neo4j-streams

neo4j-streams copied to clipboard

Make the timeout in LEADER discovery parametrized

conker84

After the #99 we need to manage the following use-case: if we have two Neo4j instances that are configured as Source and Sink with the CDC schema implementation we need...

conker84

Provide a way to validate refuse data into ingestion strategies

3

We need to highlight into the ingestion strategy if the data is refused. We can attach the `dead letter queue` mechanism in order to simplify the debug process. Thanks @moxious...

conker84

Parallelism

2

1. Connector.taskConfigs() MUST be able to divide the work into the requested maximum number of tasks. Connectors MAY choose to use fewer tasks than allowed.

conker84

Error Handling

1

1. Connectors MUST NOT throw RuntimeException unless it encounters a truly fatal error. 2. Connectors SHOULD catch exceptions and either handle them gracefully to continue operation or rethrow them as...

conker84

provide procedures in order to change configuration at runtime

2

Some procedure like: `CALL streams.config.setValue(‘streams.source.topic.nodes.neo4j-outbound’, ‘SomeNewValue’)` Thanks to @moxious

conker84

Add a procedure to change configuration setting

1

to allow run-time configuration changes

omarlarus

Allow the streams.produce procedure to have a more fine-graded control over the ProducerRecord

1

We should allow the streams.produce to have a more fine-graded control over the ProducerRecord, in particular, we should manage: * the ProducerRecord id in case someone wants to update a...

conker84

investigate streams.consume timeout behaviour

2

Starting with an empty topic, I did this: CALL streams.consume('test', { timeout: 2000 }) YIELD event; the query returned in 800ms Thanks to @moxious

conker84

provide a procedure streams.status

1

The procedure should provide the status of the consumer/producer Thanks to @moxious for the feedback

conker84