Merge branch 'caalberts-use-crystalball-with-knapsack' into 'master'

Use test detection based on crystalball mapping in rspec jobs (take 3)

See merge request gitlab-org/gitlab!49446

.gitlab/ci/rails.gitlab-ci.yml

@@ -22,7 +22,7 @@
     RUBY_GC_MALLOC_LIMIT_MAX: 134217728
     CRYSTALBALL: "true"
     RECORD_DEPRECATIONS: "true"
-  needs: ["setup-test-env", "retrieve-tests-metadata", "compile-test-assets"]
+  needs: ["setup-test-env", "retrieve-tests-metadata", "compile-test-assets", "detect-tests"]
   script:
     - *base-script
     - rspec_paralellized_job "--tag ~quarantine --tag ~geo --tag ~level:migration"
@@ -64,7 +64,7 @@
     - .rspec-base
     - .as-if-foss
     - .use-pg11
-  needs: ["setup-test-env", "retrieve-tests-metadata", "compile-test-assets as-if-foss"]
+  needs: ["setup-test-env", "retrieve-tests-metadata", "compile-test-assets as-if-foss", "detect-tests"]
 
 .rspec-ee-base-pg11:
   extends:

.gitlab/ci/rules.gitlab-ci.yml

@@ -604,10 +604,9 @@
   rules:
     - <<: *if-not-ee
       when: never
-    - <<: *if-security-merge-request
-      changes: *code-backstage-patterns
-    - <<: *if-dot-com-gitlab-org-merge-request
+    - <<: *if-default-refs
       changes: *code-backstage-patterns
+    - <<: *if-merge-request-title-run-all-rspec
 
 .rails:rules:rspec-foss-impact:
   rules:

.gitlab/ci/setup.gitlab-ci.yml

@@ -61,15 +61,17 @@ verify-tests-yml:
     - scripts/verify-tff-mapping
 
 .detect-test-base:
-  image: ruby:2.7-alpine
+  image: ruby:2.7
   needs: []
   stage: prepare
   script:
-    - source scripts/utils.sh
+    - source ./scripts/utils.sh
+    - source ./scripts/rspec_helpers.sh
     - install_gitlab_gem
     - install_tff_gem
-    - tooling/bin/find_foss_tests ${MATCHED_TESTS_FILE}
-    - 'echo "test files affected: $(cat $MATCHED_TESTS_FILE)"'
+    - retrieve_tests_mapping
+    - 'if [ -n "$CI_MERGE_REQUEST_IID" ]; then tooling/bin/find_tests ${MATCHED_TESTS_FILE}; fi'
+    - 'if [ -n "$CI_MERGE_REQUEST_IID" ]; then echo "test files affected: $(cat $MATCHED_TESTS_FILE)"; fi'
   artifacts:
     expire_in: 7d
     paths:

scripts/api/get_job_id

@@ -20,6 +20,7 @@ class JobFinder
     @job_query = options.delete(:job_query)
     @pipeline_id = options.delete(:pipeline_id)
     @job_name = options.delete(:job_name)
+    @artifact_path = options.delete(:artifact_path)
 
     # Force the token to be a string so that if api_token is nil, it's set to '', allowing unauthenticated requests (for forks).
     api_token = options.delete(:api_token).to_s
@@ -33,19 +34,31 @@ class JobFinder
   end
 
   def execute
-    find_job_with_filtered_pipelines || find_job_in_pipeline
+    find_job_with_artifact || find_job_with_filtered_pipelines || find_job_in_pipeline
   end
 
   private
 
-  attr_reader :project, :pipeline_query, :job_query, :pipeline_id, :job_name
+  attr_reader :project, :pipeline_query, :job_query, :pipeline_id, :job_name, :artifact_path
+
+  def find_job_with_artifact
+    return if artifact_path.nil?
+
+    Gitlab.pipelines(project, pipeline_query_params).auto_paginate do |pipeline|
+      Gitlab.pipeline_jobs(project, pipeline.id, job_query_params).auto_paginate do |job|
+        return job if found_job_with_artifact?(job) # rubocop:disable Cop/AvoidReturnFromBlocks
+      end
+    end
+
+    raise 'Job not found!'
+  end
 
   def find_job_with_filtered_pipelines
     return if pipeline_query.empty?
 
     Gitlab.pipelines(project, pipeline_query_params).auto_paginate do |pipeline|
       Gitlab.pipeline_jobs(project, pipeline.id, job_query_params).auto_paginate do |job|
-        return job if job.name == job_name # rubocop:disable Cop/AvoidReturnFromBlocks
+        return job if found_job_by_name?(job) # rubocop:disable Cop/AvoidReturnFromBlocks
       end
     end
 
@@ -56,12 +69,22 @@ class JobFinder
     return unless pipeline_id
 
     Gitlab.pipeline_jobs(project, pipeline_id, job_query_params).auto_paginate do |job|
-      return job if job.name == job_name # rubocop:disable Cop/AvoidReturnFromBlocks
+      return job if found_job_by_name?(job) # rubocop:disable Cop/AvoidReturnFromBlocks
     end
 
     raise 'Job not found!'
   end
 
+  def found_job_with_artifact?(job)
+    artifact_url = "https://gitlab.com/api/v4/projects/#{CGI.escape(project)}/jobs/#{job.id}/artifacts/#{artifact_path}"
+    response = HTTParty.head(artifact_url) # rubocop:disable Gitlab/HTTParty
+    response.success?
+  end
+
+  def found_job_by_name?(job)
+    job.name == job_name
+  end
+
   def pipeline_query_params
     @pipeline_query_params ||= { per_page: 100, **pipeline_query }
   end
@@ -95,6 +118,10 @@ if $0 == __FILE__
       options[:job_name] = value
     end
 
+    opts.on("-a", "--artifact-path ARTIFACT_PATH", String, "A valid artifact path") do |value|
+      options[:artifact_path] = value
+    end
+
     opts.on("-t", "--api-token API_TOKEN", String, "A value API token with the `read_api` scope") do |value|
       options[:api_token] = value
     end

scripts/rspec_helpers.sh

 #!/usr/bin/env bash
 
 function retrieve_tests_metadata() {
-  mkdir -p crystalball/ knapsack/ rspec_flaky/ rspec_profiling/
+  mkdir -p knapsack/ rspec_flaky/ rspec_profiling/
 
   local project_path="gitlab-org/gitlab"
   local test_metadata_job_id
@@ -16,13 +16,6 @@ function retrieve_tests_metadata() {
   if [[ ! -f "${FLAKY_RSPEC_SUITE_REPORT_PATH}" ]]; then
     scripts/api/download_job_artifact --project "${project_path}" --job-id "${test_metadata_job_id}" --artifact-path "${FLAKY_RSPEC_SUITE_REPORT_PATH}" || echo "{}" > "${FLAKY_RSPEC_SUITE_REPORT_PATH}"
   fi
-
-  # FIXME: We will need to find a pipeline where the $RSPEC_PACKED_TESTS_MAPPING_PATH.gz actually exists (Crystalball only runs every two-hours, but the `update-tests-metadata` runs for all `master` pipelines...).
-  # if [[ ! -f "${RSPEC_PACKED_TESTS_MAPPING_PATH}" ]]; then
-  #   (scripts/api/download_job_artifact --project "${project_path}" --job-id "${test_metadata_job_id}" --artifact-path "${RSPEC_PACKED_TESTS_MAPPING_PATH}.gz" && gzip -d "${RSPEC_PACKED_TESTS_MAPPING_PATH}.gz") || echo "{}" > "${RSPEC_PACKED_TESTS_MAPPING_PATH}"
-  # fi
-  #
-  # scripts/unpack-test-mapping "${RSPEC_PACKED_TESTS_MAPPING_PATH}" "${RSPEC_TESTS_MAPPING_PATH}"
 }
 
 function update_tests_metadata() {
@@ -43,6 +36,21 @@ function update_tests_metadata() {
   fi
 }
 
+function retrieve_tests_mapping() {
+  mkdir -p crystalball/
+
+  local project_path="gitlab-org/gitlab"
+  local test_metadata_with_mapping_job_id
+
+  test_metadata_with_mapping_job_id=$(scripts/api/get_job_id --project "${project_path}" -q "status=success" -q "ref=master" -q "username=gitlab-bot" -Q "scope=success" --job-name "update-tests-metadata" --artifact-path "${RSPEC_PACKED_TESTS_MAPPING_PATH}.gz")
+
+  if [[ ! -f "${RSPEC_PACKED_TESTS_MAPPING_PATH}" ]]; then
+   (scripts/api/download_job_artifact --project "${project_path}" --job-id "${test_metadata_with_mapping_job_id}" --artifact-path "${RSPEC_PACKED_TESTS_MAPPING_PATH}.gz" && gzip -d "${RSPEC_PACKED_TESTS_MAPPING_PATH}.gz") || echo "{}" > "${RSPEC_PACKED_TESTS_MAPPING_PATH}"
+  fi
+
+  scripts/unpack-test-mapping "${RSPEC_PACKED_TESTS_MAPPING_PATH}" "${RSPEC_TESTS_MAPPING_PATH}"
+}
+
 function update_tests_mapping() {
   if ! crystalball_rspec_data_exists; then
     echo "No crystalball rspec data found."
@@ -119,8 +127,8 @@ function rspec_paralellized_job() {
 
   local rspec_args="-Ispec -rspec_helper --color --format documentation --format RspecJunitFormatter --out junit_rspec.xml ${rspec_opts}"
 
-  if [[ -n $RSPEC_MATCHING_TESTS_ENABLED ]]; then
-    tooling/bin/parallel_rspec --rspec_args "${rspec_args}" --filter tmp/matching_tests.txt
+  if [[ -n $RSPEC_TESTS_MAPPING_ENABLED ]]; then
+    tooling/bin/parallel_rspec --rspec_args "${rspec_args}" --filter "tmp/matching_tests.txt"
   else
     tooling/bin/parallel_rspec --rspec_args "${rspec_args}"
   fi

scripts/utils.sh

@@ -36,7 +36,7 @@ function install_gitlab_gem() {
 }
 
 function install_tff_gem() {
-  gem install test_file_finder --version 0.1.0
+  gem install test_file_finder --version 0.1.1
 }
 
 function run_timed_command() {

tooling/bin/find_foss_tests → tooling/bin/find_tests

@@ -19,7 +19,12 @@ mr_iid = ENV.fetch('CI_MERGE_REQUEST_IID')
 mr_changes = Gitlab.merge_request_changes(mr_project_path, mr_iid)
 changed_files = mr_changes.changes.map { |change| change['new_path'] }
 
-mapping = TestFileFinder::Mapping.load('tests.yml')
-test_files = TestFileFinder::FileFinder.new(paths: changed_files, mapping: mapping).test_files
+tff = TestFileFinder::FileFinder.new(paths: changed_files).tap do |file_finder|
+  file_finder.use TestFileFinder::MappingStrategies::PatternMatching.load('tests.yml')
 
-File.write(output_file, test_files.uniq.join(' '))
+  if ENV['RSPEC_TESTS_MAPPING_ENABLED']
+    file_finder.use TestFileFinder::MappingStrategies::DirectMatching.load_json(ENV['RSPEC_TESTS_MAPPING_PATH'])
+  end
+end
+
+File.write(output_file, tff.test_files.uniq.join(' '))

tooling/lib/tooling/parallel_rspec_runner.rb

@@ -60,7 +60,10 @@ module Tooling
     end
 
     def tests_to_run
-      return node_tests if filter_tests.empty?
+      if filter_tests.empty?
+        Knapsack.logger.info 'Running all node tests without filter'
+        return node_tests
+      end
 
       @tests_to_run ||= node_tests & filter_tests
     end