[
https://issues.apache.org/jira/browse/FLINK-16998?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17100403#comment-17100403
]
Jingsong Lee commented on FLINK-16998:
--------------------------------------
+1 to keep nested flag.
When introducing data structure, Kurt and I discussed whether NestedRow should
keep header information. The final conclusion is that we don't need to treat
external rows differently from nested rows, which will make our design and
implementation very complicated. The same binary data between BinaryRow and
NestedRow makes NestedRow's processing easy.
And since we choose to save the flag information in Kafka, filesystem and other
connectors, the flag information may be nested. may be stored in list, may be
used anywhere.
> Add a changeflag to Row type
> ----------------------------
>
> Key: FLINK-16998
> URL: https://issues.apache.org/jira/browse/FLINK-16998
> Project: Flink
> Issue Type: Sub-task
> Components: API / Core
> Reporter: Timo Walther
> Assignee: Timo Walther
> Priority: Major
>
> In Blink planner, the change flag of records travelling through the pipeline
> are part of the record itself but not part of the logical schema. This
> simplifies the architecture and API in many cases.
> Which is why we aim adopt the same mechanism for
> {{org.apache.flink.types.Row}}.
> Take {{tableEnv.toRetractStream()}} as an example that returns either Scala
> or Java {{Tuple2<Boolean, Row>}}. For FLIP-95 we need to support more update
> kinds than just a binary boolean.
> This means:
> - Add a changeflag {{RowKind}} to to {{Row}}
> - Update the {{Row.toString()}} method
> - Update serializers in backwards compatible way
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
