jonvex commented on code in PR #9624: URL: https://github.com/apache/hudi/pull/9624#discussion_r1346202791
########## hudi-client/hudi-spark-client/src/main/scala/org/apache/hudi/util/CloseableScalaIterator.scala: ########## Review Comment: this should probably implement [CachingIterator](https://github.com/apache/hudi/blob/master/hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/util/CachingIterator.scala) ########## hudi-common/src/main/java/org/apache/hudi/common/table/read/HoodieFileGroupReader.java: ########## @@ -0,0 +1,233 @@ +/* + * 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.hudi.common.table.read; + +import org.apache.hudi.common.config.HoodieMemoryConfig; +import org.apache.hudi.common.config.HoodieReaderConfig; +import org.apache.hudi.common.config.TypedProperties; +import org.apache.hudi.common.engine.HoodieReaderContext; +import org.apache.hudi.common.model.HoodieBaseFile; +import org.apache.hudi.common.model.HoodieRecord; +import org.apache.hudi.common.model.HoodieRecordMerger; +import org.apache.hudi.common.model.HoodieTableQueryType; +import org.apache.hudi.common.table.HoodieTableMetaClient; +import org.apache.hudi.common.table.log.HoodieMergedLogRecordReader; +import org.apache.hudi.common.table.timeline.HoodieTimeline; +import org.apache.hudi.common.util.Option; +import org.apache.hudi.common.util.collection.ClosableIterator; +import org.apache.hudi.common.util.collection.EmptyIterator; +import org.apache.hudi.common.util.collection.Pair; + +import org.apache.avro.Schema; +import org.apache.hadoop.conf.Configuration; +import org.apache.hadoop.fs.FileSystem; +import org.apache.hadoop.fs.Path; + +import java.io.Closeable; +import java.io.IOException; +import java.util.Collections; +import java.util.HashMap; +import java.util.Iterator; +import java.util.List; +import java.util.Map; + +import static org.apache.hudi.common.fs.FSUtils.getRelativePartitionPath; +import static org.apache.hudi.common.table.HoodieTableConfig.RECORD_MERGER_STRATEGY; +import static org.apache.hudi.common.util.ConfigUtils.getBooleanWithAltKeys; +import static org.apache.hudi.common.util.ConfigUtils.getIntWithAltKeys; +import static org.apache.hudi.common.util.ConfigUtils.getStringWithAltKeys; + +/** + * A file group reader that iterates through the records in a single file group. + * <p> + * This should be used by the every engine integration, by plugging in a + * {@link HoodieReaderContext<T>} implementation. + * + * @param <T> The type of engine-specific record representation, e.g.,{@code InternalRow} + * in Spark and {@code RowData} in Flink. + */ +public final class HoodieFileGroupReader<T> implements Closeable { + private final HoodieReaderContext<T> readerContext; + private final Option<HoodieBaseFile> baseFilePath; + private final Option<List<String>> logFilePathList; + private final Configuration hadoopConf; + private final TypedProperties props; + // Byte offset to start reading from the base file + private final long start; + // Length of bytes to read from the base file + private final long length; + // Key to record and metadata mapping from log files + private final Map<String, Pair<Option<T>, Map<String, Object>>> logFileRecordMapping = new HashMap<>(); + private final FileGroupReaderState readerState = new FileGroupReaderState(); + private ClosableIterator<T> baseFileIterator; + // This is only initialized and used after all records from the base file are iterated + private Iterator<Pair<Option<T>, Map<String, Object>>> logRecordIterator; + private HoodieRecordMerger recordMerger; + + T nextRecord; + + public HoodieFileGroupReader(HoodieReaderContext<T> readerContext, + HoodieTableMetaClient metaClient, + String fileGroupId, + TypedProperties props, + HoodieTimeline timeline, + HoodieTableQueryType queryType, + Option<String> instantTime, + Option<String> startInstantTime) { + // This constructor is a placeholder now to allow automatically fetching the correct list of + // base and log files for a file group. + // Derive base and log files and call the corresponding constructor. + this.readerContext = readerContext; + this.hadoopConf = metaClient.getHadoopConf(); + this.baseFilePath = Option.empty(); + this.logFilePathList = Option.empty(); + this.props = props; + this.start = 0; + this.length = Long.MAX_VALUE; + this.baseFileIterator = new EmptyIterator<>(); + } + + public HoodieFileGroupReader(HoodieReaderContext<T> readerContext, + Configuration hadoopConf, + String tablePath, + String latestCommitTime, + Option<HoodieBaseFile> baseFilePath, + Option<List<String>> logFilePathList, + Schema avroSchema, Review Comment: Do we want to restrict ourselves to avro schema? Can we make generic? ########## hudi-common/src/main/java/org/apache/hudi/common/table/read/FileGroupReaderState.java: ########## @@ -0,0 +1,36 @@ +/* + * 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.hudi.common.table.read; + +import org.apache.hudi.common.config.TypedProperties; + +import org.apache.avro.Schema; + +/** + * A class holding the state that is needed by {@code HoodieFileGroupReader}, + * e.g., schema, merging strategy, etc. + */ +public class FileGroupReaderState { + public String tablePath; + public String latestCommitTime; + public Schema baseFileAvroSchema; + public Schema logRecordAvroSchema; Review Comment: log records can have multiple schemas ########## hudi-common/src/main/java/org/apache/hudi/common/table/read/HoodieFileGroupReader.java: ########## @@ -0,0 +1,233 @@ +/* + * 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.hudi.common.table.read; + +import org.apache.hudi.common.config.HoodieMemoryConfig; +import org.apache.hudi.common.config.HoodieReaderConfig; +import org.apache.hudi.common.config.TypedProperties; +import org.apache.hudi.common.engine.HoodieReaderContext; +import org.apache.hudi.common.model.HoodieBaseFile; +import org.apache.hudi.common.model.HoodieRecord; +import org.apache.hudi.common.model.HoodieRecordMerger; +import org.apache.hudi.common.model.HoodieTableQueryType; +import org.apache.hudi.common.table.HoodieTableMetaClient; +import org.apache.hudi.common.table.log.HoodieMergedLogRecordReader; +import org.apache.hudi.common.table.timeline.HoodieTimeline; +import org.apache.hudi.common.util.Option; +import org.apache.hudi.common.util.collection.ClosableIterator; +import org.apache.hudi.common.util.collection.EmptyIterator; +import org.apache.hudi.common.util.collection.Pair; + +import org.apache.avro.Schema; +import org.apache.hadoop.conf.Configuration; +import org.apache.hadoop.fs.FileSystem; +import org.apache.hadoop.fs.Path; + +import java.io.Closeable; +import java.io.IOException; +import java.util.Collections; +import java.util.HashMap; +import java.util.Iterator; +import java.util.List; +import java.util.Map; + +import static org.apache.hudi.common.fs.FSUtils.getRelativePartitionPath; +import static org.apache.hudi.common.table.HoodieTableConfig.RECORD_MERGER_STRATEGY; +import static org.apache.hudi.common.util.ConfigUtils.getBooleanWithAltKeys; +import static org.apache.hudi.common.util.ConfigUtils.getIntWithAltKeys; +import static org.apache.hudi.common.util.ConfigUtils.getStringWithAltKeys; + +/** + * A file group reader that iterates through the records in a single file group. + * <p> + * This should be used by the every engine integration, by plugging in a + * {@link HoodieReaderContext<T>} implementation. + * + * @param <T> The type of engine-specific record representation, e.g.,{@code InternalRow} + * in Spark and {@code RowData} in Flink. + */ +public final class HoodieFileGroupReader<T> implements Closeable { + private final HoodieReaderContext<T> readerContext; + private final Option<HoodieBaseFile> baseFilePath; + private final Option<List<String>> logFilePathList; + private final Configuration hadoopConf; + private final TypedProperties props; + // Byte offset to start reading from the base file + private final long start; + // Length of bytes to read from the base file + private final long length; + // Key to record and metadata mapping from log files + private final Map<String, Pair<Option<T>, Map<String, Object>>> logFileRecordMapping = new HashMap<>(); + private final FileGroupReaderState readerState = new FileGroupReaderState(); + private ClosableIterator<T> baseFileIterator; + // This is only initialized and used after all records from the base file are iterated + private Iterator<Pair<Option<T>, Map<String, Object>>> logRecordIterator; + private HoodieRecordMerger recordMerger; + + T nextRecord; + + public HoodieFileGroupReader(HoodieReaderContext<T> readerContext, Review Comment: I don't see any opportunity for pushed down filters? Have we considered that for the design? -- 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]
