[GitHub] [spark] amaliujia commented on a diff in pull request #40070: [SPARK-42481][CONNECT] Implement agg.{max,min,mean,count,avg,sum}

["amaliujia (via GitHub)" <gi...@apache.org>]

amaliujia commented on code in PR #40070:
URL: https://github.com/apache/spark/pull/40070#discussion_r1110349585


##########
connector/connect/client/jvm/src/main/scala/org/apache/spark/sql/RelationalGroupedDataset.scala:
##########
@@ -109,44 +109,131 @@ class RelationalGroupedDataset protected[sql] (
     agg(exprs.asScala.toMap)
   }
 
-  private[this] def strToExpr(expr: String, inputExpr: proto.Expression): proto.Expression = {
+  private[this] def strToExpr(expr: String, columnName: String): proto.Expression = {
     val builder = proto.Expression.newBuilder()
 
     expr.toLowerCase(Locale.ROOT) match {
       // We special handle a few cases that have alias that are not in function registry.
       case "avg" | "average" | "mean" =>
-        builder.getUnresolvedFunctionBuilder
-          .setFunctionName("avg")
-          .addArguments(inputExpr)
-          .setIsDistinct(false)
+        functions.avg(columnName)
       case "stddev" | "std" =>
-        builder.getUnresolvedFunctionBuilder
-          .setFunctionName("stddev")
-          .addArguments(inputExpr)
-          .setIsDistinct(false)
+        functions.stddev(columnName)
       // Also special handle count because we need to take care count(*).
       case "count" | "size" =>
-        // Turn count(*) into count(1)
-        inputExpr match {
-          case s if s.hasUnresolvedStar =>
-            val exprBuilder = proto.Expression.newBuilder
-            exprBuilder.getLiteralBuilder.setInteger(1)
-            builder.getUnresolvedFunctionBuilder
-              .setFunctionName("count")
-              .addArguments(exprBuilder)
-              .setIsDistinct(false)
-          case _ =>
-            builder.getUnresolvedFunctionBuilder
-              .setFunctionName("count")
-              .addArguments(inputExpr)
-              .setIsDistinct(false)
-        }
+        functions.col(columnName)
       case name =>
         builder.getUnresolvedFunctionBuilder
           .setFunctionName(name)
-          .addArguments(inputExpr)
+          .addArguments(df(columnName).expr)

Review Comment:
   Hold on that I will revert this part. It seems hit an issue somewhere if I switch to use functions API. I will debug that separately.



-- 
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: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org