[Proposal] Launch a new project for sink connector of Apache Flink

[whhe]

As obconnector-j has been open sourced, and we has Flink CDC as source connector of Apache Flink, we can now launch a new project for sink connector of Apache Flink.

We can do some performance optimization referencing datax, and support the MySQL mode of OceanBase in the first release.

[longdafeng]

If just for Flink, why not contribute to Flink project.

But if we plan to implement multiple sink connector, not only for Flink, but also for Chunjun/TapData, setting up one project make sense.

[whhe]

According to the disscussion in the mail list, Flink community is going to move most of connectors outside the Flink repository, just like flink-connector-jdbc.
And for now it is a common practice for database brands to maintain the Flink/Spark connector under their own organizations. In this way, we can have a strong control over the iteration cycle, requirements priority and testing standard, which is very meaningful for a new project.

[whhe]

The repository is public now.

https://github.com/oceanbase/flink-connector-oceanbase