rdblue commented on a change in pull request #710: Parquet changes for vectorized reads URL: https://github.com/apache/incubator-iceberg/pull/710#discussion_r362050520
########## File path: parquet/src/main/java/org/apache/iceberg/parquet/ReadConf.java ########## @@ -0,0 +1,203 @@ +/* + * 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.parquet; + +import com.google.common.collect.ImmutableList; +import com.google.common.collect.ImmutableMap; +import java.io.IOException; +import java.util.List; +import java.util.Map; +import java.util.Set; +import java.util.function.Function; +import java.util.stream.Collectors; +import javax.annotation.Nullable; +import org.apache.iceberg.Schema; +import org.apache.iceberg.exceptions.RuntimeIOException; +import org.apache.iceberg.expressions.Expression; +import org.apache.iceberg.io.InputFile; +import org.apache.parquet.ParquetReadOptions; +import org.apache.parquet.hadoop.ParquetFileReader; +import org.apache.parquet.hadoop.metadata.BlockMetaData; +import org.apache.parquet.hadoop.metadata.ColumnChunkMetaData; +import org.apache.parquet.hadoop.metadata.ColumnPath; +import org.apache.parquet.schema.MessageType; + +/** + * Configuration for Parquet readers. + * + * @param <T> type of value to read + */ +class ReadConf<T> { + private final ParquetFileReader reader; + private final InputFile file; + private final ParquetReadOptions options; + private final MessageType projection; + @Nullable + private final ParquetValueReader<T> model; + @Nullable + private final VectorizedReader<T> vectorizedModel; + private final List<BlockMetaData> rowGroups; + private final boolean[] shouldSkip; + private final long totalValues; + private final boolean reuseContainers; + @Nullable + private final Integer batchSize; + + // List of column chunk metadata for each row group + private final List<Map<ColumnPath, ColumnChunkMetaData>> columnChunkMetaDataForRowGroups; + + @SuppressWarnings("unchecked") + ReadConf(InputFile file, ParquetReadOptions options, Schema expectedSchema, Expression filter, + Function<MessageType, ParquetValueReader<?>> readerFunc, Function<MessageType, + VectorizedReader<?>> batchedReaderFunc, boolean reuseContainers, + boolean caseSensitive, Integer bSize) { + this.file = file; + this.options = options; + this.reader = newReader(file, options); + MessageType fileSchema = reader.getFileMetaData().getSchema(); + + boolean hasIds = ParquetSchemaUtil.hasIds(fileSchema); + MessageType typeWithIds = hasIds ? fileSchema : ParquetSchemaUtil.addFallbackIds(fileSchema); + + this.projection = hasIds ? + ParquetSchemaUtil.pruneColumns(fileSchema, expectedSchema) : + ParquetSchemaUtil.pruneColumnsFallback(fileSchema, expectedSchema); + this.rowGroups = reader.getRowGroups(); + this.shouldSkip = new boolean[rowGroups.size()]; + + ParquetMetricsRowGroupFilter statsFilter = null; + ParquetDictionaryRowGroupFilter dictFilter = null; + if (filter != null) { + statsFilter = new ParquetMetricsRowGroupFilter(expectedSchema, filter, caseSensitive); + dictFilter = new ParquetDictionaryRowGroupFilter(expectedSchema, filter, caseSensitive); + } + + long computedTotalValues = 0L; + for (int i = 0; i < shouldSkip.length; i += 1) { + BlockMetaData rowGroup = rowGroups.get(i); + boolean shouldRead = filter == null || ( + statsFilter.shouldRead(typeWithIds, rowGroup) && + dictFilter.shouldRead(typeWithIds, rowGroup, reader.getDictionaryReader(rowGroup))); + this.shouldSkip[i] = !shouldRead; + if (shouldRead) { + computedTotalValues += rowGroup.getRowCount(); + } + } + + this.totalValues = computedTotalValues; + if (readerFunc != null) { + this.model = (ParquetValueReader<T>) readerFunc.apply(typeWithIds); + this.vectorizedModel = null; + this.columnChunkMetaDataForRowGroups = null; + } else { + this.model = null; + this.vectorizedModel = (VectorizedReader<T>) batchedReaderFunc.apply(typeWithIds); + this.columnChunkMetaDataForRowGroups = getColumnChunkMetadataForRowGroups(); + } + + this.reuseContainers = reuseContainers; + this.batchSize = bSize; + } + + private ReadConf(ReadConf<T> toCopy) { + this.reader = null; + this.file = toCopy.file; + this.options = toCopy.options; + this.projection = toCopy.projection; + this.model = toCopy.model; + this.rowGroups = toCopy.rowGroups; + this.shouldSkip = toCopy.shouldSkip; + this.totalValues = toCopy.totalValues; + this.reuseContainers = toCopy.reuseContainers; + this.batchSize = toCopy.batchSize; + this.vectorizedModel = toCopy.vectorizedModel; + this.columnChunkMetaDataForRowGroups = toCopy.columnChunkMetaDataForRowGroups; + } + + ParquetFileReader reader() { + if (reader != null) { + reader.setRequestedSchema(projection); + return reader; + } + + ParquetFileReader newReader = newReader(file, options); + newReader.setRequestedSchema(projection); + return newReader; + } + + ParquetValueReader<T> model() { + return model; + } + + VectorizedReader<T> vectorizedModel() { + return vectorizedModel; + } + + boolean[] shouldSkip() { + return shouldSkip; + } + + long totalValues() { + return totalValues; + } + + boolean reuseContainers() { + return reuseContainers; + } + + Integer batchSize() { Review comment: I just noticed this is an `Integer` instead of `int`. Can we change this to the primitive so that we don't have to worry about null pointer exceptions? Just set it to a default for non-batch cases, or use a precondition in this method to check whether the batch reader function is set. ---------------------------------------------------------------- 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 With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
