[
https://issues.apache.org/jira/browse/FLINK-38271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18022948#comment-18022948
]
ouyangwulin commented on FLINK-38271:
-------------------------------------
[~catorleader]
Thanks for your reply, debezium PostgresChangeRecordEmitter#
synchronizeTableSchema have this logic, I'm going to reform implementation
according to this logic. The core is to determine two logic, 1. Whether it is
schemachange or not 2. Production schema change datachangeevent. I understood
that at least one of the two steps needed to read the schema, and I didn't
think about performance at first. And I will call you review code.
> PostgresSQL pipeline supports table schema evolution.
> -----------------------------------------------------
>
> Key: FLINK-38271
> URL: https://issues.apache.org/jira/browse/FLINK-38271
> Project: Flink
> Issue Type: Improvement
> Components: Flink CDC
> Affects Versions: cdc-3.5.0
> Reporter: ouyangwulin
> Assignee: ouyangwulin
> Priority: Critical
>
> PostgreSQL obtains incremental data from the log log parsing, the current
> parser does not support DDL, but we can deduce the table structure from the
> schema information, so it is also possible to implement table structure
> evolution.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
