danny0405 commented on code in PR #9624: URL: https://github.com/apache/hudi/pull/9624#discussion_r1340788640
########## 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; + 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; + // 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(); + + 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, + TypedProperties props, + long start, + long length) { + this.readerContext = readerContext; + this.hadoopConf = hadoopConf; + this.baseFilePath = baseFilePath; + this.logFilePathList = logFilePathList; + this.props = props; + this.start = start; + this.length = length; + this.recordMerger = readerContext.getRecordMerger( + getStringWithAltKeys(props, RECORD_MERGER_STRATEGY, RECORD_MERGER_STRATEGY.defaultValue())); + this.readerState.tablePath = tablePath; + this.readerState.latestCommitTime = latestCommitTime; + this.readerState.baseFileAvroSchema = avroSchema; + this.readerState.logRecordAvroSchema = avroSchema; + this.readerState.mergeProps.putAll(props); + } + + /** + * Initialize internal iterators on the base and log files. + */ + public void initRecordIterators() { + this.baseFileIterator = baseFilePath.isPresent() + ? readerContext.getFileRecordIterator( + baseFilePath.get().getHadoopPath(), 0, baseFilePath.get().getFileLen(), + readerState.baseFileAvroSchema, readerState.baseFileAvroSchema, hadoopConf) + : new EmptyIterator<>(); + scanLogFiles(); + } + + /** + * @return {@code true} if the next record exists; {@code false} otherwise. + * @throws IOException on reader error. + */ + public boolean hasNext() throws IOException { + while (baseFileIterator.hasNext()) { + T baseRecord = baseFileIterator.next(); + String recordKey = readerContext.getRecordKey(baseRecord, readerState.baseFileAvroSchema); + Pair<Option<T>, Map<String, Object>> logRecordInfo = logFileRecordMapping.remove(recordKey); + Option<T> resultRecord = logRecordInfo != null + ? merge(Option.of(baseRecord), Collections.emptyMap(), logRecordInfo.getLeft(), logRecordInfo.getRight()) + : merge(Option.empty(), Collections.emptyMap(), Option.of(baseRecord), Collections.emptyMap()); + if (resultRecord.isPresent()) { + nextRecord = readerContext.seal(resultRecord.get()); + return true; + } + } + + if (logRecordIterator == null) { + logRecordIterator = logFileRecordMapping.values().iterator(); + } + + while (logRecordIterator.hasNext()) { + Pair<Option<T>, Map<String, Object>> nextRecordInfo = logRecordIterator.next(); + Option<T> resultRecord = merge(Option.empty(), Collections.emptyMap(), + nextRecordInfo.getLeft(), nextRecordInfo.getRight()); + if (resultRecord.isPresent()) { + nextRecord = readerContext.seal(resultRecord.get()); + return true; + } + } + + return false; + } + + /** + * @return The next record after calling {@link #hasNext}. + */ + public T next() { + T result = nextRecord; + nextRecord = null; + return result; + } + + private void scanLogFiles() { + if (logFilePathList.isPresent()) { + FileSystem fs = readerContext.getFs(logFilePathList.get().get(0), hadoopConf); + HoodieMergedLogRecordReader<T> logRecordReader = HoodieMergedLogRecordReader.newBuilder() + .withHoodieReaderContext(readerContext) + .withFileSystem(fs) + .withBasePath(readerState.tablePath) + .withLogFilePaths(logFilePathList.get()) + .withLatestInstantTime(readerState.latestCommitTime) + .withReaderSchema(readerState.logRecordAvroSchema) + .withReadBlocksLazily(getBooleanWithAltKeys(props, HoodieReaderConfig.COMPACTION_LAZY_BLOCK_READ_ENABLE)) + .withReverseReader(false) + .withBufferSize(getIntWithAltKeys(props, HoodieMemoryConfig.MAX_DFS_STREAM_BUFFER_SIZE)) + .withPartition(getRelativePartitionPath( Review Comment: Do we handle the `_hoodie_operation` field correctly? For example, when the `_hoodie_operation` is `D`, 1. the record needs to be ignored for snapshot queries. 2. for streaming queries, we should emit it out for downstream operators to do retractions. -- 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]
