JesseAtSZ edited a comment on issue #1461: URL: https://github.com/apache/incubator-seatunnel/issues/1461#issuecomment-1064156814
@ruanwenjun Thank you very much for your advice, but I still have the following questions: 1. Is it inefficient to convert SourceRecord to DataStream frequently? I wonder if Flink has any more suitable API to implement this conversion. For example, I found a structure called ChangeLogStream (just an example), although I don't know what it is used for, it seems to be very suitable for conversion from the perspective of naming. Is there any more efficient method to do conversion? In addition, I also saw the description of Avro on the Debezium official website, is there any method to store Schema and deal with schema changes through Avro? I think this is very helpful to improve efficiency. 2. In fact, I'm confused about how to write into mysql, the official website only introduces the use of Flink SQL, and I have seen many examples by writing SQL and then executing JDBC as Sink, I wonder if there is a more elegant way, or I think writing to MySQL is a very common way, Is there any library that can directly process SourceRecord? 3. I have also tried to use RowDataDebeziumdeSerializationSchema, hoping to reduce the conversion overhead, but I found that using this serialization method requires manually specifying the schema. I think this means that I cannot cope with the scenario of schema change, so it is not suitable for me; And I found that the RowData converted in this way carries very little information. Is my understanding correct? In addition, I'd like to ask your opinion on the difference between RowData and Row, I think Row carries more information than RowData. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
