Correctly set search quote analyzer for keyword fields (#68315)

Keyword fields with split_queries_on_whitespace=true were also setting whitespace analyzers to be used for quoted queries. Instead, keyword fields should always set their searchQuoteAnalyzer to be the same as the index-time normalizer. Fixes #68313
elastic · Feb 2, 2021 · 439cdbc · 439cdbc
1 parent 8849a18
commit 439cdbc
Show file tree

Hide file tree

Showing 3 changed files with 20 additions and 5 deletions.
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/KeywordFieldMapper.java b/server/src/main/java/org/elasticsearch/index/mapper/KeywordFieldMapper.java
@@ -144,23 +144,23 @@ protected List<Parameter<?>> getParameters() {
         private KeywordFieldType buildFieldType(ContentPath contentPath, FieldType fieldType) {
             NamedAnalyzer normalizer = Lucene.KEYWORD_ANALYZER;
             NamedAnalyzer searchAnalyzer = Lucene.KEYWORD_ANALYZER;
+            NamedAnalyzer quoteAnalyzer = Lucene.KEYWORD_ANALYZER;
             String normalizerName = this.normalizer.getValue();
             if (Objects.equals(normalizerName, "default") == false) {
                 assert indexAnalyzers != null;
                 normalizer = indexAnalyzers.getNormalizer(normalizerName);
                 if (normalizer == null) {
                     throw new MapperParsingException("normalizer [" + normalizerName + "] not found for field [" + name + "]");
                 }
+                searchAnalyzer = quoteAnalyzer = normalizer;
                 if (splitQueriesOnWhitespace.getValue()) {
                     searchAnalyzer = indexAnalyzers.getWhitespaceNormalizer(normalizerName);
-                } else {
-                    searchAnalyzer = normalizer;
                 }
             }
             else if (splitQueriesOnWhitespace.getValue()) {
                 searchAnalyzer = Lucene.WHITESPACE_ANALYZER;
             }
-            return new KeywordFieldType(buildFullName(contentPath), fieldType, normalizer, searchAnalyzer, this);
+            return new KeywordFieldType(buildFullName(contentPath), fieldType, normalizer, searchAnalyzer, quoteAnalyzer, this);
         }
 
         @Override
@@ -184,12 +184,13 @@ public static final class KeywordFieldType extends StringFieldType {
         private final NamedAnalyzer normalizer;
 
         public KeywordFieldType(String name, FieldType fieldType,
-                                NamedAnalyzer normalizer, NamedAnalyzer searchAnalyzer, Builder builder) {
+                                NamedAnalyzer normalizer, NamedAnalyzer searchAnalyzer, NamedAnalyzer quoteAnalyzer,
+                                Builder builder) {
             super(name,
                 fieldType.indexOptions() != IndexOptions.NONE,
                 fieldType.stored(),
                 builder.hasDocValues.getValue(),
-                new TextSearchInfo(fieldType, builder.similarity.getValue(), searchAnalyzer, searchAnalyzer),
+                new TextSearchInfo(fieldType, builder.similarity.getValue(), searchAnalyzer, quoteAnalyzer),
                 builder.meta.getValue());
             setEagerGlobalOrdinals(builder.eagerGlobalOrdinals.getValue());
             setBoost(builder.boost.getValue());

diff --git a/server/src/test/java/org/elasticsearch/index/mapper/KeywordFieldMapperTests.java b/server/src/test/java/org/elasticsearch/index/mapper/KeywordFieldMapperTests.java
@@ -435,6 +435,8 @@ public void testSplitQueriesOnWhitespace() throws IOException {
             ft.getTextSearchInfo().getSearchAnalyzer().analyzer().tokenStream("", "Hello World"),
             new String[] { "hello", "world" }
         );
+        Analyzer q = ft.getTextSearchInfo().getSearchQuoteAnalyzer();
+        assertTokenStreamContents(q.tokenStream("", "Hello World"), new String[] { "hello world" });
 
         mapperService = createMapperService(mapping(b -> {
             b.startObject("field").field("type", "keyword").field("split_queries_on_whitespace", true).endObject();

diff --git a/server/src/test/java/org/elasticsearch/index/query/QueryStringQueryBuilderTests.java b/server/src/test/java/org/elasticsearch/index/query/QueryStringQueryBuilderTests.java
@@ -98,6 +98,10 @@ protected void initializeAdditionalMappings(MapperService mapperService) throws
             .field("type", "text")
             .startObject("index_prefixes").endObject()
             .endObject()
+            .startObject("ww_keyword")
+            .field("type", "keyword")
+            .field("split_queries_on_whitespace", true)
+            .endObject()
             .endObject().endObject().endObject();
 
         mapperService.merge("_doc",
@@ -1658,4 +1662,12 @@ private void assertQueryCachability(QueryStringQueryBuilder qb, boolean cachingE
         assertEquals("query should " + (cachingExpected ? "" : "not") + " be cacheable: " + qb.toString(), cachingExpected,
                 context.isCacheable());
     }
+
+    public void testWhitespaceKeywordQueries() throws IOException {
+        String query = "\"query with spaces\"";
+        QueryStringQueryBuilder b = new QueryStringQueryBuilder(query);
+        b.field("ww_keyword");
+        Query q = b.doToQuery(createSearchExecutionContext());
+        assertEquals(new TermQuery(new Term("ww_keyword", "query with spaces")), q);
+    }
 }