Shuiqiang Chen created FLINK-20647:
--------------------------------------
Summary: Use yield to generate output datas in ProcessFunction for
Python DataStream
Key: FLINK-20647
URL: https://issues.apache.org/jira/browse/FLINK-20647
Project: Flink
Issue Type: Improvement
Components: API / Python
Reporter: Shuiqiang Chen
Fix For: 1.13.0
Currently, users need to use a `collector` to collect the output of their
ProcessFunction. We should use yield to make it more efficient that avoid using
an array buffer to store the collected data, and it would be more pythonic as
well.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
