unify ICUNameProcessorRules and ICURuleLoader

There is no need for the additional layer of indirection that
the ICUNameProcessorRules class adds. The ICURuleLoader can
fill the database properties directly.

test/python/dummy_tokenizer.py

@@ -2,6 +2,7 @@
 Tokenizer for testing.
 """
 from nominatim.indexer.place_info import PlaceInfo
+from nominatim.config import Configuration
 
 def create(dsn, data_dir):
     """ Create a new instance of the tokenizer provided by this module.
@@ -22,7 +23,8 @@ class DummyTokenizer:
         self.init_state = "new"
 
 
-    def init_from_project(self):
+    def init_from_project(self, config):
+        assert isinstance(config, Configuration)
         assert self.init_state is None
         self.init_state = "loaded"
 

test/python/test_tokenizer_icu.py

@@ -7,7 +7,6 @@ import yaml
 import pytest
 
 from nominatim.tokenizer import icu_tokenizer
-from nominatim.tokenizer.icu_name_processor import ICUNameProcessorRules
 from nominatim.tokenizer.icu_rule_loader import ICURuleLoader
 from nominatim.db import properties
 from nominatim.db.sql_preprocessor import SQLPreprocessor
@@ -72,7 +71,8 @@ def analyzer(tokenizer_factory, test_config, monkeypatch,
         cfgstr = {'normalization' : list(norm),
                    'transliteration' : list(trans),
                    'variants' : [ {'words': list(variants)}]}
-        tok.naming_rules = ICUNameProcessorRules(loader=ICURuleLoader(cfgstr))
+        (test_config.project_dir / 'icu_tokenizer.yaml').write_text(yaml.dump(cfgstr))
+        tok.loader = ICURuleLoader(test_config)
 
         return tok.name_analyzer()
 
@@ -178,9 +178,9 @@ def test_init_from_project(monkeypatch, test_config, tokenizer_factory):
     monkeypatch.undo()
 
     tok = tokenizer_factory()
-    tok.init_from_project()
+    tok.init_from_project(test_config)
 
-    assert tok.naming_rules is not None
+    assert tok.loader is not None
     assert tok.term_normalization == ':: lower();'
 
 

test/python/test_tokenizer_icu_name_processor.py

@@ -4,15 +4,17 @@ Tests for import name normalisation and variant generation.
 from textwrap import dedent
 
 import pytest
-import yaml
 
 from nominatim.tokenizer.icu_rule_loader import ICURuleLoader
-from nominatim.tokenizer.icu_name_processor import ICUNameProcessor, ICUNameProcessorRules
 
 from nominatim.errors import UsageError
 
 @pytest.fixture
-def cfgfile():
+def cfgfile(def_config, tmp_path):
+    project_dir = tmp_path / 'project_dir'
+    project_dir.mkdir()
+    def_config.project_dir = project_dir
+
     def _create_config(*variants, **kwargs):
         content = dedent("""\
         normalization:
@@ -30,7 +32,9 @@ def cfgfile():
         content += '\n'.join(("      - " + s for s in variants)) + '\n'
         for k, v in kwargs:
             content += "    {}: {}\n".format(k, v)
-        return yaml.safe_load(content)
+        (project_dir / 'icu_tokenizer.yaml').write_text(content)
+
+        return def_config
 
     return _create_config
 
@@ -40,10 +44,9 @@ def get_normalized_variants(proc, name):
 
 
 def test_variants_empty(cfgfile):
-    fpath = cfgfile('saint -> 🜵', 'street -> st')
+    config = cfgfile('saint -> 🜵', 'street -> st')
 
-    rules = ICUNameProcessorRules(loader=ICURuleLoader(fpath))
-    proc = ICUNameProcessor(rules)
+    proc = ICURuleLoader(config).make_token_analysis()
 
     assert get_normalized_variants(proc, '🜵') == []
     assert get_normalized_variants(proc, '🜳') == []
@@ -83,8 +86,8 @@ VARIANT_TESTS = [
 
 @pytest.mark.parametrize("rules,name,variants", VARIANT_TESTS)
 def test_variants(cfgfile, rules, name, variants):
-    fpath = cfgfile(*rules)
-    proc = ICUNameProcessor(ICUNameProcessorRules(loader=ICURuleLoader(fpath)))
+    config = cfgfile(*rules)
+    proc = ICURuleLoader(config).make_token_analysis()
 
     result = get_normalized_variants(proc, name)
 
@@ -93,10 +96,8 @@ def test_variants(cfgfile, rules, name, variants):
 
 
 def test_search_normalized(cfgfile):
-    fpath = cfgfile('~street => s,st', 'master => mstr')
-
-    rules = ICUNameProcessorRules(loader=ICURuleLoader(fpath))
-    proc = ICUNameProcessor(rules)
+    config = cfgfile('~street => s,st', 'master => mstr')
+    proc = ICURuleLoader(config).make_token_analysis()
 
     assert proc.get_search_normalized('Master Street') == 'master street'
     assert proc.get_search_normalized('Earnes St') == 'earnes st'

test/python/test_tokenizer_icu_rule_loader.py

@@ -12,7 +12,16 @@ from nominatim.errors import UsageError
 from icu import Transliterator
 
 @pytest.fixture
-def cfgrules():
+def test_config(def_config, tmp_path):
+    project_dir = tmp_path / 'project_dir'
+    project_dir.mkdir()
+    def_config.project_dir = project_dir
+
+    return def_config
+
+
+@pytest.fixture
+def cfgrules(test_config):
     def _create_config(*variants, **kwargs):
         content = dedent("""\
         normalization:
@@ -29,19 +38,21 @@ def cfgrules():
         content += '\n'.join(("      - " + s for s in variants)) + '\n'
         for k, v in kwargs:
             content += "    {}: {}\n".format(k, v)
-        return yaml.safe_load(content)
+        (test_config.project_dir / 'icu_tokenizer.yaml').write_text(content)
+
+        return test_config
 
     return _create_config
 
 
-def test_empty_rule_set():
-    rule_cfg = yaml.safe_load(dedent("""\
+def test_empty_rule_set(test_config):
+    (test_config.project_dir / 'icu_tokenizer.yaml').write_text(dedent("""\
         normalization:
         transliteration:
         variants:
         """))
 
-    rules = ICURuleLoader(rule_cfg)
+    rules = ICURuleLoader(test_config)
     assert rules.get_search_rules() == ''
     assert rules.get_normalization_rules() == ''
     assert rules.get_transliteration_rules() == ''
@@ -50,11 +61,12 @@ def test_empty_rule_set():
 CONFIG_SECTIONS = ('normalization', 'transliteration', 'variants')
 
 @pytest.mark.parametrize("section", CONFIG_SECTIONS)
-def test_missing_section(section):
+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))
 
     with pytest.raises(UsageError):
-        ICURuleLoader(rule_cfg)
+        ICURuleLoader(test_config)
 
 
 def test_get_search_rules(cfgrules):
@@ -88,9 +100,8 @@ def test_get_transliteration_rules(cfgrules):
     assert trans.transliterate(" проспект-Prospekt ") == " prospekt Prospekt "
 
 
-def test_transliteration_rules_from_file(def_config, tmp_path):
-    def_config.project_dir = tmp_path
-    cfgpath = tmp_path / ('test_config.yaml')
+def test_transliteration_rules_from_file(test_config):
+    cfgpath = test_config.project_dir / ('icu_tokenizer.yaml')
     cfgpath.write_text(dedent("""\
         normalization:
         transliteration:
@@ -98,10 +109,10 @@ def test_transliteration_rules_from_file(def_config, tmp_path):
             - !include transliteration.yaml
         variants:
         """))
-    transpath = tmp_path / ('transliteration.yaml')
+    transpath = test_config.project_dir / ('transliteration.yaml')
     transpath.write_text('- "x > y"')
 
-    loader = ICURuleLoader(def_config.load_sub_configuration('test_config.yaml'))
+    loader = ICURuleLoader(test_config)
     rules = loader.get_transliteration_rules()
     trans = Transliterator.createFromRules("test", rules)
 

test/python/test_tokenizer_legacy.py

@@ -132,10 +132,10 @@ def test_init_module_custom(tokenizer_factory, test_config,
     assert not (test_config.project_dir / 'module').exists()
 
 
-def test_init_from_project(tokenizer_setup, tokenizer_factory):
+def test_init_from_project(tokenizer_setup, tokenizer_factory, test_config):
     tok = tokenizer_factory()
 
-    tok.init_from_project()
+    tok.init_from_project(test_config)
 
     assert tok.normalization is not None
 

test/python/test_tools_check_database.py

@@ -53,7 +53,7 @@ def test_check_tokenizer(temp_db_conn, def_config, monkeypatch,
                          check_result, state):
     class _TestTokenizer:
         @staticmethod
-        def check_database():
+        def check_database(_):
             return check_result
 
     monkeypatch.setattr(chkdb.tokenizer_factory, 'get_tokenizer_for_db',