import psycopg2
import pytest
-SRC_DIR = (Path(__file__) / '..' / '..' / '..').resolve()
-
# always test against the source
-sys.path.insert(0, str(SRC_DIR.resolve()))
+SRC_DIR = (Path(__file__) / '..' / '..' / '..').resolve()
+sys.path.insert(0, str(SRC_DIR))
from nominatim.config import Configuration
from nominatim.db import connection
from cursor import CursorForTesting
+@pytest.fixture
+def src_dir():
+ return SRC_DIR
+
+
@pytest.fixture
def temp_db(monkeypatch):
""" Create an empty database for the test. The database name is also
@pytest.fixture
-def def_config():
- cfg = Configuration(None, SRC_DIR.resolve() / 'settings')
+def def_config(src_dir):
+ cfg = Configuration(None, src_dir / 'settings')
cfg.set_libdirs(module='.', osm2pgsql='.',
- php=SRC_DIR / 'lib-php',
- sql=SRC_DIR / 'lib-sql',
- data=SRC_DIR / 'data')
+ php=src_dir / 'lib-php',
+ sql=src_dir / 'lib-sql',
+ data=src_dir / 'data')
return cfg
@pytest.fixture
-def src_dir():
- return SRC_DIR.resolve()
+def project_env(src_dir, tmp_path):
+ projdir = tmp_path / 'project'
+ projdir.mkdir()
+ cfg = Configuration(projdir, src_dir / 'settings')
+ cfg.set_libdirs(module='.', osm2pgsql='.',
+ php=src_dir / 'lib-php',
+ sql=src_dir / 'lib-sql',
+ data=src_dir / 'data')
+ return cfg
@pytest.fixture
@pytest.fixture
-def test_tokenizer(tokenizer_mock, def_config, tmp_path):
- def_config.project_dir = tmp_path
- return factory.create_tokenizer(def_config)
+def test_tokenizer(tokenizer_mock, project_env):
+ return factory.create_tokenizer(project_env)
@pytest.mark.parametrize("threads", [1, 15])
from nominatim.errors import UsageError
from dummy_tokenizer import DummyTokenizer
-@pytest.fixture
-def test_config(def_config, tmp_path, property_table, tokenizer_mock):
- def_config.project_dir = tmp_path
- return def_config
+def test_setup_bad_tokenizer_name(project_env, monkeypatch):
+ monkeypatch.setenv('NOMINATIM_TOKENIZER', 'dummy')
-def test_setup_dummy_tokenizer(temp_db_conn, test_config):
- tokenizer = factory.create_tokenizer(test_config)
+ with pytest.raises(UsageError):
+ factory.create_tokenizer(project_env)
- assert isinstance(tokenizer, DummyTokenizer)
- assert tokenizer.init_state == "new"
- assert (test_config.project_dir / 'tokenizer').is_dir()
- assert properties.get_property(temp_db_conn, 'tokenizer') == 'dummy'
+class TestFactory:
+ @pytest.fixture(autouse=True)
+ def init_env(self, project_env, property_table, tokenizer_mock):
+ self.config = project_env
-def test_setup_tokenizer_dir_exists(test_config):
- (test_config.project_dir / 'tokenizer').mkdir()
+ def test_setup_dummy_tokenizer(self, temp_db_conn):
+ tokenizer = factory.create_tokenizer(self.config)
- tokenizer = factory.create_tokenizer(test_config)
+ assert isinstance(tokenizer, DummyTokenizer)
+ assert tokenizer.init_state == "new"
+ assert (self.config.project_dir / 'tokenizer').is_dir()
- assert isinstance(tokenizer, DummyTokenizer)
- assert tokenizer.init_state == "new"
+ assert properties.get_property(temp_db_conn, 'tokenizer') == 'dummy'
-def test_setup_tokenizer_dir_failure(test_config):
- (test_config.project_dir / 'tokenizer').write_text("foo")
+ def test_setup_tokenizer_dir_exists(self):
+ (self.config.project_dir / 'tokenizer').mkdir()
- with pytest.raises(UsageError):
- factory.create_tokenizer(test_config)
+ tokenizer = factory.create_tokenizer(self.config)
+ assert isinstance(tokenizer, DummyTokenizer)
+ assert tokenizer.init_state == "new"
-def test_setup_bad_tokenizer_name(def_config, tmp_path, monkeypatch):
- def_config.project_dir = tmp_path
- monkeypatch.setenv('NOMINATIM_TOKENIZER', 'dummy')
- with pytest.raises(UsageError):
- factory.create_tokenizer(def_config)
+ def test_setup_tokenizer_dir_failure(self):
+ (self.config.project_dir / 'tokenizer').write_text("foo")
+ with pytest.raises(UsageError):
+ factory.create_tokenizer(self.config)
-def test_load_tokenizer(test_config):
- factory.create_tokenizer(test_config)
- tokenizer = factory.get_tokenizer_for_db(test_config)
+ def test_load_tokenizer(self):
+ factory.create_tokenizer(self.config)
- assert isinstance(tokenizer, DummyTokenizer)
- assert tokenizer.init_state == "loaded"
+ tokenizer = factory.get_tokenizer_for_db(self.config)
+ assert isinstance(tokenizer, DummyTokenizer)
+ assert tokenizer.init_state == "loaded"
-def test_load_no_tokenizer_dir(test_config):
- factory.create_tokenizer(test_config)
- test_config.project_dir = test_config.project_dir / 'foo'
+ def test_load_no_tokenizer_dir(self):
+ factory.create_tokenizer(self.config)
- with pytest.raises(UsageError):
- factory.get_tokenizer_for_db(test_config)
+ self.config.project_dir = self.config.project_dir / 'foo'
+ with pytest.raises(UsageError):
+ factory.get_tokenizer_for_db(self.config)
-def test_load_missing_propoerty(temp_db_cursor, test_config):
- factory.create_tokenizer(test_config)
- temp_db_cursor.execute("TRUNCATE TABLE nominatim_properties")
+ def test_load_missing_property(self, temp_db_cursor):
+ factory.create_tokenizer(self.config)
- with pytest.raises(UsageError):
- factory.get_tokenizer_for_db(test_config)
+ temp_db_cursor.execute("TRUNCATE TABLE nominatim_properties")
+
+ with pytest.raises(UsageError):
+ factory.get_tokenizer_for_db(self.config)
@pytest.fixture
-def test_config(def_config, tmp_path):
- def_config.project_dir = tmp_path / 'project'
- def_config.project_dir.mkdir()
-
+def test_config(project_env, tmp_path):
sqldir = tmp_path / 'sql'
sqldir.mkdir()
(sqldir / 'tokenizer').mkdir()
(sqldir / 'tokenizer' / 'icu_tokenizer.sql').write_text("SELECT 'a'")
- shutil.copy(str(def_config.lib_dir.sql / 'tokenizer' / 'icu_tokenizer_tables.sql'),
+ shutil.copy(str(project_env.lib_dir.sql / 'tokenizer' / 'icu_tokenizer_tables.sql'),
str(sqldir / 'tokenizer' / 'icu_tokenizer_tables.sql'))
- def_config.lib_dir.sql = sqldir
+ project_env.lib_dir.sql = sqldir
- return def_config
+ return project_env
@pytest.fixture
from icu import Transliterator
-@pytest.fixture
-def test_config(def_config, tmp_path):
- project_dir = tmp_path / 'project_dir'
- project_dir.mkdir()
- def_config.project_dir = project_dir
+CONFIG_SECTIONS = ('normalization', 'transliteration', 'token-analysis')
+
+class TestIcuRuleLoader:
+
+ @pytest.fixture(autouse=True)
+ def init_env(self, project_env):
+ self.project_env = project_env
- return def_config
+ def write_config(self, content):
+ (self.project_env.project_dir / 'icu_tokenizer.yaml').write_text(dedent(content))
-@pytest.fixture
-def cfgrules(test_config):
- def _create_config(*variants, **kwargs):
+
+ def config_rules(self, *variants):
content = dedent("""\
normalization:
- ":: NFD ()"
transliteration:
- ":: Latin ()"
- "[[:Punctuation:][:Space:]]+ > ' '"
- """)
- content += "token-analysis:\n - analyzer: generic\n variants:\n - words:\n"
- content += '\n'.join((" - " + s for s in variants)) + '\n'
- for k, v in kwargs:
- content += " {}: {}\n".format(k, v)
- (test_config.project_dir / 'icu_tokenizer.yaml').write_text(content)
-
- return test_config
-
- return _create_config
-
-
-def test_empty_rule_set(test_config):
- (test_config.project_dir / 'icu_tokenizer.yaml').write_text(dedent("""\
- normalization:
- transliteration:
token-analysis:
- - analyzer: generic
- variants:
- """))
+ - analyzer: generic
+ variants:
+ - words:
+ """)
+ content += '\n'.join((" - " + s for s in variants)) + '\n'
+ self.write_config(content)
- rules = ICURuleLoader(test_config)
- assert rules.get_search_rules() == ''
- assert rules.get_normalization_rules() == ''
- assert rules.get_transliteration_rules() == ''
-CONFIG_SECTIONS = ('normalization', 'transliteration', 'token-analysis')
+ def get_replacements(self, *variants):
+ self.config_rules(*variants)
+ loader = ICURuleLoader(self.project_env)
+ rules = loader.analysis[None].config['replacements']
-@pytest.mark.parametrize("section", CONFIG_SECTIONS)
-def test_missing_section(section, test_config):
- rule_cfg = { s: [] for s in CONFIG_SECTIONS if s != section}
- (test_config.project_dir / 'icu_tokenizer.yaml').write_text(yaml.dump(rule_cfg))
+ return sorted((k, sorted(v)) for k,v in rules)
- with pytest.raises(UsageError):
- ICURuleLoader(test_config)
+ def test_empty_rule_set(self):
+ self.write_config("""\
+ normalization:
+ transliteration:
+ token-analysis:
+ - analyzer: generic
+ variants:
+ """)
-def test_get_search_rules(cfgrules):
- loader = ICURuleLoader(cfgrules())
+ rules = ICURuleLoader(self.project_env)
+ assert rules.get_search_rules() == ''
+ assert rules.get_normalization_rules() == ''
+ assert rules.get_transliteration_rules() == ''
- rules = loader.get_search_rules()
- trans = Transliterator.createFromRules("test", rules)
- assert trans.transliterate(" Baum straße ") == " baum straße "
- assert trans.transliterate(" Baumstraße ") == " baumstraße "
- assert trans.transliterate(" Baumstrasse ") == " baumstrasse "
- assert trans.transliterate(" Baumstr ") == " baumstr "
- assert trans.transliterate(" Baumwegstr ") == " baumwegstr "
- assert trans.transliterate(" Αθήνα ") == " athēna "
- assert trans.transliterate(" проспект ") == " prospekt "
+ @pytest.mark.parametrize("section", CONFIG_SECTIONS)
+ def test_missing_section(self, section):
+ rule_cfg = { s: [] for s in CONFIG_SECTIONS if s != section}
+ self.write_config(yaml.dump(rule_cfg))
+ with pytest.raises(UsageError):
+ ICURuleLoader(self.project_env)
-def test_get_normalization_rules(cfgrules):
- loader = ICURuleLoader(cfgrules())
- rules = loader.get_normalization_rules()
- trans = Transliterator.createFromRules("test", rules)
- assert trans.transliterate(" проспект-Prospekt ") == " проспект prospekt "
+ def test_get_search_rules(self):
+ self.config_rules()
+ loader = ICURuleLoader(self.project_env)
+ rules = loader.get_search_rules()
+ trans = Transliterator.createFromRules("test", rules)
-def test_get_transliteration_rules(cfgrules):
- loader = ICURuleLoader(cfgrules())
- rules = loader.get_transliteration_rules()
- trans = Transliterator.createFromRules("test", rules)
+ assert trans.transliterate(" Baum straße ") == " baum straße "
+ assert trans.transliterate(" Baumstraße ") == " baumstraße "
+ assert trans.transliterate(" Baumstrasse ") == " baumstrasse "
+ assert trans.transliterate(" Baumstr ") == " baumstr "
+ assert trans.transliterate(" Baumwegstr ") == " baumwegstr "
+ assert trans.transliterate(" Αθήνα ") == " athēna "
+ assert trans.transliterate(" проспект ") == " prospekt "
- assert trans.transliterate(" проспект-Prospekt ") == " prospekt Prospekt "
+ def test_get_normalization_rules(self):
+ self.config_rules()
+ loader = ICURuleLoader(self.project_env)
+ rules = loader.get_normalization_rules()
+ trans = Transliterator.createFromRules("test", rules)
-def test_transliteration_rules_from_file(test_config):
- cfgpath = test_config.project_dir / ('icu_tokenizer.yaml')
- cfgpath.write_text(dedent("""\
- normalization:
- transliteration:
- - "'ax' > 'b'"
- - !include transliteration.yaml
- token-analysis:
- - analyzer: generic
- variants:
- """))
- transpath = test_config.project_dir / ('transliteration.yaml')
- transpath.write_text('- "x > y"')
+ assert trans.transliterate(" проспект-Prospekt ") == " проспект prospekt "
- loader = ICURuleLoader(test_config)
- rules = loader.get_transliteration_rules()
- trans = Transliterator.createFromRules("test", rules)
- assert trans.transliterate(" axxt ") == " byt "
+ def test_get_transliteration_rules(self):
+ self.config_rules()
+ loader = ICURuleLoader(self.project_env)
+ rules = loader.get_transliteration_rules()
+ trans = Transliterator.createFromRules("test", rules)
+ assert trans.transliterate(" проспект-Prospekt ") == " prospekt Prospekt "
-def test_search_rules(cfgrules):
- config = cfgrules('~street => s,st', 'master => mstr')
- proc = ICURuleLoader(config).make_token_analysis()
- assert proc.search.transliterate('Master Street').strip() == 'master street'
- assert proc.search.transliterate('Earnes St').strip() == 'earnes st'
- assert proc.search.transliterate('Nostreet').strip() == 'nostreet'
+ def test_transliteration_rules_from_file(self):
+ self.write_config("""\
+ normalization:
+ transliteration:
+ - "'ax' > 'b'"
+ - !include transliteration.yaml
+ token-analysis:
+ - analyzer: generic
+ variants:
+ """)
+ transpath = self.project_env.project_dir / ('transliteration.yaml')
+ transpath.write_text('- "x > y"')
+ loader = ICURuleLoader(self.project_env)
+ rules = loader.get_transliteration_rules()
+ trans = Transliterator.createFromRules("test", rules)
-class TestGetReplacements:
+ assert trans.transliterate(" axxt ") == " byt "
- @pytest.fixture(autouse=True)
- def setup_cfg(self, cfgrules):
- self.cfgrules = cfgrules
- def get_replacements(self, *variants):
- loader = ICURuleLoader(self.cfgrules(*variants))
- rules = loader.analysis[None].config['replacements']
+ def test_search_rules(self):
+ self.config_rules('~street => s,st', 'master => mstr')
+ proc = ICURuleLoader(self.project_env).make_token_analysis()
- return sorted((k, sorted(v)) for k,v in rules)
+ assert proc.search.transliterate('Master Street').strip() == 'master street'
+ assert proc.search.transliterate('Earnes St').strip() == 'earnes st'
+ assert proc.search.transliterate('Nostreet').strip() == 'nostreet'
@pytest.mark.parametrize("variant", ['foo > bar', 'foo -> bar -> bar',
'~foo~ -> bar', 'fo~ o -> bar'])
def test_invalid_variant_description(self, variant):
+ self.config_rules(variant)
with pytest.raises(UsageError):
- ICURuleLoader(self.cfgrules(variant))
+ ICURuleLoader(self.project_env)
def test_add_full(self):
repl = self.get_replacements("foo -> bar")
from nominatim.errors import UsageError
@pytest.fixture
-def test_config(def_config, tmp_path):
- def_config.project_dir = tmp_path / 'project'
- def_config.project_dir.mkdir()
-
+def test_config(project_env, tmp_path):
module_dir = tmp_path / 'module_src'
module_dir.mkdir()
(module_dir / 'nominatim.so').write_text('TEST nomiantim.so')
- def_config.lib_dir.module = module_dir
+ project_env.lib_dir.module = module_dir
sqldir = tmp_path / 'sql'
sqldir.mkdir()
(sqldir / 'tokenizer').mkdir()
(sqldir / 'tokenizer' / 'legacy_tokenizer.sql').write_text("SELECT 'a'")
(sqldir / 'words.sql').write_text("SELECT 'a'")
- shutil.copy(str(def_config.lib_dir.sql / 'tokenizer' / 'legacy_tokenizer_tables.sql'),
+ shutil.copy(str(project_env.lib_dir.sql / 'tokenizer' / 'legacy_tokenizer_tables.sql'),
str(sqldir / 'tokenizer' / 'legacy_tokenizer_tables.sql'))
- def_config.lib_dir.sql = sqldir
- def_config.lib_dir.data = sqldir
+ project_env.lib_dir.sql = sqldir
+ project_env.lib_dir.data = sqldir
- return def_config
+ return project_env
@pytest.fixture
assert temp_db_cursor.table_rows('address_levels') > 0
-def test_load_ranks_from_project_dir(def_config, temp_db_conn, temp_db_cursor,
- tmp_path):
- test_file = tmp_path / 'address-levels.json'
+def test_load_ranks_from_project_dir(project_env, temp_db_conn, temp_db_cursor):
+ test_file = project_env.project_dir / 'address-levels.json'
test_file.write_text('[{"tags":{"place":{"sea":2}}}]')
- def_config.project_dir = tmp_path
- load_address_levels_from_config(temp_db_conn, def_config)
+ load_address_levels_from_config(temp_db_conn, project_env)
assert temp_db_cursor.table_rows('address_levels') == 1
-def test_load_ranks_from_broken_file(def_config, temp_db_conn, tmp_path):
- test_file = tmp_path / 'address-levels.json'
+def test_load_ranks_from_broken_file(project_env, temp_db_conn):
+ test_file = project_env.project_dir / 'address-levels.json'
test_file.write_text('[{"tags":"place":{"sea":2}}}]')
- def_config.project_dir = tmp_path
with pytest.raises(json.decoder.JSONDecodeError):
- load_address_levels_from_config(temp_db_conn, def_config)
+ load_address_levels_from_config(temp_db_conn, project_env)
def test_load_ranks_country(temp_db_conn, temp_db_cursor):
from nominatim.tools.refresh import create_functions
-@pytest.fixture
-def sql_tmp_path(tmp_path, def_config):
- def_config.lib_dir.sql = tmp_path
- return tmp_path
-
-@pytest.fixture
-def conn(sql_preprocessor, temp_db_conn):
- return temp_db_conn
-
-
-def test_create_functions(temp_db_cursor, conn, def_config, sql_tmp_path):
- sqlfile = sql_tmp_path / 'functions.sql'
- sqlfile.write_text("""CREATE OR REPLACE FUNCTION test() RETURNS INTEGER
- AS $$
- BEGIN
- RETURN 43;
- END;
- $$ LANGUAGE plpgsql IMMUTABLE;
- """)
-
- create_functions(conn, def_config)
-
- assert temp_db_cursor.scalar('SELECT test()') == 43
-
-
-@pytest.mark.parametrize("dbg,ret", ((True, 43), (False, 22)))
-def test_create_functions_with_template(temp_db_cursor, conn, def_config, sql_tmp_path,
- dbg, ret):
- sqlfile = sql_tmp_path / 'functions.sql'
- sqlfile.write_text("""CREATE OR REPLACE FUNCTION test() RETURNS INTEGER
- AS $$
- BEGIN
- {% if debug %}
- RETURN 43;
- {% else %}
- RETURN 22;
- {% endif %}
- END;
- $$ LANGUAGE plpgsql IMMUTABLE;
- """)
-
- create_functions(conn, def_config, enable_debug=dbg)
-
- assert temp_db_cursor.scalar('SELECT test()') == ret
+class TestCreateFunctions:
+ @pytest.fixture(autouse=True)
+ def init_env(self, sql_preprocessor, temp_db_conn, def_config, tmp_path):
+ self.conn = temp_db_conn
+ self.config = def_config
+ def_config.lib_dir.sql = tmp_path
+
+
+ def write_functions(self, content):
+ sqlfile = self.config.lib_dir.sql / 'functions.sql'
+ sqlfile.write_text(content)
+
+
+ def test_create_functions(self, temp_db_cursor):
+ self.write_functions("""CREATE OR REPLACE FUNCTION test() RETURNS INTEGER
+ AS $$
+ BEGIN
+ RETURN 43;
+ END;
+ $$ LANGUAGE plpgsql IMMUTABLE;
+ """)
+
+ create_functions(self.conn, self.config)
+
+ assert temp_db_cursor.scalar('SELECT test()') == 43
+
+
+ @pytest.mark.parametrize("dbg,ret", ((True, 43), (False, 22)))
+ def test_create_functions_with_template(self, temp_db_cursor, dbg, ret):
+ self.write_functions("""CREATE OR REPLACE FUNCTION test() RETURNS INTEGER
+ AS $$
+ BEGIN
+ {% if debug %}
+ RETURN 43;
+ {% else %}
+ RETURN 22;
+ {% endif %}
+ END;
+ $$ LANGUAGE plpgsql IMMUTABLE;
+ """)
+
+ create_functions(self.conn, self.config, enable_debug=dbg)
+
+ assert temp_db_cursor.scalar('SELECT test()') == ret
@pytest.fixture
-def run_website_script(tmp_path, def_config, temp_db_conn):
- def_config.lib_dir.php = tmp_path / 'php'
- def_config.project_dir = tmp_path
+def run_website_script(tmp_path, project_env, temp_db_conn):
+ project_env.lib_dir.php = tmp_path / 'php'
def _runner():
- refresh.setup_website(tmp_path, def_config, temp_db_conn)
+ refresh.setup_website(tmp_path, project_env, temp_db_conn)
proc = subprocess.run(['/usr/bin/env', 'php', '-Cq',
tmp_path / 'search.php'], check=False)
projects / nominatim.git / commitdiff