[
https://issues.apache.org/jira/browse/FLINK-29222?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jark Wu updated FLINK-29222:
----------------------------
Fix Version/s: 1.17.0
> Wrong behavior for Hive's load data inpath
> ------------------------------------------
>
> Key: FLINK-29222
> URL: https://issues.apache.org/jira/browse/FLINK-29222
> Project: Flink
> Issue Type: Bug
> Components: Connectors / Hive
> Affects Versions: 1.16.0
> Reporter: luoyuxia
> Priority: Critical
> Labels: pull-request-available
> Fix For: 1.16.0, 1.17.0
>
>
> In hive, `load data inpath` will remove src file, and `load data local
> inpath` won't remove the src file.
> But When using the following sql with Hive dialect:
> {code:java}
> load data local inpath 'test.txt' INTO TABLE tab2 {code}
> The file `test.txt` will be removed, although the expected is not to remove
> the `test.txt`.
> The reason is the parameter order is not right when try to call
> `HiveCatalog#loadTable(..., isOverWrite, isSourceLocal)`,
> It'll call it with
> {code:java}
> hiveCatalog.loadTable(
> ...,
> hiveLoadDataOperation.isSrcLocal(), // should be isOverwrite
> hiveLoadDataOperation.isOverwrite()); // should be isSrcLocal{code}
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
