dionysus-rb 1.0.0 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0c0362ffd0074956ac7ace1f3c4a392eaa7b5b09a79ecf113a14f9da8f9f5b22
-  data.tar.gz: 44cdab85d128017c09dc0535c1225d9d68515addc5c2d1d4b28b50dbb2fc0150
+  metadata.gz: d190bbc3fd4b2ca0aa7c7d1be7c0ddaa6a3b6bdcbd6556b90e4752c2e875bfc6
+  data.tar.gz: 33379f5ccc97bceee7caad1d9b3383e84c3073e48b15bda1d0f592e765c39352
 SHA512:
-  metadata.gz: 63905f2eb238cdca13e750e728498449d6238be26f30bbe68c4709c61218c989d33dd8c7da553c4c4ec6e36dc6b05cc84a8cc44067048752b88018338d1809ee
-  data.tar.gz: cabf8edf52bd2ba5ffaf26608110a7f110ba4a7c15ff2ead082c74cd780c5018bd92b0b44a15e847edcf65c39a898bdd537f2c9c85c3c191c1a1803c1a657e96
+  metadata.gz: cafb4cb76ad00f5dbbc45f9c3a578284ef554aaf680310286961d1f8251620bc67c10e190f301ac06fca1c8e55be9fab2e75977cc85ecf80f21503504558da20
+  data.tar.gz: 78e8c07d652f3dcc46cb3a527834cb72b2290b119266e4362cde3ddbf1d5679c147cb03cce34acb6682451c6453fb58532faab8a7bf5be43a6db6be791e7e1e2

data/CHANGELOG.md

@@ -1,5 +1,11 @@
 ## [Unreleased]
 
+## [1.1.0]
+
+### Fixed
+
+- Keep the last published message when removing duplicates from a consumer batch. `RemoveDuplicatesStrategy` kept the message with the highest payload `updated_at` per event and id, but payloads are serialized at publish time, so freshness follows publish order rather than `updated_at`: a payload carrying values computed on read from associated records changes without the record's own `updated_at` changing, and can then arrive with an older `updated_at` than a staler message. Those messages were silently discarded, losing the change in every consumer with no error and no trace. The message with the highest Kafka offset is kept instead - a Karafka batch comes from a single topic-partition, so offsets strictly increase in publish order.
+
 ## [1.0.0]
 
 Brings `dionysus-rb` back in line with the private gem it was extracted from, which had moved on

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    dionysus-rb (1.0.0)
+    dionysus-rb (1.1.0)
       activerecord (>= 5)
       activesupport (>= 3.2)
       concurrent-ruby

data/lib/dionysus/consumer/params_batch_transformations/remove_duplicates_strategy.rb

@@ -10,18 +10,16 @@ class Dionysus::Consumer::ParamsBatchTransformations::RemoveDuplicatesStrategy
   private
 
   # the idea is following:
-  # 1. group messages by event and id - for a given model we can expect unique messages for _created and _deleted ecents
+  # 1. group messages by event and id - for a given model we can expect unique messages for _created and _deleted events
   #   but we can have multiple _updated events, so this is where we are interested in removing duplicates
-  # 2. sort each group by updated_at, reverse (because the sort is ascending) and take the first one -
-  #   this way we will have the most recent update
+  # 2. from each group keep the message published last (highest offset) - payloads are serialized at
+  #   publish time, so that is the freshest state; "updated_at" does not track it
   # 3. flatten the array of arrays as all groups will have a single item
-  # It is safer to apply sorting just for the _updated events to a given model as otherwise we could change the order of
-  #   the messages for a different type which might be not desirable
   def transform_messages_array(params_batch)
     params_batch
       .to_a
       .group_by { |batch| grouping_key_by_event_and_id(batch) }
-      .map { |_, group| group.max_by { |batch| updated_at_from_batch(batch) } }
+      .map { |_, group| group.max_by(&:offset) }
       .flatten
   end
 
@@ -32,14 +30,6 @@ class Dionysus::Consumer::ParamsBatchTransformations::RemoveDuplicatesStrategy
     ].join
   end
 
-  def updated_at_from_batch(batch)
-    timestamp = batch.payload.fetch("message", []).first.to_h.fetch("data", []).first.to_h.fetch("updated_at", nil)
-
-    return timestamp.to_datetime if timestamp.respond_to?(:to_datetime)
-
-    Time.current
-  end
-
   def duplicates_removal_not_applicable?(params_batch)
     any_message_containing_more_than_one_event?(params_batch) || any_event_containing_more_than_one_item(params_batch)
   end

data/lib/dionysus/version.rb

@@ -3,5 +3,5 @@
 module Dionysus
   module Version
   end
-  VERSION = "1.0.0"
+  VERSION = "1.1.0"
 end

metadata

@@ -1,13 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: dionysus-rb
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.1.0
 platform: ruby
 authors:
 - Karol Galanciak
+autorequire:
 bindir: bin
 cert_chain: []
-date: 1980-01-02 00:00:00.000000000 Z
+date: 2026-08-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activerecord
@@ -563,6 +564,7 @@ metadata:
   source_code_uri: https://github.com/BookingSync/dionysus-rb
   changelog_uri: https://github.com/BookingSync/dionysus-rb/blob/master/CHANGELOG.md
   rubygems_mfa_required: 'true'
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -577,7 +579,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 4.0.18
+rubygems_version: 3.5.22
+signing_key:
 specification_version: 4
 summary: A framework on top of Karafka for Change Data Capture on the model level.
 test_files: []