[SPARK-40707][CONNECT] Add groupby to connect DSL and test more than one grouping expressions #38155
Conversation
|
R: @cloud-fan |
amaliujia
changed the title
|
Can one of the admins verify this patch? |
connector/connect/src/main/scala/org/apache/spark/sql/connect/dsl/package.scala
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
why do we add group by expression to aggregate expressions?
There was a problem hiding this comment.
renamed the proto to make this clear.
|
@cloud-fan PR updated. PLAT. |
There was a problem hiding this comment.
@cloud-fan I still keep this as AggregateFunction. proto.Expression is a too general type for now.
connect does not have a NamedExpression. I will follow up on this to improve.
This PR is to improve the grouping_expressions anyway
There was a problem hiding this comment.
followup improvement SGTM. I don't think we even need AggregateFunction. The SQL parser usually just generate UnresolvedFunction, and the analyzer will look up the function and figure out if it's scalar/aggregate/window/table value function.
amaliujia
force-pushed
the
support_more_than_one_grouping_set
branch
from
cc48ed7 to
33f59ed
Compare
| * This object offers methods to convert to/from connect proto to catalyst types. | ||
| */ | ||
| object TypeProtoConverter { | ||
| def toCatalystType(t: proto.Type): DataType = { |
There was a problem hiding this comment.
shall we name it proto.DataType? And rename this object to DateTypeProtoConverter
|
thanks, merging to master! |
3 participants
What changes were proposed in this pull request?
groupbyto connect DSL and test more than one grouping expressionsTraitin the testing code.Why are the changes needed?
Enhance connect's support for GROUP BY.
Does this PR introduce any user-facing change?
No
How was this patch tested?
UT