[
https://issues.apache.org/jira/browse/HUDI-7441?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ethan Guo updated HUDI-7441:
----------------------------
Fix Version/s: 0.15.0
1.0.0
> Decouple hive dependencies during flink streaming read hudi
> -----------------------------------------------------------
>
> Key: HUDI-7441
> URL: https://issues.apache.org/jira/browse/HUDI-7441
> Project: Apache Hudi
> Issue Type: Improvement
> Reporter: Rascal Wu
> Priority: Major
> Labels: pull-request-available
> Fix For: 0.15.0, 1.0.0
>
> Attachments: image-2024-02-24-22-50-44-306.png,
> image-2024-02-24-22-51-01-355.png
>
>
> Currently, the `IncrementalInputSplits#getReadPartitions` method will call `
> HoodieInputFormatUtils.getWritePartitionPaths`, but `HoodieInputFormatUtils`
> class imports hive classes. So if flink engine streaming read a hudi table,
> and without any case need to integrate with hive, there is no need to include
> hive jar, because the `hudi-flink-bundle` doesn't contains hive classes
> defaultly.
>
> It's better to move the required mehtod from
> HoodieInputFormatUtils(hudi-hadoop-mr module) to common module to decouple
> the hive dependencies.
>
> !image-2024-02-24-22-50-44-306.png!
>
> !image-2024-02-24-22-51-01-355.png!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
