connector/read/PartitionReaderFactory.java

0001 /*
0002  * Licensed to the Apache Software Foundation (ASF) under one or more
0003  * contributor license agreements.  See the NOTICE file distributed with
0004  * this work for additional information regarding copyright ownership.
0005  * The ASF licenses this file to You under the Apache License, Version 2.0
0006  * (the "License"); you may not use this file except in compliance with
0007  * the License.  You may obtain a copy of the License at
0008  *
0009  *    http://www.apache.org/licenses/LICENSE-2.0
0010  *
0011  * Unless required by applicable law or agreed to in writing, software
0012  * distributed under the License is distributed on an "AS IS" BASIS,
0013  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0014  * See the License for the specific language governing permissions and
0015  * limitations under the License.
0016  */
0017
0018 package org.apache.spark.sql.connector.read;
0019
0020 import java.io.Serializable;
0021
0022 import org.apache.spark.annotation.Evolving;
0023 import org.apache.spark.sql.catalyst.InternalRow;
0024 import org.apache.spark.sql.vectorized.ColumnarBatch;
0025
0026 /**
0027  * A factory used to create {@link PartitionReader} instances.
0028  *
0029  * If Spark fails to execute any methods in the implementations of this interface or in the returned
0030  * {@link PartitionReader} (by throwing an exception), corresponding Spark task would fail and
0031  * get retried until hitting the maximum retry times.
0032  *
0033  * @since 3.0.0
0034  */
0035 @Evolving
0036 public interface PartitionReaderFactory extends Serializable {
0037
0038   /**
0039    * Returns a row-based partition reader to read data from the given {@link InputPartition}.
0040    *
0041    * Implementations probably need to cast the input partition to the concrete
0042    * {@link InputPartition} class defined for the data source.
0043    */
0044   PartitionReader<InternalRow> createReader(InputPartition partition);
0045
0046   /**
0047    * Returns a columnar partition reader to read data from the given {@link InputPartition}.
0048    *
0049    * Implementations probably need to cast the input partition to the concrete
0050    * {@link InputPartition} class defined for the data source.
0051    */
0052   default PartitionReader<ColumnarBatch> createColumnarReader(InputPartition partition) {
0053     throw new UnsupportedOperationException("Cannot create columnar reader.");
0054   }
0055
0056   /**
0057    * Returns true if the given {@link InputPartition} should be read by Spark in a columnar way.
0058    * This means, implementations must also implement {@link #createColumnarReader(InputPartition)}
0059    * for the input partitions that this method returns true.
0060    *
0061    * As of Spark 2.4, Spark can only read all input partition in a columnar way, or none of them.
0062    * Data source can't mix columnar and row-based partitions. This may be relaxed in future
0063    * versions.
0064    */
0065   default boolean supportColumnarReads(InputPartition partition) {
0066     return false;
0067   }
0068 }