datadog 2.0.0.beta1

data/lib/datadog/tracing/contrib/racecar/patcher.rb

@@ -0,0 +1,29 @@
+# frozen_string_literal: true
+
+require_relative '../patcher'
+require_relative 'ext'
+require_relative 'events'
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Racecar
+        # Patcher enables patching of 'racecar' module.
+        module Patcher
+          include Contrib::Patcher
+
+          module_function
+
+          def target_version
+            Integration.version
+          end
+
+          def patch
+            # Subscribe to Racecar events
+            Events.subscribe!
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/configuration/settings.rb

@@ -0,0 +1,59 @@
+# frozen_string_literal: true
+
+require_relative '../../configuration/settings'
+require_relative '../ext'
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Rack
+        module Configuration
+          # Custom settings for the Rack integration
+          # @public_api
+          class Settings < Contrib::Configuration::Settings
+            DEFAULT_HEADERS = {
+              response: %w[
+                Content-Type
+                X-Request-ID
+              ]
+            }.freeze
+
+            option :enabled do |o|
+              o.type :bool
+              o.env Ext::ENV_ENABLED
+              o.default true
+            end
+
+            # @!visibility private
+            option :analytics_enabled do |o|
+              o.type :bool, nilable: true
+              o.env Ext::ENV_ANALYTICS_ENABLED
+            end
+
+            option :analytics_sample_rate do |o|
+              o.type :float
+              o.env Ext::ENV_ANALYTICS_SAMPLE_RATE
+              o.default 1.0
+            end
+
+            option :application
+            option :distributed_tracing, default: true, type: :bool
+            option :headers, default: DEFAULT_HEADERS, type: :hash
+            option :middleware_names, default: false, type: :bool
+            option :quantize, default: {}, type: :hash
+            option :request_queuing do |o|
+              o.type :bool
+              o.default false
+            end
+
+            option :service_name do |o|
+              o.type :string, nilable: true
+            end
+
+            option :web_service_name, default: Ext::DEFAULT_PEER_WEBSERVER_SERVICE_NAME, type: :string
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/ext.rb

@@ -0,0 +1,30 @@
+# frozen_string_literal: true
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Rack
+        # Rack integration constants
+        # @public_api Changing resource names, tag names, or environment variables creates breaking changes.
+        module Ext
+          ENV_ENABLED = 'DD_TRACE_RACK_ENABLED'
+          # @!visibility private
+          ENV_ANALYTICS_ENABLED = 'DD_TRACE_RACK_ANALYTICS_ENABLED'
+          ENV_ANALYTICS_SAMPLE_RATE = 'DD_TRACE_RACK_ANALYTICS_SAMPLE_RATE'
+          RACK_ENV_REQUEST_SPAN = 'datadog.rack_request_span'
+          SPAN_HTTP_PROXY_REQUEST = 'http.proxy.request'
+          SPAN_HTTP_PROXY_QUEUE = 'http.proxy.queue'
+          SPAN_REQUEST = 'rack.request'
+          TAG_COMPONENT = 'rack'
+          TAG_COMPONENT_HTTP_PROXY = 'http_proxy'
+          TAG_OPERATION_REQUEST = 'request'
+          TAG_OPERATION_HTTP_PROXY_REQUEST = 'request'
+          TAG_OPERATION_HTTP_PROXY_QUEUE = 'queue'
+          TAG_OPERATION_HTTP_SERVER_QUEUE = 'queue'
+          WEBSERVER_APP = 'webserver'
+          DEFAULT_PEER_WEBSERVER_SERVICE_NAME = 'web-server'
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/header_collection.rb

@@ -0,0 +1,40 @@
+# frozen_string_literal: true
+
+require_relative '../../../core/header_collection'
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Rack
+        # Classes and utilities for handling headers in Rack.
+        module Header
+          # An implementation of a header collection that looks up headers from a Rack environment.
+          class RequestHeaderCollection < Datadog::Core::HeaderCollection
+            # Creates a header collection from a rack environment.
+            def initialize(env)
+              super()
+              @env = env
+            end
+
+            # Gets the value of the header with the given name.
+            def get(header_name)
+              @env[Header.to_rack_header(header_name)]
+            end
+
+            # Allows this class to have a similar API to a {Hash}.
+            alias [] get
+
+            # Tests whether a header with the given name exists in the environment.
+            def key?(header_name)
+              @env.key?(Header.to_rack_header(header_name))
+            end
+          end
+
+          def self.to_rack_header(name)
+            "HTTP_#{name.to_s.upcase.gsub(/[-\s]/, '_')}"
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/header_tagging.rb

@@ -0,0 +1,63 @@
+# frozen_string_literal: true
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Rack
+        # Matches Rack-style headers with a matcher and sets matching headers into a span.
+        module HeaderTagging
+          def self.tag_request_headers(span, env, configuration)
+            # Wrap env in a case-insensitive Rack-style accessor.
+            headers = env.is_a?(Header::RequestHeaderCollection) ? env : Header::RequestHeaderCollection.new(env)
+
+            # Use global DD_TRACE_HEADER_TAGS if integration-level configuration is not provided
+            tags = if configuration.using_default?(:headers) && !Datadog.configuration.tracing.using_default?(:header_tags)
+                     Datadog.configuration.tracing.header_tags.request_tags(headers)
+                   else
+                     whitelist = configuration[:headers][:request] || []
+                     whitelist.each_with_object({}) do |header, result|
+                       header_value = headers.get(header)
+                       unless header_value.nil?
+                         header_tag = Tracing::Metadata::Ext::HTTP::RequestHeaders.to_tag(header)
+                         result[header_tag] = header_value
+                       end
+                     end
+                   end
+
+            span.set_tags(tags)
+          end
+
+          def self.tag_response_headers(span, headers, configuration)
+            headers = Core::Utils::Hash::CaseInsensitiveWrapper.new(headers) # Make header access case-insensitive
+
+            # Use global DD_TRACE_HEADER_TAGS if integration-level configuration is not provided
+            tags = if configuration.using_default?(:headers) && !Datadog.configuration.tracing.using_default?(:header_tags)
+                     Datadog.configuration.tracing.header_tags.response_tags(headers)
+                   else
+                     whitelist = configuration[:headers][:response] || []
+                     whitelist.each_with_object({}) do |header, result|
+                       header_value = headers[header]
+
+                       next if header_value.nil?
+
+                       header_tag = Tracing::Metadata::Ext::HTTP::ResponseHeaders.to_tag(header)
+
+                       # Maintain the value format between Rack 2 and 3
+                       #
+                       # Rack 2.x => { 'foo' => 'bar,baz' }
+                       # Rack 3.x => { 'foo' => ['bar', 'baz'] }
+                       result[header_tag] = if header_value.is_a? Array
+                                              header_value.join(',')
+                                            else
+                                              header_value
+                                            end
+                     end
+                   end
+
+            span.set_tags(tags)
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/integration.rb

@@ -0,0 +1,50 @@
+# frozen_string_literal: true
+
+require_relative '../integration'
+require_relative 'configuration/settings'
+require_relative 'patcher'
+require_relative '../rails/utils'
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Rack
+        # Description of Rack integration
+        class Integration
+          include Contrib::Integration
+
+          MINIMUM_VERSION = Gem::Version.new('1.1.0')
+
+          # @public_api Changing the integration name or integration options can cause breaking changes
+          register_as :rack, auto_patch: false
+
+          def self.version
+            Gem.loaded_specs['rack'] && Gem.loaded_specs['rack'].version
+          end
+
+          def self.loaded?
+            !defined?(::Rack).nil?
+          end
+
+          def self.compatible?
+            super && version >= MINIMUM_VERSION
+          end
+
+          # enabled by rails integration so should only auto instrument
+          # if detected that it is being used without rails
+          def auto_instrument?
+            !Contrib::Rails::Utils.railtie_supported?
+          end
+
+          def new_configuration
+            Configuration::Settings.new
+          end
+
+          def patcher
+            Patcher
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/middlewares.rb

@@ -0,0 +1,265 @@
+# frozen_string_literal: true
+
+require 'date'
+
+require_relative '../../../core/environment/variable_helpers'
+require_relative '../../../core/remote/tie/tracing'
+require_relative '../../client_ip'
+require_relative '../../metadata/ext'
+require_relative '../http'
+require_relative '../analytics'
+require_relative '../utils/quantization/http'
+require_relative 'ext'
+require_relative 'header_collection'
+require_relative 'header_tagging'
+require_relative 'request_queue'
+require_relative 'trace_proxy_middleware'
+
+module Datadog
+  module Tracing
+    module Contrib
+      # Rack module includes middlewares that are required to trace any framework
+      # and application built on top of Rack.
+      module Rack
+        # TraceMiddleware ensures that the Rack Request is properly traced
+        # from the beginning to the end. The middleware adds the request span
+        # in the Rack environment so that it can be retrieved by the underlying
+        # application. If request tags are not set by the app, they will be set using
+        # information available at the Rack level.
+        class TraceMiddleware
+          def initialize(app)
+            @app = app
+          end
+
+          def call(env)
+            # Find out if this is rack within rack
+            previous_request_span = env[Ext::RACK_ENV_REQUEST_SPAN]
+
+            return @app.call(env) if previous_request_span
+
+            boot = Datadog::Core::Remote::Tie.boot
+
+            # Extract distributed tracing context before creating any spans,
+            # so that all spans will be added to the distributed trace.
+            if configuration[:distributed_tracing]
+              trace_digest = Contrib::HTTP.extract(env)
+              Tracing.continue_trace!(trace_digest)
+            end
+
+            TraceProxyMiddleware.call(env, configuration) do
+              trace_options = { type: Tracing::Metadata::Ext::HTTP::TYPE_INBOUND }
+              trace_options[:service] = configuration[:service_name] if configuration[:service_name]
+
+              # start a new request span and attach it to the current Rack environment;
+              # we must ensure that the span `resource` is set later
+              request_span = Tracing.trace(Ext::SPAN_REQUEST, **trace_options)
+              request_span.resource = nil
+
+              # When tracing and distributed tracing are both disabled, `.active_trace` will be `nil`,
+              # Return a null object to continue operation
+              request_trace = Tracing.active_trace || TraceOperation.new
+              env[Ext::RACK_ENV_REQUEST_SPAN] = request_span
+
+              Datadog::Core::Remote::Tie::Tracing.tag(boot, request_span)
+
+              # Copy the original env, before the rest of the stack executes.
+              # Values may change; we want values before that happens.
+              original_env = env.dup
+
+              # call the rest of the stack
+              status, headers, response = @app.call(env)
+
+              [status, headers, response]
+
+            # Here we really want to catch *any* exception, not only StandardError,
+            # as we really have no clue of what is in the block,
+            # and it is user code which should be executed no matter what.
+            # It's not a problem since we re-raise it afterwards so for example a
+            # SignalException::Interrupt would still bubble up.
+            rescue Exception => e # rubocop:disable Lint/RescueException
+              # catch exceptions that may be raised in the middleware chain
+              # Note: if a middleware catches an Exception without re raising,
+              # the Exception cannot be recorded here.
+              request_span.set_error(e) unless request_span.nil?
+              raise e
+            ensure
+              env[Ext::RACK_ENV_REQUEST_SPAN] = previous_request_span if previous_request_span
+
+              if request_span
+                # Rack is a really low level interface and it doesn't provide any
+                # advanced functionality like routers. Because of that, we assume that
+                # the underlying framework or application has more knowledge about
+                # the result for this request; `resource` and `tags` are expected to
+                # be set in another level but if they're missing, reasonable defaults
+                # are used.
+                set_request_tags!(request_trace, request_span, env, status, headers, response, original_env || env)
+
+                # ensure the request_span is finished and the context reset;
+                # this assumes that the Rack middleware creates a root span
+                request_span.finish
+              end
+            end
+          end
+
+          # rubocop:disable Metrics/AbcSize
+          # rubocop:disable Metrics/CyclomaticComplexity
+          # rubocop:disable Metrics/PerceivedComplexity
+          # rubocop:disable Metrics/MethodLength
+          def set_request_tags!(trace, request_span, env, status, headers, response, original_env)
+            request_header_collection = Header::RequestHeaderCollection.new(env)
+
+            # Since it could be mutated, it would be more accurate to fetch from the original env,
+            # e.g. ActionDispatch::ShowExceptions middleware with Rails exceptions_app configuration
+            original_request_method = original_env['REQUEST_METHOD']
+
+            # request_headers is subject to filtering and configuration so we
+            # get the user agent separately
+            user_agent = parse_user_agent_header(request_header_collection)
+
+            # The priority
+            # 1. User overrides span.resource
+            # 2. Configuration
+            # 3. Nested App override trace.resource
+            # 4. Fallback with verb + status, eq `GET 200`
+            request_span.resource ||=
+              if configuration[:middleware_names] && env['RESPONSE_MIDDLEWARE']
+                "#{env['RESPONSE_MIDDLEWARE']}##{original_request_method}"
+              elsif trace.resource_override?
+                trace.resource
+              else
+                "#{original_request_method} #{status}".strip
+              end
+
+            # Overrides the trace resource if it never been set
+            # Otherwise, the getter method would delegate to its root span
+            trace.resource = request_span.resource unless trace.resource_override?
+
+            request_span.set_tag(Tracing::Metadata::Ext::TAG_COMPONENT, Ext::TAG_COMPONENT)
+            request_span.set_tag(Tracing::Metadata::Ext::TAG_OPERATION, Ext::TAG_OPERATION_REQUEST)
+            request_span.set_tag(Tracing::Metadata::Ext::TAG_KIND, Tracing::Metadata::Ext::SpanKind::TAG_SERVER)
+
+            # Set analytics sample rate
+            if Contrib::Analytics.enabled?(configuration[:analytics_enabled])
+              Contrib::Analytics.set_sample_rate(request_span, configuration[:analytics_sample_rate])
+            end
+
+            # Measure service stats
+            Contrib::Analytics.set_measured(request_span)
+
+            if request_span.get_tag(Tracing::Metadata::Ext::HTTP::TAG_METHOD).nil?
+              request_span.set_tag(Tracing::Metadata::Ext::HTTP::TAG_METHOD, original_request_method)
+            end
+
+            url = parse_url(env, original_env)
+
+            if request_span.get_tag(Tracing::Metadata::Ext::HTTP::TAG_URL).nil?
+              options = configuration[:quantize] || {}
+
+              # Quantization::HTTP.url base defaults to :show, but we are transitioning
+              options[:base] ||= :exclude
+
+              request_span.set_tag(
+                Tracing::Metadata::Ext::HTTP::TAG_URL,
+                Contrib::Utils::Quantization::HTTP.url(url, options)
+              )
+            end
+
+            if request_span.get_tag(Tracing::Metadata::Ext::HTTP::TAG_BASE_URL).nil?
+              options = configuration[:quantize]
+
+              unless options[:base] == :show
+                base_url = Contrib::Utils::Quantization::HTTP.base_url(url)
+
+                unless base_url.empty?
+                  request_span.set_tag(
+                    Tracing::Metadata::Ext::HTTP::TAG_BASE_URL,
+                    base_url
+                  )
+                end
+              end
+            end
+
+            if request_span.get_tag(Tracing::Metadata::Ext::HTTP::TAG_CLIENT_IP).nil?
+              Tracing::ClientIp.set_client_ip_tag(
+                request_span,
+                headers: request_header_collection,
+                remote_ip: env['REMOTE_ADDR']
+              )
+            end
+
+            if request_span.get_tag(Tracing::Metadata::Ext::HTTP::TAG_STATUS_CODE).nil? && status
+              request_span.set_tag(Tracing::Metadata::Ext::HTTP::TAG_STATUS_CODE, status)
+            end
+
+            if request_span.get_tag(Tracing::Metadata::Ext::HTTP::TAG_USER_AGENT).nil? && user_agent
+              request_span.set_tag(Tracing::Metadata::Ext::HTTP::TAG_USER_AGENT, user_agent)
+            end
+
+            HeaderTagging.tag_request_headers(request_span, request_header_collection, configuration)
+            HeaderTagging.tag_response_headers(request_span, headers, configuration) if headers
+
+            # detect if the status code is a 5xx and flag the request span as an error
+            # unless it has been already set by the underlying framework
+            request_span.status = 1 if status.to_s.start_with?('5') && request_span.status.zero?
+          end
+          # rubocop:enable Metrics/AbcSize
+          # rubocop:enable Metrics/CyclomaticComplexity
+          # rubocop:enable Metrics/PerceivedComplexity
+          # rubocop:enable Metrics/MethodLength
+
+          private
+
+          def configuration
+            Datadog.configuration.tracing[:rack]
+          end
+
+          def parse_url(env, original_env)
+            request_obj = ::Rack::Request.new(env)
+
+            # scheme, host, and port
+            base_url = if request_obj.respond_to?(:base_url)
+                         request_obj.base_url
+                       else
+                         # Compatibility for older Rack versions
+                         request_obj.url.chomp(request_obj.fullpath)
+                       end
+
+            # https://github.com/rack/rack/blob/main/SPEC.rdoc
+            #
+            # The source of truth in Rack is the PATH_INFO key that holds the
+            # URL for the current request; but some frameworks may override that
+            # value, especially during exception handling.
+            #
+            # Because of this, we prefer to use REQUEST_URI, if available, which is the
+            # relative path + query string, and doesn't mutate.
+            #
+            # REQUEST_URI is only available depending on what web server is running though.
+            # So when its not available, we want the original, unmutated PATH_INFO, which
+            # is just the relative path without query strings.
+            #
+            # SCRIPT_NAME is the first part of the request URL path, so that
+            # the application can know its virtual location. It should be
+            # prepended to PATH_INFO to reflect the correct user visible path.
+            request_uri = env['REQUEST_URI'].to_s
+            fullpath = if request_uri.empty?
+                         query_string = original_env['QUERY_STRING'].to_s
+                         path = original_env['SCRIPT_NAME'].to_s + original_env['PATH_INFO'].to_s
+
+                         query_string.empty? ? path : "#{path}?#{query_string}"
+                       else
+                         # normally REQUEST_URI starts at the path, but it
+                         # might contain the full URL in some cases (e.g WEBrick)
+                         request_uri.delete_prefix(base_url)
+                       end
+
+            base_url + fullpath
+          end
+
+          def parse_user_agent_header(headers)
+            headers.get(Tracing::Metadata::Ext::HTTP::HEADER_USER_AGENT)
+          end
+        end
+      end
+    end
+  end
+end

data/lib/datadog/tracing/contrib/rack/patcher.rb

@@ -0,0 +1,119 @@
+# frozen_string_literal: true
+
+module Datadog
+  module Tracing
+    module Contrib
+      module Rack
+        # Provides instrumentation for `rack`
+        module MiddlewarePatcher
+          include Contrib::Patcher
+
+          module_function
+
+          def target_version
+            Integration.version
+          end
+
+          def patch
+            # Patch middleware
+            require_relative 'middlewares'
+          end
+        end
+
+        # Provides instrumentation for Rack middleware names
+        module MiddlewareNamePatcher
+          include Contrib::Patcher
+
+          module_function
+
+          def target_version
+            Integration.version
+          end
+
+          def patch
+            patch_middleware_names
+          end
+
+          def patch_middleware_names
+            retain_middleware_name(get_option(:application))
+          rescue => e
+            # We can safely ignore these exceptions since they happen only in the
+            # context of middleware patching outside a Rails server process (eg. a
+            # process that doesn't serve HTTP requests but has Rails environment
+            # loaded such as a Resque master process)
+            Datadog.logger.debug("Error patching middleware stack: #{e}")
+          end
+
+          def retain_middleware_name(middleware)
+            return unless middleware && middleware.respond_to?(:call)
+
+            middleware.singleton_class.class_eval do
+              alias_method :__call, :call
+
+              def call(env)
+                env['RESPONSE_MIDDLEWARE'] = self.class.to_s
+                __call(env)
+              end
+            end
+
+            following = (middleware.instance_variable_get('@app') if middleware.instance_variable_defined?('@app'))
+
+            retain_middleware_name(following)
+          end
+
+          def get_option(option)
+            Datadog.configuration.tracing[:rack].get_option(option)
+          end
+        end
+
+        # Applies multiple patches
+        module Patcher
+          PATCHERS = [
+            MiddlewarePatcher,
+            MiddlewareNamePatcher
+          ].freeze
+
+          module_function
+
+          def patched?
+            PATCHERS.all?(&:patched?)
+          end
+
+          def target_version
+            Integration.version
+          end
+
+          def patch
+            MiddlewarePatcher.patch unless MiddlewarePatcher.patched?
+
+            # Patch middleware names
+            if !MiddlewareNamePatcher.patched? && get_option(:middleware_names)
+              if get_option(:application)
+                MiddlewareNamePatcher.patch
+              else
+                Datadog.logger.warn(
+                  %(
+                Rack :middleware_names requires you to also pass :application.
+                Middleware names have NOT been patched; please provide :application.
+                e.g. use: :rack, middleware_names: true, application: my_rack_app)
+                )
+              end
+            end
+          end
+
+          def get_option(option)
+            Datadog.configuration.tracing[:rack].get_option(option)
+          end
+
+          def patch_successful
+            MiddlewarePatcher.patch_successful || MiddlewareNamePatcher.patch_successful
+          end
+
+          def patch_error_result
+            MiddlewarePatcher.patch_error_result || MiddlewareNamePatcher.patch_error_result
+          end
+        end
+      end
+    end
+  end
+end