gitlab_quality-test_tooling 3.0.0 → 3.7.0

data/lib/gitlab_quality/test_tooling/report/generate_test_session.rb

@@ -1,288 +0,0 @@
-# frozen_string_literal: true
-
-require 'erb'
-require 'date'
-
-module GitlabQuality
-  module TestTooling
-    module Report
-      class GenerateTestSession < ReportAsIssue
-        def initialize(ci_project_token:, pipeline_stages: nil, **kwargs)
-          super
-          @ci_project_token = ci_project_token
-          @pipeline_stages = Set.new(pipeline_stages)
-          @issue_type = 'issue'
-        end
-
-        private
-
-        attr_reader :ci_project_token, :pipeline_stages
-
-        # rubocop:disable Metrics/AbcSize
-        def run!
-          puts "Generating test results in `#{files.join(',')}` as issues in project `#{project}` via the API at `#{Runtime::Env.gitlab_api_base}`."
-
-          tests = Dir.glob(files).flat_map do |path|
-            puts "Loading tests in #{path}"
-
-            TestResults::JsonTestResults.new(path: path).to_a
-          end
-
-          tests = tests.select { |test| pipeline_stages.include? test.report["stage"] } unless pipeline_stages.empty?
-
-          issue = gitlab.create_issue(
-            title: "#{Time.now.to_date.iso8601} Test session report | #{Runtime::Env.qa_run_type}",
-            description: generate_description(tests),
-            labels: ['automation:bot-authored', 'E2E', 'triage report', pipeline_name_label, 'suppress-contributor-links'],
-            confidential: confidential
-          )
-
-          # Workaround for https://gitlab.com/gitlab-org/gitlab/-/issues/295493
-          unless Runtime::Env.qa_issue_url.to_s.empty?
-            gitlab.create_issue_note(
-              iid: issue.iid,
-              note: "/relate #{Runtime::Env.qa_issue_url}")
-          end
-
-          issue&.web_url # Issue isn't created in dry-run mode
-        end
-        # rubocop:enable Metrics/AbcSize
-
-        def generate_description(tests)
-          <<~MARKDOWN.rstrip
-          ## Session summary
-
-          * Deploy version: #{Runtime::Env.deploy_version}
-          * Deploy environment: #{Runtime::Env.deploy_environment}
-          * Pipeline: #{Runtime::Env.pipeline_from_project_name} [#{Runtime::Env.ci_pipeline_id}](#{Runtime::Env.ci_pipeline_url})
-          #{generate_summary(tests: tests)}
-
-          #{generate_failed_jobs_listing}
-
-          #{generate_stages_listing(tests)}
-
-          #{generate_qa_issue_relation}
-
-          #{generate_link_to_dashboard}
-          MARKDOWN
-        end
-
-        def generate_summary(tests:, tests_by_status: nil)
-          tests_by_status ||= tests.group_by(&:status)
-          total = tests.size
-          passed = tests_by_status['passed']&.size || 0
-          failed = tests_by_status['failed']&.size || 0
-          others = total - passed - failed
-
-          <<~MARKDOWN.chomp
-          * Total #{total} tests
-          * Passed #{passed} tests
-          * Failed #{failed} tests
-          * #{others} other tests (usually skipped)
-          MARKDOWN
-        end
-
-        def generate_failed_jobs_listing
-          failed_jobs = fetch_pipeline_failed_jobs
-          listings = failed_jobs.filter_map do |job|
-            next if pipeline_stages.any? && !pipeline_stages.include?(job.stage)
-
-            allowed_to_fail = ' (allowed to fail)' if job.allow_failure
-
-            "* [#{job.name}](#{job.web_url})#{allowed_to_fail}"
-          end.join("\n")
-
-          <<~MARKDOWN.chomp if failed_jobs.any?
-          ## Failed jobs
-
-          #{listings}
-          MARKDOWN
-        end
-
-        def generate_stages_listing(tests)
-          generate_tests_by_stage(tests).map do |stage, tests_for_stage|
-            tests_by_status = tests_for_stage.group_by(&:status)
-
-            <<~MARKDOWN.chomp
-            ### #{stage&.capitalize || 'Unknown'}
-
-            #{generate_summary(
-              tests: tests_for_stage, tests_by_status: tests_by_status)}
-
-            #{generate_testcase_listing_by_status(
-              tests: tests_for_stage, tests_by_status: tests_by_status)}
-            MARKDOWN
-          end.join("\n\n")
-        end
-
-        def generate_tests_by_stage(tests)
-          # https://about.gitlab.com/handbook/product/product-categories/#devops-stages
-          ordering = %w[
-            manage
-            plan
-            create
-            verify
-            package
-            release
-            configure
-            monitor
-            secure
-            defend
-            growth
-            fulfillment
-            enablement
-            self-managed
-            saas
-          ]
-
-          tests.sort_by do |test|
-            ordering.index(test.stage) || ordering.size
-          end.group_by(&:stage)
-        end
-
-        def generate_testcase_listing_by_status(tests:, tests_by_status:)
-          failed_tests = tests_by_status['failed']
-          passed_tests = tests_by_status['passed']
-          other_tests = tests.reject do |test|
-            test.status == 'failed' || test.status == 'passed'
-          end
-
-          [
-            (failed_listings(failed_tests) if failed_tests),
-            (passed_listings(passed_tests) if passed_tests),
-            (other_listings(other_tests) if other_tests.any?)
-          ].compact.join("\n\n")
-        end
-
-        def failed_listings(failed_tests)
-          generate_testcase_listing(failed_tests)
-        end
-
-        def passed_listings(passed_tests)
-          <<~MARKDOWN.chomp
-            <details><summary>Passed tests:</summary>
-
-            #{generate_testcase_listing(passed_tests, passed: true)}
-
-            </details>
-          MARKDOWN
-        end
-
-        def other_listings(other_tests)
-          <<~MARKDOWN.chomp
-            <details><summary>Other tests:</summary>
-
-            #{generate_testcase_listing(other_tests)}
-
-            </details>
-          MARKDOWN
-        end
-
-        def generate_testcase_listing(tests, passed: false)
-          body = tests.group_by(&:testcase).map do |testcase, tests_with_same_testcase|
-            tests_with_same_testcase.sort_by!(&:name)
-            [
-              generate_test_text(testcase, tests_with_same_testcase, passed),
-              generate_test_job(tests_with_same_testcase),
-              generate_test_status(tests_with_same_testcase),
-              generate_test_actions(tests_with_same_testcase)
-            ].join(' | ')
-          end.join("\n")
-
-          <<~MARKDOWN.chomp
-          | Test | Job | Status | Action |
-          | - | - | - | - |
-          #{body}
-          MARKDOWN
-        end
-
-        def generate_test_text(testcase, tests_with_same_testcase, passed)
-          text = tests_with_same_testcase.map(&:name).uniq.join(', ')
-
-          if testcase && !passed
-            "[#{text}](#{testcase})"
-          else
-            text
-          end
-        end
-
-        def generate_test_job(tests_with_same_testcase)
-          tests_with_same_testcase.map do |test|
-            ci_job_id = test.ci_job_url[/\d+\z/]
-
-            "[#{ci_job_id}](#{test.ci_job_url})#{' ~"quarantine"' if test.quarantine?}"
-          end.uniq.join(', ')
-        end
-
-        def generate_test_status(tests_with_same_testcase)
-          tests_with_same_testcase.map(&:status).uniq.map do |status|
-            %(~"#{status}")
-          end.join(', ')
-        end
-
-        def generate_test_actions(tests_with_same_testcase)
-          # All failed tests would be grouped together, meaning that
-          # if one failed, all the tests here would be failed too.
-          # So this check is safe. Same applies to 'passed'.
-          # But all other status might be mixing together,
-          # we cannot assume other statuses.
-          if tests_with_same_testcase.first.status == 'failed'
-            tests_having_failure_issue =
-              tests_with_same_testcase.select(&:failure_issue)
-
-            if tests_having_failure_issue.any?
-              items = tests_having_failure_issue.uniq(&:failure_issue).map do |test|
-                "<li>[ ] [failure issue](#{test.failure_issue})</li>"
-              end.join(' ')
-
-              "<ul>#{items}</ul>"
-            else
-              '<ul><li>[ ] failure issue exists or was created</li></ul>'
-            end
-          else
-            '-'
-          end
-        end
-
-        def generate_qa_issue_relation
-          return unless Runtime::Env.qa_issue_url
-
-          <<~MARKDOWN.chomp
-          ## Release QA issue
-
-          * #{Runtime::Env.qa_issue_url}
-
-          /relate #{Runtime::Env.qa_issue_url}
-          MARKDOWN
-        end
-
-        def generate_link_to_dashboard
-          return unless Runtime::Env.qa_run_type
-
-          <<~MARKDOWN.chomp
-          ## Link to Grafana dashboard for run-type of #{Runtime::Env.qa_run_type}
-
-          * https://dashboards.quality.gitlab.net/d/tR_SmBDVk/main-runs?orgId=1&refresh=1m&var-run_type=#{Runtime::Env.qa_run_type}
-          MARKDOWN
-        end
-
-        def fetch_pipeline_failed_jobs
-          failed_jobs = []
-
-          ci_project_client = Gitlab.client(
-            endpoint: Runtime::Env.ci_api_v4_url,
-            private_token: ci_project_token)
-
-          gitlab.handle_gitlab_client_exceptions do
-            failed_jobs = ci_project_client.pipeline_jobs(
-              Runtime::Env.ci_project_id,
-              Runtime::Env.ci_pipeline_id,
-              scope: 'failed')
-          end
-
-          failed_jobs
-        end
-      end
-    end
-  end
-end

data/lib/gitlab_quality/test_tooling/report/test_health_issue_finder.rb

@@ -1,79 +0,0 @@
-# frozen_string_literal: true
-
-require 'http'
-require 'json'
-
-module GitlabQuality
-  module TestTooling
-    module Report
-      class TestHealthIssueFinder < ReportAsIssue
-        HEALTH_PROBLEM_TYPE_TO_LABEL = {
-          'pass-after-retry' => 'test-health:pass-after-retry',
-          'slow' => 'test-health:slow',
-          'failures' => 'test-health:failures'
-        }.freeze
-
-        def initialize(health_problem_type: [], **kwargs)
-          super(**kwargs)
-
-          @health_problem_type = health_problem_type
-        end
-
-        def found_existing_unhealthy_test_issue?
-          issue_url = invoke!
-
-          !issue_url.nil? && !issue_url.empty?
-        end
-
-        def run!
-          existing_issue_found = nil
-
-          applicable_tests.each do |test|
-            issues = find_issues_by_hash(test_hash(test), state: 'opened', labels: search_labels)
-            next if issues.empty?
-
-            existing_issue_found = issues.first.web_url
-            puts "Found an existing test health issue of type #{health_problem_type} for test #{test.file}:#{test.line_number}: #{existing_issue_found}."
-            break
-          end
-
-          puts "Did not find an existing test health issue of type #{health_problem_type}." unless existing_issue_found
-
-          existing_issue_found
-        end
-
-        def applicable_tests
-          applicable_tests = []
-
-          TestResults::Builder.new(file_glob: files, token: token, project: project).test_results_per_file do |test_results|
-            applicable_tests = test_results.select { |test| test_is_applicable?(test) }
-          end
-
-          applicable_tests
-        end
-
-        private
-
-        attr_reader :health_problem_type
-
-        # Be mindful about the number of tests this method would return,
-        # as we will make at least one API request per test.
-        def test_is_applicable?(test)
-          expected_test_status =
-            case health_problem_type
-            when 'failures'
-              'failed'
-            else
-              'passed'
-            end
-
-          test.status == expected_test_status
-        end
-
-        def search_labels
-          ['test', HEALTH_PROBLEM_TYPE_TO_LABEL[health_problem_type]]
-        end
-      end
-    end
-  end
-end