Releases: neo4j-contrib/neo4j-streams
4.0.1
New release
First Release for Neo4j v 4.0
Hi community!
We are very proud to share the first release of Streams Project for Neo4j 4.0 (here the docs).
This release supports the Multi database feature for sink and source in the following way.
Neo4j Streams: Source Module
You can set for each database instance a configuration suffix with the following pattern from.<DB_NAME> to the properties in your neo4j.conf file.
Following the list of new properties that allows to support multi-tenancy:
streams.source.topic.nodes.<TOPIC_NAME>.from.<DB_NAME>=<PATTERN>
streams.source.topic.relationships.<TOPIC_NAME>.from.<DB_NAME>=<PATTERN>
streams.source.enabled.from.<DB_NAME>=<true/false, default=true>
streams.procedures.enabled.<DB_NAME>=<true/false, default=true>
For more in-depth description, please got to the following doc page
Neo4j Streams: Sink Module
You can set for each database instance a configuration suffix with the following pattern to.<DB_NAME> to the properties in your neo4j.conf file.
Following the list of new properties that allows to support multi-tenancy:
streams.sink.topic.cypher.<TOPIC_NAME>.to.<DB_NAME>=<CYPHER_QUERY>
streams.sink.topic.cdc.sourceId.to.<DB_NAME>=<LIST_OF_TOPICS_SEPARATE_BY_SEMICOLON>
streams.sink.topic.cdc.schema.to.<DB_NAME>=<LIST_OF_TOPICS_SEPARATE_BY_SEMICOLON>
streams.sink.topic.pattern.node.<TOPIC_NAME>.to.<DB_NAME>=<NODE_EXTRACTION_PATTERN>
streams.sink.topic.pattern.relationship.<TOPIC_NAME>.to.<DB_NAME>=<RELATIONSHIP_EXTRACTION_PATTERN>
streams.sink.enabled.to.<DB_NAME>=<true/false, default=true>
streams.procedures.enabled.<DB_NAME>=<true/false, default=true>
For more in-depth description, please got to the following doc page
Multi Database Support
You can define into the json (or via the Confluent UI) a param named neo4j.database
which is the targeted database name.
N.b. If no value is specified the connector will use the Neo4j's default db.
Minor release 3.5.6
This minor release contains the following fixes:
- fixes #284: Core server will not rejoin cluster if Kafka plugin is enabled.
3.5.5
Hello Neo4j Community!
We're very happy to present the new 3.5.5 release of the Neo4j Kafka integrations, that is the result of great community feedback!
Thanks a lot to everyone who contributed to this: @moxious, @jexp, @conker84 and @F-Guardian
Following a list of improvements:
-
with
fixes #252: Source with non-existant topic hangs indefinitely (#257)
we improved the behaviour in case the BROKER configauto.create.topics.enable
is false and you're sending an event to a topic that is not created in the Kafka cluster. If you're publish events to a topic that is not created theProducer#send
can block until the timeout defined bymax.block.ms
, so we add a system that checks every topic defined in the Kafka Cluster and log if the topic that you defined into yourneo4j.conf
file exist. -
with
fixes #247: Schema Ingestion Strategy: double node creation (#248)
we fixed a bug where, under particular circumstances, can lead to a double node creation in case you're ingesting data with theSchema ingestion strategy
-
with
Fixes an issue with wrong ids value in cdc when updating relationships
we fixed a bug where that could lead to stream wrong ids for the target node while you're updating relationships
Other fixes:
- fix typo in documentation section 4.2 (#253)
- fixes #249: Correct typo in DLQ property (#262)
- fixes #259: Fix Kafka plugin version (#260)
- fixes #264: Neo4j Sink Connector: CDC ingestion with schema strategy
To get more info about it please look at the documentation.
3.5.4
Hello Neo4j Community!
We're very happy to present the new 3.5.4 release of the Neo4j Kafka integrations, that is the result of great community feedback!
-
We improved the management of the DLQ. Please look at the documentation
-
We introduced the support for the transient errors into the Kafka Connect Plugin
-
With #211 we introduced the AVRO support to the Neo4j Sink Plugin, the
streams.consume
procedure manages it as well! Please look at the Supported Kafka deserializers -
With #188 we introduced the support for a new file interexchange format, we called it CUD File Format and is a JSON file that represents Graph Entities (Nodes/Relationships) and how to mange them in term of Create/Update/Delete operations. i.e.:
{
"op": "merge",
"properties": {
"foo": "value",
"key": 1
},
"ids": {"key": 1, "otherKey": "foo"},
"labels": ["Foo","Bar"],
"type": "node",
"detach": true
}
will be transformed into the follwing Cypher Query:
UNWIND [..., {
"op": "merge",
"properties": {
"foo": "value",
"key": 1
},
"ids": {"key": 1, "otherKey": "foo"},
"labels": ["Foo","Bar"],
"type": "node",
"detach": true
}, ...] AS event
MERGE (n:Foo:Bar {key: event.ids.key, otherKey: event.ids.otherKey})
SET n += event.properties
For more details please look into the documentation
- With #181 we validate a minimun amount of configuration params in orde to make the streams plugin work.
To get more info about it please look at the documentation.
3.5.3
We're very happy to present the new 3.5.3 release of the Neo4j Kafka integrations.
Our focus since the last release was on making our integration easier to use and fixing some issues.
We introduced several new features in the Sink:
With fixes #177: Create a Dead Letter Queue you can manage the Dead Lettere Queue in case of bad data, this feature is initially available only into the Streams Plugin, we plan to also release it for the Kafka Connect Sink Plugin in the near future.
With fixes #198: Tombstone record Management both the Streams Sink and the Kafka Connect Sink Plugin can get improved, in particular, the pattern strategy come out with the support to the Tombstone Record,
in order to leverage it your event should contain as key the record that you want to delete and null
for the value.
To get more info about it please look at the documentation.
3.4.4
We're very happy to present the new 3.4.4 release of the Neo4j Kafka integrations.
Our focus since the last release was on making our integration easier to use and fixing some issues.
We introduced several new features in the Sink:
With fixes #177: Create a Dead Letter Queue you can manage the Dead Lettere Queue in case of bad data, this feature is initially available only into the Streams Plugin, we plan to also release it for the Kafka Connect Sink Plugin in the near future.
With fixes #198: Tombstone record Management both the Streams Sink and the Kafka Connect Sink Plugin can get improved, in particular, the pattern strategy come out with the support to the Tombstone Record,
in order to leverage it your event should contain as key the record that you want to delete and null
for the value.
To get more info about it please look at the documentation.
3.5.2
New release
3.4.3
New release
Release 3.5.1 of the Neo4j Kafka integration
We're very happy to present the new 3.5.1 release of the Neo4j Kafka integrations.
Our focus since the last release was on making our integration easier to use and fixing some issues.
We introduced several new features in the Sink:
The fixes #99: Provide a roundtrip-sink-config allows you to ingest data that that comes from another Neo4j instances as CDC events.
In a similar way fixes #154: provide a common pattern for ingestion allows you to define simple expressions in order to extract data from any nested event structure and transfrom that data into Nodes and/or Relationships.
For example to create a user and their purchases from the users
and orders
topics:
streams.sink.topic.pattern.node.users=User{!userId}
streams.sink.topic.pattern.relationship.orders=User{!userId} BOUGHT{purchase.price, purchase.currency} Product{!productId}
The fixes #102: Manual commit behavior for handling errors and retrievals allows you to use a manual committing consumer, moreover improves the streams.consume
procedure allowing to read data starting from a specific partition/offset.
Breaking changes
There is a little change about the Sink management, with the fixes 160: change the streams.sink.enabled to false the Streams plugin now has, for the property streams.sink.enabled
, the default value set to false
so you need to explict set it to true otherwise, if you only specify the topic mapping, you'll see a WARN
message into the neo4j.log
We also fixed several issues:
- fixes #186: Kafka even sink with manual commit fails with multiple topic subscriptions
- fixes #185: Consumer not working in Cluster environment
- Chunked docs
- fixes #102: Manual commit behavior for handling errors and retrievals
- fixes 160: change the streams.sink.enabled to false
- fixes #154: provide a common pattern for ingestion
- fixes #173: Update Documentation: Streams.Enabled and Producer.Enabled
- fixes #99: Provide a roundtrip-sink-config
- fixes #167
- fixes #164: If docker is not installed skip integration tests
- fixes #162: Conversion error in case of nested field of array records
- Fix YAML for NEO4J_ACCEPT_LICENSE_AGREEMENT and add auto-install of connector
- fixes #14: Add schema reference into Data Events
- fixes #147: Kafka Connect Sink Transient Error Mangement
- fixes #116: Connector Configuration (Sink)
- added Plugin Installation section
- Adjust config for kafka-connect-maven-plugin in Kafka connector