[
https://issues.apache.org/jira/browse/FLINK-15610?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17016703#comment-17016703
]
Jingsong Lee commented on FLINK-15610:
--------------------------------------
[~hehuiyuan] Now one way is passing a function Java instance with schema
information.
After UDF refactor, you can use type inference to get constants from context.
You can take a look to
[https://cwiki.apache.org/confluence/display/FLINK/FLIP-65%3A+New+type+inference+for+Table+API+UDFs]
[~twalthr] is working on this.
> How to achieve the udf that the number of return column is uncertain
> ---------------------------------------------------------------------
>
> Key: FLINK-15610
> URL: https://issues.apache.org/jira/browse/FLINK-15610
> Project: Flink
> Issue Type: Wish
> Components: Table SQL / API
> Reporter: hehuiyuan
> Priority: Major
>
> For
> example:[https://help.aliyun.com/knowledge_detail/98948.html?spm=a2c4g.11186631.2.3.21b81761QhpBte]
>
> {code:java}
> SELECT c1, c2
> FROM T1, lateral table(MULTI_KEYVALUE(str, split1, split2, key1, key2))
> as T(c1, c2)
> SELECT c1, c2, c3
> FROM T1, lateral table(MULTI_KEYVALUE(str, split1, split2, key1, key2, key3))
> as T(c1, c2, c3)
> {code}
> For Tablefunction:
> {code:java}
> public TypeInformation<Row> getResultType() {
> return Types.ROW(Types.STRING(),Types.STRING());
> }
> {code}
> The retrun type of `getResultType` is `TypeInformation<Row>`,i want to
> achieve the size of row is not fixed.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
