- Support for Foreign Key joins #365 (Issue [#364])
- add manifold and keep aleph in dev dependencies #360. Users of test-machine will have to add aleph to the test deps in their app.
- v0.9.10 introduced some reflection warnings. Whilst harmless seeing
.deleteTopics
appear in your logs at app startup is a little unsettling. #358 - fix to re-introduce correct handling of timestamped consumer records. Broke test asserting windowed results. #356
- Fix JSON_SR serde to correctly "Clojurize" deeply nested maps
- Trivial release to fix release tagging issue
- Trivial release to fix API publishing in cljdocs
- Fix project.edn repo typo
- Add clj-kondo and fix all lint warnings and errors #323
- Allow Kafka Schema registry library to create a client rather than do it in Jackdaw (this allows security properties to be recognised) #330
- Move away from deprecated class ConsumerRecordFactory (to prepare migration to Kafka Streams 3.2.0)
- More verbose avro serialization exception errors
- Move libraries overrides back to dependencies.
- Fixed CVE-2021-37137, CVE-2021-37136 and CVE-2021-36090.
- Fixed Circle CI build process.
This version has been tagged in the repo but unreleased due to the broken build process.
- Drop clj-time dependency.
- Added confluent schema registry support for JSON Schema.
- Removed dependency on deprecated
org.apache.kafka.streams.kstream.Serialized
class. - Added simple helpers in lambdas for using transformers more easily in a stream. #305
- Added the
flatTransform
and flatTransfromValues
calls to the core streams interface. #305 - Aded a helper function to add KV state stores to a streams builder. #305
- Add KStream-KTable inner join.
- Update Kafka to 2.8.0 (confluent 6.1.1) #292
- Improve test-machine documentation #287
- Fix CI pipeline: add -repo to repo cache names to not match with deps cache #288
- Remove codecov #289
- Bump netty related packages to latest version (related to changes in
0.7.8
which fixes CVEs).
- Added support for kafka message headers to Test Machine
- Replaced deprecated methods in avro.clj with supported methods.
-
Fix ktable
constructor to use supplied store-name
-
Bump clj-uuid
version to 0.1.9
-
Minor update to fix harmless but distracting reflection warnings
- Moved dependency on kafka_2.11 into the dev profile
- Fixed bug in Avro deserialisation, when handling a union of enum types
- as-edn/as-json functions to convert between representations of avro
- Fixed bug in
map->ProducerRecord
- Allow nullable
partition
and timestamp
in ->ProducerRecord
(previously threw NPE) - Fixed union type serialisation when members have similar fields
- Fressian Serde via clojure.data.fressian #209
- Clearer error from the command runner #214
- Documentation about Jackdaw Admin API #211
- Upgraded in #217:
- Added functions to simplify querying the test machine journal (#215)
- Upgrade the test-runner (#184)
- Added support for user-provided parameters to reset-application-fixture (#177)
- Continuing refinement of examples (#191)
- Support for
.suppress
(#23) - Support for group-options when creating rest-proxy client for the test-machine (#206)
- PR Template (#187)
- A new edn serde without the un-necessary newline (#190)
- Fail fast throwing an exception as soon as a command fails (#186)
- Throw an exception on unknown commands, useful to detect typos early (#182)
- Fixed add-key (part of publishing pipeline) (#181)
- Skip deploy_snapshots job for external contributors (#194)
- Only numbers should be coercable (#203)
- Small refactor to implementation of rest-proxy transport (#205)
- topics-ready? does not dereference the returned deferred (#193)
- Fix test machine status middleware (#157)
- Fix application reset fixture to use bootstrap servers form app-config (#157)
- Fix type-hint call to
KafkaAvroDeserializer.deseialize
to remove warning (#157)
- Allow specification of :deserialization-properties (#157)
- Back-fill a few tests of jackdaw.client.partitioning (#165)
- Upgrade Clojure version to 1.10.1 (#159)
- Partitioner in test-machine write command updated to match streams (#139)
- Reformatted all the code using cljfmt (#173)
- Delete duplicated tests (#165)
- Documentation/Examples fixes (#166, #168)
- Do not assume result of executing command is a map (#164)
- Supply
key-serde
as well as value-serde
in aggregate
methods (#172)
- Auto-coercion of clojure numbers if possible (#135)
- Added more explicit information about commit signing Contributing guide (#150)
- Merger instance required for session window aggregation (#142)
- Fixed mis-leading parameter names relating to global ktables (#147)
- Select matching record from union during serialization (#149)
- Fixed typo in one of the code examples in the streams guide (#151)
- Add Add
:do!
and :inspect
test commands (#141)
- Fix regression in multi-topic stream constructor (#143)
- Stop swallowing errors in test-machine (#144)
- Include kafka-streams-test-utils so users don't have to (#138)
- Add changelog and contributing files (#122)
- Add test-machine example to the word-count example application (#120)
- Add new arities for aggregate and reduce (#132)
- Add sign-off to contributing file (#123)
- Ensure user-supplied partitions are cast to
int
before handing
them to the underlying kafka producer (#124) - Make the test for the rest-proxy transport use keywords to identify topics (#120)
- Resolve dependency conflict reported by lein deps :tree (#125)
- Fix a typo in the
service-ready?
test fixture (#121) - Make sure mock dirver is closed after use (#128)
- Fix Jackdaw version for Word Count example (#131)
- Update changelog for 0.6.4 release (#124)
- Delete a couple of overly verbose logging statements (#124)
None
- Upgrade Kafka dependency to 2.2.0 (#123)
None
- Improvement and clarification of documentation (on-going) (#116, #117, #118)
- Support for including a literal avro schema in the topic definitions resolved by the default resolver (#109)
- Support for including a custom :partition-fn in kafka streams operations that write records (#103)
- Implement kibit recommendations (#118)
- Word-count example (#108)
- Log to file instead of stdout
- A few simplifications in the implementation
- In the streams mock driver, get-records now returns a vector of "datafied" producer-records rather than simply the k,v pairs (part of #103)
None
Can you improve this documentation? These fine people already did:
Andy Chambers, Gilles Philippart, Chris McDevitt, Matt Ford, Charles Reese, Dave Martin, Francisco Viramontes, Anastasios Panourgias, prakashkillada-fc, Andrew Keedle, Andrea Crotti, Justin Glenn Smith, Sasha Gerrand, Trevor Rawlings, Bobby Calderwood, Rodrigo Gallinares, Ray McDermott, Alexander Popov, Ted Kassen, Alexis Lee, Simon Katz, nivekuil, Marco Chan & JPolEdit on GitHub