Hi Shuiqiang, Thanks a lot for driving this discussion. Big +1 for supporting Python DataStream. In many ML scenarios, operating Object will be more natural than operating Table.
Best, Xingbo Wei Zhong <weizhong0...@gmail.com> 于2020年7月9日周四 上午10:35写道： > Hi Shuiqiang, > > Thanks for driving this. Big +1 for supporting DataStream API in PyFlink! > > Best, > Wei > > > > 在 2020年7月9日，10:29，Hequn Cheng <he...@apache.org> 写道： > > > > +1 for adding the Python DataStream API and starting with the stateless > > part. > > There are already some users that expressed their wish to have the Python > > DataStream APIs. Once we have the APIs in PyFlink, we can cover more use > > cases for our users. > > > > Best, Hequn > > > > On Wed, Jul 8, 2020 at 11:45 AM Shuiqiang Chen <acqua....@gmail.com> > wrote: > > > >> Sorry, the 3rd link is broken, please refer to this one: Support Python > >> DataStream API > >> < > >> > https://docs.google.com/document/d/1H3hz8wuk22-8cDBhQmQKNw3m1q5gDAMkwTDEwnj3FBI/edit > >>> > >> > >> Shuiqiang Chen <acqua....@gmail.com> 于2020年7月8日周三 上午11:13写道： > >> > >>> Hi everyone, > >>> > >>> As we all know, Flink provides three layered APIs: the > ProcessFunctions, > >>> the DataStream API and the SQL & Table API. Each API offers a different > >>> trade-off between conciseness and expressiveness and targets different > >> use > >>> cases[1]. > >>> > >>> Currently, the SQL & Table API has already been supported in PyFlink. > The > >>> API provides relational operations as well as user-defined functions to > >>> provide convenience for users who are familiar with python and > relational > >>> programming. > >>> > >>> Meanwhile, the DataStream API and ProcessFunctions provide more generic > >>> APIs to implement stream processing applications. The ProcessFunctions > >>> expose time and state which are the fundamental building blocks for any > >>> kind of streaming application. > >>> To cover more use cases, we are planning to cover all these APIs in > >>> PyFlink. > >>> > >>> In this discussion(FLIP-130), we propose to support the Python > DataStream > >>> API for the stateless part. For more detail, please refer to the FLIP > >> wiki > >>> page here[2]. If interested in the stateful part, you can also take a > >>> look the design doc here[3] for which we are going to discuss in a > >> separate > >>> FLIP. > >>> > >>> Any comments will be highly appreciated! > >>> > >>> [1] https://flink.apache.org/flink-applications.html#layered-apis > >>> [2] > >>> > >> > https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=158866298 > >>> [3] > >>> > >> > https://docs.google.com/document/d/1H3hz8wuk228cDBhQmQKNw3m1q5gDAMkwTDEwnj3FBI/edit?usp=sharing > >>> > >>> Best, > >>> Shuiqiang > >>> > >>> > >>> > >>> > >> > >
