Merge pull request #2341 from lonvia/cleanup-python-tests

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

diff --git a/test/python/test_tokenizer_legacy_icu.py b/test/python/test_tokenizer_legacy_icu.py
index 836f15b9e8809d50b2edf60c58cf94b710efa522..798fea37397e63e75bbe962da9310e9d660c8122 100644 (file)
--- a/test/python/test_tokenizer_legacy_icu.py
+++ b/test/python/test_tokenizer_legacy_icu.py
@@ -46,14 +46,14 @@ def db_prop(temp_db_conn):
     return _get_db_property
 
 @pytest.fixture
     return _get_db_property
 
 @pytest.fixture

-def tokenizer_setup(tokenizer_factory, test_config, monkeypatch, sql_preprocessor):
+def tokenizer_setup(tokenizer_factory, test_config):

     tok = tokenizer_factory()
     tok.init_new_db(test_config)
 
 
 @pytest.fixture
     tok = tokenizer_factory()
     tok.init_new_db(test_config)
 
 
 @pytest.fixture

-def analyzer(tokenizer_factory, test_config, monkeypatch, sql_preprocessor,
-             word_table, temp_db_with_extensions, tmp_path):
+def analyzer(tokenizer_factory, test_config, monkeypatch,
+             temp_db_with_extensions, tmp_path):

     sql = tmp_path / 'sql' / 'tokenizer' / 'legacy_icu_tokenizer.sql'
     sql.write_text("SELECT 'a';")
 
     sql = tmp_path / 'sql' / 'tokenizer' / 'legacy_icu_tokenizer.sql'
     sql.write_text("SELECT 'a';")
 


@@ -74,17 +74,18 @@ def analyzer(tokenizer_factory, test_config, monkeypatch, sql_preprocessor,
 @pytest.fixture
 def getorcreate_term_id(temp_db_cursor):
     temp_db_cursor.execute("""CREATE OR REPLACE FUNCTION getorcreate_term_id(lookup_term TEXT)
 @pytest.fixture
 def getorcreate_term_id(temp_db_cursor):
     temp_db_cursor.execute("""CREATE OR REPLACE FUNCTION getorcreate_term_id(lookup_term TEXT)

-                              RETURNS INTEGER AS $$ SELECT nextval('seq_word')::INTEGER; $$ LANGUAGE SQL""")
+                              RETURNS INTEGER AS $$
+                                SELECT nextval('seq_word')::INTEGER; $$ LANGUAGE SQL""")

 
 
 @pytest.fixture
 def getorcreate_hnr_id(temp_db_cursor):
     temp_db_cursor.execute("""CREATE OR REPLACE FUNCTION getorcreate_hnr_id(lookup_term TEXT)
 
 
 @pytest.fixture
 def getorcreate_hnr_id(temp_db_cursor):
     temp_db_cursor.execute("""CREATE OR REPLACE FUNCTION getorcreate_hnr_id(lookup_term TEXT)

-                              RETURNS INTEGER AS $$ SELECT -nextval('seq_word')::INTEGER; $$ LANGUAGE SQL""")
+                              RETURNS INTEGER AS $$
+                                SELECT -nextval('seq_word')::INTEGER; $$ LANGUAGE SQL""")

 
 
 
 

-def test_init_new(tokenizer_factory, test_config, monkeypatch, db_prop,
-                  sql_preprocessor, place_table, word_table):
+def test_init_new(tokenizer_factory, test_config, monkeypatch, db_prop):

     monkeypatch.setenv('NOMINATIM_TERM_NORMALIZATION', ':: lower();')
 
     tok = tokenizer_factory()
     monkeypatch.setenv('NOMINATIM_TERM_NORMALIZATION', ':: lower();')
 
     tok = tokenizer_factory()


@@ -105,10 +106,9 @@ def test_init_from_project(tokenizer_setup, tokenizer_factory):
     assert tok.abbreviations is not None
 
 
     assert tok.abbreviations is not None
 
 

-def test_update_sql_functions(temp_db_conn, db_prop, temp_db_cursor,
+def test_update_sql_functions(db_prop, temp_db_cursor,

                               tokenizer_factory, test_config, table_factory,
                               tokenizer_factory, test_config, table_factory,

-                              monkeypatch,
-                              sql_preprocessor, place_table, word_table):
+                              monkeypatch):

     monkeypatch.setenv('NOMINATIM_MAX_WORD_FREQUENCY', '1133')
     tok = tokenizer_factory()
     tok.init_new_db(test_config)
     monkeypatch.setenv('NOMINATIM_MAX_WORD_FREQUENCY', '1133')
     tok = tokenizer_factory()
     tok.init_new_db(test_config)


@@ -128,129 +128,146 @@ def test_update_sql_functions(temp_db_conn, db_prop, temp_db_cursor,
 
 
 def test_make_standard_word(analyzer):
 
 
 def test_make_standard_word(analyzer):

-    with analyzer(abbr=(('STREET', 'ST'), ('tiny', 't'))) as a:
-        assert a.make_standard_word('tiny street') == 'TINY ST'
+    with analyzer(abbr=(('STREET', 'ST'), ('tiny', 't'))) as anl:
+        assert anl.make_standard_word('tiny street') == 'TINY ST'

 
 

-    with analyzer(abbr=(('STRASSE', 'STR'), ('STR', 'ST'))) as a:
-        assert a.make_standard_word('Hauptstrasse') == 'HAUPTST'
+    with analyzer(abbr=(('STRASSE', 'STR'), ('STR', 'ST'))) as anl:
+        assert anl.make_standard_word('Hauptstrasse') == 'HAUPTST'

 
 
 def test_make_standard_hnr(analyzer):
 
 
 def test_make_standard_hnr(analyzer):

-    with analyzer(abbr=(('IV', '4'),)) as a:
-        assert a._make_standard_hnr('345') == '345'
-        assert a._make_standard_hnr('iv') == 'IV'
+    with analyzer(abbr=(('IV', '4'),)) as anl:
+        assert anl._make_standard_hnr('345') == '345'
+        assert anl._make_standard_hnr('iv') == 'IV'

 
 
 
 

-def test_add_postcodes_from_db(analyzer, word_table, table_factory, temp_db_cursor):
+def test_update_postcodes_from_db_empty(analyzer, table_factory, word_table):

     table_factory('location_postcode', 'postcode TEXT',
                   content=(('1234',), ('12 34',), ('AB23',), ('1234',)))
 
     table_factory('location_postcode', 'postcode TEXT',
                   content=(('1234',), ('12 34',), ('AB23',), ('1234',)))
 

-    with analyzer() as a:
-        a.add_postcodes_from_db()
+    with analyzer() as anl:
+        anl.update_postcodes_from_db()

 
 

-    assert temp_db_cursor.row_set("""SELECT word, word_token from word
-                                     """) \
-               == set((('1234', ' 1234'), ('12 34', ' 12 34'), ('AB23', ' AB23')))
+    assert word_table.count() == 3
+    assert word_table.get_postcodes() == {'1234', '12 34', 'AB23'}

 
 
 
 

-def test_update_special_phrase_empty_table(analyzer, word_table, temp_db_cursor):
-    with analyzer() as a:
-        a.update_special_phrases([
+def test_update_postcodes_from_db_add_and_remove(analyzer, table_factory, word_table):
+    table_factory('location_postcode', 'postcode TEXT',
+                  content=(('1234',), ('45BC', ), ('XX45', )))
+    word_table.add_postcode(' 1234', '1234')
+    word_table.add_postcode(' 5678', '5678')
+
+    with analyzer() as anl:
+        anl.update_postcodes_from_db()
+
+    assert word_table.count() == 3
+    assert word_table.get_postcodes() == {'1234', '45BC', 'XX45'}
+
+
+def test_update_special_phrase_empty_table(analyzer, word_table):
+    with analyzer() as anl:
+        anl.update_special_phrases([

             ("König bei", "amenity", "royal", "near"),
             ("Könige", "amenity", "royal", "-"),
             ("street", "highway", "primary", "in")
             ("König bei", "amenity", "royal", "near"),
             ("Könige", "amenity", "royal", "-"),
             ("street", "highway", "primary", "in")

-        ])
+        ], True)
+
+    assert word_table.get_special() \
+               == {(' KÖNIG BEI', 'könig bei', 'amenity', 'royal', 'near'),
+                   (' KÖNIGE', 'könige', 'amenity', 'royal', None),
+                   (' ST', 'street', 'highway', 'primary', 'in')}
+
+
+def test_update_special_phrase_delete_all(analyzer, word_table):
+    word_table.add_special(' FOO', 'foo', 'amenity', 'prison', 'in')
+    word_table.add_special(' BAR', 'bar', 'highway', 'road', None)
+
+    assert word_table.count_special() == 2
+
+    with analyzer() as anl:
+        anl.update_special_phrases([], True)

 
 

-    assert temp_db_cursor.row_set("""SELECT word_token, word, class, type, operator
-                                     FROM word WHERE class != 'place'""") \
-               == set(((' KÖNIG BEI', 'könig bei', 'amenity', 'royal', 'near'),
-                       (' KÖNIGE', 'könige', 'amenity', 'royal', None),
-                       (' ST', 'street', 'highway', 'primary', 'in')))
+    assert word_table.count_special() == 0

 
 
 
 

-def test_update_special_phrase_delete_all(analyzer, word_table, temp_db_cursor):
-    temp_db_cursor.execute("""INSERT INTO word (word_token, word, class, type, operator)
-                              VALUES (' FOO', 'foo', 'amenity', 'prison', 'in'),
-                                     (' BAR', 'bar', 'highway', 'road', null)""")
+def test_update_special_phrases_no_replace(analyzer, word_table):
+    word_table.add_special(' FOO', 'foo', 'amenity', 'prison', 'in')
+    word_table.add_special(' BAR', 'bar', 'highway', 'road', None)

 
 

-    assert 2 == temp_db_cursor.scalar("SELECT count(*) FROM word WHERE class != 'place'""")
+    assert word_table.count_special() == 2

 
 

-    with analyzer() as a:
-        a.update_special_phrases([])
+    with analyzer() as anl:
+        anl.update_special_phrases([], False)

 
 

-    assert 0 == temp_db_cursor.scalar("SELECT count(*) FROM word WHERE class != 'place'""")
+    assert word_table.count_special() == 2

 
 
 
 

-def test_update_special_phrase_modify(analyzer, word_table, temp_db_cursor):
-    temp_db_cursor.execute("""INSERT INTO word (word_token, word, class, type, operator)
-                              VALUES (' FOO', 'foo', 'amenity', 'prison', 'in'),
-                                     (' BAR', 'bar', 'highway', 'road', null)""")
+def test_update_special_phrase_modify(analyzer, word_table):
+    word_table.add_special(' FOO', 'foo', 'amenity', 'prison', 'in')
+    word_table.add_special(' BAR', 'bar', 'highway', 'road', None)

 
 

-    assert 2 == temp_db_cursor.scalar("SELECT count(*) FROM word WHERE class != 'place'""")
+    assert word_table.count_special() == 2

 
 

-    with analyzer() as a:
-        a.update_special_phrases([
-          ('prison', 'amenity', 'prison', 'in'),
-          ('bar', 'highway', 'road', '-'),
-          ('garden', 'leisure', 'garden', 'near')
-        ])
+    with analyzer() as anl:
+        anl.update_special_phrases([
+            ('prison', 'amenity', 'prison', 'in'),
+            ('bar', 'highway', 'road', '-'),
+            ('garden', 'leisure', 'garden', 'near')
+        ], True)

 
 

-    assert temp_db_cursor.row_set("""SELECT word_token, word, class, type, operator
-                                     FROM word WHERE class != 'place'""") \
-               == set(((' PRISON', 'prison', 'amenity', 'prison', 'in'),
-                       (' BAR', 'bar', 'highway', 'road', None),
-                       (' GARDEN', 'garden', 'leisure', 'garden', 'near')))
+    assert word_table.get_special() \
+               == {(' PRISON', 'prison', 'amenity', 'prison', 'in'),
+                   (' BAR', 'bar', 'highway', 'road', None),
+                   (' GARDEN', 'garden', 'leisure', 'garden', 'near')}

 
 
 def test_process_place_names(analyzer, getorcreate_term_id):
 
 
 
 def test_process_place_names(analyzer, getorcreate_term_id):
 

-    with analyzer() as a:
-        info = a.process_place({'name' : {'name' : 'Soft bAr', 'ref': '34'}})
+    with analyzer() as anl:
+        info = anl.process_place({'name' : {'name' : 'Soft bAr', 'ref': '34'}})

 
     assert info['names'] == '{1,2,3,4,5,6}'
 
 
 
     assert info['names'] == '{1,2,3,4,5,6}'
 
 

-@pytest.mark.parametrize('pc', ['12345', 'AB 123', '34-345'])
-def test_process_place_postcode(analyzer, temp_db_cursor, pc):
-    with analyzer() as a:
-        info = a.process_place({'address': {'postcode' : pc}})
+@pytest.mark.parametrize('pcode', ['12345', 'AB 123', '34-345'])
+def test_process_place_postcode(analyzer, word_table, pcode):
+    with analyzer() as anl:
+        anl.process_place({'address': {'postcode' : pcode}})

 
 

-    assert temp_db_cursor.row_set("""SELECT word FROM word
-                                     WHERE class = 'place' and type = 'postcode'""") \
-               == set(((pc, ),))
+    assert word_table.get_postcodes() == {pcode, }

 
 
 
 

-@pytest.mark.parametrize('pc', ['12:23', 'ab;cd;f', '123;836'])
-def test_process_place_bad_postcode(analyzer, temp_db_cursor, pc):
-    with analyzer() as a:
-        info = a.process_place({'address': {'postcode' : pc}})
+@pytest.mark.parametrize('pcode', ['12:23', 'ab;cd;f', '123;836'])
+def test_process_place_bad_postcode(analyzer, word_table, pcode):
+    with analyzer() as anl:
+        anl.process_place({'address': {'postcode' : pcode}})

 
 

-    assert 0 == temp_db_cursor.scalar("""SELECT count(*) FROM word
-                                         WHERE class = 'place' and type = 'postcode'""")
+    assert not word_table.get_postcodes()

 
 
 @pytest.mark.parametrize('hnr', ['123a', '1', '101'])
 def test_process_place_housenumbers_simple(analyzer, hnr, getorcreate_hnr_id):
 
 
 @pytest.mark.parametrize('hnr', ['123a', '1', '101'])
 def test_process_place_housenumbers_simple(analyzer, hnr, getorcreate_hnr_id):

-    with analyzer() as a:
-        info = a.process_place({'address': {'housenumber' : hnr}})
+    with analyzer() as anl:
+        info = anl.process_place({'address': {'housenumber' : hnr}})

 
     assert info['hnr'] == hnr.upper()
     assert info['hnr_tokens'] == "{-1}"
 
 
 def test_process_place_housenumbers_lists(analyzer, getorcreate_hnr_id):
 
     assert info['hnr'] == hnr.upper()
     assert info['hnr_tokens'] == "{-1}"
 
 
 def test_process_place_housenumbers_lists(analyzer, getorcreate_hnr_id):

-    with analyzer() as a:
-        info = a.process_place({'address': {'conscriptionnumber' : '1; 2;3'}})
+    with analyzer() as anl:
+        info = anl.process_place({'address': {'conscriptionnumber' : '1; 2;3'}})

 
     assert set(info['hnr'].split(';')) == set(('1', '2', '3'))
     assert info['hnr_tokens'] == "{-1,-2,-3}"
 
 
 def test_process_place_housenumbers_duplicates(analyzer, getorcreate_hnr_id):
 
     assert set(info['hnr'].split(';')) == set(('1', '2', '3'))
     assert info['hnr_tokens'] == "{-1,-2,-3}"
 
 
 def test_process_place_housenumbers_duplicates(analyzer, getorcreate_hnr_id):

-    with analyzer() as a:
-        info = a.process_place({'address': {'housenumber' : '134',
-                                               'conscriptionnumber' : '134',
-                                               'streetnumber' : '99a'}})
+    with analyzer() as anl:
+        info = anl.process_place({'address': {'housenumber' : '134',
+                                              'conscriptionnumber' : '134',
+                                              'streetnumber' : '99a'}})

 
     assert set(info['hnr'].split(';')) == set(('134', '99A'))
     assert info['hnr_tokens'] == "{-1,-2}"
 
     assert set(info['hnr'].split(';')) == set(('134', '99A'))
     assert info['hnr_tokens'] == "{-1,-2}"