brimzi commented on a change in pull request #775: URL: https://github.com/apache/parquet-mr/pull/775#discussion_r411692398
########## File path: parquet-hadoop/src/main/java/org/apache/parquet/hadoop/RowGroupMerger.java ########## @@ -0,0 +1,634 @@ +/* + * 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.parquet.hadoop; + +import static java.lang.String.format; +import static org.apache.parquet.column.ValuesType.DEFINITION_LEVEL; +import static org.apache.parquet.column.ValuesType.REPETITION_LEVEL; +import static org.apache.parquet.column.ValuesType.VALUES; +import static org.apache.parquet.hadoop.ParquetFileWriter.getColumnsInOrder; + +import java.io.IOException; +import java.io.UncheckedIOException; +import java.nio.ByteBuffer; +import java.util.ArrayList; +import java.util.Collections; +import java.util.HashMap; +import java.util.List; +import java.util.Map; +import java.util.Map.Entry; +import java.util.function.BiConsumer; +import java.util.function.Consumer; +import java.util.stream.Collectors; + +import org.apache.hadoop.conf.Configuration; +import org.apache.parquet.Preconditions; +import org.apache.parquet.bytes.ByteBufferInputStream; +import org.apache.parquet.bytes.BytesInput; +import org.apache.parquet.column.ColumnDescriptor; +import org.apache.parquet.column.Dictionary; +import org.apache.parquet.column.Encoding; +import org.apache.parquet.column.ParquetProperties; +import org.apache.parquet.column.page.DataPage; +import org.apache.parquet.column.page.DataPageV1; +import org.apache.parquet.column.page.DataPageV2; +import org.apache.parquet.column.page.DictionaryPage; +import org.apache.parquet.column.page.PageReadStore; +import org.apache.parquet.column.page.PageReader; +import org.apache.parquet.column.values.ValuesReader; +import org.apache.parquet.column.values.ValuesWriter; +import org.apache.parquet.compression.CompressionCodecFactory; +import org.apache.parquet.hadoop.metadata.BlockMetaData; +import org.apache.parquet.hadoop.metadata.ColumnChunkMetaData; +import org.apache.parquet.hadoop.metadata.CompressionCodecName; +import org.apache.parquet.io.InputFile; +import org.apache.parquet.io.ParquetDecodingException; +import org.apache.parquet.schema.MessageType; +import org.apache.parquet.schema.PrimitiveType; + +class RowGroupMerger { + + private final MessageType schema; + private final CodecFactory.BytesInputCompressor compressor; + private final ParquetProperties parquetProperties; + + RowGroupMerger(MessageType schema, CompressionCodecName compression, boolean useV2ValueWriter) { + this(schema, new Configuration(), compression, useV2ValueWriter); + } + + RowGroupMerger(MessageType schema, Configuration conf, CompressionCodecName compression, boolean useV2ValueWriter) { + this(schema, conf, compression, createParquetProperties(useV2ValueWriter)); + } + + RowGroupMerger(MessageType schema, Configuration conf, CompressionCodecName compression, ParquetProperties parquetProperties) { + this.schema = schema; + this.parquetProperties = parquetProperties; + this.compressor = new CodecFactory(conf, this.parquetProperties.getPageSizeThreshold()).getCompressor(compression); + } + + /** + * Merges the row groups making sure that new row groups do not exceed the supplied maxRowGroupSize + * + * @param inputFiles input files to merge + * @param maxRowGroupSize the max limit for new blocks + * @param writer writer to write the new blocks to + * @throws IOException if an IO error occurs + */ + void merge(List<InputFile> inputFiles, final long maxRowGroupSize, ParquetFileWriter writer) throws IOException { + + SizeEstimator estimator = new SizeEstimator(compressor.getCodecName() != CompressionCodecName.UNCOMPRESSED); + MutableMergedBlock mergedBlock = null; + for (InputFile file : inputFiles) { + try (ParquetFileReader reader = ParquetFileReader.open(file)) { + + for (BlockMetaData blockMeta : reader.getRowGroups()) { + PageReadStore group = reader.readNextRowGroup(); + Preconditions.checkState(group != null, + "number of groups returned by FileReader does not match metadata"); + + if (mergedBlock != null && mergedBlock.getCompressedBytesSize() + estimator.estimate(blockMeta) > maxRowGroupSize) { + saveBlockTo(mergedBlock, writer); + mergedBlock = null; + } + + if (mergedBlock == null && estimator.estimate(blockMeta) > maxRowGroupSize) { + //save it directly without re encoding it + saveBlockTo(ReadOnlyMergedBlock.of(blockMeta, group, schema, compressor), writer); + continue; + } + + if (mergedBlock == null) { + mergedBlock = new MutableMergedBlock(schema); + } + + long sizeBeforeMerge = mergedBlock.getCompressedBytesSize(); + mergedBlock.merge(blockMeta, group); + //update our estimator + long currentBlockEffect = mergedBlock.getCompressedBytesSize() - sizeBeforeMerge; + estimator.update(currentBlockEffect, blockMeta); + } + } + } + if (mergedBlock != null) + saveBlockTo(mergedBlock, writer); + } + + + private static void saveBlockTo(MergedBlock block, ParquetFileWriter writer) { + try { + writer.startBlock(block.rowCount()); + + for (MergedColumn col : block.columnsInOrder()) { + writer.startColumn(col.getColumnDesc(), col.getValueCount(), col.getCompression()); + + col.writeDictionaryPageTo(writer); + col.writeDataPagesTo(writer); + + writer.endColumn(); + } + + writer.endBlock(); + } catch (IOException e) { Review comment: Good catch, This somehow was carried over after refactoring. I have removed it ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
