indexing: precompute row counts

[nominatim.git] / test / python / tokenizer / test_icu.py

diff --git a/test/python/tokenizer/test_icu.py b/test/python/tokenizer/test_icu.py
index 7f0ffce16b5a3aa8c90e228aa9abd9f57cd6097a..a2bf676699ec9a326f59957740740899da8340dc 100644 (file)
--- a/test/python/tokenizer/test_icu.py
+++ b/test/python/tokenizer/test_icu.py
@@ -1,23 +1,22 @@
-# SPDX-License-Identifier: GPL-2.0-only
+# SPDX-License-Identifier: GPL-3.0-or-later

 #
 # This file is part of Nominatim. (https://nominatim.org)
 #
 #
 # This file is part of Nominatim. (https://nominatim.org)
 #

-# Copyright (C) 2022 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 ICU tokenizer.
 """
 # For a full list of authors see the git log.
 """
 Tests for ICU tokenizer.
 """

-import shutil

 import yaml
 import itertools
 
 import pytest
 
 import yaml
 import itertools
 
 import pytest
 

-from nominatim.tokenizer import icu_tokenizer
-import nominatim.tokenizer.icu_rule_loader
-from nominatim.db import properties
-from nominatim.db.sql_preprocessor import SQLPreprocessor
-from nominatim.data.place_info import PlaceInfo
+from nominatim_db.tokenizer import icu_tokenizer
+import nominatim_db.tokenizer.icu_rule_loader
+from nominatim_db.db import properties
+from nominatim_db.db.sql_preprocessor import SQLPreprocessor
+from nominatim_db.data.place_info import PlaceInfo

 
 from mock_icu_word_table import MockIcuWordTable
 
 
 from mock_icu_word_table import MockIcuWordTable
 


@@ -32,8 +31,6 @@ def test_config(project_env, tmp_path):
     sqldir.mkdir()
     (sqldir / 'tokenizer').mkdir()
     (sqldir / 'tokenizer' / 'icu_tokenizer.sql').write_text("SELECT 'a'")
     sqldir.mkdir()
     (sqldir / 'tokenizer').mkdir()
     (sqldir / 'tokenizer' / 'icu_tokenizer.sql').write_text("SELECT 'a'")

-    shutil.copy(str(project_env.lib_dir.sql / 'tokenizer' / 'icu_tokenizer_tables.sql'),
-                str(sqldir / 'tokenizer' / 'icu_tokenizer_tables.sql'))

 
     project_env.lib_dir.sql = sqldir
 
 
     project_env.lib_dir.sql = sqldir
 


@@ -86,7 +83,7 @@ def analyzer(tokenizer_factory, test_config, monkeypatch,
             cfgstr['token-analysis'].append({'id': '@postcode',
                                              'analyzer': 'postcodes'})
         (test_config.project_dir / 'icu_tokenizer.yaml').write_text(yaml.dump(cfgstr))
             cfgstr['token-analysis'].append({'id': '@postcode',
                                              'analyzer': 'postcodes'})
         (test_config.project_dir / 'icu_tokenizer.yaml').write_text(yaml.dump(cfgstr))

-        tok.loader = nominatim.tokenizer.icu_rule_loader.ICURuleLoader(test_config)
+        tok.loader = nominatim_db.tokenizer.icu_rule_loader.ICURuleLoader(test_config)

 
         return tok.name_analyzer()
 
 
         return tok.name_analyzer()
 


@@ -160,7 +157,7 @@ def test_init_new(tokenizer_factory, test_config, db_prop):
     tok = tokenizer_factory()
     tok.init_new_db(test_config)
 
     tok = tokenizer_factory()
     tok.init_new_db(test_config)
 

-    assert db_prop(nominatim.tokenizer.icu_rule_loader.DBCFG_IMPORT_NORM_RULES) \
+    assert db_prop(nominatim_db.tokenizer.icu_rule_loader.DBCFG_IMPORT_NORM_RULES) \

             .startswith(':: lower ();')
 
 
             .startswith(':: lower ();')
 
 


@@ -202,18 +199,16 @@ def test_update_sql_functions(db_prop, temp_db_cursor,
     assert test_content == set((('1133', ), ))
 
 
     assert test_content == set((('1133', ), ))
 
 

-def test_finalize_import(tokenizer_factory, temp_db_conn,
-                         temp_db_cursor, test_config, sql_preprocessor_cfg):
-    func_file = test_config.lib_dir.sql / 'tokenizer' / 'legacy_tokenizer_indices.sql'
-    func_file.write_text("""CREATE FUNCTION test() RETURNS TEXT
-                            AS $$ SELECT 'b'::text $$ LANGUAGE SQL""")
-
+def test_finalize_import(tokenizer_factory, temp_db_cursor,
+                         test_config, sql_preprocessor_cfg):

     tok = tokenizer_factory()
     tok.init_new_db(test_config)
 
     tok = tokenizer_factory()
     tok.init_new_db(test_config)
 

+    assert not temp_db_cursor.index_exists('word', 'idx_word_word_id')
+

     tok.finalize_import(test_config)
 
     tok.finalize_import(test_config)
 

-    temp_db_cursor.scalar('SELECT test()') == 'b'
+    assert temp_db_cursor.index_exists('word', 'idx_word_word_id')

 
 
 def test_check_database(test_config, tokenizer_factory,
 
 
 def test_check_database(test_config, tokenizer_factory,


@@ -224,23 +219,28 @@ def test_check_database(test_config, tokenizer_factory,
     assert tok.check_database(test_config) is None
 
 
     assert tok.check_database(test_config) is None
 
 

-def test_update_statistics_reverse_only(word_table, tokenizer_factory):
+def test_update_statistics_reverse_only(word_table, tokenizer_factory, test_config):

     tok = tokenizer_factory()
     tok = tokenizer_factory()

-    tok.update_statistics()
+    tok.update_statistics(test_config)

 
 
 
 

-def test_update_statistics(word_table, table_factory, temp_db_cursor, tokenizer_factory):
+def test_update_statistics(word_table, table_factory, temp_db_cursor,
+                           tokenizer_factory, test_config):

     word_table.add_full_word(1000, 'hello')
     word_table.add_full_word(1000, 'hello')

+    word_table.add_full_word(1001, 'bye')

     table_factory('search_name',
     table_factory('search_name',

-                  'place_id BIGINT, name_vector INT[]',
-                  [(12, [1000])])
+                  'place_id BIGINT, name_vector INT[], nameaddress_vector INT[]',
+                  [(12, [1000], [1001])])

     tok = tokenizer_factory()
 
     tok = tokenizer_factory()
 

-    tok.update_statistics()
+    tok.update_statistics(test_config)

 
     assert temp_db_cursor.scalar("""SELECT count(*) FROM word
 
     assert temp_db_cursor.scalar("""SELECT count(*) FROM word

-                                    WHERE type = 'W' and
-                                          (info->>'count')::int > 0""") > 0
+                                    WHERE type = 'W' and word_id = 1000 and
+                                          (info->>'count')::int > 0""") == 1
+    assert temp_db_cursor.scalar("""SELECT count(*) FROM word
+                                    WHERE type = 'W' and word_id = 1001 and
+                                          (info->>'addr_count')::int > 0""") == 1

 
 
 def test_normalize_postcode(analyzer):
 
 
 def test_normalize_postcode(analyzer):


@@ -523,7 +523,7 @@ class TestPlaceAddress:
     def test_process_place_nonexisting_street(self):
         info = self.process_address(street='Grand Road')
 
     def test_process_place_nonexisting_street(self):
         info = self.process_address(street='Grand Road')
 

-        assert 'street' not in info
+        assert info['street'] == '{}'

 
 
     def test_process_place_multiple_street_tags(self):
 
 
     def test_process_place_multiple_street_tags(self):


@@ -538,7 +538,7 @@ class TestPlaceAddress:
     def test_process_place_street_empty(self):
         info = self.process_address(street='🜵')
 
     def test_process_place_street_empty(self):
         info = self.process_address(street='🜵')
 

-        assert 'street' not in info
+        assert info['street'] == '{}'

 
 
     def test_process_place_street_from_cache(self):
 
 
     def test_process_place_street_from_cache(self):


@@ -554,7 +554,7 @@ class TestPlaceAddress:
     def test_process_place_place(self):
         info = self.process_address(place='Honu Lulu')
 
     def test_process_place_place(self):
         info = self.process_address(place='Honu Lulu')
 

-        assert eval(info['place']) == self.name_token_set('HONU', 'LULU')
+        assert eval(info['place']) == self.name_token_set('HONU', 'LULU', '#HONU LULU')

 
 
     def test_process_place_place_extra(self):
 
 
     def test_process_place_place_extra(self):


@@ -574,8 +574,8 @@ class TestPlaceAddress:
                                     suburb='Zwickau', street='Hauptstr',
                                     full='right behind the church')
 
                                     suburb='Zwickau', street='Hauptstr',
                                     full='right behind the church')
 

-        city = self.name_token_set('ZWICKAU')
-        state = self.name_token_set('SACHSEN')
+        city = self.name_token_set('ZWICKAU', '#ZWICKAU')
+        state = self.name_token_set('SACHSEN', '#SACHSEN')

 
         result = {k: eval(v) for k,v in info['addr'].items()}
 
 
         result = {k: eval(v) for k,v in info['addr'].items()}
 


@@ -587,7 +587,7 @@ class TestPlaceAddress:
 
         result = {k: eval(v) for k,v in info['addr'].items()}
 
 
         result = {k: eval(v) for k,v in info['addr'].items()}
 

-        assert result == {'city': self.name_token_set('Bruxelles')}
+        assert result == {'city': self.name_token_set('Bruxelles', '#Bruxelles')}

 
 
     def test_process_place_address_terms_empty(self):
 
 
     def test_process_place_address_terms_empty(self):