srowen commented on code in PR #37185: URL: https://github.com/apache/spark/pull/37185#discussion_r922682035
########## sql/catalyst/src/main/scala-2.13/org/apache/spark/sql/catalyst/util/GenericArrayDataHelper.scala: ########## @@ -0,0 +1,43 @@ +/* + * 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.util + +import scala.collection.immutable +import scala.collection.mutable + +object GenericArrayDataHelper { + + // SPARK-39766: Special treatment of `immutable.ArraySeq[Any]` and `mutable.ArraySeq[Any]` + // to ensure the scenes similar to `arrayOfAnyAsSeq` in `GenericArrayDataBenchmark` have + // the same performance when using Scala 2.12 and Scala 2.13 + def toArray(seq: scala.collection.Seq[Any]): Array[Any] = seq match { + case ias: immutable.ArraySeq.ofRef[_] => ias.unsafeArray.asInstanceOf[Array[Any]] Review Comment: This is the risky thing, if we are not copying the array behind it. Is this necessary for perf? ########## sql/catalyst/src/main/scala-2.12/org/apache/spark/sql/catalyst/util/GenericArrayDataHelper.scala: ########## @@ -0,0 +1,28 @@ +/* + * 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.util + +object GenericArrayDataHelper { + def toArray(seq: Seq[Any]): Array[Any] = seq.toArray + + def toArray(seqOrArray: Any): Array[Any] = seqOrArray match { + case seq: Seq[Any] => seq.toArray + case array: Array[Any] => array // array of objects, so no need to convert + case array: Array[_] => array.toSeq.toArray[Any] // array of primitives, so box them Review Comment: Does this actually box? On Scala 2.13, which I have handy, this just fails when accessing the result. I guess this is how the current code works, so hey ########## sql/catalyst/src/main/scala-2.12/org/apache/spark/sql/catalyst/util/GenericArrayDataHelper.scala: ########## @@ -0,0 +1,28 @@ +/* + * 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.util + +object GenericArrayDataHelper { + def toArray(seq: Seq[Any]): Array[Any] = seq.toArray + + def toArray(seqOrArray: Any): Array[Any] = seqOrArray match { + case seq: Seq[Any] => seq.toArray + case array: Array[Any] => array // array of objects, so no need to convert Review Comment: Wait, do you mean AnyRef here? that's the supertype of objects as opposed to primitives. But while all object arrays have the same type, each primitive array is its own type. I think this happens to 'work' as a result -- an object array would match an array of anything, while a primitive array would match only Array[_]. There's probably a cleaner way to do this, but maybe not a faster one; maybe at least say "AnyRef" here. But I guess that's how the current code works -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
