Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi blackjact, We are already working on the design of stateful part, however, it will not be supported in 1.12. We hope to support it in the later releases, e.g. 1.13. Thank you for your attention. Regards, Shuiqiang > 在 2020年10月20日，下午7:34，blackjjcat 写道： > > May I ask the plan of stateful

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

May I ask the plan of stateful part, which version is expected to be integrated? -- Sent from: http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi, It's a good idea to start with a minimum size of API and add useful ones when we find it is truly useful. >From my side, I'm also ok with the partitionCustom() method. Thanks David for your feedback! Best, Hequn On Mon, Jul 27, 2020 at 8:57 PM Aljoscha Krettek wrote: > Hi, > > I'm also

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi, I'm also not against adding that if it enables actual use cases. I don't think we need to spell out the whole API in the FLIP, though. We can add things as they come up. Best, Aljoscha On 24.07.20 14:43, Shuiqiang Chen wrote: Hi David, Thank you for your reply! I have started the vote

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi David, Thank you for your reply! I have started the vote for this FLIP, but we can keep the discussion on this thread. In my perspective, I would not against adding the DataStream.partitionCustom to Python DataStream API. However, more inputs are welcomed. Best, Shuiqiang David Anderson

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Sorry I'm coming to this rather late, but I would like to argue that DataStream.partitionCustom enables an important use case. What I have in mind is performing partitioned enrichment, where each instance can preload a slice of a static dataset that is being used for enrichment. For an example,

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi Aljoscha, Thank you for your response. I'll keep these two helper methods in the Python DataStream implementation. And thank you all for joining in the discussion. It seems that we have reached a consensus. I will start a vote for this FLIP later today. Best, Shuiqiang Hequn Cheng

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Thanks a lot for your valuable feedback and suggestions! @Aljoscha Krettek +1 to the vote. Best, Hequn On Fri, Jul 24, 2020 at 5:16 PM Aljoscha Krettek wrote: > Thanks for updating! And yes, I think it's ok to include the few helper > methods such as "readFromFile" and "print". > > I think we

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Thanks for updating! And yes, I think it's ok to include the few helper methods such as "readFromFile" and "print". I think we can now proceed to a vote! Nice work, overall! Best, Aljoscha On 16.07.20 17:16, Hequn Cheng wrote: Hi, Thanks a lot for your discussions. I think Aljoscha makes

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi, Thanks a lot for your discussions. I think Aljoscha makes good suggestions here! Those problematic APIs should not be added to the new Python DataStream API. Only one item I want to add based on the reply from Shuiqiang: I would also tend to keep the readTextFile() method. Apart from

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi Aljoscha, Thank you for your valuable comments! I agree with you that there is some optimization space for existing API and can be applied to the python DataStream API implementation. According to your comments, I have concluded them into the following parts: 1. SingleOutputStreamOperator

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Hi, thanks for the proposal! I have some comments about the API. We should not blindly copy the existing Java DataSteam because we made some mistakes with that and we now have a chance to fix them and not forward them to a new API. I don't think we need SingleOutputStreamOperator, in the Scala

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Thanks for bring up this DISCUSS Shuiqiang! +1 for the proposal! Best, Jincheng Xingbo Huang 于2020年7月9日周四 上午10:41写道： > 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

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

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 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

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

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 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

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

+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

Re: [DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

Sorry, the 3rd link is broken, please refer to this one: Support Python DataStream API Shuiqiang Chen 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

[DISCUSS] FLIP-130: Support for Python DataStream API (Stateless Part)

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