cloud-fan commented on code in PR #38276:
URL: https://github.com/apache/spark/pull/38276#discussion_r997170213
##########
connector/connect/src/main/scala/org/apache/spark/sql/connect/planner/SparkConnectPlanner.scala:
##########
@@ -71,6 +73,42 @@ class SparkConnectPlanner(plan: proto.Relation, session:
SparkSession) {
session.sessionState.sqlParser.parsePlan(sql.getQuery)
}
+ private def transformDeduplicate(rel: proto.Deduplicate): LogicalPlan = {
+ if (!rel.hasInput) {
+ throw InvalidPlanInput("Deduplicate needs a plan input")
+ }
+ if (rel.getIsAll && rel.getDeduplicateColumnsCount > 0) {
+ throw InvalidPlanInput("Cannot deduplicate on both all columns and a
subset of columns")
+ }
+ if (!rel.getIsAll && rel.getDeduplicateColumnsCount == 0) {
+ throw InvalidPlanInput(
+ "Deduplicate requires to either deduplicate on all columns or a subset
of columns")
+ }
+ val queryExecution = new QueryExecution(session,
transformRelation(rel.getInput))
+ val resolver = session.sessionState.analyzer.resolver
+ val allColumns = queryExecution.analyzed.output
+ val toGroupColumnNames = {
+ if (rel.getIsAll) {
+ allColumns.map(t => t.name).toSeq
Review Comment:
We don't need to do name lookup. We can just create `Deduplicate(allColumns,
queryExecution.analyzed)`
--
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]
