[3.2][Spark] Execute MERGE using Dataframe API in Scala #3585
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
(cherrypick of delta-io#3456) Due to Spark unfortunate behavior of resolving plan nodes it doesn't know, the `DeltaMergeInto` plan created when using the MERGE scala API needs to be manually resolved to ensure spark doesn't interfere with its analysis. This currently completely bypasses Spark's analysis as we then manually execute the MERGE command which has negatiev effects, e.g. the execution is not visible in QueryExecutionListener. This change addresses this issue, by executing the plan using the Dataframe API after it's manually resolved so that the command goes through the regular code path. Resolves delta-io#1521 Covered by existing tests.
This was referenced Jun 25, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
(cherrypick of #3456)
Due to Spark unfortunate behavior of resolving plan nodes it doesn't know, the
DeltaMergeIntoplan created when using the MERGE scala API needs to be manually resolved to ensure spark doesn't interfere with its analysis.This currently completely bypasses Spark's analysis as we then manually execute the MERGE command which has negatiev effects, e.g. the execution is not visible in QueryExecutionListener.
This change addresses this issue, by executing the plan using the Dataframe API after it's manually resolved so that the command goes through the regular code path.
Resolves #1521 Covered by existing tests.