search: improve with CompositeSuggestQueryParser

inveniosoftware · Feb 21, 2025 · 87d182e · 87d182e
1 parent 66b8d13
commit 87d182e
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 7 deletions.
diff --git a/invenio_users_resources/services/users/config.py b/invenio_users_resources/services/users/config.py
@@ -26,10 +26,10 @@
     SortParam,
 )
 from invenio_records_resources.services.records.queryparser import (
+    CompositeSuggestQueryParser,
     FieldValueMapper,
     QueryParser,
     SearchFieldTransformer,
-    SuggestQueryParser,
 )
 from luqum.tree import Word
 
@@ -68,9 +68,15 @@ class UserSearchOptions(SearchOptions, SearchOptionsMixin):
     # The user search needs to be highly restricted to avoid leaking
     # account information, hence do not edit here unless you are
     # absolutely sure what you're doing.
-    suggest_parser_cls = SuggestQueryParser.factory(
+    suggest_parser_cls = CompositeSuggestQueryParser.factory(
         tree_transformer_cls=SearchFieldTransformer,
-        fields=["username^2", "email^2", "profile.full_name^3", "profile.affiliations"],
+        fields=[
+            "username^2",
+            "email.keyword^2",
+            "email^2",
+            "profile.full_name^3",
+            "profile.affiliations",
+        ],
         # Only public emails because hidden emails are stored in email_hidden field.
         allow_list=["username", "email"],
         mapping={
@@ -81,7 +87,6 @@ class UserSearchOptions(SearchOptions, SearchOptionsMixin):
             "name": "profile.full_name",
         },
         type="most_fields",  # https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-multi-match-query.html#multi-match-types
-        fuzziness="AUTO",  # https://www.elastic.co/guide/en/elasticsearch/reference/current/common-options.html#fuzziness
     )
 
     params_interpreters_cls = [

diff --git a/tests/services/users/test_service_users.py b/tests/services/users/test_service_users.py
@@ -111,10 +111,9 @@ def test_user_search_field_not_searchable(user_service, user_pub, query):
         ("CERN", USERNAME_BOTH),
         ("Jose", USERNAME_JOSE),
         ("Jos", USERNAME_JOSE),
-        ("Jose CERN", USERNAME_BOTH),
-        ("Jose AND CERN", USERNAME_BOTH),
+        ("Jose CERN", USERNAME_JOSE),
         ("Tim", USERNAME_TIM),
-        ("Tim CERN", USERNAME_BOTH),
+        ("Tim CERN", USERNAME_TIM),
         ("pub@inveniosoftware.org", USERNAME_JOSE),
         ("pub@inveniosoftware.or", USERNAME_JOSE),
         ("pub@inveniosoft", USERNAME_JOSE),