michaelmitchell-bit opened a new pull request, #4823: URL: https://github.com/apache/datafusion-comet/pull/4823
## Which issue does this PR close? Closes #4820. Depends on #4819. ## Rationale for this change After #4819, `approx_count_distinct` stores HyperLogLogPlusPlus buffers in the same packed Long layout as Spark. That makes the partial/final aggregate state compatible across Spark and Comet, so Comet can safely enable mixed partial/final execution for this aggregate. This PR is draft while #4819 is still open because the branch is stacked on that work. ## What changes are included in this PR? - Enable mixed partial/final execution for `CometApproxCountDistinct`. - Add planner coverage for Comet partial with Spark final execution. - Add planner coverage for Spark partial with Comet final execution. - Add SQL-file coverage for `approx_count_distinct` through a partial-merge aggregate path. ## How are these changes tested? - `make core` - `./mvnw test -Dtest=none -Dsuites="org.apache.comet.rules.CometExecRuleSuite approx_count_distinct" -Dscalastyle.skip=true` - `./mvnw test -Dtest=none -Dsuites="org.apache.comet.CometSqlFileTestSuite partial_merge" -Dscalastyle.skip=true` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
