From: Sarah Hoffmann <lonvia@denofr.de>
Date: Tue, 8 Aug 2023 12:59:33 +0000 (+0200)
Subject: block queries with lots of one and two letter terms
X-Git-Tag: deploy~59
X-Git-Url: https://git.openstreetmap.org./nominatim.git/commitdiff_plain/0a19cc18e5060e02eb76850cf428a4d9c48ec0c1?ds=inline

block queries with lots of one and two letter terms
---

diff --git a/nominatim/api/search/icu_tokenizer.py b/nominatim/api/search/icu_tokenizer.py
index f259995d..ad08294e 100644
--- a/nominatim/api/search/icu_tokenizer.py
+++ b/nominatim/api/search/icu_tokenizer.py
@@ -192,7 +192,12 @@ 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]: