X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/08dcd05d7bc09ad4e70ff47ff6f393a1b91ad21a..d42e2e391f1e0ac252f8251561d984a72bd3b1ff:/nominatim/api/search/icu_tokenizer.py?ds=inline

diff --git a/nominatim/api/search/icu_tokenizer.py b/nominatim/api/search/icu_tokenizer.py
index 9bd16e1d..f259995d 100644
--- a/nominatim/api/search/icu_tokenizer.py
+++ b/nominatim/api/search/icu_tokenizer.py
@@ -153,7 +153,7 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
         """
         log().section('Analyze query (using ICU tokenizer)')
         normalized = list(filter(lambda p: p.text,
-                                 (qmod.Phrase(p.ptype, self.normalizer.transliterate(p.text))
+                                 (qmod.Phrase(p.ptype, self.normalize_text(p.text))
                                   for p in phrases)))
         query = qmod.QueryStruct(normalized)
         log().var_dump('Normalized query', query.source)
@@ -187,6 +187,14 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
         return query
 
 
+    def normalize_text(self, text: str) -> str:
+        """ Bring the given text into a normalized form. That is the
+            standardized form search will work with. All information removed
+            at this stage is inevitably lost.
+        """
+        return cast(str, self.normalizer.transliterate(text))
+
+
     def split_query(self, query: qmod.QueryStruct) -> Tuple[QueryParts, WordDict]:
         """ Transliterate the phrases and split them into tokens.