Handle SynonymQuery extraction for the FastVectorHighlighter (#20829)

SynonymQuery was ignored by the FastVectorHighlighter.
This change adds the support for SynonymQuery in the FVH.
Although this change should be implemented in Lucene directly which is why https://issues.apache.org/jira/browse/LUCENE-7484 has been opened.
In the meantime this PR handles the issue on ES side and could be removed when LUCENE-7484 gets merged.

Fixes #20781
This commit is contained in:
Jim Ferenczi 2016-10-10 17:46:19 +02:00 committed by GitHub
parent c80a563a71
commit e874dee3bc
2 changed files with 49 additions and 0 deletions

View File

@ -28,6 +28,7 @@ import org.apache.lucene.search.ConstantScoreQuery;
import org.apache.lucene.search.MultiPhraseQuery;
import org.apache.lucene.search.PhraseQuery;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.SynonymQuery;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.search.join.ToParentBlockJoinQuery;
import org.apache.lucene.search.spans.SpanTermQuery;
@ -88,6 +89,13 @@ public class CustomFieldQuery extends FieldQuery {
flatten(boostingQuery.getMatch(), reader, flatQueries, boost);
//flatten negative query with negative boost
flatten(boostingQuery.getContext(), reader, flatQueries, boostingQuery.getBoost());
} else if (sourceQuery instanceof SynonymQuery) {
// SynonymQuery should be handled by the parent class directly.
// This statement should be removed when https://issues.apache.org/jira/browse/LUCENE-7484 is merged.
SynonymQuery synQuery = (SynonymQuery) sourceQuery;
for (Term term : synQuery.getTerms()) {
flatten(new TermQuery(term), reader, flatQueries, boost);
}
} else {
super.flatten(sourceQuery, reader, flatQueries, boost);
}

View File

@ -2907,4 +2907,45 @@ public class HighlighterSearchIT extends ESIntegTestCase {
assertThat(field.getFragments().length, equalTo(1));
assertThat(field.getFragments()[0].string(), equalTo("<em>brown</em>"));
}
public void testSynonyms() throws IOException {
Builder builder = Settings.builder()
.put(indexSettings())
.put("index.analysis.analyzer.synonym.tokenizer", "whitespace")
.putArray("index.analysis.analyzer.synonym.filter", "synonym", "lowercase")
.put("index.analysis.filter.synonym.type", "synonym")
.putArray("index.analysis.filter.synonym.synonyms", "fast,quick");
assertAcked(prepareCreate("test").setSettings(builder.build())
.addMapping("type1", "field1",
"type=text,term_vector=with_positions_offsets,search_analyzer=synonym," +
"analyzer=english,index_options=offsets"));
ensureGreen();
client().prepareIndex("test", "type1", "0").setSource(
"field1", "The quick brown fox jumps over the lazy dog").get();
refresh();
for (String highlighterType : new String[] {"plain", "postings", "fvh"}) {
logger.info("--> highlighting (type=" + highlighterType + ") and searching on field1");
SearchSourceBuilder source = searchSource()
.query(matchQuery("field1", "quick brown fox").operator(Operator.AND))
.highlighter(
highlight()
.field("field1")
.order("score")
.preTags("<x>")
.postTags("</x>")
.highlighterType(highlighterType));
SearchResponse searchResponse = client().search(searchRequest("test").source(source)).actionGet();
assertHighlight(searchResponse, 0, "field1", 0, 1,
equalTo("The <x>quick</x> <x>brown</x> <x>fox</x> jumps over the lazy dog"));
source = searchSource()
.query(matchQuery("field1", "fast brown fox").operator(Operator.AND))
.highlighter(highlight().field("field1").order("score").preTags("<x>").postTags("</x>"));
searchResponse = client().search(searchRequest("test").source(source)).actionGet();
assertHighlight(searchResponse, 0, "field1", 0, 1,
equalTo("The <x>quick</x> <x>brown</x> <x>fox</x> jumps over the lazy dog"));
}
}
}