rdblue commented on code in PR #12774: URL: https://github.com/apache/iceberg/pull/12774#discussion_r2162696513
########## core/src/main/java/org/apache/iceberg/io/FileAccessFactory.java: ########## @@ -0,0 +1,139 @@ +/* + * 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.io; + +import org.apache.iceberg.FileFormat; +import org.apache.iceberg.data.Record; +import org.apache.iceberg.deletes.PositionDelete; + +/** + * Interface that provides a unified abstraction for converting between data file formats and + * input/output data representations. + * + * <p>FileAccessFactory serves as a bridge between storage formats ({@link FileFormat}) and expected + * input/output data structures, optimizing performance through direct conversion without + * intermediate representations. File format implementations handle the low-level parsing details + * while the object model determines the in-memory representation used for the parsed data. + * Together, these provide a consistent API for consuming data files while optimizing for specific + * processing engines. + * + * <p>Iceberg provides these built-in object models: + * + * <ul> + * <li><strong>generic</strong> - for {@link Record} objects (engine-agnostic) + * <li><strong>spark</strong> - for Apache Spark InternalRow objects + * <li><strong>spark-vectorized</strong> - for columnar batch processing in Spark (not supported + * with {@link FileFormat#AVRO}) + * <li><strong>flink</strong> - for Apache Flink RowData objects + * <li><strong>arrow</strong> - for Apache Arrow columnar format (only supported with {@link + * FileFormat#PARQUET}) + * </ul> + * + * <p>Processing engines can implement custom object models to integrate with Iceberg's file reading + * and writing capabilities. + * + * @param <E> input schema type used when converting input data to the file format + * @param <D> output type used for reading data, and input type for writing data and deletes + */ +public interface FileAccessFactory<E, D> { + /** The file format which is read/written by the object model. */ + FileFormat format(); + + /** + * Returns the unique identifier for the object model implementation processed by this factory. + * + * <p>The object model names (such as "generic", "spark", "spark-vectorized", "flink", "arrow") + * act as a contract specifying the expected data structures for both reading (converting file + * formats into output objects) and writing (converting input objects into file formats). This + * ensures proper integration between Iceberg's storage layer and processing engines. + * + * @return string identifier for this object model implementation + */ + String objectModelName(); + + /** + * Creates a writer builder for standard data files. + * + * <p>The returned {@link WriteBuilder} configures and creates a writer that converts input + * objects into the file format supported by this factory for regular data content. + * + * <p>The builder follows the fluent pattern for configuring writer properties like compression, + * encryption, row group size, and other format-specific options. + * + * @param outputFile destination for the written data + * @return configured writer builder for standard data files + * @param <B> the concrete builder type for method chaining + */ + <B extends WriteBuilder<B, E, D>> B dataWriteBuilder(OutputFile outputFile); Review Comment: The existing builders are created using `writeData` and `writeDeletes`. Why are the names longer here? -- 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: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
