Merge remote-tracking branch 'upstream/master'

[nominatim.git] / test / python / api / search / test_icu_query_analyzer.py

diff --git a/test/python/api/search/test_icu_query_analyzer.py b/test/python/api/search/test_icu_query_analyzer.py
index faf8137526106a0e5db77b325c0fa73f30b94a05..8e5480fcfb9ce49693fea8dc23e2f5bd5e43b476 100644 (file)
--- a/test/python/api/search/test_icu_query_analyzer.py
+++ b/test/python/api/search/test_icu_query_analyzer.py
@@ -2,7 +2,7 @@
 #
 # This file is part of Nominatim. (https://nominatim.org)
 #
-# Copyright (C) 2023 by the Nominatim developer community.
+# Copyright (C) 2024 by the Nominatim developer community.
 # For a full list of authors see the git log.
 """
 Tests for query analyzer for ICU tokenizer.
@@ -12,10 +12,10 @@ from pathlib import Path
 import pytest
 import pytest_asyncio
 
-from nominatim.api import NominatimAPIAsync
-from nominatim.api.search.query import Phrase, PhraseType, TokenType, BreakType
-import nominatim.api.search.icu_tokenizer as tok
-from nominatim.api.logging import set_log_output, get_and_disable
+from nominatim_api import NominatimAPIAsync
+from nominatim_api.search.query import Phrase, PhraseType, TokenType, BreakType
+import nominatim_api.search.icu_tokenizer as tok
+from nominatim_api.logging import set_log_output, get_and_disable
 
 async def add_word(conn, word_id, word_token, wtype, word, info = None):
     t = conn.t.meta.tables['word']
@@ -134,10 +134,23 @@ async def test_category_words_only_at_beginning(conn):
 
     assert query.num_token_slots() == 3
     assert len(query.nodes[0].starting) == 1
-    assert query.nodes[0].starting[0].ttype == TokenType.CATEGORY
+    assert query.nodes[0].starting[0].ttype == TokenType.NEAR_ITEM
     assert not query.nodes[2].starting
 
 
+@pytest.mark.asyncio
+async def test_freestanding_qualifier_words_become_category(conn):
+    ana = await tok.create_query_analyzer(conn)
+
+    await add_word(conn, 1, 'foo', 'S', 'FOO', {'op': '-'})
+
+    query = await ana.analyze_query(make_phrase('foo'))
+
+    assert query.num_token_slots() == 1
+    assert len(query.nodes[0].starting) == 1
+    assert query.nodes[0].starting[0].ttype == TokenType.NEAR_ITEM
+
+
 @pytest.mark.asyncio
 async def test_qualifier_words(conn):
     ana = await tok.create_query_analyzer(conn)
@@ -148,9 +161,9 @@ async def test_qualifier_words(conn):
     query = await ana.analyze_query(make_phrase('foo BAR foo BAR foo'))
 
     assert query.num_token_slots() == 5
-    assert set(t.ttype for t in query.nodes[0].starting) == {TokenType.CATEGORY, TokenType.QUALIFIER}
+    assert set(t.ttype for t in query.nodes[0].starting) == {TokenType.QUALIFIER}
     assert set(t.ttype for t in query.nodes[2].starting) == {TokenType.QUALIFIER}
-    assert set(t.ttype for t in query.nodes[4].starting) == {TokenType.CATEGORY, TokenType.QUALIFIER}
+    assert set(t.ttype for t in query.nodes[4].starting) == {TokenType.QUALIFIER}
 
 
 @pytest.mark.asyncio