[jira] [Updated] (HUDI-3095) Abstract partition filter logic of clustering plan to enable code reuse

["Ethan Guo (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HUDI-3095?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ethan Guo updated HUDI-3095:
----------------------------
    Component/s: clustering
                 table-service

> Abstract partition filter logic of clustering plan to enable code reuse
> -----------------------------------------------------------------------
>
>                 Key: HUDI-3095
>                 URL: https://issues.apache.org/jira/browse/HUDI-3095
>             Project: Apache Hudi
>          Issue Type: Improvement
>          Components: clustering, table-service
>            Reporter: Yuwei Xiao
>            Assignee: Yuwei Xiao
>            Priority: Major
>              Labels: pull-request-available
>
> The partition filter criteria, which is used when generating the clustering plan, is now embedded in specific clustering strategy implementation. However, a filter implementation (e.g., `RecentDay based filter` or `SelectedPartition based filter`) could be used for different clustering strategies (even across engines).
> This Jira plans to abstract the filter logic out of the clustering strategy implementation, making the code structure more clear and enabling code reuse.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)