[
https://issues.apache.org/jira/browse/FLINK-8577?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16701264#comment-16701264
]
ASF GitHub Bot commented on FLINK-8577:
---------------------------------------
hequn8128 commented on a change in pull request #6787: [FLINK-8577][table]
Implement proctime DataStream to Table upsert conversion
URL: https://github.com/apache/flink/pull/6787#discussion_r236911934
##########
File path:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/api/StreamTableEnvironment.scala
##########
@@ -555,18 +555,105 @@ abstract class StreamTableEnvironment(
s"But is: ${execEnv.getStreamTimeCharacteristic}")
}
+ // Can not apply key on append stream
+ if (extractUniqueKeys(fields).nonEmpty) {
+ throw new TableException(
+ s"Can not apply key on append stream, use fromUpsertStream instead.")
Review comment:
I haven't thought much about how to support key definition on append stream.
I think it may not be a no-op key definition. Maybe we have to do the following
things:
- Add similar logic in `UniqueKeyExtractor` to get keys.
- Probably have to check whether data meet the key constraint. i.e, throw an
exception when key data is not distinct.
- etc.
Given the above things. I think we would better to not support it now?
I will rephrase the error message according to your suggestions. Thanks a
lot.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Implement proctime DataStream to Table upsert conversion.
> ---------------------------------------------------------
>
> Key: FLINK-8577
> URL: https://issues.apache.org/jira/browse/FLINK-8577
> Project: Flink
> Issue Type: Sub-task
> Components: Table API & SQL
> Reporter: Hequn Cheng
> Assignee: Hequn Cheng
> Priority: Major
> Labels: pull-request-available
>
> Api will looks like:
> {code:java}
> DataStream[(String, Long, Int)] input = ???
> // upsert with keyTable
> table = tEnv.upsertFromStream(input, 'a, 'b, 'c.key)
> // upsert without key -> single row tableTable
> table = tEnv.upsertFromStream(input, 'a, 'b, 'c){code}
> A simple design
> [doc|https://docs.google.com/document/d/1yMEi9xkm3pNYNmmZykbLzqtKjFVezaWe0y7Xqd0c1zE/edit?usp=sharing]
> about this subtask.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
