Github user cloud-fan commented on a diff in the pull request:
https://github.com/apache/spark/pull/9145#discussion_r42584730
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/columnar/GenerateColumnAccessor.scala
---
@@ -0,0 +1,149 @@
+/*
+ * 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.columnar
+
+import org.apache.spark.Logging
+import org.apache.spark.sql.catalyst.InternalRow
+import org.apache.spark.sql.catalyst.expressions._
+import org.apache.spark.sql.catalyst.expressions.codegen.{CodeFormatter,
CodeGenerator}
+import org.apache.spark.sql.types._
+
+/**
+ * An Iterator to walk throught the InternalRows from a CachedBatch
+ */
+abstract class ColumnarIterator extends Iterator[InternalRow] {
+ def initialize(input: Iterator[CachedBatch], mutableRow: MutableRow,
columnTypes: Array[DataType],
+ columnIndexes: Array[Int]): Unit
+}
+
+/**
+ * Generates bytecode for an [[ColumnarIterator]] for columnar cache.
+ */
+object GenerateColumnAccessor extends CodeGenerator[Seq[DataType],
ColumnarIterator] with Logging {
+
+ protected def canonicalize(in: Seq[DataType]): Seq[DataType] = in
+ protected def bind(in: Seq[DataType], inputSchema: Seq[Attribute]):
Seq[DataType] = in
+
+ protected def create(columnTypes: Seq[DataType]): ColumnarIterator = {
+ val ctx = newCodeGenContext()
+ val (initializeAccessors, extractors) = columnTypes.zipWithIndex.map {
case (dt, index) =>
+ val accessorName = ctx.freshName("accessor")
+ val accessorCls = dt match {
+ case NullType => classOf[NullColumnAccessor].getName
+ case BooleanType => classOf[BooleanColumnAccessor].getName
+ case ByteType => classOf[ByteColumnAccessor].getName
+ case ShortType => classOf[ShortColumnAccessor].getName
+ case IntegerType | DateType => classOf[IntColumnAccessor].getName
+ case LongType | TimestampType =>
classOf[LongColumnAccessor].getName
+ case FloatType => classOf[FloatColumnAccessor].getName
+ case DoubleType => classOf[DoubleColumnAccessor].getName
+ case StringType => classOf[StringColumnAccessor].getName
+ case BinaryType => classOf[BinaryColumnAccessor].getName
+ case dt: DecimalType if dt.precision <= Decimal.MAX_LONG_DIGITS =>
+ classOf[CompactDecimalColumnAccessor].getName
+ case dt: DecimalType => classOf[DecimalColumnAccessor].getName
+ case struct: StructType => classOf[StructColumnAccessor].getName
+ case array: ArrayType => classOf[ArrayColumnAccessor].getName
+ case t: MapType => classOf[MapColumnAccessor].getName
+ }
+ ctx.addMutableState(accessorCls, accessorName, s"$accessorName =
null;")
+
+ val createCode = dt match {
+ case t if ctx.isPrimitiveType(dt) =>
+ s"$accessorName = new
$accessorCls(ByteBuffer.wrap(buffers[$index]).order(nativeOrder));"
+ case NullType | StringType | BinaryType =>
+ s"$accessorName = new
$accessorCls(ByteBuffer.wrap(buffers[$index]).order(nativeOrder));"
+ case other =>
+ s"""$accessorName = new
$accessorCls(ByteBuffer.wrap(buffers[$index]).order(nativeOrder),
+ (${dt.getClass.getName}) columnTypes[$index]);"""
+ }
+
+ val extract = s"$accessorName.extractTo(mutableRow, $index);"
+
+ (createCode, extract)
+ }.unzip
+
+ val code = s"""
+ import java.nio.ByteBuffer;
+ import java.nio.ByteOrder;
+
+ public SpecificColumnarIterator generate($exprType[] expr) {
+ return new SpecificColumnarIterator();
+ }
+
+ class SpecificColumnarIterator extends
${classOf[ColumnarIterator].getName} {
+
+ private ByteOrder nativeOrder = null;
+ private byte[][] buffers = null;
+
+ private int currentRow = 0;
+ private int numRowsInBatch = 0;
+
+ private scala.collection.Iterator input = null;
+ private MutableRow mutableRow = null;
+ private ${classOf[DataType].getName}[] columnTypes = null;
+ private int[] columnIndexes = null;
+
+ ${declareMutableStates(ctx)}
+
+ public SpecificColumnarIterator() {
+ this.nativeOrder = ByteOrder.nativeOrder();
+ this.buffers = new byte[${columnTypes.length}][];
+
+ ${initMutableStates(ctx)}
+ }
+
+ public void initialize(scala.collection.Iterator input, MutableRow
mutableRow,
+ ${classOf[DataType].getName}[] columnTypes,
int[] columnIndexes) {
--- End diff --
wrong indent here
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
