Wenchen Fan
Wenchen Fan
> This means it now relies on Spark's hash function for bucketing though, which could be different from other engines. Let's think about it this way: The v2 data source...
thanks, merging to master!
cc @viirya @MaxGekk
thanks for the review, merging to master!
can you update the PR title?
thanks, merging to master!
https://github.com/apache/spark/pull/37436#discussion_r941255226 You did not answer the question. Why it's only an issue with codegen?
thanks, merging to master/3.3!