diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ScalaUdf.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ScalaUdf.scala index 63ac2a608b6ff808291e60601c6f819a4929a94d..0b3c1df453fb867aa82e574f04ee575802762c3d 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ScalaUdf.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ScalaUdf.scala @@ -18,10 +18,14 @@ package org.apache.spark.sql.catalyst.expressions import org.apache.spark.sql.catalyst.types.DataType +import org.apache.spark.util.ClosureCleaner case class ScalaUdf(function: AnyRef, dataType: DataType, children: Seq[Expression]) extends Expression { + // Clean function when not called with default no-arg constructor. + if (function != null) { ClosureCleaner.clean(function) } + type EvaluatedType = Any def nullable = true diff --git a/sql/core/src/main/scala/org/apache/spark/sql/parquet/ParquetFilters.scala b/sql/core/src/main/scala/org/apache/spark/sql/parquet/ParquetFilters.scala index 2298a9b933df52e91763284e0bddd53a957ac82d..fe28e0d7269e0bfbc93524027ccbfffdfe076d53 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/parquet/ParquetFilters.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/parquet/ParquetFilters.scala @@ -17,6 +17,8 @@ package org.apache.spark.sql.parquet +import java.nio.ByteBuffer + import org.apache.hadoop.conf.Configuration import parquet.filter._ @@ -25,6 +27,7 @@ import parquet.column.ColumnReader import com.google.common.io.BaseEncoding +import org.apache.spark.SparkEnv import org.apache.spark.sql.catalyst.types._ import org.apache.spark.sql.catalyst.expressions.{Predicate => CatalystPredicate} import org.apache.spark.sql.catalyst.expressions._ @@ -237,7 +240,8 @@ object ParquetFilters { */ def serializeFilterExpressions(filters: Seq[Expression], conf: Configuration): Unit = { if (filters.length > 0) { - val serialized: Array[Byte] = SparkSqlSerializer.serialize(filters) + val serialized: Array[Byte] = + SparkEnv.get.closureSerializer.newInstance().serialize(filters).array() val encoded: String = BaseEncoding.base64().encode(serialized) conf.set(PARQUET_FILTER_DATA, encoded) } @@ -252,7 +256,7 @@ object ParquetFilters { val data = conf.get(PARQUET_FILTER_DATA) if (data != null) { val decoded: Array[Byte] = BaseEncoding.base64().decode(data) - SparkSqlSerializer.deserialize(decoded) + SparkEnv.get.closureSerializer.newInstance().deserialize(ByteBuffer.wrap(decoded)) } else { Seq() }