switch to a more flexible variant description format

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

diff --git a/test/python/test_tokenizer_icu_name_processor.py b/test/python/test_tokenizer_icu_name_processor.py
index 817f9072aae965151ab66aa084b21d08f742b116..c1ad767577a67569c10119629e3d3a1df5b2e207 100644 (file)
--- a/test/python/test_tokenizer_icu_name_processor.py
+++ b/test/python/test_tokenizer_icu_name_processor.py
@@ -12,21 +12,23 @@ from nominatim.errors import UsageError
 
 @pytest.fixture
 def cfgfile(tmp_path, suffix='.yaml'):
-    def _create_config(suffixes, abbr):
+    def _create_config(*variants, **kwargs):
         content = dedent("""\
         normalization:
             - ":: NFD ()"
+            - "'🜳' > ' '"
             - "[[:Nonspacing Mark:] [:Cf:]] >"
             - ":: lower ()"
             - "[[:Punctuation:][:Space:]]+ > ' '"
             - ":: NFC ()"
         transliteration:
             - "::  Latin ()"
+            - "'🜵' > ' '"
         """)
-        content += "compound_suffixes:\n"
-        content += '\n'.join(("    - " + s for s in suffixes)) + '\n'
-        content += "abbreviations:\n"
-        content += '\n'.join(("    - " + s for s in abbr)) + '\n'
+        content += "variants:\n  - words:\n"
+        content += '\n'.join(("      - " + s for s in variants)) + '\n'
+        for k, v in kwargs:
+            content += "    {}: {}\n".format(k, v)
         fpath = tmp_path / ('test_config' + suffix)
         fpath.write_text(dedent(content))
         return fpath
@@ -38,22 +40,36 @@ def get_normalized_variants(proc, name):
     return proc.get_variants_ascii(proc.get_normalized(name))
 
 def test_simple_variants(cfgfile):
-    fpath = cfgfile(['strasse', 'straße', 'weg'],
-                    ['strasse,straße => str',
-                     'prospekt => pr'])
+    fpath = cfgfile('~strasse,~straße -> str',
+                    '~weg => weg',
+                    'prospekt -> pr')
 
     rules = ICUNameProcessorRules(loader=ICURuleLoader(fpath))
     proc = ICUNameProcessor(rules)
 
     assert set(get_normalized_variants(proc, "Bauwegstraße")) \
-            == {'bauweg straße', 'bauweg str'}
-    assert get_normalized_variants(proc, "Bauwegstr") == ['bauweg str']
-    assert get_normalized_variants(proc, "holzweg") == ['holz weg']
+            == {'bauweg straße', 'bauweg str', 'bauwegstraße', 'bauwegstr'}
+    assert get_normalized_variants(proc, "Bauwegstr") == ['bauwegstr']
+    assert set(get_normalized_variants(proc, "holzweg")) \
+            == {'holz weg', 'holzweg'}
+    assert set(get_normalized_variants(proc, "Meier Weg")) \
+            == {'meier weg', 'meierweg'}
     assert get_normalized_variants(proc, "hallo") == ['hallo']
 
 
+def test_variants_empty(cfgfile):
+    fpath = cfgfile('saint -> 🜵', 'street -> st')
+
+    rules = ICUNameProcessorRules(loader=ICURuleLoader(fpath))
+    proc = ICUNameProcessor(rules)
+
+    assert get_normalized_variants(proc, '🜵') == []
+    assert get_normalized_variants(proc, '🜳') == []
+    assert get_normalized_variants(proc, 'saint') == ['saint']
+
+
 def test_multiple_replacements(cfgfile):
-    fpath = cfgfile([], ['saint => s,st', 'street => st'])
+    fpath = cfgfile('saint -> s,st', 'street -> st')
 
     rules = ICUNameProcessorRules(loader=ICURuleLoader(fpath))
     proc = ICUNameProcessor(rules)
@@ -64,11 +80,11 @@ def test_multiple_replacements(cfgfile):
 
 
 def test_search_normalized(cfgfile):
-    fpath = cfgfile(['street'], ['street => s,st', 'master => mstr'])
+    fpath = cfgfile('~street => s,st', 'master => mstr')
 
     rules = ICUNameProcessorRules(loader=ICURuleLoader(fpath))
     proc = ICUNameProcessor(rules)
 
-    assert proc.get_search_normalized('Master Street') == 'master  street'
-    assert proc.get_search_normalized('Earnes St') == 'earne s  st'
-    assert proc.get_search_normalized('Nostreet') == 'no street'
+    assert proc.get_search_normalized('Master Street') == 'master street'
+    assert proc.get_search_normalized('Earnes St') == 'earnes st'
+    assert proc.get_search_normalized('Nostreet') == 'nostreet'