LibJS: Use a Utf8View on the subject if the regex has the unicode flag

This makes LibRegex behave (more) correctly with regards to matching
unicode code points.

1 files changed, 6 insertions, 1 deletions

diff --git a/Userland/Libraries/LibJS/Runtime/RegExpPrototype.cpp b/Userland/Libraries/LibJS/Runtime/RegExpPrototype.cpp
index 5881153ce9..c3c18e9251 100644
--- a/Userland/Libraries/LibJS/Runtime/RegExpPrototype.cpp
+++ b/Userland/Libraries/LibJS/Runtime/RegExpPrototype.cpp
@@ -240,7 +240,12 @@ static Value regexp_builtin_exec(GlobalObject& global_object, RegExpObject& rege
         }
 
         regex.start_offset = last_index;
-        result = regex.match(string);
+        // FIXME: JavaScript strings are UTF-16, update this if the backing storage
+        //        encoding changes for spec compliance reasons.
+        if (unicode)
+            result = regex.match(Utf8View { string });
+        else
+            result = regex.match(string);
 
         if (result.success)
             break;