Re: Flink SQL support array transform function

Hi Agree with @yuxia, you should do the operation of lambda function in your own udf Best, Shammon On Wed, Feb 22, 2023 at 10:32 AM yuxia wrote: > May be you can try with a non-lambda function. > But TBH, I haven't seen any Flink UDF that accepts function as parameter > in my previous experie

RE: Re: Inconsistent data format of flink-training-repo and learn-flink doc

I have made a pull request (https://github.com/apache/flink/pull/21989) to resolve this issue. Can anybody review it? On 2023/02/22 02:40:23 yuxia wrote: > Seems in this pr[1], startTime and endTIme have been replaced with a single eveentTime, but the doc forget to update. > Coud you please help c

Re: Inconsistent data format of flink-training-repo and learn-flink doc

Seems in this pr[1], startTime and endTIme have been replaced with a single eveentTime, but the doc forget to update. Coud you please help create a JIRA[2] for it? And if you're interested to fix it, weclome to contribute. [1] https://github.com/apache/flink-training/pull/36 [2] https://issue

Re: Flink SQL support array transform function

May be you can try with a non-lambda function. But TBH, I haven't seen any Flink UDF that accepts function as parameter in my previous experience. I'm afraid that it's no allowed to pass a function as parameter. Best regards, Yuxia 发件人: "Xuekui" 收件人: "yuxia" 抄送: "fskmine" , "Caizhi Wen

Re: Issue with de-serializing CompiledPlan and UNNEST_ROWS in Table API

Hi, Daniel Henneberger. Thanks for reporting. It seems a bug to me. Could you please help create a Jira[1] for it? As a workaround, is it possible not to use UNNEST? May be you can try to use EXPLODE function for the Flink planner will rewrites UNNEST to explode function in implementation[2].

Issue with de-serializing CompiledPlan and UNNEST_ROWS in Table API

Dear Apache Flink community, I could use some help with a serialization issue I'm having while using the Table API. Specifically, I'm trying to deserialize a serialized CompiledPlan, but I'm running into trouble with the UNNEST_ROWS operation. It seems that the CompilePlan deserializer isn't looki

Re: Calculation of UI's maximum non-heap memory

Very useful, thanks a lot. Regards, Alexis. Am Di., 21. Feb. 2023 um 12:04 Uhr schrieb Weihua Hu : > Hi Alexis, > > The maximum Non-Heap is the sum of the memory pool (which is non-hep) max > size. There are 3 memory pools(based on jdk11): > 1. Metaspace, we can control the size with JVM parame

Re: Job Cancellation Failing

I noticed a test instability that sounds quite similar to what you're experiencing. I created FLINK-31168 [1] to follow-up on this one. [1] https://issues.apache.org/jira/browse/FLINK-31168 On Mon, Feb 20, 2023 at 4:50 PM Matthias Pohl wrote: > What do you mean by "earlier it used to fail due t

Inconsistent data format of flink-training-repo and learn-flink doc

Hi, The data format specified in flink-training-repo ( https://github.com/apache/flink-training/tree/release-1.16) shows that a TaixRide is either a start or an end one with the eventTime. However, the Java code in "Data Pipelines & ETL" ( https://nightlies.apache.org/flink/flink-docs-release-1.16

Re: Whether Flink SQL window operations support "Allow Lateness and SideOutput"?

Hi, Maybe you can use CURRENT_WATERMARK()[1] to handle some late data. [1] https://nightlies.apache.org/flink/flink-docs-master/docs/dev/table/functions/systemfunctions/ Best, Weihua On Tue, Feb 21, 2023 at 1:46 PM wang <24248...@163.com> wrote: > Hi dear engineers, > > One question as t

Re: Calculation of UI's maximum non-heap memory

Hi Alexis, The maximum Non-Heap is the sum of the memory pool (which is non-hep) max size. There are 3 memory pools(based on jdk11): 1. Metaspace, we can control the size with JVM parameter -XX:MaxMetaspaceSize or Flink configuration: jobmanager.memory.jvm-metaspace.size. For your job, this pool