flyrain commented on a change in pull request #3516: URL: https://github.com/apache/iceberg/pull/3516#discussion_r746952039
########## File path: spark/v3.2/spark/src/main/java/org/apache/iceberg/spark/source/SparkBatch.java ########## @@ -0,0 +1,118 @@ +/* + * 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.iceberg.spark.source; + +import java.util.List; +import org.apache.iceberg.CombinedScanTask; +import org.apache.iceberg.FileFormat; +import org.apache.iceberg.Schema; +import org.apache.iceberg.SchemaParser; +import org.apache.iceberg.SerializableTable; +import org.apache.iceberg.Table; +import org.apache.iceberg.spark.SparkReadConf; +import org.apache.iceberg.spark.source.SparkBatchScan.ReadTask; +import org.apache.iceberg.spark.source.SparkBatchScan.ReaderFactory; +import org.apache.iceberg.util.TableScanUtil; +import org.apache.iceberg.util.Tasks; +import org.apache.iceberg.util.ThreadPools; +import org.apache.spark.api.java.JavaSparkContext; +import org.apache.spark.broadcast.Broadcast; +import org.apache.spark.sql.connector.read.Batch; +import org.apache.spark.sql.connector.read.InputPartition; +import org.apache.spark.sql.connector.read.PartitionReaderFactory; + +class SparkBatch implements Batch { + + private final JavaSparkContext sparkContext; + private final Table table; + private final SparkReadConf readConf; + private final List<CombinedScanTask> tasks; + private final Schema expectedSchema; + private final boolean caseSensitive; + private final boolean localityEnabled; + + SparkBatch(JavaSparkContext sparkContext, Table table, SparkReadConf readConf, + List<CombinedScanTask> tasks, Schema expectedSchema) { + this.sparkContext = sparkContext; + this.table = table; + this.readConf = readConf; + this.tasks = tasks; + this.expectedSchema = expectedSchema; + this.caseSensitive = readConf.caseSensitive(); + this.localityEnabled = readConf.localityEnabled(); + } + + @Override + public InputPartition[] planInputPartitions() { + // broadcast the table metadata as input partitions will be sent to executors + Broadcast<Table> tableBroadcast = sparkContext.broadcast(SerializableTable.copyOf(table)); + String expectedSchemaString = SchemaParser.toJson(expectedSchema); + + InputPartition[] readTasks = new InputPartition[tasks.size()]; + + Tasks.range(readTasks.length) + .stopOnFailure() + .executeWith(localityEnabled ? ThreadPools.getWorkerPool() : null) + .run(index -> readTasks[index] = new ReadTask( + tasks.get(index), tableBroadcast, expectedSchemaString, + caseSensitive, localityEnabled)); + + return readTasks; + } + + @Override + public PartitionReaderFactory createReaderFactory() { + return new ReaderFactory(batchSize()); + } + + private int batchSize() { + if (parquetOnly() && parquetBatchReadsEnabled()) { Review comment: I somehow prefer to consider `parquetOnly()` as a part of method `parquetBatchReadsEnabled`. The same to `orcOnly`. So the method is like ``` if(parquetBatchReadsEnabled()){return parquetBatchSize} else if(orcBatchReadsEnabled()) {return orcBatchsize} ``` But it's just my preference, I'm OK with the current one as well. -- 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]
