Skip to content
/ flink Public

Flink Connectors, Formats and other plugins

Notifications You must be signed in to change notification settings

exness/flink

Repository files navigation

Flink Utils

Various tools and improvements for Flink

Formats

json-schema (Json With Schema Registry)

'json-schema' format has been created in order to support Schema Registry Json messages.
This format is based on the Apicurio Json SerDe
All the Apicurio configs can be used with this format.
Connector kafka will supply the default json-schema.apicurio.registry.artifact.artifact-id equals to topic.name-value, but it can be changed in the DDL.

This format has the same configuration as Json Format with addition of all the Apicurio configs and the following ones:

Parameter Description Example Default
json-schema.schema-registry.schema-name title field in Json Schema definition. By default equals to the artifact-id in Upper Camel Case SomeSchemaName TopicNameValue

Default Apicurio configs for this format:

Parameter Description Example Default
json-schema.apicurio.registry.as-confluent Use Confluent Id format false true
json-schema.apicurio.registry.headers.enabled Boolean to indicate whether serde classes should pass Global Id information via message headers instead of in the message payload. false true
json-schema.apicurio.registry.auto-register Boolean to indicate whether serializer classes should attempt to create an artifact in the registry. false true
json-schema.apicurio.registry.use-id Configures the serdes to use the specified IdOption as the identifier for the artifacts. Instructs the serializer to write the specified id into the kafka records and instructs the deserializer to read and use the specified id from the kafka records (to find the schema). globalId contentId

Schema Registry URL should be configured via json-schema.apicurio.registry.url parameter.

Keep in mind that the URL should contain only https://schema-registry.host.name value and should NOT contain /apis/ccompat7 part of it

avro-schema (Avro With Schema Registry)

'avro-schema' format has been created in order to support Schema Registry Avro messages.
This format is based on the Apicurio Avro SerDe
All the Apicurio configs can be used with this format.
Connector kafka will supply the default avro-schema.apicurio.registry.artifact.artifact-id equals to topic.name-value, but it can be changed in the DDL.

This format has all the Apicurio configs and the following ones:

Parameter Description Example Default
avro-schema.schema-registry.schema-name name field in Avro Schema definition. By default equals to the artifact-id in Upper Camel Case SomeSchemaName TopicNameValue
avro-schema.schema-registry.schema-namespace namespace field in Avro Schema definition. By default equals to com.exness.sdp.flink. custom.namespace com.exness.sdp.flink

Default Apicurio configs for this format:

Parameter Description Example Default
avro-schema.apicurio.registry.as-confluent Use Confluent Id format false true
avro-schema.apicurio.registry.headers.enabled Boolean to indicate whether serde classes should pass Global Id information via message headers instead of in the message payload. false false
avro-schema.apicurio.registry.auto-register Boolean to indicate whether serializer classes should attempt to create an artifact in the registry. false true
avro-schema.apicurio.registry.use-id Configures the serdes to use the specified IdOption as the identifier for the artifacts. Instructs the serializer to write the specified id into the kafka records and instructs the deserializer to read and use the specified id from the kafka records (to find the schema). globalId contentId

Schema Registry URL should be configured via avro-schema.apicurio.registry.url parameter.

Keep in mind that the URL should contain only https://schema-registry.host.name value and should NOT contain /apis/ccompat7 part of it

About

Flink Connectors, Formats and other plugins

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages