Add merge requests to Project Migration

  - Add merge requests to Project Migration when
    using Group Migration tool

lib/bulk_imports/projects/pipelines/merge_requests_pipeline.rb

+# frozen_string_literal: true
+
+module BulkImports
+  module Projects
+    module Pipelines
+      class MergeRequestsPipeline
+        include NdjsonPipeline
+
+        relation_name 'merge_requests'
+
+        extractor ::BulkImports::Common::Extractors::NdjsonExtractor, relation: relation
+
+        def after_run(_)
+          context.portable.merge_requests.set_latest_merge_request_diff_ids!
+        end
+      end
+    end
+  end
+end

lib/bulk_imports/projects/pipelines/repository_pipeline.rb

@@ -17,10 +17,18 @@ module BulkImports
         def load(context, data)
           url = data['httpUrlToRepo']
           url = url.sub("://", "://oauth2:#{context.configuration.access_token}@")
+          project = context.portable
 
           Gitlab::UrlBlocker.validate!(url, allow_local_network: allow_local_requests?, allow_localhost: allow_local_requests?)
 
-          context.portable.repository.import_repository(url)
+          project.ensure_repository
+          project.repository.fetch_as_mirror(url)
+        end
+
+        # The initial fetch can bring in lots of loose refs and objects.
+        # Running a `git gc` will make importing merge requests faster.
+        def after_run(_)
+          ::Repositories::HousekeepingService.new(context.portable, :gc).execute
         end
 
         private

lib/bulk_imports/projects/stage.rb

@@ -27,6 +27,10 @@ module BulkImports
             pipeline: BulkImports::Common::Pipelines::BoardsPipeline,
             stage: 4
           },
+          merge_requests: {
+            pipeline: BulkImports::Projects::Pipelines::MergeRequestsPipeline,
+            stage: 4
+          },
           uploads: {
             pipeline: BulkImports::Common::Pipelines::UploadsPipeline,
             stage: 5

spec/lib/bulk_imports/projects/pipelines/merge_requests_pipeline_spec.rb

+# frozen_string_literal: true
+
+require 'spec_helper'
+
+RSpec.describe BulkImports::Projects::Pipelines::MergeRequestsPipeline do
+  let_it_be(:user) { create(:user) }
+  let_it_be(:group) { create(:group) }
+  let_it_be(:project) { create(:project, :repository, group: group) }
+  let_it_be(:bulk_import) { create(:bulk_import, user: user) }
+  let_it_be(:entity) do
+    create(
+      :bulk_import_entity,
+      :project_entity,
+      project: project,
+      bulk_import: bulk_import,
+      source_full_path: 'source/full/path',
+      destination_name: 'My Destination Project',
+      destination_namespace: group.full_path
+    )
+  end
+
+  let_it_be(:tracker) { create(:bulk_import_tracker, entity: entity) }
+  let_it_be(:context) { BulkImports::Pipeline::Context.new(tracker) }
+
+  let(:mr) do
+    {
+      'iid' => 7,
+      'author_id' => 22,
+      'source_project_id' => 1234,
+      'target_project_id' => 1234,
+      'title' => 'Imported MR',
+      'description' => 'Description',
+      'state' => 'opened',
+      'source_branch' => 'feature',
+      'target_branch' => 'main',
+      'source_branch_sha' => 'ABCD',
+      'target_branch_sha' => 'DCBA',
+      'created_at' => '2020-06-14T15:02:47.967Z',
+      'updated_at' => '2020-06-14T15:03:47.967Z',
+      'merge_request_diff' => {
+        'state' => 'collected',
+        'base_commit_sha' => 'ae73cb07c9eeaf35924a10f713b364d32b2dd34f',
+        'head_commit_sha' => 'a97f74ddaa848b707bea65441c903ae4bf5d844d',
+        'start_commit_sha' => '9eea46b5c72ead701c22f516474b95049c9d9462',
+        'merge_request_diff_commits' => [
+          {
+            'sha' => 'COMMIT1',
+            'relative_order' => 0,
+            'message' => 'commit message',
+            'authored_date' => '2014-08-06T08:35:52.000+02:00',
+            'committed_date' => '2014-08-06T08:35:52.000+02:00',
+            'commit_author' => {
+              'name' => 'Commit Author',
+              'email' => 'gitlab@example.com'
+            },
+            'committer' => {
+              'name' => 'Committer',
+              'email' => 'committer@example.com'
+            }
+          }
+        ],
+        'merge_request_diff_files' => [
+          {
+            'relative_order' => 0,
+            'utf8_diff' => '--- a/.gitignore\n+++ b/.gitignore\n@@ -1 +1 @@ test\n',
+            'new_path' => '.gitignore',
+            'old_path' => '.gitignore',
+            'a_mode' => '100644',
+            'b_mode' => '100644',
+            'new_file' => false,
+            'renamed_file' => false,
+            'deleted_file' => false,
+            'too_large' => false
+          }
+        ]
+      }
+    }.merge(attributes)
+  end
+
+  let(:attributes) { {} }
+  let(:imported_mr) { project.merge_requests.find_by_title(mr['title']) }
+
+  subject(:pipeline) { described_class.new(context) }
+
+  describe '#run' do
+    before do
+      group.add_owner(user)
+
+      allow_next_instance_of(BulkImports::Common::Extractors::NdjsonExtractor) do |extractor|
+        allow(extractor).to receive(:remove_tmp_dir)
+        allow(extractor).to receive(:extract).and_return(BulkImports::Pipeline::ExtractedData.new(data: [[mr, 0]]))
+      end
+
+      allow(project.repository).to receive(:fetch_source_branch!).and_return(true)
+      allow(project.repository).to receive(:branch_exists?).and_return(false)
+      allow(project.repository).to receive(:create_branch)
+
+      pipeline.run
+    end
+
+    it 'imports a merge request' do
+      expect(project.merge_requests.count).to eq(1)
+      expect(imported_mr.title).to eq(mr['title'])
+      expect(imported_mr.description).to eq(mr['description'])
+      expect(imported_mr.state).to eq(mr['state'])
+      expect(imported_mr.iid).to eq(mr['iid'])
+      expect(imported_mr.created_at).to eq(mr['created_at'])
+      expect(imported_mr.updated_at).to eq(mr['updated_at'])
+      expect(imported_mr.author).to eq(user)
+    end
+
+    context 'merge request state' do
+      context 'when mr is closed' do
+        let(:attributes) { { 'state' => 'closed' } }
+
+        it 'imported mr as closed' do
+          expect(imported_mr.state).to eq(attributes['state'])
+        end
+      end
+
+      context 'when mr is merged' do
+        let(:attributes) { { 'state' => 'merged' } }
+
+        it 'imported mr as merged' do
+          expect(imported_mr.state).to eq(attributes['state'])
+        end
+      end
+    end
+
+    context 'source & target project' do
+      it 'has the new project as target' do
+        expect(imported_mr.target_project).to eq(project)
+      end
+
+      it 'has the new project as source' do
+        expect(imported_mr.source_project).to eq(project)
+      end
+
+      context 'when source/target projects differ' do
+        let(:attributes) { { 'source_project_id' => 4321 } }
+
+        it 'has no source' do
+          expect(imported_mr.source_project).to be_nil
+        end
+
+        context 'when diff_head_sha is present' do
+          let(:attributes) { { 'diff_head_sha' => 'HEAD', 'source_project_id' => 4321 } }
+
+          it 'has the new project as source' do
+            expect(imported_mr.source_project).to eq(project)
+          end
+        end
+      end
+    end
+
+    context 'resource label events' do
+      let(:attributes) { { 'resource_label_events' => [{ 'action' => 'add', 'user_id' => 1 }] } }
+
+      it 'restores resource label events' do
+        expect(imported_mr.resource_label_events.first.action).to eq('add')
+      end
+    end
+
+    context 'award emoji' do
+      let(:attributes) { { 'award_emoji' => [{ 'name' => 'tada', 'user_id' => 22 }] } }
+
+      it 'has award emoji' do
+        expect(imported_mr.award_emoji.first.name).to eq(attributes['award_emoji'].first['name'])
+      end
+    end
+
+    context 'notes' do
+      let(:note) { imported_mr.notes.first }
+      let(:attributes) do
+        {
+          'notes' => [
+            {
+              'note' => 'Issue note',
+              'note_html' => '<p>something else entirely</p>',
+              'cached_markdown_version' => 917504,
+              'author_id' => 22,
+              'author' => { 'name' => 'User 22' },
+              'created_at' => '2016-06-14T15:02:56.632Z',
+              'updated_at' => '2016-06-14T15:02:47.770Z',
+              'award_emoji' => [{ 'name' => 'clapper', 'user_id' => 22 }]
+            }
+          ]
+        }
+      end
+
+      it 'imports mr note' do
+        expect(note).to be_present
+        expect(note.note).to include('By User 22')
+        expect(note.note).to include(attributes['notes'].first['note'])
+        expect(note.author).to eq(user)
+      end
+
+      it 'has award emoji' do
+        emoji = note.award_emoji.first
+
+        expect(emoji.name).to eq('clapper')
+        expect(emoji.user).to eq(user)
+      end
+
+      it 'does not import note_html' do
+        expect(note.note_html).to match(attributes['notes'].first['note'])
+        expect(note.note_html).not_to match(attributes['notes'].first['note_html'])
+      end
+    end
+
+    context 'system note metadata' do
+      let(:attributes) do
+        {
+          'notes' => [
+            {
+              'note' => 'added 3 commits',
+              'system' => true,
+              'author_id' => 22,
+              'author' => { 'name' => 'User 22' },
+              'created_at' => '2016-06-14T15:02:56.632Z',
+              'updated_at' => '2016-06-14T15:02:47.770Z',
+              'system_note_metadata' => { 'action' => 'commit', 'commit_count' => 3 }
+            }
+          ]
+        }
+      end
+
+      it 'restores system note metadata' do
+        note = imported_mr.notes.first
+
+        expect(note.system).to eq(true)
+        expect(note.noteable_type).to eq('MergeRequest')
+        expect(note.system_note_metadata.action).to eq('commit')
+        expect(note.system_note_metadata.commit_count).to eq(3)
+      end
+    end
+
+    context 'diffs' do
+      it 'imports merge request diff' do
+        expect(imported_mr.merge_request_diff).to be_present
+      end
+
+      it 'has the correct data for merge request latest_merge_request_diff' do
+        expect(imported_mr.latest_merge_request_diff_id).to eq(imported_mr.merge_request_diffs.maximum(:id))
+      end
+
+      it 'imports diff files' do
+        expect(imported_mr.merge_request_diff.merge_request_diff_files.count).to eq(1)
+      end
+
+      context 'diff commits' do
+        it 'imports diff commits' do
+          expect(imported_mr.merge_request_diff.merge_request_diff_commits.count).to eq(1)
+        end
+
+        it 'assigns committer and author details to diff commits' do
+          commit = imported_mr.merge_request_diff.merge_request_diff_commits.first
+
+          expect(commit.commit_author_id).not_to be_nil
+          expect(commit.committer_id).not_to be_nil
+        end
+
+        it 'assigns the correct commit users to diff commits' do
+          commit = MergeRequestDiffCommit.find_by(sha: 'COMMIT1')
+
+          expect(commit.commit_author.name).to eq('Commit Author')
+          expect(commit.commit_author.email).to eq('gitlab@example.com')
+          expect(commit.committer.name).to eq('Committer')
+          expect(commit.committer.email).to eq('committer@example.com')
+        end
+      end
+    end
+
+    context 'labels' do
+      let(:attributes) do
+        {
+          'label_links' => [
+            { 'label' => { 'title' => 'imported label 1', 'type' => 'ProjectLabel' } },
+            { 'label' => { 'title' => 'imported label 2', 'type' => 'ProjectLabel' } }
+          ]
+        }
+      end
+
+      it 'imports labels' do
+        expect(imported_mr.labels.pluck(:title)).to contain_exactly('imported label 1', 'imported label 2')
+      end
+    end
+
+    context 'milestone' do
+      let(:attributes) { { 'milestone' => { 'title' => 'imported milestone' } } }
+
+      it 'imports milestone' do
+        expect(imported_mr.milestone.title).to eq(attributes.dig('milestone', 'title'))
+      end
+    end
+  end
+end

spec/lib/bulk_imports/projects/pipelines/repository_pipeline_spec.rb

@@ -3,71 +3,72 @@
 require 'spec_helper'
 
 RSpec.describe BulkImports::Projects::Pipelines::RepositoryPipeline do
-  describe '#run' do
-    let_it_be(:user) { create(:user) }
-    let_it_be(:parent) { create(:project) }
-    let_it_be(:bulk_import) { create(:bulk_import, user: user) }
-    let_it_be(:bulk_import_configuration) { create(:bulk_import_configuration, bulk_import: bulk_import) }
-
-    let_it_be(:entity) do
-      create(
-        :bulk_import_entity,
-        :project_entity,
-        bulk_import: bulk_import,
-        source_full_path: 'source/full/path',
-        destination_name: 'My Destination Repository',
-        destination_namespace: parent.full_path,
-        project: parent
-      )
-    end
+  let_it_be(:user) { create(:user) }
+  let_it_be(:parent) { create(:project) }
+  let_it_be(:bulk_import) { create(:bulk_import, user: user) }
+  let_it_be(:bulk_import_configuration) { create(:bulk_import_configuration, bulk_import: bulk_import) }
+
+  let_it_be(:entity) do
+    create(
+      :bulk_import_entity,
+      :project_entity,
+      bulk_import: bulk_import,
+      source_full_path: 'source/full/path',
+      destination_name: 'My Destination Repository',
+      destination_namespace: parent.full_path,
+      project: parent
+    )
+  end
 
-    let_it_be(:tracker) { create(:bulk_import_tracker, entity: entity) }
-    let_it_be(:context) { BulkImports::Pipeline::Context.new(tracker) }
+  let_it_be(:tracker) { create(:bulk_import_tracker, entity: entity) }
+  let_it_be(:context) { BulkImports::Pipeline::Context.new(tracker) }
 
-    context 'successfully imports repository' do
-      let(:project_data) do
-        {
-          'httpUrlToRepo' => 'http://test.git'
-        }
-      end
+  let(:extracted_data) { BulkImports::Pipeline::ExtractedData.new(data: project_data) }
 
-      subject { described_class.new(context) }
+  subject(:pipeline) { described_class.new(context) }
+
+  before do
+    allow_next_instance_of(BulkImports::Common::Extractors::GraphqlExtractor) do |extractor|
+      allow(extractor).to receive(:extract).and_return(extracted_data)
+    end
+  end
+
+  describe '#run' do
+    context 'successfully imports repository' do
+      let(:project_data) { { 'httpUrlToRepo' => 'http://test.git' } }
 
       it 'imports new repository into destination project' do
-        allow_next_instance_of(BulkImports::Common::Extractors::GraphqlExtractor) do |extractor|
-          allow(extractor).to receive(:extract).and_return(BulkImports::Pipeline::ExtractedData.new(data: project_data))
-        end
+        url = project_data['httpUrlToRepo'].sub("://", "://oauth2:#{bulk_import_configuration.access_token}@")
 
-        expect_next_instance_of(Gitlab::GitalyClient::RepositoryService) do |repository_service|
-          url = project_data['httpUrlToRepo'].sub("://", "://oauth2:#{bulk_import_configuration.access_token}@")
-          expect(repository_service).to receive(:import_repository).with(url).and_return 0
-        end
+        expect(context.portable).to receive(:ensure_repository)
+        expect(context.portable.repository).to receive(:fetch_as_mirror).with(url)
 
-        subject.run
+        pipeline.run
       end
     end
 
     context 'blocked local networks' do
-      let(:project_data) do
-        {
-          'httpUrlToRepo' => 'http://localhost/foo.git'
-        }
-      end
+      let(:project_data) { { 'httpUrlToRepo' => 'http://localhost/foo.git' } }
 
-      before do
+      it 'imports new repository into destination project' do
         allow(Gitlab.config.gitlab).to receive(:host).and_return('notlocalhost.gitlab.com')
         allow(Gitlab::CurrentSettings).to receive(:allow_local_requests_from_web_hooks_and_services?).and_return(false)
-        allow_next_instance_of(BulkImports::Common::Extractors::GraphqlExtractor) do |extractor|
-          allow(extractor).to receive(:extract).and_return(BulkImports::Pipeline::ExtractedData.new(data: project_data))
-        end
-      end
 
-      subject { described_class.new(context) }
+        pipeline.run
 
-      it 'imports new repository into destination project' do
-        subject.run
-        expect(context.entity.failed?).to be_truthy
+        expect(context.entity.failed?).to eq(true)
       end
     end
   end
+
+  describe '#after_run' do
+    it 'executes housekeeping service after import' do
+      service = instance_double(Repositories::HousekeepingService)
+
+      expect(Repositories::HousekeepingService).to receive(:new).with(context.portable, :gc).and_return(service)
+      expect(service).to receive(:execute)
+
+      pipeline.after_run(context)
+    end
+  end
 end

spec/lib/bulk_imports/projects/stage_spec.rb

@@ -10,6 +10,7 @@ RSpec.describe BulkImports::Projects::Stage do
       [2, BulkImports::Common::Pipelines::LabelsPipeline],
       [3, BulkImports::Projects::Pipelines::IssuesPipeline],
       [4, BulkImports::Common::Pipelines::BoardsPipeline],
+      [4, BulkImports::Projects::Pipelines::MergeRequestsPipeline],
       [5, BulkImports::Common::Pipelines::UploadsPipeline],
       [6, BulkImports::Common::Pipelines::EntityFinisher]
     ]