cpoerschke commented on a change in pull request #300: SOLR-11831: Skip second grouping step if group.limit is 1 (aka Las Vegas Patch) URL: https://github.com/apache/lucene-solr/pull/300#discussion_r276777359
########## File path: solr/core/src/java/org/apache/solr/search/grouping/distributed/responseprocessor/SkipSecondStepSearchGroupShardResponseProcessor.java ########## @@ -0,0 +1,118 @@ +/* + * 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.solr.search.grouping.distributed.responseprocessor; + +import java.util.Collection; +import java.util.HashMap; +import java.util.Map; + +import org.apache.lucene.search.Sort; +import org.apache.lucene.search.TotalHits; +import org.apache.lucene.search.grouping.GroupDocs; +import org.apache.lucene.search.grouping.SearchGroup; +import org.apache.lucene.search.grouping.TopGroups; +import org.apache.lucene.util.BytesRef; +import org.apache.solr.handler.component.ResponseBuilder; +import org.apache.solr.handler.component.ShardDoc; +import org.apache.solr.handler.component.ShardResponse; +import org.apache.solr.search.SolrIndexSearcher; +import org.apache.solr.search.grouping.GroupingSpecification; +import org.apache.solr.search.grouping.distributed.shardresultserializer.SearchGroupsResultTransformer; + +public class SkipSecondStepSearchGroupShardResponseProcessor extends SearchGroupShardResponseProcessor { + + @Override + protected SearchGroupsResultTransformer newSearchGroupsResultTransformer(SolrIndexSearcher solrIndexSearcher) { + return new SearchGroupsResultTransformer.SkipSecondStepSearchResultResultTransformer(solrIndexSearcher); + } + + @Override + protected SearchGroupsContainer newSearchGroupsContainer(ResponseBuilder rb) { + return new SkipSecondStepSearchGroupsContainer(rb.getGroupingSpec().getFields()); + } + + protected static class SkipSecondStepSearchGroupsContainer extends SearchGroupsContainer { + + private final Map<Object, String> docIdToShard = new HashMap<>(); + + public SkipSecondStepSearchGroupsContainer(String[] fields) { + super(fields); + } + + @Override + public void addSearchGroups(ShardResponse srsp, String field, Collection<SearchGroup<BytesRef>> searchGroups) { + super.addSearchGroups(srsp, field, searchGroups); + for (SearchGroup<BytesRef> searchGroup : searchGroups) { + assert(srsp.getShard() != null); + docIdToShard.put(searchGroup.topDocSolrId, srsp.getShard()); + } + } + + @Override + public void addMergedSearchGroups(ResponseBuilder rb, String groupField, Collection<SearchGroup<BytesRef>> mergedTopGroups ) { + // TODO: add comment or javadoc re: why this method is overridden as a no-op + } + + @Override + public void addSearchGroupToShards(ResponseBuilder rb, String groupField, Collection<SearchGroup<BytesRef>> mergedTopGroups) { + super.addSearchGroupToShards(rb, groupField, mergedTopGroups); + + final GroupingSpecification groupingSpecification = rb.getGroupingSpec(); + final Sort groupSort = groupingSpecification.getGroupSort(); + final String[] fields = groupingSpecification.getFields(); + + GroupDocs<BytesRef>[] groups = new GroupDocs[mergedTopGroups.size()]; + Map<Object, ShardDoc> resultsId = new HashMap<>(mergedTopGroups.size()); + + // This is the max score found in any document on any group + float maxScore = 0; + int index = 0; + + for (SearchGroup<BytesRef> group : mergedTopGroups) { + maxScore = Math.max(maxScore, group.topDocScore); + final String shard = docIdToShard.get(group.topDocSolrId); + assert(shard != null); + ShardDoc sdoc = new ShardDoc(group.topDocScore, + fields, Review comment: The passing of `fields` to the `ShardDoc` here seems incorrect to me: * `ShardDoc` is a `FieldDoc` and per https://github.com/apache/lucene-solr/blob/releases/lucene-solr/8.0.0/lucene/core/src/java/org/apache/lucene/search/FieldDoc.java#L43-L51 it expects "... values which are used to sort ..." to be stored in that field (e.g. 2019). * Instead here `GroupingSpecification.getFields()` is passed (e.g. `publicationYear`) i.e. a field name rather than a field value. On the `group.skip.second.step=true` code path that particular argument is not actually used in practice, hence this name-vs-value difference can easily go unnoticed. Something like commit https://github.com/cpoerschke/lucene-solr/commit/2b3069b52ac514478e4f6363d18b485a93eae566 would be a way to use the other `ShardDoc` constructor. ---------------------------------------------------------------- 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: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
