HeartSaVioR edited a comment on issue #24738: [WIP][SPARK-23098][SQL] Migrate Kafka Batch source to v2. URL: https://github.com/apache/spark/pull/24738#issuecomment-497904088 @rdblue Maybe I haven't explained properly. Logically, Kafka has a fixed schema (key/value), but Kafka source and sink are also providing/receiving kinds of metadata as well, which are useful to apply some logic based on that information. For example, Kafka has different use cases compared to read from specific table: reading from multiple topics, which end users want to know which topic the row comes from and apply some logic based on that. Please refer Structured Streaming + Kafka Integration Guide. http://spark.apache.org/docs/latest/structured-streaming-kafka-integration.html > source (reader) Column | Type -- | -- key | binary value | binary topic | string partition | int offset | long timestamp | long timestampType | int > sink (writer) Column | Type -- | -- key (optional) | string or binary value (required) | string or binary topic (*optional) | string * The topic column is required if the “topic” configuration option is not specified. Kafka writer's query validation has been done in KafkaWriter. https://github.com/apache/spark/blob/aec0869fb2ae1ace93056ee1f9ea50b1bdbae9ad/external/kafka-0-10-sql/src/main/scala/org/apache/spark/sql/kafka010/KafkaWriter.scala#L45-L78
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
