https://github.com/apache/spark
Revision 15ebd56de6ae37587d750bb1e106c5dcb3e22958 authored by ulysses-you on 03 August 2022, 17:03:45 UTC, committed by Wenchen Fan on 03 August 2022, 17:04:10 UTC
### What changes were proposed in this pull request?

recacheByPlan the result relation inside `SaveIntoDataSourceCommand`

### Why are the changes needed?

The behavior of `SaveIntoDataSourceCommand` is similar with `InsertIntoDataSourceCommand` which supports append or overwirte data. In order to keep data consistent,  we should always do recacheByPlan the relation on post hoc.

### Does this PR introduce _any_ user-facing change?

yes, bug fix

### How was this patch tested?

add test

Closes #37380 from ulysses-you/refresh.

Authored-by: ulysses-you <ulyssesyou18@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
(cherry picked from commit 5fe0b245f7891a05bc4e1e641fd0aa9130118ea4)
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
1 parent 6e9a58f
History
Tip revision: 15ebd56de6ae37587d750bb1e106c5dcb3e22958 authored by ulysses-you on 03 August 2022, 17:03:45 UTC
[SPARK-39952][SQL] SaveIntoDataSourceCommand should recache result relation
Tip revision: 15ebd56

README.md

back to top