[
https://issues.apache.org/jira/browse/FLINK-13389?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17336459#comment-17336459
]
Flink Jira Bot commented on FLINK-13389:
----------------------------------------
This issue was labeled "stale-major" 7 ago and has not received any updates so
it is being deprioritized. If this ticket is actually Major, please raise the
priority and ask a committer to assign you the issue or revive the public
discussion.
> Setting DataStream return type breaks some type conversion between Table and
> DataStream
> ---------------------------------------------------------------------------------------
>
> Key: FLINK-13389
> URL: https://issues.apache.org/jira/browse/FLINK-13389
> Project: Flink
> Issue Type: Bug
> Components: API / DataStream, Table SQL / API
> Reporter: Rong Rong
> Priority: Major
> Labels: stale-major
>
> When converting between data stream and table, there are situations where
> only GenericTypeInfo can be successfully applied, but not directly setting
> the specific RowTypeInfo.
> For example the following code doesn't work
> {code:java}
> TypeInformation<?>[] types = {
> BasicTypeInfo.INT_TYPE_INFO,
> TimeIndicatorTypeInfo.ROWTIME_INDICATOR(),
> BasicTypeInfo.STRING_TYPE_INFO};
> String[] names = {"a", "b", "c"};
> RowTypeInfo typeInfo = new RowTypeInfo(types, names);
> DataStream<Row> ds = env.fromCollection(data).returns(typeInfo);
> Table sourceTable = tableEnv.fromDataStream(ds, "a,b,c");
> tableEnv.registerTable("MyTableRow", sourceTable);
> DataStream<Row> stream = tableEnv.toAppendStream(sourceTable,
> Row.class)
> .map(a -> a)
> // this line breaks the conversion, it sets the
> typeinfo to RowTypeInfo.
> // without this line the output type is
> GenericTypeInfo(Row)
> .returns(sourceTable.getSchema().toRowType());
> stream.addSink(new StreamITCase.StringSink<Row>());
> env.execute();
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
