[
https://issues.apache.org/jira/browse/FLINK-23764?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17405707#comment-17405707
]
Hang Ruan commented on FLINK-23764:
-----------------------------------
Now we have to provide a group.id property when using _KafkaSourceBuilder_ to
build _KafkaSource._ There is a gap between original Table/SQL API and the
builder, as the Table/SQL API doesn't need the group.id property.
To automatically generate a meaningful group.id property for the _KafkaReader_,
I think we should make use of the operator id which is not contained in the
_SourceReaderContext_.
[~fabian.paul] , do you have any other suggestion about automatically
generating a meaningful group.id property?
> add RuntimeContext to SourceReaderContext
> -----------------------------------------
>
> Key: FLINK-23764
> URL: https://issues.apache.org/jira/browse/FLINK-23764
> Project: Flink
> Issue Type: Improvement
> Components: Connectors / Common
> Reporter: Hang Ruan
> Priority: Major
> Labels: connector
>
> RuntimeContext is the important information for sourceReader. Not only the
> subtask index, we sometimes need to get other information in RuntimeContext
> like the operator id.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
