Jark Wu created FLINK-20348:
-------------------------------
Summary: Make "schema-registry.subject" optional for Kafka sink with avro-confluent format
Key: FLINK-20348
URL:
https://issues.apache.org/jira/browse/FLINK-20348 Project: Flink
Issue Type: Improvement
Components: Formats (JSON, Avro, Parquet, ORC, SequenceFile), Table SQL / Ecosystem
Reporter: Jark Wu
Fix For: 1.12.0
Currently, configuration "schema-registry.subject" in avro-confluent format is required by sink. However, this is quite verbose set it manually. By default, it can be to set to {{<topic_name>-key}} and {{<topic_name>-value}} if it works with kafka or upsert-kafka connector. This can also makes 'avro-confluent' format to be more handy and works better with Kafka/Confluent ecosystem.
{code:sql}
CREATE TABLE kafka_gmv (
day_str STRING,
gmv BIGINT,
PRIMARY KEY (day_str) NOT ENFORCED
) WITH (
'connector' = 'upsert-kafka',
'topic' = 'kafka_gmv',
'properties.bootstrap.servers' = 'localhost:9092',
-- 'key.format' = 'raw',
'key.format' = 'avro-confluent',
'key.avro-confluent.schema-registry.url' = '<a href="http://localhost:8181'">http://localhost:8181',
'key.avro-confluent.schema-registry.subject' = 'kafka_gmv-key',
'value.format' = 'avro-confluent',
'value.avro-confluent.schema-registry.url' = '<a href="http://localhost:8181'">http://localhost:8181',
'value.avro-confluent.schema-registry.subject' = 'kafka_gmv-value'
);
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)