Fix case insensitive query on wildcard field (opensearch-project#15882)

* fix case insensitive query on wildcard field Signed-off-by: gesong.samuel <gesong.samuel@bytedance.com> * fix YAML test Signed-off-by: gesong.samuel <gesong.samuel@bytedance.com> * add change log Signed-off-by: gesong.samuel <gesong.samuel@bytedance.com> --------- Signed-off-by: gesong.samuel <gesong.samuel@bytedance.com> Co-authored-by: gesong.samuel <gesong.samuel@bytedance.com>
sachinpkale · Sep 19, 2024 · 68cfd16 · 68cfd16
1 parent b93417c
commit 68cfd16
Show file tree

Hide file tree

Showing 3 changed files with 49 additions and 9 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -25,7 +25,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 ### Fixed
 - Fix wildcard query containing escaped character ([#15737](https://github.com/opensearch-project/OpenSearch/pull/15737))
-
+- Fix case-insensitive query on wildcard field ([#15882](https://github.com/opensearch-project/OpenSearch/pull/15882))
 ### Security
 
 [Unreleased 2.x]: https://github.com/opensearch-project/OpenSearch/compare/2.17...2.x
diff --git a/...-api-spec/src/main/resources/rest-api-spec/test/search/270_wildcard_fieldtype_queries.yml b/...-api-spec/src/main/resources/rest-api-spec/test/search/270_wildcard_fieldtype_queries.yml
@@ -56,6 +56,12 @@ setup:
         id: 6
         body:
           other_field: "test"
+  - do:
+      index:
+        index: test
+        id: 7
+        body:
+          my_field: "ABCD"
   - do:
       indices.refresh: {}
 
@@ -90,8 +96,9 @@ setup:
           query:
             term:
               my_field.lower: "abcd"
-  - match: { hits.total.value: 1 }
+  - match: { hits.total.value: 2 }
   - match: { hits.hits.0._id: "5" }
+  - match: { hits.hits.1._id: "7" }
 
   - do:
       search:
@@ -100,8 +107,9 @@ setup:
           query:
             term:
               my_field.lower: "ABCD"
-  - match: { hits.total.value: 1 }
+  - match: { hits.total.value: 2 }
   - match: { hits.hits.0._id: "5" }
+  - match: { hits.hits.1._id: "7" }
 
   - do:
       search:
@@ -215,7 +223,7 @@ setup:
             wildcard:
               my_field:
                 value: "*"
-  - match: { hits.total.value: 5 }
+  - match: { hits.total.value: 6 }
 ---
 "regexp match-all works":
   - do:
@@ -226,7 +234,7 @@ setup:
             regexp:
               my_field:
                 value: ".*"
-  - match: { hits.total.value: 5 }
+  - match: { hits.total.value: 6 }
 ---
 "terms query on wildcard field matches":
   - do:
@@ -237,3 +245,28 @@ setup:
             terms: { my_field: ["AbCd"] }
   - match: { hits.total.value: 1 }
   - match: { hits.hits.0._id: "5" }
+---
+"case insensitive query on wildcard field":
+  - do:
+      search:
+        index: test
+        body:
+          query:
+            wildcard:
+              my_field:
+                value: "AbCd"
+  - match: { hits.total.value: 1 }
+  - match: { hits.hits.0._id: "5" }
+
+  - do:
+      search:
+        index: test
+        body:
+          query:
+            wildcard:
+              my_field:
+                value: "AbCd"
+                case_insensitive: true
+  - match: { hits.total.value: 2 }
+  - match: { hits.hits.0._id: "5" }
+  - match: { hits.hits.1._id: "7" }
diff --git a/server/src/main/java/org/opensearch/index/mapper/WildcardFieldMapper.java b/server/src/main/java/org/opensearch/index/mapper/WildcardFieldMapper.java
@@ -40,6 +40,7 @@
 import org.apache.lucene.util.automaton.RegExp;
 import org.opensearch.common.lucene.BytesRefs;
 import org.opensearch.common.lucene.Lucene;
+import org.opensearch.common.lucene.search.AutomatonQueries;
 import org.opensearch.common.unit.Fuzziness;
 import org.opensearch.core.xcontent.XContentParser;
 import org.opensearch.index.analysis.IndexAnalyzers;
@@ -464,7 +465,7 @@ public Query wildcardQuery(String value, MultiTermQuery.RewriteMethod method, bo
                     return existsQuery(context);
                 }
             } else {
-                approximation = matchAllTermsQuery(name(), requiredNGrams);
+                approximation = matchAllTermsQuery(name(), requiredNGrams, caseInsensitive);
             }
             return new WildcardMatchingQuery(name(), approximation, matchPredicate, value, context, this);
         }
@@ -678,7 +679,7 @@ public Query termsQuery(List<?> values, QueryShardContext context) {
             StringBuilder pattern = new StringBuilder();
             for (Object value : values) {
                 String stringVal = BytesRefs.toString(value);
-                builder.add(matchAllTermsQuery(name(), getRequiredNGrams(stringVal)), BooleanClause.Occur.SHOULD);
+                builder.add(matchAllTermsQuery(name(), getRequiredNGrams(stringVal), false), BooleanClause.Occur.SHOULD);
                 expectedValues.add(stringVal);
                 if (pattern.length() > 0) {
                     pattern.append('|');
@@ -688,10 +689,16 @@ public Query termsQuery(List<?> values, QueryShardContext context) {
             return new WildcardMatchingQuery(name(), builder.build(), expectedValues::contains, pattern.toString(), context, this);
         }
 
-        private static BooleanQuery matchAllTermsQuery(String fieldName, Set<String> terms) {
+        private static BooleanQuery matchAllTermsQuery(String fieldName, Set<String> terms, boolean caseInsensitive) {
             BooleanQuery.Builder matchAllTermsBuilder = new BooleanQuery.Builder();
+            Query query;
             for (String term : terms) {
-                matchAllTermsBuilder.add(new TermQuery(new Term(fieldName, term)), BooleanClause.Occur.FILTER);
+                if (caseInsensitive) {
+                    query = AutomatonQueries.caseInsensitiveTermQuery(new Term(fieldName, term));
+                } else {
+                    query = new TermQuery(new Term(fieldName, term));
+                }
+                matchAllTermsBuilder.add(query, BooleanClause.Occur.FILTER);
             }
             return matchAllTermsBuilder.build();
         }