.to_hash)
-  end
-
   def click_stage(stage_name)
     find('.stage-nav li', text: stage_name).click
     wait_for_requests

spec/lib/gitlab/cycle_analytics/usage_data_spec.rb

+require 'spec_helper'
+
+describe Gitlab::CycleAnalytics::UsageData do
+  let(:project) { create(:project, :repository) }
+  let(:user) { create(:user, :admin) }
+  let(:issue) { create(:issue, project: project, created_at: 2.days.ago) }
+  let(:milestone) { create(:milestone, project: project) }
+  let(:mr) { create_merge_request_closing_issue(issue, commit_message: "References #{issue.to_reference}") }
+  let(:pipeline) { create(:ci_empty_pipeline, status: 'created', project: project, ref: mr.source_branch, sha: mr.source_branch_sha, head_pipeline_of: mr) }
+
+  subject { described_class.new([project]) }
+
+  describe '#to_json' do
+    before do
+      allow_any_instance_of(Gitlab::ReferenceExtractor).to receive(:issues).and_return([issue])
+
+      create_cycle(user, project, issue, mr, milestone, pipeline)
+      deploy_master
+    end
+
+    it 'returns the aggregated usage data of every selected project' do
+      result = subject.to_json
+      avg_cycle_analytics = result[:avg_cycle_analytics]
+
+      expect(result).to have_key(:avg_cycle_analytics)
+      CycleAnalytics::STAGES.each do |stage_name|
+        stage_values = avg_cycle_analytics[stage_name]
+
+        expect(avg_cycle_analytics).to have_key(stage_name)
+        expect(stage_values).to have_key(:average)
+        expect(stage_values).to have_key(:sd)
+        expect(stage_values).to have_key(:missing)
+      end
+    end
+  end
+end

spec/lib/gitlab/database/median_spec.rb

+require 'spec_helper'
+
+describe Gitlab::Database::Median do
+  describe '#extract_medians' do
+    context 'when using MySQL' do
+      it 'returns an empty hash' do
+        values = [["1", "1000"]]
+
+        allow(Gitlab::Database).to receive(:mysql?).and_return(true)
+
+        expect(described_class.new.extract_median(values)).eq({})
+      end
+    end
+  end
+end

spec/models/cycle_analytics_spec.rb

+require 'spec_helper'
+
+describe CycleAnalytics do
+  let(:project) { create(:project, :repository) }
+  let(:from_date) { 10.days.ago }
+  let(:user) { create(:user, :admin) }
+  let(:issue) { create(:issue, project: project, created_at: 2.days.ago) }
+  let(:milestone) { create(:milestone, project: project) }
+  let(:mr) { create_merge_request_closing_issue(issue, commit_message: "References #{issue.to_reference}") }
+  let(:pipeline) { create(:ci_empty_pipeline, status: 'created', project: project, ref: mr.source_branch, sha: mr.source_branch_sha, head_pipeline_of: mr) }
+
+  subject { described_class.new(project, from: from_date) }
+
+  describe '#all_medians_per_stage' do
+    before do
+      allow_any_instance_of(Gitlab::ReferenceExtractor).to receive(:issues).and_return([issue])
+
+      create_cycle(user, project, issue, mr, milestone, pipeline)
+      deploy_master
+    end
+
+    it 'returns every median for each stage for a specific project' do
+      values = described_class::STAGES.each_with_object({}) do |stage_name, hsh|
+        hsh[stage_name] = subject[stage_name].median.presence
+      end
+
+      expect(subject.all_medians_per_stage).to eq(values)
+    end
+  end
+end

spec/support/cycle_analytics_helpers.rb

@@ -26,6 +26,18 @@ module CycleAnalyticsHelpers
                        ref: 'refs/heads/master').execute
   end
 
+  def create_cycle(user, project, issue, mr, milestone, pipeline)
+    issue.update(milestone: milestone)
+    pipeline.run
+
+    ci_build = create(:ci_build, pipeline: pipeline, status: :success, author: user)
+
+    merge_merge_requests_closing_issue(issue)
+    ProcessCommitWorker.new.perform(project.id, user.id, mr.commits.last.to_hash)
+
+    ci_build
+  end
+
   def create_merge_request_closing_issue(issue, message: nil, source_branch: nil, commit_message: 'commit message')
     if !source_branch || project.repository.commit(source_branch).blank?
       source_branch = generate(:branch)

spec/support/cycle_analytics_helpers/test_generation.rb

@@ -50,7 +50,7 @@ module CycleAnalyticsHelpers
               end
 
               median_time_difference = time_differences.sort[2]
-              expect(subject[phase].median.presence).to be_within(5).of(median_time_difference)
+              expect(subject[phase].median).to be_within(5).of(median_time_difference)
             end
 
             context "when the data belongs to another project" do