X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/6e89310a9285f1ad15d8002bf68f578eada367a0..f289db9bfe21c5a725e751be98e2b824026e2310:/src/nominatim_api/search/icu_tokenizer.py?ds=sidebyside

diff --git a/src/nominatim_api/search/icu_tokenizer.py b/src/nominatim_api/search/icu_tokenizer.py
index 42665679..7bd2b092 100644
--- a/src/nominatim_api/search/icu_tokenizer.py
+++ b/src/nominatim_api/search/icu_tokenizer.py
@@ -16,8 +16,8 @@ from icu import Transliterator
 
 import sqlalchemy as sa
 
-from nominatim_core.typing import SaRow
-from nominatim_core.db.sqlalchemy_types import Json
+from ..typing import SaRow
+from ..sql.sqlalchemy_types import Json
 from ..connection import SearchConnection
 from ..logging import log
 from ..search import query as qmod
@@ -208,7 +208,12 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
             standardized form search will work with. All information removed
             at this stage is inevitably lost.
         """
-        return cast(str, self.normalizer.transliterate(text))
+        norm = cast(str, self.normalizer.transliterate(text))
+        numspaces = norm.count(' ')
+        if numspaces > 4 and len(norm) <= (numspaces + 1) * 3:
+            return ''
+
+        return norm
 
 
     def split_query(self, query: qmod.QueryStruct) -> Tuple[QueryParts, WordDict]: