Re: [PR] [SPARK-51259][SQL] Refactor and improve performance for natural and using join keys computation [spark]

["mihailotim-db (via GitHub)" <gi...@apache.org>]

mihailotim-db commented on code in PR #50009:
URL: https://github.com/apache/spark/pull/50009#discussion_r1961806667


##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala:
##########
@@ -3564,55 +3564,17 @@ class Analyzer(override val catalogManager: CatalogManager) extends RuleExecutor
       hint: JoinHint): LogicalPlan = {
     import org.apache.spark.sql.catalyst.util._
 
-    val leftKeys = joinNames.map { keyName =>
-      left.output.find(attr => resolver(attr.name, keyName)).getOrElse {
-        throw QueryCompilationErrors.unresolvedUsingColForJoinError(
-          keyName, left.schema.fieldNames.sorted.map(toSQLId).mkString(", "), "left")
-      }
-    }
-    val rightKeys = joinNames.map { keyName =>
-      right.output.find(attr => resolver(attr.name, keyName)).getOrElse {
-        throw QueryCompilationErrors.unresolvedUsingColForJoinError(
-          keyName, right.schema.fieldNames.sorted.map(toSQLId).mkString(", "), "right")
-      }
-    }
-    val joinPairs = leftKeys.zip(rightKeys)
-
-    val newCondition = (condition ++ joinPairs.map(EqualTo.tupled)).reduceOption(And)
-
-    // columns not in joinPairs
-    val lUniqueOutput = left.output.filterNot(att => leftKeys.contains(att))
-    val rUniqueOutput = right.output.filterNot(att => rightKeys.contains(att))

Review Comment:
   I changed the return type of `computeKeysForNaturalOrUsingJoin` from `Seq` to `AttributeSet` to avoid quadratic lookups here. I think it makes sense to make this change in this PR, but it can be moved to a followup. Wdyt @cloud-fan @vladimirg-db ?



-- 
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