From 996884067d44d3b2a8fb6822ca043abc2da6ff78 Mon Sep 17 00:00:00 2001
From: "opensearch-trigger-bot[bot]"
 <98922864+opensearch-trigger-bot[bot]@users.noreply.github.com>
Date: Wed, 12 Oct 2022 19:01:50 -0500
Subject: [PATCH] [Backport 2.x] Set analyzer to regex query string search
 (#4219)

Sets analyzer to regex query string search

Signed-off-by: yogev mets <yyyogev@gmail.com>
(cherry picked from commit ea4cfcc1f1a4788982e8f7bb3175b0903226ca48)

Co-authored-by: Yogev Mets <yyyogev@gmail.com>
Co-authored-by: Nicholas Walter Knize <nknize@apache.org>
---
 CHANGELOG.md                                  |  1 +
 .../test/search/190_index_prefix_search.yml   | 22 +++++++++++++++++++
 .../index/search/QueryStringQueryParser.java  | 17 ++++++++------
 3 files changed, 33 insertions(+), 7 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index a21b1277155d2..d948d1bc0e27f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -76,6 +76,7 @@ Inspired from [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 - Updated jackson to 2.13.4 and snakeyml to 1.32 ([#4563](https://github.com/opensearch-project/OpenSearch/pull/4563))
 - Add BWC version 1.3.7 ([#4709](https://github.com/opensearch-project/OpenSearch/pull/4709))
 - Bump `jettison` from 1.4.1 to 1.5.1 ([#4717](https://github.com/opensearch-project/OpenSearch/pull/4717))
+- Set analyzer to regex query string search ([4219](https://github.com/opensearch-project/OpenSearch/pull/4219))
 
 ### Changed
 
diff --git a/rest-api-spec/src/main/resources/rest-api-spec/test/search/190_index_prefix_search.yml b/rest-api-spec/src/main/resources/rest-api-spec/test/search/190_index_prefix_search.yml
index 6f276f669f815..ca8e89aa2d5b5 100644
--- a/rest-api-spec/src/main/resources/rest-api-spec/test/search/190_index_prefix_search.yml
+++ b/rest-api-spec/src/main/resources/rest-api-spec/test/search/190_index_prefix_search.yml
@@ -17,6 +17,12 @@ setup:
           id:     1
           body:   { text: some short words with a stupendously long one }
 
+  - do:
+      index:
+        index:  test
+        id:     2
+        body:   { text: sentence with UPPERCASE WORDS }
+
   - do:
       indices.refresh:
         index: [test]
@@ -76,6 +82,22 @@ setup:
   - match: {hits.max_score: 1}
   - match: {hits.hits.0._score: 1}
 
+---
+"search with uppercase regex":
+  - do:
+      search:
+        rest_total_hits_as_int: true
+        index: test
+        body:
+          query:
+            query_string:
+              default_field: text
+              query: /UPPERCASE/
+
+  - match: {hits.total: 1}
+  - match: {hits.max_score: 1}
+  - match: {hits.hits.0._score: 1}
+
 ---
 "search index prefixes with span_multi":
   - do:
diff --git a/server/src/main/java/org/opensearch/index/search/QueryStringQueryParser.java b/server/src/main/java/org/opensearch/index/search/QueryStringQueryParser.java
index cdb7464ff250a..6d59e861eb32f 100644
--- a/server/src/main/java/org/opensearch/index/search/QueryStringQueryParser.java
+++ b/server/src/main/java/org/opensearch/index/search/QueryStringQueryParser.java
@@ -56,7 +56,6 @@
 import org.apache.lucene.search.SynonymQuery;
 import org.apache.lucene.search.WildcardQuery;
 import org.apache.lucene.util.BytesRef;
-import org.apache.lucene.util.automaton.RegExp;
 import org.opensearch.common.lucene.search.Queries;
 import org.opensearch.common.regex.Regex;
 import org.opensearch.common.unit.Fuzziness;
@@ -565,7 +564,7 @@ private Query getPrefixQuerySingle(String field, String termStr) throws ParseExc
             if (currentFieldType == null || currentFieldType.getTextSearchInfo() == TextSearchInfo.NONE) {
                 return newUnmappedFieldQuery(field);
             }
-            setAnalyzer(forceAnalyzer == null ? queryBuilder.context.getSearchAnalyzer(currentFieldType) : forceAnalyzer);
+            setAnalyzer(getSearchAnalyzer(currentFieldType));
             Query query = null;
             if (currentFieldType.getTextSearchInfo().isTokenized() == false) {
                 query = currentFieldType.prefixQuery(termStr, getMultiTermRewriteMethod(), context);
@@ -741,6 +740,13 @@ private Query getWildcardQuerySingle(String field, String termStr) throws ParseE
         }
     }
 
+    private Analyzer getSearchAnalyzer(MappedFieldType currentFieldType) {
+        if (forceAnalyzer == null) {
+            return queryBuilder.context.getSearchAnalyzer(currentFieldType);
+        }
+        return forceAnalyzer;
+    }
+
     @Override
     protected Query getRegexpQuery(String field, String termStr) throws ParseException {
         final int maxAllowedRegexLength = context.getIndexSettings().getMaxRegexLength();
@@ -781,11 +787,8 @@ private Query getRegexpQuerySingle(String field, String termStr) throws ParseExc
             if (currentFieldType == null) {
                 return newUnmappedFieldQuery(field);
             }
-            if (forceAnalyzer != null) {
-                setAnalyzer(forceAnalyzer);
-                return super.getRegexpQuery(field, termStr);
-            }
-            return currentFieldType.regexpQuery(termStr, RegExp.ALL, 0, getDeterminizeWorkLimit(), getMultiTermRewriteMethod(), context);
+            setAnalyzer(getSearchAnalyzer(currentFieldType));
+            return super.getRegexpQuery(field, termStr);
         } catch (RuntimeException e) {
             if (lenient) {
                 return newLenientFieldQuery(field, e);