import org.apache.spark.sql.catalyst.dsl.expressions._
import org.apache.spark.sql.catalyst.dsl.plans._
// Using hacks to disable two Catalyst DSL implicits
implicit def symbolToColumn(ack: ThatWasABadIdea) = ack
implicit class StringToColumn(val sc: StringContext) {}
import org.apache.spark.sql.catalyst.plans.logical.LocalRelation
val t1 = LocalRelation('a.int, 'b.int)
val t2 = LocalRelation('C.int, 'D.int)
val plan = t1.except(t2, isAll = false)
import org.apache.spark.sql.catalyst.optimizer.ReplaceExceptWithAntiJoin
val optimizedPlan = ReplaceExceptWithAntiJoin(plan)
scala> println(optimizedPlan.numberedTreeString)
00 Distinct
01 +- Join LeftAnti, ((a#14 <=> C#18) && (b#15 <=> D#19))
02 :- LocalRelation <empty>, [a#14, b#15]
03 +- LocalRelation <empty>, [C#18, D#19]ReplaceExceptWithAntiJoin Logical Optimization Rule — Rewriting Except (DISTINCT) Operators
ReplaceExceptWithAntiJoin is a Catalyst rule for transforming logical plans (i.e. Rule[LogicalPlan]).
When executed, ReplaceExceptWithAntiJoin transforms an Except (distinct) logical operator to a Distinct unary logical operator with a left-anti Join operator. The output columns of the left and right child logical operators of the Except operator are used to build a logical AND join condition of EqualNullSafe expressions.
ReplaceExceptWithAntiJoin requires that the number of columns of the left- and right-side of the Except operator are the same or throws an AssertionError.
ReplaceExceptWithAntiJoin is a part of the Replace Operators fixed-point rule batch of the base Catalyst Optimizer.
Demo: ReplaceExceptWithAntiJoin