Hi Soheil, I assume that you are using `DataStream` API. Please check the document [1] to get more information. Other guys said a lot about this.
Regardless the interfaces, I'm just wondering how could you read a Mysql table "continuously"? Kafka can be used as a message queue which is convenient to get the incremental messages. How do you plan to do that based on Mysql table? Through binary log? 1. https://ci.apache.org/projects/flink/flink-docs-release-1.8/dev/datastream_api.html#data-sources Caizhi Weng <tsreape...@gmail.com> 于2019年7月16日周二 上午10:12写道： > Hi Soheil, > > It's not recommended to implement a streaming source using `InputFormat` > (it's mainly used for batch source). To implement a streaming source, > `SourceFunction` is recommended. > > It's clearly written (with examples) in the java docs in `SourceFucntion` > how to write a `run` and `cancel` method. You can refer to that to write > your own MySQL streaming source. > > Soheil Pourbafrani <soheil.i...@gmail.com> 于2019年7月16日周二 上午7:29写道： > >> Hi, >> >> Extending the "RichInputFormat" class I could create my own MySQL input. >> I want to use it for reading data continuously from a table but I observed >> that the "RichInputFormat" class read all data and finish the job. >> >> I guess for reading data continuously I need to extend the >> "SourceFunction" but I observed that it has only two methods: the run() and >> the cancel() >> >> So I was wondering is it possible to implement a new class to read data >> from MySQL tables continuously? Like what we can do with Kafka connector >> >> Thanks >> >
