[jira] [Commented] (FLINK-20386) ClassCastException when lookup join a JDBC table on INT UNSIGNED column

["jiawen xiao (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-20386?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17243872#comment-17243872 ] 

jiawen xiao commented on FLINK-20386:
-------------------------------------

hi,[~jark] .this problem is the same as FLINK-20481.Does this issue need to be discussed before optimization? I am interested in this question。

> ClassCastException when lookup join a JDBC table on INT UNSIGNED column
> -----------------------------------------------------------------------
>
>                 Key: FLINK-20386
>                 URL: https://issues.apache.org/jira/browse/FLINK-20386
>             Project: Flink
>          Issue Type: Bug
>          Components: Connectors / JDBC, Table SQL / Ecosystem
>    Affects Versions: 1.12.0, 1.11.2
>            Reporter: Jark Wu
>            Priority: Major
>
> The primary key of the MySQL is an INT UNSIGNED column, but declared INT in Flink. 
> I know the [docs|https://ci.apache.org/projects/flink/flink-docs-release-1.11/dev/table/connectors/jdbc.html#data-type-mapping] say it should be decalred BIGINT in Flink, however, would be better not fail the job. 
> At least, the exception is hard to understand for users. We can also check the schema before start the job. 
> {code}
> java.lang.ClassCastException: java.lang.Long cannot be cast to java.lang.Integer
> 	at org.apache.flink.table.data.GenericRowData.getInt(GenericRowData.java:149) ~[flink-table-blink_2.11-1.11-vvr-2.1.1-SNAPSHOT.jar:1.11-vvr-2.1.1-SNAPSHOT]
> 	at JoinTableFuncCollector$6460.collect(Unknown Source) ~[?:?]
> 	at org.apache.flink.table.functions.TableFunction.collect(TableFunction.java:203) ~[flink-table-blink_2.11-1.11-vvr-2.1.1-SNAPSHOT.jar:1.11-vvr-2.1.1-SNAPSHOT]
> 	at org.apache.flink.connector.jdbc.table.JdbcRowDataLookupFunction.eval(JdbcRowDataLookupFunction.java:162) ~[?:?]
> {code}



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