You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Jark Wu (Jira)" <ji...@apache.org> on 2020/05/26 10:25:00 UTC

[jira] [Assigned] (FLINK-17753) watermark defined in ddl does not work in Table api

     [ https://issues.apache.org/jira/browse/FLINK-17753?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jark Wu reassigned FLINK-17753:
-------------------------------

    Assignee: godfrey he

> watermark defined in ddl does not work in Table api
> ---------------------------------------------------
>
>                 Key: FLINK-17753
>                 URL: https://issues.apache.org/jira/browse/FLINK-17753
>             Project: Flink
>          Issue Type: Bug
>          Components: Table SQL / Planner
>            Reporter: godfrey he
>            Assignee: godfrey he
>            Priority: Major
>             Fix For: 1.11.0
>
>
> the following code will get {{org.apache.flink.table.api.ValidationException: A group window expects a time attribute for grouping in a stream environment.}}
> {code:java}
> @Test
>   def testRowTimeTableSourceGroupWindow(): Unit = {
>     val ddl =
>       s"""
>          |CREATE TABLE rowTimeT (
>          |  id int,
>          |  rowtime timestamp(3),
>          |  val bigint,
>          |  name varchar(32),
>          |  watermark for rowtime as rowtime
>          |) WITH (
>          |  'connector' = 'projectable-values',
>          |  'bounded' = 'false'
>          |)
>        """.stripMargin
>     util.tableEnv.executeSql(ddl)
>     val t = util.tableEnv.from("rowTimeT")
>       .where($"val" > 100)
>       .window(Tumble over 10.minutes on 'rowtime as 'w)
>       .groupBy('name, 'w)
>       .select('name, 'w.end, 'val.avg)
>     util.verifyPlan(t)
>   }
> {code}
> The reason is planner does not convert {{watermarkSpecs}} in {{TableSchema}} to correct type when calling {{tableEnv.from}}



--
This message was sent by Atlassian Jira
(v8.3.4#803005)