From 7cb7cf848d69e818cb3cccd656c4a34713316264 Mon Sep 17 00:00:00 2001 From: Sarah Hoffmann Date: Tue, 27 Apr 2021 21:50:35 +0200 Subject: [PATCH] move amenity creation to tokenizer The BDD tests still use the old-style amenity creation scripts because we don't have simple means to import a hand-crafted test file of special phrases right now. --- lib-sql/tokenizer/legacy_tokenizer.sql | 61 ---- nominatim/clicmd/special_phrases.py | 5 +- nominatim/tokenizer/legacy_tokenizer.py | 57 +++- .../special_phrases/importer_statistics.py | 37 --- .../special_phrases_importer.py | 107 +------ test/python/dummy_tokenizer.py | 2 + test/python/test_cli.py | 2 +- .../test_tools_import_special_phrases.py | 179 +---------- test/testdb/specialphrases_testdb.sql | 289 +++++++++++------- 9 files changed, 256 insertions(+), 483 deletions(-) diff --git a/lib-sql/tokenizer/legacy_tokenizer.sql b/lib-sql/tokenizer/legacy_tokenizer.sql index 2bbf3ed7..fe82762e 100644 --- a/lib-sql/tokenizer/legacy_tokenizer.sql +++ b/lib-sql/tokenizer/legacy_tokenizer.sql @@ -202,57 +202,6 @@ $$ LANGUAGE plpgsql; -CREATE OR REPLACE FUNCTION getorcreate_amenity(lookup_word TEXT, normalized_word TEXT, - lookup_class text, lookup_type text) - RETURNS INTEGER - AS $$ -DECLARE - lookup_token TEXT; - return_word_id INTEGER; -BEGIN - lookup_token := ' '||trim(lookup_word); - SELECT min(word_id) FROM word - WHERE word_token = lookup_token and word = normalized_word - and class = lookup_class and type = lookup_type - INTO return_word_id; - IF return_word_id IS NULL THEN - return_word_id := nextval('seq_word'); - INSERT INTO word VALUES (return_word_id, lookup_token, normalized_word, - lookup_class, lookup_type, null, 0); - END IF; - RETURN return_word_id; -END; -$$ -LANGUAGE plpgsql; - - -CREATE OR REPLACE FUNCTION getorcreate_amenityoperator(lookup_word TEXT, - normalized_word TEXT, - lookup_class text, - lookup_type text, - op text) - RETURNS INTEGER - AS $$ -DECLARE - lookup_token TEXT; - return_word_id INTEGER; -BEGIN - lookup_token := ' '||trim(lookup_word); - SELECT min(word_id) FROM word - WHERE word_token = lookup_token and word = normalized_word - and class = lookup_class and type = lookup_type and operator = op - INTO return_word_id; - IF return_word_id IS NULL THEN - return_word_id := nextval('seq_word'); - INSERT INTO word VALUES (return_word_id, lookup_token, normalized_word, - lookup_class, lookup_type, null, 0, op); - END IF; - RETURN return_word_id; -END; -$$ -LANGUAGE plpgsql; - - CREATE OR REPLACE FUNCTION getorcreate_name_id(lookup_word TEXT, src_word TEXT) RETURNS INTEGER AS $$ @@ -276,16 +225,6 @@ $$ LANGUAGE plpgsql; -CREATE OR REPLACE FUNCTION getorcreate_name_id(lookup_word TEXT) - RETURNS INTEGER - AS $$ -DECLARE -BEGIN - RETURN getorcreate_name_id(lookup_word, ''); -END; -$$ -LANGUAGE plpgsql; - -- Normalize a string and lookup its word ids (partial words). CREATE OR REPLACE FUNCTION addr_ids_from_name(lookup_word TEXT) RETURNS INTEGER[] diff --git a/nominatim/clicmd/special_phrases.py b/nominatim/clicmd/special_phrases.py index 1eb25bdc..5b322158 100644 --- a/nominatim/clicmd/special_phrases.py +++ b/nominatim/clicmd/special_phrases.py @@ -22,10 +22,13 @@ class ImportSpecialPhrases: @staticmethod def run(args): + from ..tokenizer import factory as tokenizer_factory + if args.import_from_wiki: LOG.warning('Special phrases importation starting') + tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config) with connect(args.config.get_libpq_dsn()) as db_connection: SpecialPhrasesImporter( args.config, args.phplib_dir, db_connection - ).import_from_wiki() + ).import_from_wiki(tokenizer) return 0 diff --git a/nominatim/tokenizer/legacy_tokenizer.py b/nominatim/tokenizer/legacy_tokenizer.py index d4068aea..00b9f50e 100644 --- a/nominatim/tokenizer/legacy_tokenizer.py +++ b/nominatim/tokenizer/legacy_tokenizer.py @@ -6,6 +6,7 @@ import logging import re import shutil +from icu import Transliterator import psycopg2 import psycopg2.extras @@ -158,7 +159,9 @@ class LegacyTokenizer: Analyzers are not thread-safe. You need to instantiate one per thread. """ - return LegacyNameAnalyzer(self.dsn) + normalizer = Transliterator.createFromRules("phrase normalizer", + self.normalization) + return LegacyNameAnalyzer(self.dsn, normalizer) def _init_db_tables(self, config): @@ -182,7 +185,6 @@ class LegacyTokenizer: properties.set_property(conn, DBCFG_MAXWORDFREQ, config.MAX_WORD_FREQUENCY) - class LegacyNameAnalyzer: """ The legacy analyzer uses the special Postgresql module for splitting names. @@ -191,9 +193,10 @@ class LegacyNameAnalyzer: normalization. """ - def __init__(self, dsn): + def __init__(self, dsn, normalizer): self.conn = connect(dsn).connection self.conn.autocommit = True + self.normalizer = normalizer psycopg2.extras.register_hstore(self.conn) self._cache = _TokenCache(self.conn) @@ -215,6 +218,13 @@ class LegacyNameAnalyzer: self.conn = None + def normalize(self, phrase): + """ Normalize the given phrase, i.e. remove all properties that + are irrelevant for search. + """ + return self.normalizer.transliterate(phrase) + + def add_postcodes_from_db(self): """ Add postcodes from the location_postcode table to the word table. """ @@ -224,6 +234,47 @@ class LegacyNameAnalyzer: FROM location_postcode) x""") + def update_special_phrases(self, phrases): + """ Replace the search index for special phrases with the new phrases. + """ + norm_phrases = set(((self.normalize(p[0]), p[1], p[2], p[3]) + for p in phrases)) + + with self.conn.cursor() as cur: + # Get the old phrases. + existing_phrases = set() + cur.execute("""SELECT word, class, type, operator FROM word + WHERE class != 'place' + OR (type != 'house' AND type != 'postcode')""") + for label, cls, typ, oper in cur: + existing_phrases.add((label, cls, typ, oper or '-')) + + to_add = norm_phrases - existing_phrases + to_delete = existing_phrases - norm_phrases + + if to_add: + psycopg2.extras.execute_values( + cur, + """ INSERT INTO word (word_id, word_token, word, class, type, + search_name_count, operator) + (SELECT nextval('seq_word'), make_standard_name(name), name, + class, type, 0, + CASE WHEN op in ('in', 'near') THEN op ELSE null END + FROM (VALUES %s) as v(name, class, type, op))""", + to_add) + + if to_delete: + psycopg2.extras.execute_values( + cur, + """ DELETE FROM word USING (VALUES %s) as v(name, in_class, in_type, op) + WHERE word = name and class = in_class and type = in_type + and ((op = '-' and operator is null) or op = operator)""", + to_delete) + + LOG.info("Total phrases: %s. Added: %s. Deleted: %s", + len(norm_phrases), len(to_add), len(to_delete)) + + def add_country_names(self, country_code, names): """ Add names for the given country to the search index. """ diff --git a/nominatim/tools/special_phrases/importer_statistics.py b/nominatim/tools/special_phrases/importer_statistics.py index 46965c4b..9b97bca6 100644 --- a/nominatim/tools/special_phrases/importer_statistics.py +++ b/nominatim/tools/special_phrases/importer_statistics.py @@ -24,9 +24,6 @@ class SpecialPhrasesImporterStatistics(): self.tables_deleted = 0 self.tables_ignored = 0 self.global_phrases_invalid = 0 - self.global_phrases_added = 0 - self.global_phrases_ignored = 0 - self.global_phrases_deleted = 0 def _set_lang_values_to_0(self): """ @@ -34,8 +31,6 @@ class SpecialPhrasesImporterStatistics(): lang to 0. """ self.lang_phrases_invalid = 0 - self.lang_phrases_added = 0 - self.lang_phrases_ignored = 0 def notify_one_phrase_invalid(self): """ @@ -45,29 +40,6 @@ class SpecialPhrasesImporterStatistics(): self.lang_phrases_invalid += 1 self.global_phrases_invalid += 1 - def notify_one_phrase_added(self): - """ - Add +1 to the count of entries - added to the db. - """ - self.lang_phrases_added += 1 - self.global_phrases_added += 1 - - def notify_one_phrase_ignored(self): - """ - Add +1 to the count of ignored - entries as it was already in the db. - """ - self.lang_phrases_ignored += 1 - self.global_phrases_ignored += 1 - - def notify_one_phrase_deleted(self): - """ - Add +1 to the count of phrases deleted - from the database. - """ - self.global_phrases_deleted += 1 - def notify_one_table_created(self): """ Add +1 to the count of created tables. @@ -97,12 +69,6 @@ class SpecialPhrasesImporterStatistics(): LOG.info('- %s phrases were invalid.', self.global_phrases_invalid) if self.global_phrases_invalid > 0: LOG.info(' Those invalid phrases have been skipped.') - LOG.info('- %s phrases were ignored as they are already in the database', - self.global_phrases_ignored) - LOG.info('- %s phrases were added to the database', self.global_phrases_added) - LOG.info('- %s phrases were deleted from the database', self.global_phrases_deleted) - if self.global_phrases_deleted > 0: - LOG.info(' They were deleted as they are not valid anymore.') LOG.info('- %s tables were ignored as they already exist on the database', self.tables_ignored) LOG.info('- %s tables were created', self.tables_created) @@ -126,9 +92,6 @@ class SpecialPhrasesImporterStatistics(): LOG.info('- %s phrases were invalid.', self.lang_phrases_invalid) if self.lang_phrases_invalid > 0: LOG.info(' Those invalid phrases have been skipped.') - LOG.info('- %s phrases were ignored as they are already in the database', - self.lang_phrases_ignored) - LOG.info('- %s phrases were added to the database', self.lang_phrases_added) LOG.info('====================================================================') if self.lang_phrases_invalid > 0: diff --git a/nominatim/tools/special_phrases/special_phrases_importer.py b/nominatim/tools/special_phrases/special_phrases_importer.py index a8e780c3..9649f94b 100644 --- a/nominatim/tools/special_phrases/special_phrases_importer.py +++ b/nominatim/tools/special_phrases/special_phrases_importer.py @@ -9,7 +9,6 @@ import re import subprocess import json -from icu import Transliterator from psycopg2.sql import Identifier, Literal, SQL from nominatim.tools.exec_utils import get_url @@ -33,21 +32,14 @@ class SpecialPhrasesImporter(): r'\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([\-YN])' ) self.sanity_check_pattern = re.compile(r'^\w+$') - self.transliterator = Transliterator.createFromRules("special-phrases normalizer", - self.config.TERM_NORMALIZATION) - #This set will contain all existing phrases from the word table which - #no longer exist on the wiki. - #It contain tuples with the following format: (normalized_word, class, type, operator) - self.words_phrases_to_delete = set() - #This set will contain the phrases which still exist from the wiki. - #It is used to prevent duplicates on the wiki by removing them from - #the word_phrases_to_delete only at the end. - self.words_phrases_still_exist = set() + # This set will contain all existing phrases to be added. + # It contains tuples with the following format: (lable, class, type, operator) + self.word_phrases = set() #This set will contain all existing place_classtype tables which doesn't match any #special phrases class/type on the wiki. self.table_phrases_to_delete = set() - def import_from_wiki(self, languages=None): + def import_from_wiki(self, tokenizer, languages=None): """ Iterate through all specified languages and extract corresponding special phrases from the wiki. @@ -55,7 +47,6 @@ class SpecialPhrasesImporter(): if languages is not None and not isinstance(languages, list): raise TypeError('The \'languages\' argument should be of type list.') - self._fetch_existing_words_phrases() self._fetch_existing_place_classtype_tables() #Get all languages to process. @@ -71,30 +62,15 @@ class SpecialPhrasesImporter(): self.statistics_handler.notify_current_lang_done(lang) self._create_place_classtype_table_and_indexes(class_type_pairs) - self._remove_non_existent_phrases_from_db() + self._remove_non_existent_tables_from_db() self.db_connection.commit() + + with tokenizer.name_analyzer() as analyzer: + analyzer.update_special_phrases(self.word_phrases) + LOG.warning('Import done.') self.statistics_handler.notify_import_done() - def _fetch_existing_words_phrases(self): - """ - Fetch existing special phrases from the word table. - Fill the word_phrases_to_delete set of the class. - """ - #Only extract special phrases terms: - #If class=place and type=house then it is a housenumber term. - #If class=place and type=postcode then it is a postcode term. - word_query = """ - SELECT word, class, type, operator FROM word - WHERE class != 'place' OR (type != 'house' AND type != 'postcode') - """ - with self.db_connection.cursor() as db_cursor: - db_cursor.execute(SQL(word_query)) - for row in db_cursor: - row[3] = '-' if row[3] is None else row[3] - self.words_phrases_to_delete.add( - (row[0], row[1], row[2], row[3]) - ) def _fetch_existing_place_classtype_tables(self): """ @@ -176,7 +152,6 @@ class SpecialPhrasesImporter(): for match in matches: phrase_label = match[0].strip() - normalized_label = self.transliterator.transliterate(phrase_label) phrase_class = match[1].strip() phrase_type = match[2].strip() phrase_operator = match[3].strip() @@ -198,20 +173,6 @@ class SpecialPhrasesImporter(): ): continue - #Check if the phrase already exists in the database. - if ( - (normalized_label, phrase_class, phrase_type, phrase_operator) - in self.words_phrases_to_delete - ): - #Remove this phrase from the ones to delete as it still exist on the wiki. - self.words_phrases_still_exist.add( - (normalized_label, phrase_class, phrase_type, phrase_operator) - ) - class_type_pairs.add((phrase_class, phrase_type)) - self.statistics_handler.notify_one_phrase_ignored() - #Dont need to add this phrase as it already exists in the word table. - continue - #sanity check, in case somebody added garbage in the wiki if not self._check_sanity(lang, phrase_class, phrase_type): self.statistics_handler.notify_one_phrase_invalid() @@ -219,35 +180,11 @@ class SpecialPhrasesImporter(): class_type_pairs.add((phrase_class, phrase_type)) - self._process_amenity( - phrase_label, normalized_label, phrase_class, - phrase_type, phrase_operator - ) - self.statistics_handler.notify_one_phrase_added() + self.word_phrases.add((phrase_label, phrase_class, + phrase_type, phrase_operator)) return class_type_pairs - def _process_amenity(self, phrase_label, normalized_label, - phrase_class, phrase_type, phrase_operator): - # pylint: disable-msg=too-many-arguments - """ - Add phrase lookup and corresponding class and - type to the word table based on the operator. - """ - with self.db_connection.cursor() as db_cursor: - if phrase_operator == 'near': - db_cursor.execute("""SELECT getorcreate_amenityoperator( - make_standard_name(%s), %s, %s, %s, 'near')""", - (phrase_label, normalized_label, phrase_class, phrase_type)) - elif phrase_operator == 'in': - db_cursor.execute("""SELECT getorcreate_amenityoperator( - make_standard_name(%s), %s, %s, %s, 'in')""", - (phrase_label, normalized_label, phrase_class, phrase_type)) - else: - db_cursor.execute("""SELECT getorcreate_amenity( - make_standard_name(%s), %s, %s, %s)""", - (phrase_label, normalized_label, phrase_class, phrase_type)) - def _create_place_classtype_table_and_indexes(self, class_type_pairs): """ @@ -339,33 +276,15 @@ class SpecialPhrasesImporter(): .format(Identifier(table_name), Identifier(self.config.DATABASE_WEBUSER))) - def _remove_non_existent_phrases_from_db(self): + def _remove_non_existent_tables_from_db(self): """ Remove special phrases which doesn't exist on the wiki anymore. - Delete from the word table and delete the place_classtype tables. + Delete the place_classtype tables. """ LOG.warning('Cleaning database...') - self.words_phrases_to_delete = self.words_phrases_to_delete - self.words_phrases_still_exist #Array containing all queries to execute. Contain tuples of format (query, parameters) queries_parameters = [] - #Delete phrases from the word table which are not on the wiki anymore. - for phrase_to_delete in self.words_phrases_to_delete: - self.statistics_handler.notify_one_phrase_deleted() - if phrase_to_delete[3] == '-': - query = """ - DELETE FROM word WHERE word = %s AND class = %s AND type = %s AND operator IS null - """ - parameters = (phrase_to_delete[0], phrase_to_delete[1], phrase_to_delete[2], ) - queries_parameters.append((query, parameters)) - else: - query = """ - DELETE FROM word WHERE word = %s AND class = %s AND type = %s AND operator = %s - """ - parameters = (phrase_to_delete[0], phrase_to_delete[1], - phrase_to_delete[2], phrase_to_delete[3], ) - queries_parameters.append((query, parameters)) - #Delete place_classtype tables corresponding to class/type which are not on the wiki anymore for table in self.table_phrases_to_delete: self.statistics_handler.notify_one_table_deleted() diff --git a/test/python/dummy_tokenizer.py b/test/python/dummy_tokenizer.py index 79197dfb..0868b57d 100644 --- a/test/python/dummy_tokenizer.py +++ b/test/python/dummy_tokenizer.py @@ -50,6 +50,8 @@ class DummyNameAnalyzer: def add_postcodes_from_db(self): pass + def update_special_phrases(self, phrases): + self.analyser_cache['special_phrases'] = phrases def add_country_names(self, code, names): self.analyser_cache['countries'].append((code, names)) diff --git a/test/python/test_cli.py b/test/python/test_cli.py index 4cca3080..e0d4fb86 100644 --- a/test/python/test_cli.py +++ b/test/python/test_cli.py @@ -248,7 +248,7 @@ def test_index_command(mock_func_factory, temp_db_cursor, tokenizer_mock, assert bnd_mock.called == do_bnds assert rank_mock.called == do_ranks -def test_special_phrases_command(temp_db, mock_func_factory): +def test_special_phrases_command(temp_db, mock_func_factory, tokenizer_mock): func = mock_func_factory(nominatim.clicmd.special_phrases.SpecialPhrasesImporter, 'import_from_wiki') call_nominatim('special-phrases', '--import-from-wiki') diff --git a/test/python/test_tools_import_special_phrases.py b/test/python/test_tools_import_special_phrases.py index cb82f6b7..24b3318d 100644 --- a/test/python/test_tools_import_special_phrases.py +++ b/test/python/test_tools_import_special_phrases.py @@ -11,41 +11,6 @@ from nominatim.tools import SpecialPhrasesImporter TEST_BASE_DIR = Path(__file__) / '..' / '..' -def test_fetch_existing_words_phrases_basic(special_phrases_importer, word_table, - temp_db_cursor): - """ - Check for the fetch_existing_words_phrases() method. - It should return special phrase term added to the word - table. - """ - query =""" - INSERT INTO word VALUES(99999, 'lookup_token', 'normalized_word', - 'class', 'type', null, 0, 'near'); - """ - temp_db_cursor.execute(query) - - assert not special_phrases_importer.words_phrases_to_delete - special_phrases_importer._fetch_existing_words_phrases() - contained_phrase = special_phrases_importer.words_phrases_to_delete.pop() - assert contained_phrase == ('normalized_word', 'class', 'type', 'near') - -@pytest.mark.parametrize("house_type", ['house', 'postcode']) -def test_fetch_existing_words_phrases_special_cases(special_phrases_importer, word_table, - house_type, temp_db_cursor): - """ - Check for the fetch_existing_words_phrases() method. - It should return nothing as the terms added correspond - to a housenumber and postcode term. - """ - query =""" - INSERT INTO word VALUES(99999, 'lookup_token', 'normalized_word', - 'place', %s, null, 0, 'near'); - """ - temp_db_cursor.execute(query, (house_type,)) - - special_phrases_importer._fetch_existing_words_phrases() - assert not special_phrases_importer.words_phrases_to_delete - def test_fetch_existing_place_classtype_tables(special_phrases_importer, temp_db_cursor): """ Check for the fetch_existing_place_classtype_tables() method. @@ -118,41 +83,11 @@ def test_convert_settings_giving_json(special_phrases_importer): the same path is directly returned """ json_file = (TEST_BASE_DIR / 'testfiles' / 'phrase_settings.json').resolve() - + returned = special_phrases_importer._convert_php_settings_if_needed(json_file) assert returned == json_file -def test_process_amenity_with_operator(special_phrases_importer, getorcreate_amenityoperator_funcs, - temp_db_conn, word_table): - """ - Test that _process_amenity() execute well the - getorcreate_amenityoperator() SQL function and that - the 2 differents operators are well handled. - """ - special_phrases_importer._process_amenity('', '', '', '', 'near') - special_phrases_importer._process_amenity('', '', '', '', 'in') - - with temp_db_conn.cursor() as temp_db_cursor: - temp_db_cursor.execute("SELECT * FROM word WHERE operator='near' OR operator='in'") - results = temp_db_cursor.fetchall() - - assert len(results) == 2 - -def test_process_amenity_without_operator(special_phrases_importer, getorcreate_amenity_funcs, - temp_db_conn, word_table): - """ - Test that _process_amenity() execute well the - getorcreate_amenity() SQL function. - """ - special_phrases_importer._process_amenity('', '', '', '', '') - - with temp_db_conn.cursor() as temp_db_cursor: - temp_db_cursor.execute("SELECT * FROM word WHERE operator='no_operator'") - result = temp_db_cursor.fetchone() - - assert result - def test_create_place_classtype_indexes(temp_db_conn, special_phrases_importer): """ Test that _create_place_classtype_indexes() create the @@ -215,8 +150,7 @@ def test_create_place_classtype_table_and_indexes( assert check_placeid_and_centroid_indexes(temp_db_conn, pair[0], pair[1]) assert check_grant_access(temp_db_conn, def_config.DATABASE_WEBUSER, pair[0], pair[1]) -def test_process_xml_content(temp_db_conn, def_config, special_phrases_importer, word_table, - getorcreate_amenity_funcs, getorcreate_amenityoperator_funcs): +def test_process_xml_content(temp_db_conn, def_config, special_phrases_importer): """ Test that _process_xml_content() process the given xml content right by executing the right SQL functions for amenities and @@ -228,11 +162,9 @@ def test_process_xml_content(temp_db_conn, def_config, special_phrases_importer, #Converted output set to a dict for easy assert further. results = dict(special_phrases_importer._process_xml_content(get_test_xml_wiki_content(), 'en')) - assert check_amenities_with_op(temp_db_conn) - assert check_amenities_without_op(temp_db_conn) assert results[class_test] and type_test in results.values() -def test_remove_non_existent_phrases_from_db(special_phrases_importer, default_phrases, +def test_remove_non_existent_tables_from_db(special_phrases_importer, default_phrases, temp_db_conn): """ Check for the remove_non_existent_phrases_from_db() method. @@ -245,22 +177,10 @@ def test_remove_non_existent_phrases_from_db(special_phrases_importer, default_p be deleted. """ with temp_db_conn.cursor() as temp_db_cursor: - to_delete_phrase_tuple = ('normalized_word', 'class', 'type', 'near') - to_keep_phrase_tuple = ( - 'normalized_word_exists', 'class_exists', 'type_exists', 'near' - ) - special_phrases_importer.words_phrases_to_delete = { - to_delete_phrase_tuple, - to_keep_phrase_tuple - } - special_phrases_importer.words_phrases_still_exist = { - to_keep_phrase_tuple - } special_phrases_importer.table_phrases_to_delete = { 'place_classtype_testclasstypetable_to_delete' } - query_words = 'SELECT word, class, type, operator FROM word;' query_tables = """ SELECT table_name FROM information_schema.tables @@ -268,21 +188,16 @@ def test_remove_non_existent_phrases_from_db(special_phrases_importer, default_p AND table_name like 'place_classtype_%'; """ - special_phrases_importer._remove_non_existent_phrases_from_db() + special_phrases_importer._remove_non_existent_tables_from_db() - temp_db_cursor.execute(query_words) - words_result = temp_db_cursor.fetchall() temp_db_cursor.execute(query_tables) tables_result = temp_db_cursor.fetchall() - assert len(words_result) == 1 and words_result[0] == [ - 'normalized_word_exists', 'class_exists', 'type_exists', 'near' - ] assert (len(tables_result) == 1 and tables_result[0][0] == 'place_classtype_testclasstypetable_to_keep' ) -def test_import_from_wiki(monkeypatch, temp_db_conn, def_config, special_phrases_importer, placex_table, - getorcreate_amenity_funcs, getorcreate_amenityoperator_funcs, word_table): +def test_import_from_wiki(monkeypatch, temp_db_conn, def_config, special_phrases_importer, + placex_table, tokenizer_mock): """ Check that the main import_from_wiki() method is well executed. It should create the place_classtype table, the place_id and centroid indexes, @@ -294,17 +209,14 @@ def test_import_from_wiki(monkeypatch, temp_db_conn, def_config, special_phrases #what is deleted and what is preserved. with temp_db_conn.cursor() as temp_db_cursor: temp_db_cursor.execute(""" - INSERT INTO word VALUES(99999, ' animal shelter', 'animal shelter', - 'amenity', 'animal_shelter', null, 0, null); - - INSERT INTO word VALUES(99999, ' wrong_lookup_token', 'wrong_normalized_word', - 'wrong_class', 'wrong_type', null, 0, 'near'); - CREATE TABLE place_classtype_amenity_animal_shelter(); CREATE TABLE place_classtype_wrongclass_wrongtype();""") monkeypatch.setattr('nominatim.tools.SpecialPhrasesImporter._get_wiki_content', mock_get_wiki_content) - special_phrases_importer.import_from_wiki(['en']) + tokenizer = tokenizer_mock() + special_phrases_importer.import_from_wiki(tokenizer, ['en']) + + assert len(tokenizer.analyser_cache['special_phrases']) == 18 class_test = 'aerialway' type_test = 'zip_line' @@ -312,22 +224,12 @@ def test_import_from_wiki(monkeypatch, temp_db_conn, def_config, special_phrases assert check_table_exist(temp_db_conn, class_test, type_test) assert check_placeid_and_centroid_indexes(temp_db_conn, class_test, type_test) assert check_grant_access(temp_db_conn, def_config.DATABASE_WEBUSER, class_test, type_test) - assert check_amenities_with_op(temp_db_conn) - assert check_amenities_without_op(temp_db_conn) assert check_table_exist(temp_db_conn, 'amenity', 'animal_shelter') assert not check_table_exist(temp_db_conn, 'wrong_class', 'wrong_type') #Format (query, should_return_something_bool) use to easily execute all asserts queries_tests = set() - #Used to check that the correct phrase already in the word table before is still there. - query_correct_word = "SELECT * FROM word WHERE word = 'animal shelter'" - queries_tests.add((query_correct_word, True)) - - #Used to check if wrong phrase was deleted from the word table of the database. - query_wrong_word = "SELECT word FROM word WHERE word = 'wrong_normalized_word'" - queries_tests.add((query_wrong_word, False)) - #Used to check that correct place_classtype table already in the datase before is still there. query_existing_table = """ SELECT table_name @@ -412,24 +314,6 @@ def check_placeid_and_centroid_indexes(temp_db_conn, phrase_class, phrase_type): temp_db_conn.index_exists(index_prefix + 'place_id') ) -def check_amenities_with_op(temp_db_conn): - """ - Check that the test table for the SQL function getorcreate_amenityoperator() - contains more than one value (so that the SQL function was call more than one time). - """ - with temp_db_conn.cursor() as temp_db_cursor: - temp_db_cursor.execute("SELECT * FROM word WHERE operator != 'no_operator'") - return len(temp_db_cursor.fetchall()) > 1 - -def check_amenities_without_op(temp_db_conn): - """ - Check that the test table for the SQL function getorcreate_amenity() - contains more than one value (so that the SQL function was call more than one time). - """ - with temp_db_conn.cursor() as temp_db_cursor: - temp_db_cursor.execute("SELECT * FROM word WHERE operator = 'no_operator'") - return len(temp_db_cursor.fetchall()) > 1 - @pytest.fixture def special_phrases_importer(temp_db_conn, def_config, temp_phplib_dir_with_migration): """ @@ -453,48 +337,7 @@ def temp_phplib_dir_with_migration(): yield Path(phpdir) @pytest.fixture -def default_phrases(word_table, temp_db_cursor): +def default_phrases(temp_db_cursor): temp_db_cursor.execute(""" - INSERT INTO word VALUES(99999, 'lookup_token', 'normalized_word', - 'class', 'type', null, 0, 'near'); - - INSERT INTO word VALUES(99999, 'lookup_token', 'normalized_word_exists', - 'class_exists', 'type_exists', null, 0, 'near'); - CREATE TABLE place_classtype_testclasstypetable_to_delete(); CREATE TABLE place_classtype_testclasstypetable_to_keep();""") - -@pytest.fixture -def make_strandard_name_func(temp_db_cursor): - temp_db_cursor.execute(""" - CREATE OR REPLACE FUNCTION make_standard_name(name TEXT) RETURNS TEXT AS $$ - BEGIN - RETURN trim(name); --Basically return only the trimed name for the tests - END; - $$ LANGUAGE plpgsql IMMUTABLE;""") - -@pytest.fixture -def getorcreate_amenity_funcs(temp_db_cursor, make_strandard_name_func): - temp_db_cursor.execute(""" - CREATE OR REPLACE FUNCTION getorcreate_amenity(lookup_word TEXT, normalized_word TEXT, - lookup_class text, lookup_type text) - RETURNS void as $$ - BEGIN - INSERT INTO word VALUES(null, lookup_word, normalized_word, - lookup_class, lookup_type, null, 0, 'no_operator'); - END; - $$ LANGUAGE plpgsql""") - -@pytest.fixture -def getorcreate_amenityoperator_funcs(temp_db_cursor, make_strandard_name_func): - temp_db_cursor.execute(""" - CREATE TABLE temp_with_operator(op TEXT); - - CREATE OR REPLACE FUNCTION getorcreate_amenityoperator(lookup_word TEXT, normalized_word TEXT, - lookup_class text, lookup_type text, op text) - RETURNS void as $$ - BEGIN - INSERT INTO word VALUES(null, lookup_word, normalized_word, - lookup_class, lookup_type, null, 0, op); - END; - $$ LANGUAGE plpgsql""") \ No newline at end of file diff --git a/test/testdb/specialphrases_testdb.sql b/test/testdb/specialphrases_testdb.sql index b3b5d76d..7e72076e 100644 --- a/test/testdb/specialphrases_testdb.sql +++ b/test/testdb/specialphrases_testdb.sql @@ -1,120 +1,170 @@ -SELECT getorcreate_amenity(make_standard_name('Aerodrome'), 'aerodrome', 'aeroway', 'aerodrome'); -SELECT getorcreate_amenity(make_standard_name('Aerodromes'), 'aerodromes', 'aeroway', 'aerodrome'); -SELECT getorcreate_amenityoperator(make_standard_name('Aerodrome in'), 'aerodrome in', 'aeroway', 'aerodrome', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Aerodromes in'), 'aerodromes in', 'aeroway', 'aerodrome', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Aerodrome near'), 'aerodrome near', 'aeroway', 'aerodrome', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Aerodromes near'), 'aerodromes near', 'aeroway', 'aerodrome', 'near'); -SELECT getorcreate_amenity(make_standard_name('Airport'), 'airport', 'aeroway', 'aerodrome'); -SELECT getorcreate_amenity(make_standard_name('Airports'), 'airports', 'aeroway', 'aerodrome'); -SELECT getorcreate_amenityoperator(make_standard_name('Airport in'), 'airport in', 'aeroway', 'aerodrome', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Airports in'), 'airports in', 'aeroway', 'aerodrome', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Airport near'), 'airport near', 'aeroway', 'aerodrome', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Airports near'), 'airports near', 'aeroway', 'aerodrome', 'near'); -SELECT getorcreate_amenity(make_standard_name('Bar'), 'bar', 'amenity', 'bar'); -SELECT getorcreate_amenity(make_standard_name('Bars'), 'bars', 'amenity', 'bar'); -SELECT getorcreate_amenityoperator(make_standard_name('Bar in'), 'bar in', 'amenity', 'bar', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Bars in'), 'bars in', 'amenity', 'bar', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Bar near'), 'bar near', 'amenity', 'bar', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Bars near'), 'bars near', 'amenity', 'bar', 'near'); -SELECT getorcreate_amenity(make_standard_name('Bar'), 'bar', 'amenity', 'pub'); -SELECT getorcreate_amenity(make_standard_name('Bars'), 'bars', 'amenity', 'pub'); -SELECT getorcreate_amenityoperator(make_standard_name('Bar in'), 'bar in', 'amenity', 'pub', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Bars in'), 'bars in', 'amenity', 'pub', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Bar near'), 'bar near', 'amenity', 'pub', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Bars near'), 'bars near', 'amenity', 'pub', 'near'); -SELECT getorcreate_amenity(make_standard_name('Food'), 'food', 'amenity', 'restaurant'); -SELECT getorcreate_amenity(make_standard_name('Food'), 'food', 'amenity', 'restaurant'); -SELECT getorcreate_amenityoperator(make_standard_name('Food in'), 'food in', 'amenity', 'restaurant', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Food in'), 'food in', 'amenity', 'restaurant', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Food near'), 'food near', 'amenity', 'restaurant', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Food near'), 'food near', 'amenity', 'restaurant', 'near'); -SELECT getorcreate_amenity(make_standard_name('Pub'), 'pub', 'amenity', 'bar'); -SELECT getorcreate_amenity(make_standard_name('Pubs'), 'pubs', 'amenity', 'bar'); -SELECT getorcreate_amenityoperator(make_standard_name('Pub in'), 'pub in', 'amenity', 'bar', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Pubs in'), 'pubs in', 'amenity', 'bar', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Pub near'), 'pub near', 'amenity', 'bar', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Pubs near'), 'pubs near', 'amenity', 'bar', 'near'); -SELECT getorcreate_amenity(make_standard_name('Pub'), 'pub', 'amenity', 'pub'); -SELECT getorcreate_amenity(make_standard_name('Pubs'), 'pubs', 'amenity', 'pub'); -SELECT getorcreate_amenityoperator(make_standard_name('Pub in'), 'pub in', 'amenity', 'pub', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Pubs in'), 'pubs in', 'amenity', 'pub', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Pub near'), 'pub near', 'amenity', 'pub', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Pubs near'), 'pubs near', 'amenity', 'pub', 'near'); -SELECT getorcreate_amenity(make_standard_name('Restaurant'), 'restaurant', 'amenity', 'restaurant'); -SELECT getorcreate_amenity(make_standard_name('Restaurants'), 'restaurants', 'amenity', 'restaurant'); -SELECT getorcreate_amenityoperator(make_standard_name('Restaurant in'), 'restaurant in', 'amenity', 'restaurant', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Restaurants in'), 'restaurants in', 'amenity', 'restaurant', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Restaurant near'), 'restaurant near', 'amenity', 'restaurant', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Restaurants near'), 'restaurants near', 'amenity', 'restaurant', 'near'); -SELECT getorcreate_amenity(make_standard_name('Mural'), 'mural', 'artwork_type', 'mural'); -SELECT getorcreate_amenity(make_standard_name('Murals'), 'murals', 'artwork_type', 'mural'); -SELECT getorcreate_amenityoperator(make_standard_name('Mural in'), 'mural in', 'artwork_type', 'mural', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Murals in'), 'murals in', 'artwork_type', 'mural', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Mural near'), 'mural near', 'artwork_type', 'mural', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Murals near'), 'murals near', 'artwork_type', 'mural', 'near'); -SELECT getorcreate_amenity(make_standard_name('Sculpture'), 'sculpture', 'artwork_type', 'sculpture'); -SELECT getorcreate_amenity(make_standard_name('Sculptures'), 'sculptures', 'artwork_type', 'sculpture'); -SELECT getorcreate_amenityoperator(make_standard_name('Sculpture in'), 'sculpture in', 'artwork_type', 'sculpture', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Sculptures in'), 'sculptures in', 'artwork_type', 'sculpture', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Sculpture near'), 'sculpture near', 'artwork_type', 'sculpture', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Sculptures near'), 'sculptures near', 'artwork_type', 'sculpture', 'near'); -SELECT getorcreate_amenity(make_standard_name('Statue'), 'statue', 'artwork_type', 'statue'); -SELECT getorcreate_amenity(make_standard_name('Statues'), 'statues', 'artwork_type', 'statue'); -SELECT getorcreate_amenityoperator(make_standard_name('Statue in'), 'statue in', 'artwork_type', 'statue', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Statues in'), 'statues in', 'artwork_type', 'statue', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Statue near'), 'statue near', 'artwork_type', 'statue', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Statues near'), 'statues near', 'artwork_type', 'statue', 'near'); -SELECT getorcreate_amenity(make_standard_name('ATM'), 'atm', 'atm', 'yes'); -SELECT getorcreate_amenity(make_standard_name('ATMs'), 'atms', 'atm', 'yes'); -SELECT getorcreate_amenityoperator(make_standard_name('ATM in'), 'atm in', 'atm', 'yes', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('ATMs in'), 'atms in', 'atm', 'yes', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('ATM near'), 'atm near', 'atm', 'yes', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('ATMs near'), 'atms near', 'atm', 'yes', 'near'); -SELECT getorcreate_amenity(make_standard_name('National Park'), 'national park', 'boundary', 'national_park'); -SELECT getorcreate_amenity(make_standard_name('National Parks'), 'national parks', 'boundary', 'national_park'); -SELECT getorcreate_amenityoperator(make_standard_name('National Park in'), 'national park in', 'boundary', 'national_park', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('National Parks in'), 'national parks in', 'boundary', 'national_park', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('National Park near'), 'national park near', 'boundary', 'national_park', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('National Parks near'), 'national parks near', 'boundary', 'national_park', 'near'); -SELECT getorcreate_amenity(make_standard_name('Changing table'), 'changing table', 'changing_table', 'yes'); -SELECT getorcreate_amenity(make_standard_name('Changing tables'), 'changing tables', 'changing_table', 'yes'); -SELECT getorcreate_amenityoperator(make_standard_name('Changing table in'), 'changing table in', 'changing_table', 'yes', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Changing tables in'), 'changing tables in', 'changing_table', 'yes', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Changing table near'), 'changing table near', 'changing_table', 'yes', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Changing tables near'), 'changing tables near', 'changing_table', 'yes', 'near'); -SELECT getorcreate_amenity(make_standard_name('Roundabout'), 'roundabout', 'junction', 'roundabout'); -SELECT getorcreate_amenity(make_standard_name('Roundabouts'), 'roundabouts', 'junction', 'roundabout'); -SELECT getorcreate_amenityoperator(make_standard_name('Roundabout in'), 'roundabout in', 'junction', 'roundabout', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Roundabouts in'), 'roundabouts in', 'junction', 'roundabout', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Roundabout near'), 'roundabout near', 'junction', 'roundabout', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Roundabouts near'), 'roundabouts near', 'junction', 'roundabout', 'near'); -SELECT getorcreate_amenity(make_standard_name('Plaque'), 'plaque', 'memorial', 'plaque'); -SELECT getorcreate_amenity(make_standard_name('Plaques'), 'plaques', 'memorial', 'plaque'); -SELECT getorcreate_amenityoperator(make_standard_name('Plaque in'), 'plaque in', 'memorial', 'plaque', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Plaques in'), 'plaques in', 'memorial', 'plaque', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Plaque near'), 'plaque near', 'memorial', 'plaque', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Plaques near'), 'plaques near', 'memorial', 'plaque', 'near'); -SELECT getorcreate_amenity(make_standard_name('Statue'), 'statue', 'memorial', 'statue'); -SELECT getorcreate_amenity(make_standard_name('Statues'), 'statues', 'memorial', 'statue'); -SELECT getorcreate_amenityoperator(make_standard_name('Statue in'), 'statue in', 'memorial', 'statue', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Statues in'), 'statues in', 'memorial', 'statue', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Statue near'), 'statue near', 'memorial', 'statue', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Statues near'), 'statues near', 'memorial', 'statue', 'near'); -SELECT getorcreate_amenity(make_standard_name('Stolperstein'), 'stolperstein', 'memorial', 'stolperstein'); -SELECT getorcreate_amenity(make_standard_name('Stolpersteins'), 'stolpersteins', 'memorial', 'stolperstein'); -SELECT getorcreate_amenity(make_standard_name('Stolpersteine'), 'stolpersteine', 'memorial', 'stolperstein'); -SELECT getorcreate_amenityoperator(make_standard_name('Stolperstein in'), 'stolperstein in', 'memorial', 'stolperstein', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Stolpersteins in'), 'stolpersteins in', 'memorial', 'stolperstein', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Stolpersteine in'), 'stolpersteine in', 'memorial', 'stolperstein', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('Stolperstein near'), 'stolperstein near', 'memorial', 'stolperstein', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Stolpersteins near'), 'stolpersteins near', 'memorial', 'stolperstein', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('Stolpersteine near'), 'stolpersteine near', 'memorial', 'stolperstein', 'near'); -SELECT getorcreate_amenity(make_standard_name('War Memorial'), 'war memorial', 'memorial', 'war_memorial'); -SELECT getorcreate_amenity(make_standard_name('War Memorials'), 'war memorials', 'memorial', 'war_memorial'); -SELECT getorcreate_amenityoperator(make_standard_name('War Memorial in'), 'war memorial in', 'memorial', 'war_memorial', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('War Memorials in'), 'war memorials in', 'memorial', 'war_memorial', 'in'); -SELECT getorcreate_amenityoperator(make_standard_name('War Memorial near'), 'war memorial near', 'memorial', 'war_memorial', 'near'); -SELECT getorcreate_amenityoperator(make_standard_name('War Memorials near'), 'war memorials near', 'memorial', 'war_memorial', 'near'); +CREATE OR REPLACE FUNCTION test_getorcreate_amenity(lookup_word TEXT, normalized_word TEXT, + lookup_class text, lookup_type text) + RETURNS INTEGER + AS $$ +DECLARE + lookup_token TEXT; + return_word_id INTEGER; +BEGIN + lookup_token := ' '||trim(lookup_word); + SELECT min(word_id) FROM word + WHERE word_token = lookup_token and word = normalized_word + and class = lookup_class and type = lookup_type + INTO return_word_id; + IF return_word_id IS NULL THEN + return_word_id := nextval('seq_word'); + INSERT INTO word VALUES (return_word_id, lookup_token, normalized_word, + lookup_class, lookup_type, null, 0); + END IF; + RETURN return_word_id; +END; +$$ +LANGUAGE plpgsql; + + +CREATE OR REPLACE FUNCTION test_getorcreate_amenityoperator(lookup_word TEXT, + normalized_word TEXT, + lookup_class text, + lookup_type text, + op text) + RETURNS INTEGER + AS $$ +DECLARE + lookup_token TEXT; + return_word_id INTEGER; +BEGIN + lookup_token := ' '||trim(lookup_word); + SELECT min(word_id) FROM word + WHERE word_token = lookup_token and word = normalized_word + and class = lookup_class and type = lookup_type and operator = op + INTO return_word_id; + IF return_word_id IS NULL THEN + return_word_id := nextval('seq_word'); + INSERT INTO word VALUES (return_word_id, lookup_token, normalized_word, + lookup_class, lookup_type, null, 0, op); + END IF; + RETURN return_word_id; +END; +$$ +LANGUAGE plpgsql; + +SELECT test_getorcreate_amenity(make_standard_name('Aerodrome'), 'aerodrome', 'aeroway', 'aerodrome'); +SELECT test_getorcreate_amenity(make_standard_name('Aerodromes'), 'aerodromes', 'aeroway', 'aerodrome'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Aerodrome in'), 'aerodrome in', 'aeroway', 'aerodrome', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Aerodromes in'), 'aerodromes in', 'aeroway', 'aerodrome', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Aerodrome near'), 'aerodrome near', 'aeroway', 'aerodrome', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Aerodromes near'), 'aerodromes near', 'aeroway', 'aerodrome', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Airport'), 'airport', 'aeroway', 'aerodrome'); +SELECT test_getorcreate_amenity(make_standard_name('Airports'), 'airports', 'aeroway', 'aerodrome'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Airport in'), 'airport in', 'aeroway', 'aerodrome', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Airports in'), 'airports in', 'aeroway', 'aerodrome', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Airport near'), 'airport near', 'aeroway', 'aerodrome', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Airports near'), 'airports near', 'aeroway', 'aerodrome', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Bar'), 'bar', 'amenity', 'bar'); +SELECT test_getorcreate_amenity(make_standard_name('Bars'), 'bars', 'amenity', 'bar'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bar in'), 'bar in', 'amenity', 'bar', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bars in'), 'bars in', 'amenity', 'bar', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bar near'), 'bar near', 'amenity', 'bar', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bars near'), 'bars near', 'amenity', 'bar', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Bar'), 'bar', 'amenity', 'pub'); +SELECT test_getorcreate_amenity(make_standard_name('Bars'), 'bars', 'amenity', 'pub'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bar in'), 'bar in', 'amenity', 'pub', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bars in'), 'bars in', 'amenity', 'pub', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bar near'), 'bar near', 'amenity', 'pub', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Bars near'), 'bars near', 'amenity', 'pub', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Food'), 'food', 'amenity', 'restaurant'); +SELECT test_getorcreate_amenity(make_standard_name('Food'), 'food', 'amenity', 'restaurant'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Food in'), 'food in', 'amenity', 'restaurant', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Food in'), 'food in', 'amenity', 'restaurant', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Food near'), 'food near', 'amenity', 'restaurant', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Food near'), 'food near', 'amenity', 'restaurant', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Pub'), 'pub', 'amenity', 'bar'); +SELECT test_getorcreate_amenity(make_standard_name('Pubs'), 'pubs', 'amenity', 'bar'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pub in'), 'pub in', 'amenity', 'bar', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pubs in'), 'pubs in', 'amenity', 'bar', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pub near'), 'pub near', 'amenity', 'bar', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pubs near'), 'pubs near', 'amenity', 'bar', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Pub'), 'pub', 'amenity', 'pub'); +SELECT test_getorcreate_amenity(make_standard_name('Pubs'), 'pubs', 'amenity', 'pub'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pub in'), 'pub in', 'amenity', 'pub', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pubs in'), 'pubs in', 'amenity', 'pub', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pub near'), 'pub near', 'amenity', 'pub', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Pubs near'), 'pubs near', 'amenity', 'pub', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Restaurant'), 'restaurant', 'amenity', 'restaurant'); +SELECT test_getorcreate_amenity(make_standard_name('Restaurants'), 'restaurants', 'amenity', 'restaurant'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Restaurant in'), 'restaurant in', 'amenity', 'restaurant', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Restaurants in'), 'restaurants in', 'amenity', 'restaurant', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Restaurant near'), 'restaurant near', 'amenity', 'restaurant', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Restaurants near'), 'restaurants near', 'amenity', 'restaurant', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Mural'), 'mural', 'artwork_type', 'mural'); +SELECT test_getorcreate_amenity(make_standard_name('Murals'), 'murals', 'artwork_type', 'mural'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Mural in'), 'mural in', 'artwork_type', 'mural', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Murals in'), 'murals in', 'artwork_type', 'mural', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Mural near'), 'mural near', 'artwork_type', 'mural', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Murals near'), 'murals near', 'artwork_type', 'mural', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Sculpture'), 'sculpture', 'artwork_type', 'sculpture'); +SELECT test_getorcreate_amenity(make_standard_name('Sculptures'), 'sculptures', 'artwork_type', 'sculpture'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Sculpture in'), 'sculpture in', 'artwork_type', 'sculpture', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Sculptures in'), 'sculptures in', 'artwork_type', 'sculpture', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Sculpture near'), 'sculpture near', 'artwork_type', 'sculpture', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Sculptures near'), 'sculptures near', 'artwork_type', 'sculpture', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Statue'), 'statue', 'artwork_type', 'statue'); +SELECT test_getorcreate_amenity(make_standard_name('Statues'), 'statues', 'artwork_type', 'statue'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statue in'), 'statue in', 'artwork_type', 'statue', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statues in'), 'statues in', 'artwork_type', 'statue', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statue near'), 'statue near', 'artwork_type', 'statue', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statues near'), 'statues near', 'artwork_type', 'statue', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('ATM'), 'atm', 'atm', 'yes'); +SELECT test_getorcreate_amenity(make_standard_name('ATMs'), 'atms', 'atm', 'yes'); +SELECT test_getorcreate_amenityoperator(make_standard_name('ATM in'), 'atm in', 'atm', 'yes', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('ATMs in'), 'atms in', 'atm', 'yes', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('ATM near'), 'atm near', 'atm', 'yes', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('ATMs near'), 'atms near', 'atm', 'yes', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('National Park'), 'national park', 'boundary', 'national_park'); +SELECT test_getorcreate_amenity(make_standard_name('National Parks'), 'national parks', 'boundary', 'national_park'); +SELECT test_getorcreate_amenityoperator(make_standard_name('National Park in'), 'national park in', 'boundary', 'national_park', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('National Parks in'), 'national parks in', 'boundary', 'national_park', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('National Park near'), 'national park near', 'boundary', 'national_park', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('National Parks near'), 'national parks near', 'boundary', 'national_park', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Changing table'), 'changing table', 'changing_table', 'yes'); +SELECT test_getorcreate_amenity(make_standard_name('Changing tables'), 'changing tables', 'changing_table', 'yes'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Changing table in'), 'changing table in', 'changing_table', 'yes', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Changing tables in'), 'changing tables in', 'changing_table', 'yes', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Changing table near'), 'changing table near', 'changing_table', 'yes', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Changing tables near'), 'changing tables near', 'changing_table', 'yes', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Roundabout'), 'roundabout', 'junction', 'roundabout'); +SELECT test_getorcreate_amenity(make_standard_name('Roundabouts'), 'roundabouts', 'junction', 'roundabout'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Roundabout in'), 'roundabout in', 'junction', 'roundabout', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Roundabouts in'), 'roundabouts in', 'junction', 'roundabout', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Roundabout near'), 'roundabout near', 'junction', 'roundabout', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Roundabouts near'), 'roundabouts near', 'junction', 'roundabout', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Plaque'), 'plaque', 'memorial', 'plaque'); +SELECT test_getorcreate_amenity(make_standard_name('Plaques'), 'plaques', 'memorial', 'plaque'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Plaque in'), 'plaque in', 'memorial', 'plaque', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Plaques in'), 'plaques in', 'memorial', 'plaque', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Plaque near'), 'plaque near', 'memorial', 'plaque', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Plaques near'), 'plaques near', 'memorial', 'plaque', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Statue'), 'statue', 'memorial', 'statue'); +SELECT test_getorcreate_amenity(make_standard_name('Statues'), 'statues', 'memorial', 'statue'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statue in'), 'statue in', 'memorial', 'statue', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statues in'), 'statues in', 'memorial', 'statue', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statue near'), 'statue near', 'memorial', 'statue', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Statues near'), 'statues near', 'memorial', 'statue', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('Stolperstein'), 'stolperstein', 'memorial', 'stolperstein'); +SELECT test_getorcreate_amenity(make_standard_name('Stolpersteins'), 'stolpersteins', 'memorial', 'stolperstein'); +SELECT test_getorcreate_amenity(make_standard_name('Stolpersteine'), 'stolpersteine', 'memorial', 'stolperstein'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Stolperstein in'), 'stolperstein in', 'memorial', 'stolperstein', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Stolpersteins in'), 'stolpersteins in', 'memorial', 'stolperstein', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Stolpersteine in'), 'stolpersteine in', 'memorial', 'stolperstein', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Stolperstein near'), 'stolperstein near', 'memorial', 'stolperstein', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Stolpersteins near'), 'stolpersteins near', 'memorial', 'stolperstein', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('Stolpersteine near'), 'stolpersteine near', 'memorial', 'stolperstein', 'near'); +SELECT test_getorcreate_amenity(make_standard_name('War Memorial'), 'war memorial', 'memorial', 'war_memorial'); +SELECT test_getorcreate_amenity(make_standard_name('War Memorials'), 'war memorials', 'memorial', 'war_memorial'); +SELECT test_getorcreate_amenityoperator(make_standard_name('War Memorial in'), 'war memorial in', 'memorial', 'war_memorial', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('War Memorials in'), 'war memorials in', 'memorial', 'war_memorial', 'in'); +SELECT test_getorcreate_amenityoperator(make_standard_name('War Memorial near'), 'war memorial near', 'memorial', 'war_memorial', 'near'); +SELECT test_getorcreate_amenityoperator(make_standard_name('War Memorials near'), 'war memorials near', 'memorial', 'war_memorial', 'near'); CREATE INDEX idx_placex_classtype ON placex (class, type);CREATE TABLE place_classtype_aeroway_aerodrome AS SELECT place_id AS place_id,st_centroid(geometry) AS centroid FROM placex WHERE class = 'aeroway' AND type = 'aerodrome'; CREATE INDEX idx_place_classtype_aeroway_aerodrome_centroid ON place_classtype_aeroway_aerodrome USING GIST (centroid); CREATE INDEX idx_place_classtype_aeroway_aerodrome_place_id ON place_classtype_aeroway_aerodrome USING btree(place_id); @@ -175,4 +225,7 @@ CREATE TABLE place_classtype_memorial_war_memorial AS SELECT place_id AS place_i CREATE INDEX idx_place_classtype_memorial_war_memorial_centroid ON place_classtype_memorial_war_memorial USING GIST (centroid); CREATE INDEX idx_place_classtype_memorial_war_memorial_place_id ON place_classtype_memorial_war_memorial USING btree(place_id); GRANT SELECT ON place_classtype_memorial_war_memorial TO "www-data"; -DROP INDEX idx_placex_classtype; \ No newline at end of file +DROP INDEX idx_placex_classtype; + +DROP FUNCTION test_getorcreate_amenity; +DROP FUNCTION test_getorcreate_amenityoperator;