rdblue commented on code in PR #15006: URL: https://github.com/apache/iceberg/pull/15006#discussion_r2850191395
########## core/src/main/java/org/apache/iceberg/DVUtil.java: ########## @@ -0,0 +1,184 @@ +/* + * 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; + +import java.io.IOException; +import java.io.UncheckedIOException; +import java.util.Comparator; +import java.util.List; +import java.util.Map; +import java.util.Objects; +import java.util.concurrent.ExecutorService; +import java.util.function.Supplier; +import java.util.stream.Collectors; +import org.apache.iceberg.deletes.BaseDVFileWriter; +import org.apache.iceberg.deletes.DVFileWriter; +import org.apache.iceberg.deletes.Deletes; +import org.apache.iceberg.deletes.PositionDeleteIndex; +import org.apache.iceberg.io.FileIO; +import org.apache.iceberg.io.OutputFile; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.apache.iceberg.relocated.com.google.common.collect.Maps; +import org.apache.iceberg.types.Comparators; +import org.apache.iceberg.util.Tasks; + +class DVUtil { + private DVUtil() {} + + /** + * Merges duplicate DVs for the same data file and writes the merged DV Puffin files. If there is + * exactly 1 DV for a given data file then it is return as is + * + * @param dvsByFile map of data file location to DVs + * @return a list containing both any newly merged DVs and any DVs that are already valid + */ + static List<DeleteFile> mergeAndWriteDVsIfRequired( + Map<String, List<DeleteFile>> dvsByFile, + Supplier<OutputFile> dvOutputFile, + FileIO fileIO, + Map<Integer, PartitionSpec> specs, + ExecutorService pool) { + Map<String, List<DeleteFile>> duplicateDVsByFile = + dvsByFile.entrySet().stream() + .filter(e -> e.getValue().size() > 1) + .collect(Collectors.toMap(Map.Entry::getKey, Map.Entry::getValue)); + + if (duplicateDVsByFile.isEmpty()) { + return dvsByFile.values().stream().flatMap(List::stream).collect(Collectors.toList()); + } + + validateCanMerge(duplicateDVsByFile, specs); + List<DeleteFile> duplicateDVs = + duplicateDVsByFile.values().stream().flatMap(List::stream).collect(Collectors.toList()); + Map<String, PositionDeleteIndex> mergedPositionsForFile = + readAndMergeDVs(duplicateDVs, fileIO, pool); + + List<DeleteFile> mergedDVs = Lists.newArrayList(); + Map<String, PartitionSpec> specByFile = Maps.newHashMap(); + Map<String, StructLike> partitionByFile = Maps.newHashMap(); + for (List<DeleteFile> dvs : dvsByFile.values()) { + if (!dvs.isEmpty()) { + DeleteFile firstDV = dvs.get(0); + if (dvs.size() == 1) { + mergedDVs.add(firstDV); + } else { + specByFile.put(firstDV.referencedDataFile(), specs.get(firstDV.specId())); + partitionByFile.put(firstDV.referencedDataFile(), firstDV.partition()); + } + } + } + + mergedDVs.addAll(writeDVs(mergedPositionsForFile, specByFile, partitionByFile, dvOutputFile)); + return mergedDVs; + } + + private static void validateCanMerge( Review Comment: I think this can be simpler if you accumulate partition map in the caller while looping over its inputs just once. Here's the version that does that and reuses partition data from the caller: ```java private static void validateCanMerge( Multimap<String, DeleteFile> duplicates, Map<String, Pair<PartitionSpec, StructLike>> partitions, Map<String, Long> sequenceNumbers) { Map<Integer, Comparator<StructLike>> comparatorsBySpecId = Maps.newHashMap(); for (Map.Entry<String, Collection<DeleteFile>> entry : duplicates.asMap().entrySet()) { String referencedFile = entry.getKey(); // validate that each file matches the expected partition Pair<PartitionSpec, StructLike> partition = partitions.get(referencedFile); long sequenceNumber = sequenceNumbers.get(referencedFile); PartitionSpec spec = partition.first(); StructLike tuple = partition.second(); Comparator<StructLike> comparator = comparatorsBySpecId.computeIfAbsent( spec.specId(), id -> Comparators.forType(spec.partitionType())); for (DeleteFile dv : entry.getValue()) { Preconditions.checkArgument( sequenceNumber == dv.dataSequenceNumber(), "Cannot merge DVs, mismatched sequence numbers (%s, %s) for %s", sequenceNumber, dv.dataSequenceNumber(), referencedFile); Preconditions.checkArgument( spec.specId() == dv.specId(), "Cannot merge DVs, mismatched partition specs (%s, %s) for %s", spec.specId(), dv.specId(), referencedFile); Preconditions.checkArgument( comparator.compare(tuple, dv.partition()) == 0, "Cannot merge DVs, mismatched partition tuples (%s, %s) for %s", tuple, dv.partition(), referencedFile); } } } ``` I also combined this with `validateCanMerge` since having a separate method for the inner loop didn't seem useful. The addition of a data sequence number map is optional. You could extract it here since this is the only place where it's used. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
