karafka 1.4.4 → 2.1.10

data/lib/karafka/pro/processing/strategies/mom/ftr.rb

@@ -0,0 +1,53 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      module Strategies
+        module Mom
+          # Filtering support for MoM
+          module Ftr
+            include Strategies::Ftr::Default
+            include Strategies::Mom::Default
+
+            # MoM + Ftr
+            FEATURES = %i[
+              filtering
+              manual_offset_management
+            ].freeze
+
+            # When mom is enabled, we do not mark messages as consumed after processing
+            # but we also need to keep in mind throttling here
+            def handle_after_consume
+              coordinator.on_finished do
+                return if revoked?
+
+                if coordinator.success?
+                  coordinator.pause_tracker.reset
+
+                  # Do not throttle if paused
+                  return if coordinator.manual_pause?
+
+                  handle_post_filtering
+                else
+                  retry_after_pause
+                end
+              end
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/processing/strategies/mom/ftr_vp.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      module Strategies
+        module Mom
+          # - Mom enabled
+          # - Ftr enabled
+          # - Vp enabled
+          module FtrVp
+            include Strategies::Mom::Vp
+            include Strategies::Mom::Ftr
+
+            # Features of this strategy
+            FEATURES = %i[
+              filtering
+              manual_offset_management
+              virtual_partitions
+            ].freeze
+          end
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/processing/strategies/mom/vp.rb

@@ -0,0 +1,35 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      module Strategies
+        module Mom
+          # - Mom enabled
+          # - Vp enabled
+          module Vp
+            include Strategies::Vp::Default
+            include Strategies::Mom::Default
+
+            # Features of this strategy
+            FEATURES = %i[
+              manual_offset_management
+              virtual_partitions
+            ].freeze
+          end
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/processing/strategies/vp/default.rb

@@ -0,0 +1,104 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      module Strategies
+        # VP starting strategies
+        module Vp
+          # Just Virtual Partitions enabled
+          module Default
+            # This flow is exactly the same as the default one because the default one is wrapper
+            # with `coordinator#on_finished`
+            include Strategies::Default
+
+            # Features for this strategy
+            FEATURES = %i[
+              virtual_partitions
+            ].freeze
+
+            # @param message [Karafka::Messages::Message] marks message as consumed
+            # @note This virtual offset management uses a regular default marking API underneath.
+            #   We do not alter the "real" marking API, as VPs are just one of many cases we want
+            #   to support and we do not want to impact them with collective offsets management
+            def mark_as_consumed(message)
+              return super if collapsed?
+
+              manager = coordinator.virtual_offset_manager
+
+              coordinator.synchronize do
+                manager.mark(message)
+                # If this is last marking on a finished flow, we can use the original
+                # last message and in order to do so, we need to mark all previous messages as
+                # consumed as otherwise the computed offset could be different
+                # We mark until our offset just in case of a DLQ flow or similar, where we do not
+                # want to mark all but until the expected location
+                manager.mark_until(message) if coordinator.finished?
+
+                return revoked? unless manager.markable?
+              end
+
+              manager.markable? ? super(manager.markable) : revoked?
+            end
+
+            # @param message [Karafka::Messages::Message] blocking marks message as consumed
+            def mark_as_consumed!(message)
+              return super if collapsed?
+
+              manager = coordinator.virtual_offset_manager
+
+              coordinator.synchronize do
+                manager.mark(message)
+                manager.mark_until(message) if coordinator.finished?
+              end
+
+              manager.markable? ? super(manager.markable) : revoked?
+            end
+
+            # @return [Boolean] is the virtual processing collapsed in the context of given
+            #   consumer.
+            def collapsed?
+              coordinator.collapsed?
+            end
+
+            # @return [Boolean] true if any of virtual partition we're operating in the entangled
+            #   mode has already failed and we know we are failing collectively.
+            #   Useful for early stop to minimize number of things processed twice.
+            #
+            # @note We've named it `#failing?` instead of `#failure?`  because it aims to be used
+            #   from within virtual partitions where we want to have notion of collective failing
+            #   not just "local" to our processing. We "are" failing with other virtual partitions
+            #   raising an error, but locally we are still processing.
+            def failing?
+              coordinator.failure?
+            end
+
+            private
+
+            # Prior to adding work to the queue, registers all the messages offsets into the
+            # virtual offset group.
+            #
+            # @note This can be done without the mutex, because it happens from the same thread
+            #   for all the work (listener thread)
+            def handle_before_enqueue
+              coordinator.virtual_offset_manager.register(
+                messages.map(&:offset)
+              )
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/processing/strategies.rb

@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      # Pro processing strategies namespace
+      module Strategies
+      end
+    end
+  end
+end

data/lib/karafka/pro/processing/strategy_selector.rb

@@ -0,0 +1,84 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      # Selector of appropriate processing strategy matching topic combinations
+      # When using Karafka Pro, there is a different set of strategies than for regular, as there
+      # are different features.
+      class StrategySelector
+        attr_reader :strategies
+
+        # Strategies that we support in the Pro offering
+        # They can be combined
+        SUPPORTED_FEATURES = %i[
+          active_job
+          long_running_job
+          manual_offset_management
+          virtual_partitions
+          dead_letter_queue
+          filtering
+        ].freeze
+
+        def initialize
+          # Preload the strategies
+          # We load them once for performance reasons not to do too many lookups
+          @strategies = find_all
+        end
+
+        # @param topic [Karafka::Routing::Topic] topic with settings based on which we find
+        #   the strategy
+        # @return [Module] module with proper strategy
+        def find(topic)
+          feature_set = SUPPORTED_FEATURES.map do |feature_name|
+            topic.public_send("#{feature_name}?") ? feature_name : nil
+          end
+
+          feature_set.compact!
+          feature_set.sort!
+
+          @strategies.find do |strategy|
+            strategy::FEATURES.sort == feature_set
+          end || raise(Errors::StrategyNotFoundError, topic.name)
+        end
+
+        private
+
+        # @return [Array<Module>] all available strategies
+        def find_all
+          scopes = [Strategies]
+          modules = Strategies.constants
+
+          modules.each do |const|
+            scopes << Strategies.const_get(const)
+            modules += scopes.last.constants
+          end
+
+          scopes.flat_map do |scope|
+            modules.map do |const|
+              next if const == :FEATURES
+              next unless scope.const_defined?(const)
+
+              candidate = scope.const_get(const)
+
+              next unless candidate.const_defined?(:FEATURES)
+
+              candidate
+            end
+          end.uniq.compact
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/processing/virtual_offset_manager.rb

@@ -0,0 +1,147 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Processing
+      # Manager that keeps track of our offsets with the virtualization layer that are local
+      # to given partition assignment. It allows for easier offset management for virtual
+      # virtual partition cases as it provides us ability to mark as consumed and move the
+      # real offset behind as expected.
+      #
+      # @note We still use the regular coordinator "real" offset management as we want to have
+      #   them as separated as possible because the real seek offset management is also used for
+      #   pausing, filtering and others and should not be impacted by the virtual one
+      #
+      # @note This manager is **not** thread-safe by itself. It should operate from coordinator
+      #   locked locations.
+      class VirtualOffsetManager
+        attr_reader :groups
+
+        # @param topic [String]
+        # @param partition [Integer]
+        #
+        # @note We need topic and partition because we use a seek message (virtual) for real offset
+        #   management. We could keep real message reference but this can be memory consuming
+        #   and not worth it.
+        def initialize(topic, partition)
+          @topic = topic
+          @partition = partition
+          @groups = []
+          @marked = {}
+          @real_offset = -1
+        end
+
+        # Clears the manager for a next collective operation
+        def clear
+          @groups.clear
+          @marked = {}
+          @real_offset = -1
+        end
+
+        # Registers an offset group coming from one virtual consumer. In order to move the real
+        # underlying offset accordingly, we need to make sure to track the virtual consumers
+        # offsets groups independently and only materialize the end result.
+        #
+        # @param offsets_group [Array<Integer>] offsets from one virtual consumer
+        def register(offsets_group)
+          @groups << offsets_group
+
+          offsets_group.each { |offset| @marked[offset] = false }
+        end
+
+        # Marks given message as marked (virtually consumed).
+        # We mark given message offset and other earlier offsets from the same group as done
+        # and we can refresh our real offset representation based on that as it might have changed
+        # to a newer real offset.
+        # @param message [Karafka::Messages::Message] message coming from VP we want to mark
+        def mark(message)
+          offset = message.offset
+
+          group = @groups.find { |reg_group| reg_group.include?(offset) }
+
+          # This case can happen when someone uses MoM and wants to mark message from a previous
+          # batch as consumed. We can add it, since the real offset refresh will point to it
+          unless group
+            group = [offset]
+            @groups << group
+          end
+
+          position = group.index(offset)
+
+          # Mark all previous messages from the same group also as virtually consumed
+          group[0..position].each do |markable_offset|
+            @marked[markable_offset] = true
+          end
+
+          # Recompute the real offset representation
+          materialize_real_offset
+        end
+
+        # Mark all from all groups including the `message`.
+        # Useful when operating in a collapsed state for marking
+        # @param message [Karafka::Messages::Message]
+        def mark_until(message)
+          mark(message)
+
+          @groups.each do |group|
+            group.each do |offset|
+              next if offset > message.offset
+
+              @marked[offset] = true
+            end
+          end
+
+          materialize_real_offset
+        end
+
+        # @return [Array<Integer>] Offsets of messages already marked as consumed virtually
+        def marked
+          @marked.select { |_, status| status }.map(&:first).sort
+        end
+
+        # Is there a real offset we can mark as consumed
+        # @return [Boolean]
+        def markable?
+          !@real_offset.negative?
+        end
+
+        # @return [Messages::Seek] markable message for real offset marking
+        def markable
+          raise Errors::InvalidRealOffsetUsageError unless markable?
+
+          Messages::Seek.new(
+            @topic,
+            @partition,
+            @real_offset
+          )
+        end
+
+        private
+
+        # Recomputes the biggest possible real offset we can have.
+        # It picks the the biggest offset that has uninterrupted stream of virtually marked as
+        # consumed because this will be the collective offset.
+        def materialize_real_offset
+          @marked.to_a.sort_by(&:first).each do |offset, marked|
+            break unless marked
+
+            @real_offset = offset
+          end
+
+          @real_offset = (@marked.keys.min - 1) if @real_offset.negative?
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/routing/features/base.rb

@@ -0,0 +1,24 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Routing
+      module Features
+        # Base class for all the Pro features
+        class Base < Karafka::Routing::Features::Base
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/routing/features/dead_letter_queue/contract.rb

@@ -0,0 +1,50 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Routing
+      module Features
+        class DeadLetterQueue < Base
+          # Extended rules for dead letter queue settings
+          class Contract < Contracts::Base
+            configure do |config|
+              config.error_messages = YAML.safe_load(
+                File.read(
+                  File.join(Karafka.gem_root, 'config', 'locales', 'pro_errors.yml')
+                )
+              ).fetch('en').fetch('validations').fetch('topic')
+            end
+
+            # Make sure that when we use virtual partitions with DLQ, at least one retry is set
+            # We cannot use VP with DLQ without retries as we in order to provide ordering
+            # warranties on errors with VP, we need to collapse the VPs concurrency and retry
+            # without any indeterministic work
+            virtual do |data, errors|
+              next unless errors.empty?
+
+              dead_letter_queue = data[:dead_letter_queue]
+              virtual_partitions = data[:virtual_partitions]
+
+              next unless dead_letter_queue[:active]
+              next unless virtual_partitions[:active]
+              next if dead_letter_queue[:max_retries].positive?
+
+              [[%i[dead_letter_queue], :with_virtual_partitions]]
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/routing/features/dead_letter_queue.rb

@@ -0,0 +1,27 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Routing
+      module Features
+        # Enhanced Dead Letter Queue
+        #
+        # This version behaves similar to the regular version but provides ordering warranties and
+        # allows for batch skipping
+        class DeadLetterQueue < Base
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/routing/features/delaying/config.rb

@@ -0,0 +1,27 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Routing
+      module Features
+        class Delaying < Base
+          # Delaying feature configuration
+          Config = Struct.new(:active, :delay, keyword_init: true) do
+            alias_method :active?, :active
+          end
+        end
+      end
+    end
+  end
+end

data/lib/karafka/pro/routing/features/delaying/contract.rb

@@ -0,0 +1,38 @@
+# frozen_string_literal: true
+
+# This Karafka component is a Pro component under a commercial license.
+# This Karafka component is NOT licensed under LGPL.
+#
+# All of the commercial components are present in the lib/karafka/pro directory of this
+# repository and their usage requires commercial license agreement.
+#
+# Karafka has also commercial-friendly license, commercial support and commercial components.
+#
+# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
+# your code to Maciej Mensfeld.
+
+module Karafka
+  module Pro
+    module Routing
+      module Features
+        class Delaying < Base
+          # Contract to validate configuration of the expiring feature
+          class Contract < Contracts::Base
+            configure do |config|
+              config.error_messages = YAML.safe_load(
+                File.read(
+                  File.join(Karafka.gem_root, 'config', 'locales', 'pro_errors.yml')
+                )
+              ).fetch('en').fetch('validations').fetch('topic')
+            end
+
+            nested(:delaying) do
+              required(:active) { |val| [true, false].include?(val) }
+              required(:delay) { |val| val.nil? || (val.is_a?(Integer) && val.positive?) }
+            end
+          end
+        end
+      end
+    end
+  end
+end