datadog-ci 1.16.0 → 1.18.0

data/lib/datadog/ci/configuration/components.rb

@@ -4,6 +4,7 @@ require "datadog/core/telemetry/ext"
 
 require_relative "../ext/settings"
 require_relative "../git/tree_uploader"
+require_relative "../impacted_tests_detection/component"
 require_relative "../logs/component"
 require_relative "../logs/transport"
 require_relative "../remote/component"
@@ -36,7 +37,7 @@ module Datadog
       # Adds CI behavior to Datadog trace components
       module Components
         attr_reader :test_visibility, :test_optimisation, :git_tree_upload_worker, :ci_remote, :test_retries,
-          :test_management, :agentless_logs_submission
+          :test_management, :agentless_logs_submission, :impacted_tests_detection
 
         def initialize(settings)
           @test_optimisation = nil
@@ -45,6 +46,7 @@ module Datadog
           @ci_remote = nil
           @test_retries = TestRetries::NullComponent.new
           @test_management = TestManagement::NullComponent.new
+          @impacted_tests_detection = nil
 
           # Activate CI mode if enabled
           if settings.ci.enabled
@@ -138,6 +140,8 @@ module Datadog
           )
 
           @agentless_logs_submission = build_agentless_logs_component(settings, test_visibility_api)
+
+          @impacted_tests_detection = ImpactedTestsDetection::Component.new(enabled: settings.ci.impacted_tests_detection_enabled)
         end
 
         def build_test_optimisation(settings, test_visibility_api)
@@ -237,7 +241,7 @@ module Datadog
 
         def build_git_upload_worker(settings, api)
           if settings.ci.git_metadata_upload_enabled
-            git_tree_uploader = Git::TreeUploader.new(api: api)
+            git_tree_uploader = Git::TreeUploader.new(api: api, force_unshallow: settings.ci.impacted_tests_detection_enabled)
             Worker.new do |repository_url|
               git_tree_uploader.call(repository_url)
             end
@@ -323,13 +327,21 @@ module Datadog
           settings.telemetry.shutdown_timeout_seconds = 60.0
 
           begin
-            require "datadog/core/telemetry/http/adapters/net"
+            require "datadog/core/transport/http/adapters/net"
 
-            # patch gem's telemetry transport layer to use Net::HTTP instead of WebMock's Net::HTTP
-            Core::Telemetry::Http::Adapters::Net.include(CI::Transport::Adapters::TelemetryWebmockSafeAdapter)
+            # patch gem's core transport layer to use Net::HTTP instead of WebMock's Net::HTTP
+            Core::Transport::HTTP::Adapters::Net.include(CI::Transport::Adapters::TelemetryWebmockSafeAdapter)
           rescue LoadError, StandardError => e
             Datadog.logger.warn("Failed to patch Datadog gem's telemetry layer: #{e}")
           end
+
+          # for compatibility with old telemetry transport
+          begin
+            require "datadog/core/telemetry/http/adapters/net"
+            Core::Telemetry::Http::Adapters::Net.include(CI::Transport::Adapters::TelemetryWebmockSafeAdapter)
+          rescue LoadError, StandardError => e
+            Datadog.logger.debug("The old telemetry transport layer is not available: #{e}")
+          end
         end
 
         # When timecop is present:

data/lib/datadog/ci/configuration/settings.rb

@@ -151,6 +151,12 @@ module Datadog
                 o.env CI::Ext::Settings::ENV_TEST_VISIBILITY_DRB_SERVER_URI
               end
 
+              option :impacted_tests_detection_enabled do |o|
+                o.type :bool
+                o.env CI::Ext::Settings::ENV_IMPACTED_TESTS_DETECTION_ENABLED
+                o.default false
+              end
+
               define_method(:instrument) do |integration_name, options = {}, &block|
                 return unless enabled
 

data/lib/datadog/ci/contrib/knapsack/patcher.rb

@@ -9,9 +9,7 @@ module Datadog
         module Patcher
           include Datadog::CI::Contrib::Patcher
 
-          module_function
-
-          def patch
+          def self.patch
             if ::RSpec::Core::Runner.ancestors.include?(::KnapsackPro::Extensions::RSpecExtension::Runner)
               # knapsack already patched rspec runner
               require_relative "runner"

data/lib/datadog/ci/contrib/knapsack/runner.rb

@@ -19,6 +19,7 @@ module Datadog
               return super if ::RSpec.configuration.dry_run? && !datadog_configuration[:dry_run_enabled]
               return super unless datadog_configuration[:enabled]
 
+              # @type var test_session: Datadog::CI::TestSession?
               test_session = test_visibility_component.start_test_session(
                 tags: {
                   CI::Ext::Test::TAG_FRAMEWORK => CI::Contrib::RSpec::Ext::FRAMEWORK,
@@ -27,6 +28,7 @@ module Datadog
                 service: datadog_configuration[:service_name]
               )
 
+              # @type var test_module: Datadog::CI::TestModule?
               test_module = test_visibility_component.start_test_module(CI::Contrib::RSpec::Ext::FRAMEWORK)
 
               result = super

data/lib/datadog/ci/contrib/minitest/runner.rb

@@ -38,6 +38,7 @@ module Datadog
             def run_one_method(klass, method_name)
               return super unless datadog_configuration[:enabled]
 
+              # @type var result: untyped
               result = nil
 
               test_retries_component.with_retries do

data/lib/datadog/ci/contrib/minitest/test.rb

@@ -2,6 +2,7 @@
 
 require_relative "../../ext/test"
 require_relative "../../git/local_repository"
+require_relative "../../utils/source_code"
 require_relative "../instrumentation"
 require_relative "ext"
 require_relative "helpers"
@@ -27,20 +28,33 @@ module Datadog
               end
 
               test_suite_name = Helpers.test_suite_name(self.class, name)
-              source_file, line_number = method(name).source_location
+
+              # @type var tags : Hash[String, String]
+              tags = {
+                CI::Ext::Test::TAG_FRAMEWORK => Ext::FRAMEWORK,
+                CI::Ext::Test::TAG_FRAMEWORK_VERSION => datadog_integration.version.to_s
+              }
+
+              # try to find out where test method starts and ends
+              test_method = method(name)
+              source_file, first_line_number = test_method.source_location
+              last_line_number = Utils::SourceCode.last_line(test_method)
+
+              tags[CI::Ext::Test::TAG_SOURCE_FILE] = Git::LocalRepository.relative_to_root(source_file) if source_file
+              tags[CI::Ext::Test::TAG_SOURCE_START] = first_line_number.to_s if first_line_number
+              tags[CI::Ext::Test::TAG_SOURCE_END] = last_line_number.to_s if last_line_number
 
               test_span = test_visibility_component.trace_test(
                 name,
                 test_suite_name,
-                tags: {
-                  CI::Ext::Test::TAG_FRAMEWORK => Ext::FRAMEWORK,
-                  CI::Ext::Test::TAG_FRAMEWORK_VERSION => datadog_integration.version.to_s,
-                  CI::Ext::Test::TAG_SOURCE_FILE => Git::LocalRepository.relative_to_root(source_file),
-                  CI::Ext::Test::TAG_SOURCE_START => line_number.to_s
-                },
+                tags: tags,
                 service: datadog_configuration[:service_name]
               )
+              # Steep type checker doesn't know that we patched Minitest::Test class definition
+              #
+              # steep:ignore:start
               test_span&.itr_unskippable! if self.class.dd_suite_unskippable? || self.class.dd_test_unskippable?(name)
+              # steep:ignore:end
               skip(test_span&.datadog_skip_reason) if test_span&.should_skip?
             end
 
@@ -100,9 +114,10 @@ module Datadog
             end
 
             def dd_test_unskippable?(test_name)
-              return false unless @datadog_itr_unskippable_tests
+              tests = @datadog_itr_unskippable_tests
+              return false unless tests
 
-              @datadog_itr_unskippable_tests.include?(test_name)
+              tests.include?(test_name)
             end
           end
         end

data/lib/datadog/ci/contrib/parallel_tests/patcher.rb

@@ -11,9 +11,7 @@ module Datadog
         module Patcher
           include Datadog::CI::Contrib::Patcher
 
-          module_function
-
-          def patch
+          def self.patch
             ::ParallelTests::CLI.include(CLI)
           end
         end

data/lib/datadog/ci/contrib/patcher.rb

@@ -29,9 +29,13 @@ module Datadog
             return unless defined?(super)
 
             patch_only_once.run do
+              # There is no way to tell Steep that we prepend these methods to modules that have .patch method
+              #
+              # steep:ignore:start
               super.tap do
                 @patch_successful = true
               end
+              # steep:ignore:end
             rescue => e
               on_patch_error(e)
             end

data/lib/datadog/ci/contrib/rspec/example.rb

@@ -2,6 +2,7 @@
 
 require_relative "../../ext/test"
 require_relative "../../git/local_repository"
+require_relative "../../utils/source_code"
 require_relative "../../utils/test_run"
 require_relative "../instrumentation"
 require_relative "ext"
@@ -26,17 +27,23 @@ module Datadog
               # don't report test to RSpec::Core::Reporter until retries are done
               @skip_reporting = true
 
+              # @type var tags : Hash[String, String]
+              tags = {
+                CI::Ext::Test::TAG_FRAMEWORK => Ext::FRAMEWORK,
+                CI::Ext::Test::TAG_FRAMEWORK_VERSION => datadog_integration.version.to_s,
+                CI::Ext::Test::TAG_SOURCE_FILE => Git::LocalRepository.relative_to_root(metadata[:file_path]),
+                CI::Ext::Test::TAG_SOURCE_START => metadata[:line_number].to_s,
+                CI::Ext::Test::TAG_PARAMETERS => datadog_test_parameters
+              }
+
+              end_line = Utils::SourceCode.last_line(@example_block)
+              tags[CI::Ext::Test::TAG_SOURCE_END] = end_line.to_s if end_line
+
               test_retries_component.with_retries do
                 test_visibility_component.trace_test(
                   datadog_test_name,
                   datadog_test_suite_name,
-                  tags: {
-                    CI::Ext::Test::TAG_FRAMEWORK => Ext::FRAMEWORK,
-                    CI::Ext::Test::TAG_FRAMEWORK_VERSION => datadog_integration.version.to_s,
-                    CI::Ext::Test::TAG_SOURCE_FILE => Git::LocalRepository.relative_to_root(metadata[:file_path]),
-                    CI::Ext::Test::TAG_SOURCE_START => metadata[:line_number].to_s,
-                    CI::Ext::Test::TAG_PARAMETERS => datadog_test_parameters
-                  },
+                  tags: tags,
                   service: datadog_configuration[:service_name]
                 ) do |test_span|
                   test_span&.itr_unskippable! if datadog_unskippable?

data/lib/datadog/ci/contrib/rspec/helpers.rb

@@ -6,9 +6,7 @@ module Datadog
       module RSpec
         # Helper methods for RSpec instrumentation
         module Helpers
-          module_function
-
-          def parallel_tests?
+          def self.parallel_tests?
             !!ENV.fetch("TEST_ENV_NUMBER", nil) && !!ENV.fetch("PARALLEL_TEST_GROUPS", nil)
           end
         end

data/lib/datadog/ci/ext/environment/extractor.rb

@@ -46,14 +46,12 @@ module Datadog
               Git::TAG_COMMIT_COMMITTER_EMAIL => @provider.git_commit_committer_email,
               Git::TAG_COMMIT_COMMITTER_NAME => @provider.git_commit_committer_name,
               Git::TAG_COMMIT_MESSAGE => @provider.git_commit_message,
-              Git::TAG_COMMIT_SHA => @provider.git_commit_sha
+              Git::TAG_COMMIT_SHA => @provider.git_commit_sha,
+              Git::TAG_PULL_REQUEST_BASE_BRANCH => @provider.git_pull_request_base_branch,
+              Git::TAG_PULL_REQUEST_BASE_BRANCH_SHA => @provider.git_pull_request_base_branch_sha,
+              Git::TAG_COMMIT_HEAD_SHA => @provider.git_commit_head_sha
             }
 
-            # set additional tags if provider needs them
-            @provider.additional_tags.each do |key, value|
-              @tags[key] = value
-            end
-
             # Normalize Git references and filter sensitive data
             normalize_git!
             # Expand ~

data/lib/datadog/ci/ext/environment/providers/appveyor.rb

@@ -83,6 +83,11 @@ module Datadog
               commit_message
             end
 
+            def git_pull_request_base_branch
+              # from docs: build branch. For Pull Request commits it is base branch PR is merging into
+              env["APPVEYOR_REPO_BRANCH"]
+            end
+
             private
 
             def github_repo_provider?

data/lib/datadog/ci/ext/environment/providers/base.rb

@@ -96,8 +96,13 @@ module Datadog
             def git_commit_sha
             end
 
-            def additional_tags
-              {}
+            def git_pull_request_base_branch
+            end
+
+            def git_pull_request_base_branch_sha
+            end
+
+            def git_commit_head_sha
             end
 
             private

data/lib/datadog/ci/ext/environment/providers/bitbucket.rb

@@ -59,6 +59,12 @@ module Datadog
               env["BITBUCKET_TAG"]
             end
 
+            def git_pull_request_base_branch
+              # from docs: The pull request destination branch (used in combination with BITBUCKET_BRANCH).
+              # Only available on a pull request triggered build
+              env["BITBUCKET_PR_DESTINATION_BRANCH"]
+            end
+
             private
 
             def url

data/lib/datadog/ci/ext/environment/providers/bitrise.rb

@@ -47,7 +47,7 @@ module Datadog
             end
 
             def git_branch
-              env["BITRISEIO_GIT_BRANCH_DEST"] || env["BITRISE_GIT_BRANCH"]
+              env["BITRISEIO_PULL_REQUEST_HEAD_BRANCH"] || env["BITRISE_GIT_BRANCH"]
             end
 
             def git_tag
@@ -73,6 +73,12 @@ module Datadog
             def git_commit_committer_email
               env["GIT_CLONE_COMMIT_COMMITER_EMAIL"] || env["GIT_CLONE_COMMIT_COMMITER_NAME"]
             end
+
+            def git_pull_request_base_branch
+              # from docs: Used only with builds triggered by pull requests: the destination/target branch of the pull request that triggered the build.
+              # For example, a pull request wants to merge the content of a branch into the branch main. In this case, this Env Var’s value is main.
+              env["BITRISEIO_GIT_BRANCH_DEST"]
+            end
           end
         end
       end

data/lib/datadog/ci/ext/environment/providers/buddy.rb

@@ -65,6 +65,11 @@ module Datadog
             def git_commit_committer_email
               env["BUDDY_EXECUTION_REVISION_COMMITTER_EMAIL"]
             end
+
+            def git_pull_request_base_branch
+              # from docs: The name of the Git BASE branch of the currently run Pull Request
+              env["BUDDY_RUN_PR_BASE_BRANCH"]
+            end
           end
         end
       end

data/lib/datadog/ci/ext/environment/providers/github_actions.rb

@@ -76,34 +76,53 @@ module Datadog
               }.reject { |_, v| v.nil? }.to_json
             end
 
-            def additional_tags
-              base_ref = env["GITHUB_BASE_REF"]
-              return {} if base_ref.nil? || base_ref.empty?
+            def git_pull_request_base_branch
+              return nil if github_event_json.nil?
 
-              # @type var result: Hash[String, String]
-              result = {
-                Git::TAG_PULL_REQUEST_BASE_BRANCH => base_ref
-              }
-
-              event_path = env["GITHUB_EVENT_PATH"]
-              event_json = JSON.parse(File.read(event_path))
+              env["GITHUB_BASE_REF"]
+            end
 
-              head_sha = event_json.dig("pull_request", "head", "sha")
-              result[Git::TAG_COMMIT_HEAD_SHA] = head_sha if head_sha
+            def git_pull_request_base_branch_sha
+              return nil if git_pull_request_base_branch.nil?
 
-              base_sha = event_json.dig("pull_request", "base", "sha")
-              result[Git::TAG_PULL_REQUEST_BASE_BRANCH_SHA] = base_sha if base_sha
+              event_json = github_event_json
+              return nil if event_json.nil?
 
-              result
+              event_json.dig("pull_request", "base", "sha")
             rescue => e
-              Datadog.logger.error("Failed to extract additional tags from GitHub Actions: #{e}")
-              Core::Telemetry::Logger.report(e, description: "Failed to extract additional tags from GitHub Actions")
+              Datadog.logger.error("Failed to extract pull request base branch SHA from GitHub Actions: #{e}")
+              Core::Telemetry::Logger.report(e, description: "Failed to extract pull request base branch SHA from GitHub Actions")
+              nil
+            end
 
-              {}
+            def git_commit_head_sha
+              return nil if git_pull_request_base_branch.nil?
+
+              event_json = github_event_json
+              return nil if event_json.nil?
+
+              event_json.dig("pull_request", "head", "sha")
+            rescue => e
+              Datadog.logger.error("Failed to extract commit head SHA from GitHub Actions: #{e}")
+              Core::Telemetry::Logger.report(e, description: "Failed to extract commit head SHA from GitHub Actions")
+              nil
             end
 
             private
 
+            def github_event_json
+              return @github_event_json if defined?(@github_event_json)
+
+              event_path = env["GITHUB_EVENT_PATH"]
+              return @github_event_json = nil if event_path.nil? || event_path.empty?
+
+              @github_event_json = JSON.parse(File.read(event_path))
+            rescue => e
+              Datadog.logger.error("Failed to parse GitHub event JSON: #{e}")
+              Core::Telemetry::Logger.report(e, description: "Failed to parse GitHub event JSON")
+              @github_event_json = nil
+            end
+
             def github_server_url
               return @github_server_url if defined?(@github_server_url)
 

data/lib/datadog/ci/ext/environment/providers/gitlab.rb

@@ -100,13 +100,25 @@ module Datadog
               }.to_json
             end
 
+            def git_pull_request_base_branch
+              env["CI_MERGE_REQUEST_TARGET_BRANCH_NAME"]
+            end
+
+            def git_pull_request_base_branch_sha
+              env["CI_MERGE_REQUEST_TARGET_BRANCH_SHA"]
+            end
+
+            def git_commit_head_sha
+              env["CI_MERGE_REQUEST_SOURCE_BRANCH_SHA"]
+            end
+
             private
 
             def extract_name_email
               return @name_email_tuple if defined?(@name_email_tuple)
 
               name_and_email_string = env["CI_COMMIT_AUTHOR"]
-              if name_and_email_string.include?("<") && (match = /^([^<]*)<([^>]*)>$/.match(name_and_email_string))
+              if name_and_email_string&.include?("<") && (match = /^([^<]*)<([^>]*)>$/.match(name_and_email_string))
                 name = match[1]
                 name = name.strip if name
                 email = match[2]

data/lib/datadog/ci/ext/environment/providers/user_defined_tags.rb

@@ -54,6 +54,18 @@ module Datadog
             def git_commit_committer_date
               env[Git::ENV_COMMIT_COMMITTER_DATE]
             end
+
+            def git_pull_request_base_branch
+              env[Git::ENV_PULL_REQUEST_BASE_BRANCH]
+            end
+
+            def git_pull_request_base_branch_sha
+              env[Git::ENV_PULL_REQUEST_BASE_BRANCH_SHA]
+            end
+
+            def git_commit_head_sha
+              env[Git::ENV_COMMIT_HEAD_SHA]
+            end
           end
         end
       end

data/lib/datadog/ci/ext/git.rb

@@ -36,6 +36,9 @@ module Datadog
         ENV_COMMIT_COMMITTER_NAME = "DD_GIT_COMMIT_COMMITTER_NAME"
         ENV_COMMIT_COMMITTER_EMAIL = "DD_GIT_COMMIT_COMMITTER_EMAIL"
         ENV_COMMIT_COMMITTER_DATE = "DD_GIT_COMMIT_COMMITTER_DATE"
+        ENV_PULL_REQUEST_BASE_BRANCH = "DD_GIT_PULL_REQUEST_BASE_BRANCH"
+        ENV_PULL_REQUEST_BASE_BRANCH_SHA = "DD_GIT_PULL_REQUEST_BASE_BRANCH_SHA"
+        ENV_COMMIT_HEAD_SHA = "DD_GIT_COMMIT_HEAD_SHA"
       end
     end
   end

data/lib/datadog/ci/ext/settings.rb

@@ -25,6 +25,7 @@ module Datadog
         ENV_TEST_VISIBILITY_DRB_SERVER_URI = "DD_TEST_VISIBILITY_DRB_SERVER_URI"
         ENV_AGENTLESS_LOGS_SUBMISSION_ENABLED = "DD_AGENTLESS_LOG_SUBMISSION_ENABLED"
         ENV_AGENTLESS_LOGS_SUBMISSION_URL = "DD_AGENTLESS_LOG_SUBMISSION_URL"
+        ENV_IMPACTED_TESTS_DETECTION_ENABLED = "DD_CIVISIBILITY_IMPACTED_TESTS_DETECTION_ENABLED"
 
         # Source: https://docs.datadoghq.com/getting_started/site/
         DD_SITE_ALLOWLIST = %w[

data/lib/datadog/ci/ext/telemetry.rb

@@ -90,6 +90,7 @@ module Datadog
         TAG_IS_QUARANTINED = "is_quarantined"
         TAG_IS_TEST_DISABLED = "is_disabled"
         TAG_HAS_FAILED_ALL_RETRIES = "has_failed_all_retries"
+        TAG_IS_MODIFIED = "is_modified"
         # tags for git_requests.settings_response metric
         TAG_COVERAGE_ENABLED = "coverage_enabled"
         TAG_ITR_ENABLED = "itr_enabled"
@@ -138,6 +139,8 @@ module Datadog
           GET_LOCAL_COMMITS = "get_local_commits"
           GET_OBJECTS = "get_objects"
           PACK_OBJECTS = "pack_objects"
+          DIFF = "diff"
+          BASE_COMMIT_SHA = "base_commit_sha"
         end
 
         module Provider

data/lib/datadog/ci/ext/test.rb

@@ -20,6 +20,7 @@ module Datadog
         TAG_COMMAND = "test.command"
         TAG_SOURCE_FILE = "test.source.file"
         TAG_SOURCE_START = "test.source.start"
+        TAG_SOURCE_END = "test.source.end"
         TAG_CODEOWNERS = "test.codeowners"
         TAG_PARAMETERS = "test.parameters"
 
@@ -94,7 +95,7 @@ module Datadog
             AUTO_TEST_RETRIES_VERSION = "1"
             TEST_MANAGEMENT_QUARANTINE_VERSION = "1"
             TEST_MANAGEMENT_DISABLE_VERSION = "1"
-            TEST_MANAGEMENT_ATTEMPT_TO_FIX_VERSION = "2"
+            TEST_MANAGEMENT_ATTEMPT_TO_FIX_VERSION = "4"
           end
 
           # Map of capabilities to their versions
@@ -108,6 +109,9 @@ module Datadog
           }.freeze
         end
 
+        # impacted tests detection
+        TAG_TEST_IS_MODIFIED = "test.is_modified"
+
         # internal APM tag to mark a span as a test span
         TAG_SPAN_KIND = "span.kind"
         SPAN_KIND_TEST = "test"

data/lib/datadog/ci/ext/transport.rb

@@ -48,6 +48,7 @@ module Datadog
         DD_API_SETTINGS_RESPONSE_TEST_MANAGEMENT_KEY = "test_management"
         DD_API_SETTINGS_RESPONSE_ATTEMPT_TO_FIX_RETRIES_KEY = "attempt_to_fix_retries"
         DD_API_SETTINGS_RESPONSE_DEFAULT = {DD_API_SETTINGS_RESPONSE_ITR_ENABLED_KEY => false}.freeze
+        DD_API_SETTINGS_RESPONSE_IMPACTED_TESTS_ENABLED_KEY = "impacted_tests_enabled"
 
         DD_API_GIT_SEARCH_COMMITS_PATH = "/api/v2/git/repository/search_commits"
 

data/lib/datadog/ci/git/base_branch_sha_detection/base.rb

@@ -0,0 +1,66 @@
+# frozen_string_literal: true
+
+require_relative "../cli"
+
+module Datadog
+  module CI
+    module Git
+      module BaseBranchShaDetection
+        class Base
+          attr_reader :remote_name
+          attr_reader :source_branch
+
+          def initialize(remote_name, source_branch)
+            @remote_name = remote_name
+            @source_branch = source_branch
+          end
+
+          def call
+            raise NotImplementedError, "Subclasses must implement #call"
+          end
+
+          protected
+
+          def merge_base_sha(branch, source_branch)
+            CLI.exec_git_command(["merge-base", branch, source_branch], timeout: CLI::LONG_TIMEOUT)&.strip
+          rescue CLI::GitCommandExecutionError => e
+            Datadog.logger.debug { "Merge base calculation failed for branches '#{branch}' and '#{source_branch}': #{e}" }
+            nil
+          end
+
+          def check_and_fetch_branch(branch, remote_name)
+            # @type var short_branch_name: String
+            short_branch_name = remove_remote_prefix(branch, remote_name)
+
+            # Check if branch already fetched from remote
+            CLI.exec_git_command(["show-ref", "--verify", "--quiet", "refs/remotes/#{remote_name}/#{short_branch_name}"])
+            Datadog.logger.debug { "Branch '#{remote_name}/#{short_branch_name}' already fetched from remote, skipping" }
+          rescue CLI::GitCommandExecutionError => e
+            Datadog.logger.debug { "Branch '#{remote_name}/#{short_branch_name}' doesn't exist locally, checking remote..." }
+
+            begin
+              remote_heads = CLI.exec_git_command(["ls-remote", "--heads", remote_name, short_branch_name])
+              if remote_heads.nil? || remote_heads.empty?
+                Datadog.logger.debug { "Branch '#{remote_name}/#{short_branch_name}' doesn't exist in remote" }
+                return
+              end
+
+              Datadog.logger.debug { "Branch '#{remote_name}/#{short_branch_name}' exists in remote, fetching" }
+              CLI.exec_git_command(["fetch", "--depth", "1", remote_name, short_branch_name], timeout: CLI::LONG_TIMEOUT)
+            rescue CLI::GitCommandExecutionError => e
+              Datadog.logger.debug { "Branch '#{remote_name}/#{short_branch_name}' couldn't be fetched from remote: #{e}" }
+            end
+          end
+
+          def remove_remote_prefix(branch_name, remote_name)
+            branch_name&.sub(/^#{Regexp.escape(remote_name)}\//, "")
+          end
+
+          def branches_equal?(branch_name, default_branch, remote_name)
+            remove_remote_prefix(branch_name, remote_name) == remove_remote_prefix(default_branch, remote_name)
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/ci/git/base_branch_sha_detection/branch_metric.rb

@@ -0,0 +1,34 @@
+# frozen_string_literal: true
+
+module Datadog
+  module CI
+    module Git
+      module BaseBranchShaDetection
+        # Represents metrics for a git branch comparison
+        # including how far behind/ahead it is from a source branch
+        # and the common base commit SHA
+        class BranchMetric
+          attr_reader :branch_name, :behind, :ahead, :base_sha
+
+          def initialize(branch_name:, behind:, ahead:, base_sha:)
+            @branch_name = branch_name
+            @behind = behind
+            @ahead = ahead
+            @base_sha = base_sha
+          end
+
+          # Checks if the branch is up to date with the source branch
+          def up_to_date?
+            @behind == 0 && @ahead == 0
+          end
+
+          # Used for comparison when finding the best branch
+          # Lower divergence score is better
+          def divergence_score
+            @ahead
+          end
+        end
+      end
+    end
+  end
+end