clintropolis commented on code in PR #18880: URL: https://github.com/apache/druid/pull/18880#discussion_r2672009325
########## processing/src/main/java/org/apache/druid/segment/IndexMergerV10.java: ########## @@ -0,0 +1,305 @@ +/* + * 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.druid.segment; + +import com.fasterxml.jackson.databind.ObjectMapper; +import com.google.common.collect.Lists; +import com.google.common.collect.Maps; +import org.apache.druid.java.util.common.DateTimes; +import org.apache.druid.java.util.common.FileUtils; +import org.apache.druid.java.util.common.JodaUtils; +import org.apache.druid.java.util.common.guava.Comparators; +import org.apache.druid.java.util.common.io.Closer; +import org.apache.druid.java.util.common.logger.Logger; +import org.apache.druid.segment.column.ColumnDescriptor; +import org.apache.druid.segment.column.ColumnFormat; +import org.apache.druid.segment.column.ColumnHolder; +import org.apache.druid.segment.file.SegmentFileBuilder; +import org.apache.druid.segment.file.SegmentFileBuilderV10; +import org.apache.druid.segment.file.SegmentFileChannel; +import org.apache.druid.segment.loading.SegmentizerFactory; +import org.apache.druid.segment.projections.ProjectionMetadata; +import org.apache.druid.segment.projections.Projections; +import org.apache.druid.segment.serde.NullColumnPartSerde; +import org.apache.druid.segment.writeout.SegmentWriteOutMedium; +import org.apache.druid.segment.writeout.SegmentWriteOutMediumFactory; +import org.apache.druid.utils.CollectionUtils; +import org.joda.time.DateTime; +import org.joda.time.Interval; + +import javax.annotation.Nullable; +import java.io.File; +import java.io.FileOutputStream; +import java.io.IOException; +import java.util.ArrayList; +import java.util.List; +import java.util.Map; +import java.util.TreeMap; +import java.util.function.Function; +import java.util.stream.Collectors; + +/** + * {@link IndexMerger} for creating v10 format segments with {@link SegmentFileBuilderV10} + * + * @see SegmentFileBuilderV10 + * @see org.apache.druid.segment.file.SegmentFileMapperV10 to read the resulting segment file + */ +public class IndexMergerV10 extends IndexMergerBase +{ + private static final Logger log = new Logger(IndexMergerV10.class); + + public IndexMergerV10( + ObjectMapper mapper, + IndexIO indexIO, + SegmentWriteOutMediumFactory defaultSegmentWriteOutMediumFactory + ) + { + super(mapper, indexIO, defaultSegmentWriteOutMediumFactory); + } + + @Override + protected boolean shouldStoreEmptyColumns() + { + return true; + } + + @Override + protected File makeIndexFiles( + final List<IndexableAdapter> adapters, + final @Nullable Metadata segmentMetadata, + final File outDir, + final ProgressIndicator progress, + final List<String> mergedDimensionsWithTime, // has both explicit and implicit dimensions, as well as __time + final DimensionsSpecInspector dimensionsSpecInspector, + final List<String> mergedMetrics, + final Function<List<TransformableRowIterator>, TimeAndDimsIterator> rowMergerFn, + final IndexSpec indexSpec, + final @Nullable SegmentWriteOutMediumFactory segmentWriteOutMediumFactory + ) throws IOException + { + final String basePrefix = Projections.BASE_TABLE_PROJECTION_NAME + "/"; + + progress.start(); + progress.progress(); + + // Merged dimensions without __time. + List<String> mergedDimensions = + mergedDimensionsWithTime.stream() + .filter(dim -> !ColumnHolder.TIME_COLUMN_NAME.equals(dim)) + .collect(Collectors.toList()); + Closer closer = Closer.create(); + try { + final SegmentFileBuilderV10 v10Smoosher = new SegmentFileBuilderV10( + mapper, + IndexIO.V10_FILE_NAME, + outDir, + Integer.MAX_VALUE + ); + + DateTime minTime = DateTimes.MAX; + DateTime maxTime = DateTimes.MIN; + + for (IndexableAdapter index : adapters) { + minTime = JodaUtils.minDateTime(minTime, index.getDataInterval().getStart()); + maxTime = JodaUtils.maxDateTime(maxTime, index.getDataInterval().getEnd()); + } + final Interval dataInterval = new Interval(minTime, maxTime); + v10Smoosher.addInterval(dataInterval.toString()); + v10Smoosher.addBitmapEncoding(indexSpec.getBitmapSerdeFactory()); + + FileUtils.mkdirp(outDir); + + SegmentWriteOutMediumFactory omf = segmentWriteOutMediumFactory != null + ? segmentWriteOutMediumFactory + : defaultSegmentWriteOutMediumFactory; + log.debug("Using SegmentWriteOutMediumFactory[%s]", omf.getClass().getSimpleName()); + SegmentWriteOutMedium segmentWriteOutMedium = omf.makeSegmentWriteOutMedium(outDir); + closer.register(segmentWriteOutMedium); + long startTime = System.currentTimeMillis(); + + SegmentizerFactory customSegmentLoader = indexSpec.getSegmentLoader(); + if (customSegmentLoader != null) { + try (FileOutputStream fos = new FileOutputStream(new File(outDir, "factory.json"))) { + mapper.writeValue(fos, customSegmentLoader); + log.debug("Completed factory.json in %,d millis", System.currentTimeMillis() - startTime); + } + } + + progress.progress(); + // this whole section should also just be writing a projection for the base table using the same stuff, however, + // the method that builds projections right now is missing the logic to mark mergers as 'parents' if any other + // projections are relying on them (which keeps certain resources around longer so that they are still available + // when merging the projection tables), and so it lives separately for now... will fix later + final Map<String, ColumnFormat> metricFormats = new TreeMap<>(Comparators.naturalNullsFirst()); + final List<ColumnFormat> dimFormats = Lists.newArrayListWithCapacity(mergedDimensions.size()); + mergeFormat(adapters, mergedDimensions, metricFormats, dimFormats); + + final Map<String, DimensionHandler> handlers = makeDimensionHandlers(mergedDimensions, dimFormats); + final Map<String, DimensionMergerV9> mergersMap = Maps.newHashMapWithExpectedSize(mergedDimensions.size()); + final List<DimensionMergerV9> mergers = new ArrayList<>(); + for (int i = 0; i < mergedDimensions.size(); i++) { + DimensionHandler handler = handlers.get(mergedDimensions.get(i)); + DimensionMergerV9 merger = handler.makeMerger( + basePrefix + mergedDimensions.get(i), + indexSpec, + segmentWriteOutMedium, + dimFormats.get(i).toColumnCapabilities(), + progress, + outDir, + closer + ); + mergers.add(merger); + mergersMap.put(mergedDimensions.get(i), merger); + } + + // this part right here does the parent marking + if (segmentMetadata != null && segmentMetadata.getProjections() != null) { + for (AggregateProjectionMetadata projectionMetadata : segmentMetadata.getProjections()) { + for (String dimension : projectionMetadata.getSchema().getGroupingColumns()) { + DimensionMergerV9 merger = mergersMap.get(dimension); + if (merger != null) { + merger.markAsParent(); + } + } + } + } + + /************* Setup Dim Conversions **************/ + progress.progress(); + startTime = System.currentTimeMillis(); + writeDimValuesAndSetupDimConversion(adapters, progress, mergedDimensions, mergers); + log.debug("Completed dim conversions in %,d millis.", System.currentTimeMillis() - startTime); + + /************* Walk through data sets, merge them, and write merged columns *************/ + progress.progress(); + final TimeAndDimsIterator timeAndDimsIterator = makeMergedTimeAndDimsIterator( + adapters, + mergedDimensionsWithTime, + mergedMetrics, + rowMergerFn, + handlers, + mergers + ); + closer.register(timeAndDimsIterator); + final GenericColumnSerializer timeWriter = setupTimeWriter(segmentWriteOutMedium, indexSpec); + final ArrayList<GenericColumnSerializer> metricWriters = + setupMetricsWriters(segmentWriteOutMedium, mergedMetrics, metricFormats, indexSpec, basePrefix); + IndexMergeResult indexMergeResult = mergeIndexesAndWriteColumns( + adapters, + progress, + timeAndDimsIterator, + timeWriter, + metricWriters, + mergers + ); + + /************ Create Inverted Indexes and Finalize Build Columns *************/ + final String section = "build inverted index and columns"; + progress.startSection(section); + makeTimeColumn(v10Smoosher, progress, timeWriter, indexSpec, basePrefix + ColumnHolder.TIME_COLUMN_NAME); + makeMetricsColumns( + v10Smoosher, + progress, + mergedMetrics, + metricFormats, + metricWriters, + indexSpec, + basePrefix + ); + + for (int i = 0; i < mergedDimensions.size(); i++) { + DimensionMergerV9 merger = mergers.get(i); + merger.writeIndexes(indexMergeResult.rowNumConversions); + if (!merger.hasOnlyNulls()) { + ColumnDescriptor columnDesc = merger.makeColumnDescriptor(); + makeColumn(v10Smoosher, basePrefix + mergedDimensions.get(i), columnDesc); + } else { + ColumnDescriptor columnDesc = ColumnDescriptor + .builder() + .setValueType(dimFormats.get(i).getLogicalType().getType()) + .addSerde(new NullColumnPartSerde(indexMergeResult.rowCount, indexSpec.getBitmapSerdeFactory())) + .build(); + makeColumn(v10Smoosher, basePrefix + mergedDimensions.get(i), columnDesc); + } + } + + progress.stopSection(section); + + // Recompute the projections. + final Metadata finalMetadata; + if (segmentMetadata == null || CollectionUtils.isNullOrEmpty(segmentMetadata.getProjections())) { + finalMetadata = segmentMetadata; + } else { + finalMetadata = makeProjections( + v10Smoosher, + segmentMetadata.getProjections(), + adapters, + indexSpec, + segmentWriteOutMedium, + progress, + outDir, + closer, + mergersMap, + segmentMetadata + ); + } + + List<ProjectionMetadata> projections = new ArrayList<>(); + // ingestion current builds v9 metadata... translate v9 metadata and projection stuff to v10 format + projections.add( + ProjectionMetadata.forBaseTable(indexMergeResult.rowCount, mergedDimensionsWithTime, finalMetadata) + ); + // convert v9 projections to v10 projections + for (AggregateProjectionMetadata aggMeta : finalMetadata.getProjections()) { + projections.add(new ProjectionMetadata(aggMeta.getNumRows(), aggMeta.getSchema())); + } + + v10Smoosher.addProjections(projections); + + progress.progress(); + v10Smoosher.close(); Review Comment: `IndexMergerV9` is like this too (i started this thing by copying it heh), it does look like it perhaps leaves an open filestream/channel for `currOut` of the v9 smoosher (which v10 file builder currently uses to build intermediary files to concatenate into a v10 file). Calling close looks like it would try to finish writing whatever segment it had built up so far, which seems kind of useless, so i have added an `abort` method to `SegmentFileBuilder` interface to give them a chance to cleanup resources without finishing the segment file -- 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]
