maropu commented on a change in pull request #25717: [WIP][SPARK-29013][SQL] 
Structurally equivalent subexpression elimination
URL: https://github.com/apache/spark/pull/25717#discussion_r321999396
 
 

 ##########
 File path: 
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/EquivalentExpressions.scala
 ##########
 @@ -65,35 +89,102 @@ class EquivalentExpressions {
   }
 
   /**
-   * Adds the expression to this data structure recursively. Stops if a 
matching expression
-   * is found. That is, if `expr` has already been added, its children are not 
added.
+   * Adds each expression to structural expression data structure, grouping 
them with existing
+   * structurally equivalent expressions. Non-recursive.
+   */
+  def addStructExpr(ctx: CodegenContext, expr: Expression): Unit = {
+    if (expr.deterministic) {
+      val refs = expr.collect {
+        case b: BoundReference => b
+      }
+
+      // For structural equivalent expressions, we need to pass in int type 
ordinals into
+      // split functions. If the number of ordinals is more than JVM function 
limit, we skip
+      // this expression.
+      // We calculate function parameter length by the number of ints plus 
`INPUT_ROW` plus
+      // a int type result array index.
+      val parameterLength = CodeGenerator.calculateParamLength(refs.map(_ => 
Literal(0))) + 2
+      if (CodeGenerator.isValidParamLength(parameterLength)) {
 
 Review comment:
   If the length goes over the limit, the current logic gives up eliminating 
common exprs? If so, can we fall back into the non-structural mode?

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


With regards,
Apache Git Services

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

Reply via email to