Skip to content

Conversation

@davidnavas
Copy link

@markhamstra don't know if it's worth pursuing this -- depends on if spark2.1 is ready to go soon and if this is a bug that customers have run into.
Also, should run tests and make sure I haven't broken anything, but it does at least pass a (new) test I added to sparky for this issue.

@davidnavas
Copy link
Author

Note, this DOES fail a couple of tests, and I'm not 100% sure I'm rewriting correctly (although I don't see how my changes to the Rewriter cause failures in tests that should not be subject to rewrite because there's only one distinct....)

@davidnavas
Copy link
Author

FWIW, the two tests that fail with this change also fail without it.....

@davidnavas
Copy link
Author

backporting apache/spark/apache#15668 leads to a much better place -- the resulting query loses the unnecessary columns for the aggregate children (arguments) which were literals, because that code extracts the literals earlier.

@markhamstra markhamstra merged commit d8bf9ee into alteryx:csd-1.6 Dec 4, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants