[GitHub] [spark] jackylee-ch commented on a change in pull request #35789: [SPARK-32268][SQL] Row-level Runtime Filtering

[GitBox <gi...@apache.org>]

jackylee-ch commented on a change in pull request #35789:
URL: https://github.com/apache/spark/pull/35789#discussion_r827559770



##########
File path: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/BloomFilterMightContain.scala
##########
@@ -0,0 +1,116 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.catalyst.expressions
+
+import java.io.ByteArrayInputStream
+
+import org.apache.spark.sql.catalyst.InternalRow
+import org.apache.spark.sql.catalyst.analysis.TypeCheckResult
+import org.apache.spark.sql.catalyst.expressions.codegen.{CodegenContext, CodeGenerator, ExprCode, JavaCode, TrueLiteral}
+import org.apache.spark.sql.catalyst.expressions.codegen.Block.BlockHelper
+import org.apache.spark.sql.catalyst.trees.TreePattern.OUTER_REFERENCE
+import org.apache.spark.sql.types._
+import org.apache.spark.util.sketch.BloomFilter
+
+/**
+ * An internal scalar function that returns the membership check result (either true or false)
+ * for values of `valueExpression` in the Bloom filter represented by `bloomFilterExpression`.
+ * Not that since the function is "might contain", always returning true regardless is not
+ * wrong.
+ * Note that this expression requires that `bloomFilterExpression` is either a constant value or
+ * an uncorrelated scalar subquery. This is sufficient for the Bloom filter join rewrite.
+ *
+ * @param bloomFilterExpression the Binary data of Bloom filter.
+ * @param valueExpression the Long value to be tested for the membership of `bloomFilterExpression`.
+ */
+case class BloomFilterMightContain(
+    bloomFilterExpression: Expression,
+    valueExpression: Expression) extends BinaryExpression {
+
+  override def nullable: Boolean = true
+  override def left: Expression = bloomFilterExpression
+  override def right: Expression = valueExpression
+  override def prettyName: String = "might_contain"
+  override def dataType: DataType = BooleanType
+
+  override def checkInputDataTypes(): TypeCheckResult = {
+    val typeCheckResult = (left.dataType, right.dataType) match {
+      case (BinaryType, NullType) | (NullType, LongType) | (NullType, NullType) |
+           (BinaryType, LongType) => TypeCheckResult.TypeCheckSuccess
+      case _ => TypeCheckResult.TypeCheckFailure(s"Input to function $prettyName should have " +
+        s"been ${BinaryType.simpleString} followed by a value with ${LongType.simpleString}, " +
+        s"but it's [${left.dataType.catalogString}, ${right.dataType.catalogString}].")
+    }
+    if (typeCheckResult.isFailure) {

Review comment:
       nit: the definition of typecheck is not necessary, we can return directly above. Futher more, `bloomFilterExpression ` check can also moved to upper place.
   ```
   case (BinaryType, NullType) | (NullType, LongType) | (NullType, NullType) |
              (BinaryType, LongType) =>
       bloomFilterExpression match ...
   case _ =>
       TypeCheckResult.TypeCheckFailure(...)
   ```

##########
File path: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/aggregate/BloomFilterAggregate.scala
##########
@@ -0,0 +1,197 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.catalyst.expressions.aggregate
+
+import java.io.ByteArrayInputStream
+import java.io.ByteArrayOutputStream
+
+import org.apache.spark.sql.catalyst.InternalRow
+import org.apache.spark.sql.catalyst.analysis.TypeCheckResult
+import org.apache.spark.sql.catalyst.analysis.TypeCheckResult._
+import org.apache.spark.sql.catalyst.expressions._
+import org.apache.spark.sql.catalyst.trees.TernaryLike
+import org.apache.spark.sql.types._
+import org.apache.spark.util.sketch.BloomFilter
+
+/**
+ * An internal aggregate function that creates a Bloom filter from input values.
+ *
+ * @param child                     Child expression of Long values for creating a Bloom filter.
+ * @param estimatedNumItemsExpression The number of estimated distinct items (optional).
+ * @param numBitsExpression         The number of bits to use (optional).
+ */
+case class BloomFilterAggregate(
+    child: Expression,
+    estimatedNumItemsExpression: Expression,
+    numBitsExpression: Expression,
+    override val mutableAggBufferOffset: Int,
+    override val inputAggBufferOffset: Int)
+  extends TypedImperativeAggregate[BloomFilter] with TernaryLike[Expression] {
+
+  def this(child: Expression, estimatedNumItemsExpression: Expression,
+      numBitsExpression: Expression) = {
+    this(child, estimatedNumItemsExpression, numBitsExpression, 0, 0)
+  }
+
+  def this(child: Expression, estimatedNumItemsExpression: Expression) = {
+    this(child, estimatedNumItemsExpression,
+      // 1 byte per item.
+      Multiply(estimatedNumItemsExpression, Literal(8L)))
+  }
+
+  def this(child: Expression) = {
+    this(child, Literal(BloomFilterAggregate.DEFAULT_EXPECTED_NUM_ITEMS),
+      Literal(BloomFilterAggregate.DEFAULT_NUM_BITS))
+  }
+
+  override def checkInputDataTypes(): TypeCheckResult = {
+    val typeCheckResult = (first.dataType, second.dataType, third.dataType) match {
+      case (_, NullType, _) | (_, _, NullType) =>
+        TypeCheckResult.TypeCheckFailure("Null typed values cannot be used as size arguments")
+      case (LongType, LongType, LongType) => TypeCheckResult.TypeCheckSuccess
+      case _ => TypeCheckResult.TypeCheckFailure(s"Input to function $prettyName should have " +
+        s"been a ${LongType.simpleString} value followed with two ${LongType.simpleString} size " +
+        s"arguments, but it's [${first.dataType.catalogString}, " +
+        s"${second.dataType.catalogString}, ${third.dataType.catalogString}]")
+    }
+    if (typeCheckResult.isFailure) {
+      return typeCheckResult

Review comment:
       nit: same as above.




-- 
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