Add table for merge request commits

This is an ID-less table with just three columns: an association to the merge
request diff the commit belongs to, the relative order of the commit within the
merge request diff, and the commit SHA itself.

Previously we stored much more information about the commits, so that we could
display them even when they were deleted from the repo. Since 8.0, we ensure
that those commits are kept around for as long as the target repo itself is, so
we don't need to duplicate that data in the database.

3 files changed, 33 insertions, 9 deletions

diff --git a/lib/gitlab/cycle_analytics/metrics_tables.rb b/lib/gitlab/cycle_analytics/metrics_tables.rb
index 9d25ef078e8..f5d08c0b658 100644
--- a/lib/gitlab/cycle_analytics/metrics_tables.rb
+++ b/lib/gitlab/cycle_analytics/metrics_tables.rb
@@ -13,6 +13,10 @@ module Gitlab
         MergeRequestDiff.arel_table
       end
 
+      def mr_diff_commits_table
+        MergeRequestDiffCommit.arel_table
+      end
+
       def mr_closing_issues_table
         MergeRequestsClosingIssues.arel_table
       end
diff --git a/lib/gitlab/cycle_analytics/plan_event_fetcher.rb b/lib/gitlab/cycle_analytics/plan_event_fetcher.rb
index 7d342a2d2cb..b260822788d 100644
--- a/lib/gitlab/cycle_analytics/plan_event_fetcher.rb
+++ b/lib/gitlab/cycle_analytics/plan_event_fetcher.rb
@@ -2,40 +2,59 @@ module Gitlab
   module CycleAnalytics
     class PlanEventFetcher < BaseEventFetcher
       def initialize(*args)
-        @projections = [mr_diff_table[:st_commits].as('commits'),
+        @projections = [mr_diff_table[:id],
+                        mr_diff_table[:st_commits],
                         issue_metrics_table[:first_mentioned_in_commit_at]]
 
         super(*args)
       end
 
       def events_query
-        base_query.join(mr_diff_table).on(mr_diff_table[:merge_request_id].eq(mr_table[:id]))
+        base_query
+          .join(mr_diff_table)
+          .on(mr_diff_table[:merge_request_id].eq(mr_table[:id]))
 
         super
       end
 
       private
 
+      def merge_request_diff_commits
+        @merge_request_diff_commits ||=
+          MergeRequestDiffCommit
+            .where(merge_request_diff_id: event_result.map { |event| event['id'] })
+            .group_by(&:merge_request_diff_id)
+      end
+
       def serialize(event)
-        st_commit = first_time_reference_commit(event.delete('commits'), event)
+        commit = first_time_reference_commit(event)
 
-        return unless st_commit
+        return unless commit
 
-        serialize_commit(event, st_commit, query)
+        serialize_commit(event, commit, query)
       end
 
-      def first_time_reference_commit(commits, event)
+      def first_time_reference_commit(event)
+        return nil unless event && merge_request_diff_commits
+
+        commits =
+          if event['st_commits'].present?
+            YAML.load(event['st_commits'])
+          else
+            merge_request_diff_commits[event['id'].to_i]
+          end
+
         return nil if commits.blank?
 
-        YAML.load(commits).find do |commit|
+        commits.find do |commit|
           next unless commit[:committed_date] && event['first_mentioned_in_commit_at']
 
           commit[:committed_date].to_i == DateTime.parse(event['first_mentioned_in_commit_at'].to_s).to_i
         end
       end
 
-      def serialize_commit(event, st_commit, query)
-        commit = Commit.new(Gitlab::Git::Commit.new(st_commit), @project)
+      def serialize_commit(event, commit, query)
+        commit = Commit.new(Gitlab::Git::Commit.new(commit.to_hash), @project)
 
         AnalyticsCommitSerializer.new(project: @project, total_time: event['total_time']).represent(commit)
       end
diff --git a/lib/gitlab/import_export/import_export.yml b/lib/gitlab/import_export/import_export.yml
index 1860352c96d..c8ad3a7a5e0 100644
--- a/lib/gitlab/import_export/import_export.yml
+++ b/lib/gitlab/import_export/import_export.yml
@@ -27,6 +27,7 @@ project_tree:
       - :author
       - :events
     - merge_request_diff:
+      - :merge_request_diff_commits
       - :merge_request_diff_files
     - :events
     - :timelogs