reorganize and complete tests around generic token analysis

[nominatim.git] / test / python / test_tokenizer_icu.py

diff --git a/test/python/test_tokenizer_icu.py b/test/python/test_tokenizer_icu.py
index 4b7c56d58778e577af4dc96663c1c73fca020990..6a2f2f8bd04b405f7741aca6fbaa27cbe2ce113c 100644 (file)
--- a/test/python/test_tokenizer_icu.py
+++ b/test/python/test_tokenizer_icu.py
@@ -67,10 +67,13 @@ def analyzer(tokenizer_factory, test_config, monkeypatch,
     monkeypatch.undo()
 
     def _mk_analyser(norm=("[[:Punctuation:][:Space:]]+ > ' '",), trans=(':: upper()',),
-                     variants=('~gasse -> gasse', 'street => st', )):
-        cfgstr = {'normalization' : list(norm),
-                   'transliteration' : list(trans),
-                   'variants' : [ {'words': list(variants)}]}
+                     variants=('~gasse -> gasse', 'street => st', ),
+                     sanitizers=[]):
+        cfgstr = {'normalization': list(norm),
+                  'sanitizers': sanitizers,
+                  'transliteration': list(trans),
+                  'token-analysis': [{'analyzer': 'generic',
+                                      'variants': [{'words': list(variants)}]}]}
         (test_config.project_dir / 'icu_tokenizer.yaml').write_text(yaml.dump(cfgstr))
         tok.loader = ICURuleLoader(test_config)
 
@@ -166,9 +169,7 @@ def test_init_word_table(tokenizer_factory, test_config, place_row, word_table):
     tok.init_new_db(test_config)
 
     assert word_table.get_partial_words() == {('test', 1),
-                                              ('no', 1), ('area', 2),
-                                              ('holz', 1), ('strasse', 1),
-                                              ('str', 1)}
+                                              ('no', 1), ('area', 2)}
 
 
 def test_init_from_project(monkeypatch, test_config, tokenizer_factory):
@@ -309,14 +310,15 @@ class TestPlaceNames:
 
     @pytest.fixture(autouse=True)
     def setup(self, analyzer, sql_functions):
-        with analyzer() as anl:
+        sanitizers = [{'step': 'split-name-list'},
+                      {'step': 'strip-brace-terms'}]
+        with analyzer(sanitizers=sanitizers) as anl:
             self.analyzer = anl
             yield anl
 
 
     def expect_name_terms(self, info, *expected_terms):
         tokens = self.analyzer.get_word_token_info(expected_terms)
-        print (tokens)
         for token in tokens:
             assert token[2] is not None, "No token for {0}".format(token)
 
@@ -324,9 +326,7 @@ class TestPlaceNames:
 
 
     def process_named_place(self, names):
-        place = {'name': names}
-
-        return self.analyzer.process_place(PlaceInfo(place))
+        return self.analyzer.process_place(PlaceInfo({'name': names}))
 
 
     def test_simple_names(self):