From: Sarah Hoffmann <lonvia@denofr.de>
Date: Wed, 17 Jan 2024 09:48:11 +0000 (+0100)
Subject: Merge remote-tracking branch 'upstream/master'
X-Git-Tag: deploy~23
X-Git-Url: https://git.openstreetmap.org./nominatim.git/commitdiff_plain/1260b0c95d0551a1677219f48a0651390e6357d4?hp=-c

Merge remote-tracking branch 'upstream/master'
---

1260b0c95d0551a1677219f48a0651390e6357d4
diff --combined nominatim/api/search/icu_tokenizer.py
index 6f3e09e8,1c2565d1..76a1a2e5
--- a/nominatim/api/search/icu_tokenizer.py
+++ b/nominatim/api/search/icu_tokenizer.py
@@@ -186,7 -186,10 +186,10 @@@ class ICUQueryAnalyzer(AbstractQueryAna
                          if trange.start == 0:
                              query.add_token(trange, qmod.TokenType.NEAR_ITEM, token)
                      else:
-                         query.add_token(trange, qmod.TokenType.QUALIFIER, token)
+                         if trange.start == 0 and trange.end == query.num_token_slots():
+                             query.add_token(trange, qmod.TokenType.NEAR_ITEM, token)
+                         else:
+                             query.add_token(trange, qmod.TokenType.QUALIFIER, token)
                  else:
                      query.add_token(trange, DB_TO_TOKEN_TYPE[row.type], token)
  
@@@ -203,12 -206,7 +206,12 @@@
              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]: