[
https://issues.apache.org/jira/browse/FLINK-37786?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17956487#comment-17956487
]
Yanquan Lv commented on FLINK-37786:
------------------------------------
I would like to take this.
> Infer schema information from the state of Source and emit it to avoid
> maintaining state in PreTransform
> --------------------------------------------------------------------------------------------------------
>
> Key: FLINK-37786
> URL: https://issues.apache.org/jira/browse/FLINK-37786
> Project: Flink
> Issue Type: Improvement
> Components: Flink CDC
> Affects Versions: cdc-3.5.0
> Reporter: Yanquan Lv
> Priority: Major
>
> At present, in order to maintain the schema information of the original table
> before transform processing, we have maintained a state in the PreTransform
> operator, which increases the complexity of the code and the difficulty of
> maintenance.
> In fact, such information has already been saved in Source. Taking MySQL
> Source as an example, there is a map<TableId, TableChange>saved in
> EmbeddedFlinkBaseHistory. We can derive the schema information corresponding
> to the read site from here.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
