[jira] [Created] (FLINK-14535) Fix distinct key type for DecimalType in DistinctInfo

Zhenghua Gao created FLINK-14535: Summary: Fix distinct key type for DecimalType in DistinctInfo Key: FLINK-14535 URL: https://issues.apache.org/jira/browse/FLINK-14535 Project: Flink Issue T

Re: [VOTE] Accept Stateful Functions into Apache Flink

+1 (binding) On Fri, Oct 25, 2019 at 6:39 PM Maximilian Michels wrote: > +1 (binding) > > On 25.10.19 14:31, Congxian Qiu wrote: > > +1 (non-biding) > > Best, > > Congxian > > > > > > Terry Wang 于2019年10月24日周四 上午11:15写道： > > > >> +1 (non-biding) > >> > >> Best, > >> Terry Wang > >> > >> > >> >

[jira] [Created] (FLINK-14534) FunctionCatalog.getUserDefinedFunctions() should include temp functions

Bowen Li created FLINK-14534: Summary: FunctionCatalog.getUserDefinedFunctions() should include temp functions Key: FLINK-14534 URL: https://issues.apache.org/jira/browse/FLINK-14534 Project: Flink

Re: [VOTE] FLIP-78: Flink Python UDF Environment and Dependency Management

Hi Wei, hi Jincheng, +1 on the current approach. I agree it would be nice to allow for the Beam artifact staging to use Flink's BlobServer. However, the current implementation which uses the distributed file system is more generic, since the BlobServer is only available on the TaskManager and

Re: [VOTE] Accept Stateful Functions into Apache Flink

+1 (binding) On 25.10.19 14:31, Congxian Qiu wrote: +1 (non-biding) Best, Congxian Terry Wang 于2019年10月24日周四 上午11:15写道： +1 (non-biding) Best, Terry Wang 2019年10月24日 10:31，Jingsong Li 写道： +1 (non-binding) Best, Jingsong Lee On Wed, Oct 23, 2019 at 9:02 PM Yu Li wrote: +1 (non-bin

Re: [DISCUSS] Introduce a location-oriented two-stage query mechanism toimprove the queryable state.

Hi David, I know that in some scenarios, the queryable state has some limitations. But in some scenarios, it is valuable, such as in scenarios where you need to observe "trends." It can be analogized to the two existing features provided by Flink web UI: - Metrics: An instantaneous value that

Re: [DISCUSS] FLIP-70 - Support Computed Column for Flink SQL

Hi Danny, Thanks for updating the documentation, I really like this design. But I have two more suggestions: 1) As "STORED" keyword is not in the MVP, I would like to remove the optional "[VIRTUAL]" keyword as well. Because this optional keyword is useless for now, all the computed columns are vi

[jira] [Created] (FLINK-14533) PushFilterIntoTableSourceScanRule misses predicate pushdowns

Yuval Itzchakov created FLINK-14533: --- Summary: PushFilterIntoTableSourceScanRule misses predicate pushdowns Key: FLINK-14533 URL: https://issues.apache.org/jira/browse/FLINK-14533 Project: Flink

Re: [DISCUSS] FLIP-70 - Support Computed Column for Flink SQL

The updated doc looks good to me now, +1 to kick off a voting thread. Best, Kurt On Fri, Oct 25, 2019 at 8:46 PM Danny Chan wrote: > Thanks, I have removed the contents about the “STORED” keyword, we may > introduce that in the future if there is requests. > > Also thanks all the reviewers for

Re: [DISCUSS] FLIP-70 - Support Computed Column for Flink SQL

Thanks, I have removed the contents about the “STORED” keyword, we may introduce that in the future if there is requests. Also thanks all the reviewers for the design doc, I have resolved all the questions/suggestions in the doc at this time. I will kick off a voting thread shortly as there wer

Re: [VOTE] Accept Stateful Functions into Apache Flink

+1 (non-biding) Best, Congxian Terry Wang 于2019年10月24日周四 上午11:15写道： > +1 (non-biding) > > Best, > Terry Wang > > > > > 2019年10月24日 10:31，Jingsong Li 写道： > > > > +1 (non-binding) > > > > Best, > > Jingsong Lee > > > > On Wed, Oct 23, 2019 at 9:02 PM Yu Li wrote: > > > >> +1 (non-binding) > >>

[jira] [Created] (FLINK-14532) Separate PartitionTracker into generic and JM-specific versions

Chesnay Schepler created FLINK-14532: Summary: Separate PartitionTracker into generic and JM-specific versions Key: FLINK-14532 URL: https://issues.apache.org/jira/browse/FLINK-14532 Project: Flin

[jira] [Created] (FLINK-14531) Add Kafka Metrics Reporter

chen yong created FLINK-14531: - Summary: Add Kafka Metrics Reporter Key: FLINK-14531 URL: https://issues.apache.org/jira/browse/FLINK-14531 Project: Flink Issue Type: New Feature Envir

[jira] [Created] (FLINK-14530) Capture partition meta info in PartitionTable

Chesnay Schepler created FLINK-14530: Summary: Capture partition meta info in PartitionTable Key: FLINK-14530 URL: https://issues.apache.org/jira/browse/FLINK-14530 Project: Flink Issue T

[jira] [Created] (FLINK-14529) Compile-jdk11 fails in nightly end-to-end test

Yu Li created FLINK-14529: - Summary: Compile-jdk11 fails in nightly end-to-end test Key: FLINK-14529 URL: https://issues.apache.org/jira/browse/FLINK-14529 Project: Flink Issue Type: Bug Co

Re: [DISCUSS] Introduce a location-oriented two-stage query mechanism toimprove the queryable state.

I've encountered a number of Flink users who considered using queryable state, but after investigation, decided not to. The reasons have been: (1) The current interface (point queries fetching state for one key) is too limiting. What some folks really want/need is the ability to execute SQL-querie

Re: [DISCUSS] Introduce a location-oriented two-stage query mechanism toimprove the queryable state.

Hi Jiayi, Thanks for your valuable feedback and suggestions. In our production env, we still have many applications wrote by DataStream API. Currently, we have some requirements that require Adhoc query for the runtime Flink job. The existing query interface is very difficult to use. This improv

[jira] [Created] (FLINK-14528) Add a Constructor for FlinkKafkaProducer

Peng created FLINK-14528: Summary: Add a Constructor for FlinkKafkaProducer Key: FLINK-14528 URL: https://issues.apache.org/jira/browse/FLINK-14528 Project: Flink Issue Type: Improvement Co