[
https://issues.apache.org/jira/browse/FLINK-36165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thorne updated FLINK-36165:
---------------------------
Summary: Support capturing snapshot data with where conditions (was:
Support capturing snapshot data of the scope)
> Support capturing snapshot data with where conditions
> -----------------------------------------------------
>
> Key: FLINK-36165
> URL: https://issues.apache.org/jira/browse/FLINK-36165
> Project: Flink
> Issue Type: Improvement
> Components: Flink CDC
> Reporter: Thorne
> Priority: Minor
> Attachments: image-2024-08-28-14-06-40-501.png
>
>
> 1. When I want to supplement some data (business consumption data failed)
> 2. Unable to start from checkpoint, because the binlog files has been cleared.
> 3. At this point, only the full amount of data can be synchronized from
> scratch. If the data volume is small, it is acceptable. However, if the data
> volume is large, the process of supplementing the data requires significant
> resources.
> Now：snapshot all data
> {code:java}
> select * from table{code}
> snapshot chunk data
> {code:java}
> select * from table where chunkStart>? and chunkEnd< ?{code}
> So ,I think that we can achieve this feature
> Just like Datax,users obtain data from table without all snapshot data. at
> first analysis Table.
> {code:java}
> select * from table where col >x1 and col <x2
> {code}
> [https://github.com/alibaba/DataX/blob/master/mysqlreader/doc/mysqlreader.md|http://example.com/]
>
> !image-2024-08-28-14-06-40-501.png|width=987,height=244!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
