jainankitk commented on code in PR #14439: URL: https://github.com/apache/lucene/pull/14439#discussion_r2054988914
########## lucene/sandbox/src/java/org/apache/lucene/sandbox/facet/plain/histograms/PointTreeBulkCollector.java: ########## @@ -0,0 +1,219 @@ +/* + * 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.lucene.sandbox.facet.plain.histograms; + +import static org.apache.lucene.search.DocIdSetIterator.NO_MORE_DOCS; + +import java.io.IOException; +import java.util.function.Function; +import org.apache.lucene.index.PointValues; +import org.apache.lucene.internal.hppc.LongIntHashMap; +import org.apache.lucene.search.CollectionTerminatedException; +import org.apache.lucene.search.DocIdSetIterator; +import org.apache.lucene.util.ArrayUtil; + +class PointTreeBulkCollector { + static boolean collect( + final PointValues pointValues, + final long bucketWidth, + final LongIntHashMap collectorCounts, + final int maxBuckets) + throws IOException { + // TODO: Do we really need pointValues.getDocCount() == pointValues.size() + if (pointValues == null + || pointValues.getNumDimensions() != 1 + || pointValues.getDocCount() != pointValues.size() + || ArrayUtil.getValue(pointValues.getBytesPerDimension()) == null) { + return false; + } + + final Function<byte[], Long> byteToLong = + ArrayUtil.getValue(pointValues.getBytesPerDimension()); + final long minValue = getLongFromByte(byteToLong, pointValues.getMinPackedValue()); + final long maxValue = getLongFromByte(byteToLong, pointValues.getMaxPackedValue()); + long leafMinBucket = Math.floorDiv(minValue, bucketWidth); + long leafMaxBucket = Math.floorDiv(maxValue, bucketWidth); + + // We want that # leaf nodes is more than # buckets so that we can completely skip over + // some of the leaf nodes. Higher this ratio, more efficient it is than naive approach! + if ((pointValues.size() / 512) < (leafMaxBucket - leafMinBucket)) { + return false; + } + + BucketManager collector = + new BucketManager( + collectorCounts, + minValue, + bucketWidth, + a -> getLongFromByte(byteToLong, a), + maxBuckets); + PointValues.IntersectVisitor visitor = getIntersectVisitor(collector); + try { + intersectWithRanges(visitor, pointValues.getPointTree(), collector); + } catch (CollectionTerminatedException _) { + // Early terminate since no more range to collect + } + collector.finalizePreviousBucket(null); + + return true; + } + + private static void intersectWithRanges( + PointValues.IntersectVisitor visitor, + PointValues.PointTree pointTree, + BucketManager collector) + throws IOException { + PointValues.Relation r = + visitor.compare(pointTree.getMinPackedValue(), pointTree.getMaxPackedValue()); + + switch (r) { + case CELL_INSIDE_QUERY: + collector.countNode((int) pointTree.size()); + break; + case CELL_CROSSES_QUERY: + if (pointTree.moveToChild()) { + do { + intersectWithRanges(visitor, pointTree, collector); + } while (pointTree.moveToSibling()); + pointTree.moveToParent(); + } else { + pointTree.visitDocValues(visitor); + } + break; + case CELL_OUTSIDE_QUERY: + } + } + + private static long getLongFromByte(Function<byte[], Long> f, byte[] packedValue) { + // TODO: Read long correctly instead of downcasting to int + return f.apply(packedValue).intValue(); + } + + private static PointValues.IntersectVisitor getIntersectVisitor(BucketManager collector) { + return new PointValues.IntersectVisitor() { + @Override + public void visit(int docID) { + // this branch should be unreachable + throw new UnsupportedOperationException( + "This IntersectVisitor does not perform any actions on a " + + "docID=" + + docID + + " node being visited"); + } + + @Override + public void visit(int docID, byte[] packedValue) throws IOException { + if (!collector.withinUpperBound(packedValue)) { + collector.finalizePreviousBucket(packedValue); + } + + if (collector.withinRange(packedValue)) { + collector.count(); + } + } + + @Override + public void visit(DocIdSetIterator iterator, byte[] packedValue) throws IOException { + if (!collector.withinUpperBound(packedValue)) { + collector.finalizePreviousBucket(packedValue); + } + + if (collector.withinRange(packedValue)) { + for (int doc = iterator.nextDoc(); doc != NO_MORE_DOCS; doc = iterator.nextDoc()) { + collector.count(); + } + } + } + + @Override + public PointValues.Relation compare(byte[] minPackedValue, byte[] maxPackedValue) { + // try to find the first range that may collect values from this cell + if (!collector.withinUpperBound(minPackedValue)) { + collector.finalizePreviousBucket(minPackedValue); + } + + // Not possible to have the CELL_OUTSIDE_QUERY, as bucket lower bound is updated + // while finalizing the previous bucket + if (collector.withinRange(minPackedValue) && collector.withinRange(maxPackedValue)) { + return PointValues.Relation.CELL_INSIDE_QUERY; + } + return PointValues.Relation.CELL_CROSSES_QUERY; + } + }; + } + + private static class BucketManager { + private final LongIntHashMap collectorCounts; + private int counter = 0; + private long startValue; + private long endValue; + private int nonZeroBuckets = 0; + private int maxBuckets; + private Function<byte[], Long> byteToLong; + private long bucketWidth; + + public BucketManager( + LongIntHashMap collectorCounts, + long minValue, + long bucketWidth, + Function<byte[], Long> byteToLong, + int maxBuckets) { + this.collectorCounts = collectorCounts; + this.bucketWidth = bucketWidth; + this.startValue = Math.floorDiv(minValue, bucketWidth) * bucketWidth; + this.endValue = startValue + bucketWidth; + this.byteToLong = byteToLong; + this.maxBuckets = maxBuckets; + } + + private void count() { + counter++; + } + + private void countNode(int count) { + counter += count; + } + + private void finalizePreviousBucket(byte[] packedValue) { + // TODO: Can counter ever be 0? Review Comment: I think this specific TODO, we can leave for now. More like a safety guard. I will try to add unit test in future to validate this, and remove if not needed Others TODOs were important and have been addressed -- 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: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
