clean_housenumbers: make kinds and delimiters configurable

Also adds unit tests for various options.

nominatim/tokenizer/sanitizers/clean_housenumbers.py

@@ -6,13 +6,19 @@
 # For a full list of authors see the git log.
 """
 Sanitizer that cleans and normalizes housenumbers.
+
+Arguments:
+    delimiters: Define the set of characters to be used for
+                splitting a list of housenumbers into parts. (default: ',;')
+
 """
-import re
+from nominatim.tokenizer.sanitizers.helpers import create_split_regex
 
 class _HousenumberSanitizer:
 
     def __init__(self, config):
-        pass
+        self.kinds = config.get('filter-kind', ('housenumber', ))
+        self.split_regexp = create_split_regex(config)
 
 
     def __call__(self, obj):
@@ -21,7 +27,7 @@ class _HousenumberSanitizer:
 
         new_address = []
         for item in obj.address:
-            if item.kind in ('housenumber', 'streetnumber', 'conscriptionnumber'):
+            if item.kind in self.kinds:
                 new_address.extend(item.clone(kind='housenumber', name=n) for n in self.sanitize(item.name))
             else:
                 # Don't touch other address items.
@@ -36,13 +42,9 @@ class _HousenumberSanitizer:
             The function works as a generator that yields all valid housenumbers
             that can be created from the value.
         """
-        for hnr in self._split_number(value):
+        for hnr in self.split_regexp.split(value):
-            yield from self._regularize(hnr)
+            if hnr:
-
+                yield from self._regularize(hnr)
-
-    def _split_number(self, hnr):
-        for part in re.split(r'[;,]', hnr):
-            yield part.strip()
 
 
     def _regularize(self, hnr):

nominatim/tokenizer/sanitizers/helpers.py

@@ -0,0 +1,29 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# This file is part of Nominatim. (https://nominatim.org)
+#
+# Copyright (C) 2022 by the Nominatim developer community.
+# For a full list of authors see the git log.
+"""
+Helper functions for sanitizers.
+"""
+import re
+
+from nominatim.errors import UsageError
+
+def create_split_regex(config, default=',;'):
+    """ Converts the 'delimiter' parameter in the configuration into a
+        compiled regular expression that can be used to split the names on the
+        delimiters. The regular expression makes sure that the resulting names
+        are stripped and that repeated delimiters
+        are ignored but it will still create empty fields on occasion. The
+        code needs to filter those.
+
+        The 'default' parameter defines the delimiter set to be used when
+        not explicitly configured.
+    """
+    delimiter_set = set(config.get('delimiters', default))
+    if not delimiter_set:
+        raise UsageError("Empty 'delimiter' parameter not allowed for sanitizer.")
+
+    return re.compile('\\s*[{}]+\\s*'.format(''.join('\\' + d for d in delimiter_set)))

nominatim/tokenizer/sanitizers/split_name_list.py

@@ -9,21 +9,16 @@ Sanitizer that splits lists of names into their components.
 
 Arguments:
     delimiters: Define the set of characters to be used for
-                splitting the list. (default: `,;`)
+                splitting the list. (default: ',;')
 """
-import re
-
 from nominatim.errors import UsageError
+from nominatim.tokenizer.sanitizers.helpers import create_split_regex
 
 def create(func):
     """ Create a name processing function that splits name values with
         multiple values into their components.
     """
-    delimiter_set = set(func.get('delimiters', ',;'))
+    regexp = create_split_regex(func)
-    if not delimiter_set:
-        raise UsageError("Set of delimiters in split-name-list sanitizer is empty.")
-
-    regexp = re.compile('\\s*[{}]\\s*'.format(''.join('\\' + d for d in delimiter_set)))
 
     def _process(obj):
         if not obj.names:

nominatim/tokenizer/sanitizers/tag_analyzer_by_language.py

@@ -13,7 +13,7 @@ Arguments:
 
     filter-kind: Restrict the names the sanitizer should be applied to
                  to the given tags. The parameter expects a list of
-                 regular expressions which are matched against `kind`.
+                 regular expressions which are matched against 'kind'.
                  Note that a match against the full string is expected.
     whitelist: Restrict the set of languages that should be tagged.
                Expects a list of acceptable suffixes. When unset,

test/python/tokenizer/sanitizers/test_clean_housenumbers.py

@@ -0,0 +1,44 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# This file is part of Nominatim. (https://nominatim.org)
+#
+# Copyright (C) 2022 by the Nominatim developer community.
+# For a full list of authors see the git log.
+"""
+Tests for the sanitizer that normalizes housenumbers.
+"""
+import pytest
+
+from nominatim.tokenizer.place_sanitizer import PlaceSanitizer
+from nominatim.indexer.place_info import PlaceInfo
+
+@pytest.fixture
+def sanitize(request):
+    sanitizer_args = {'step': 'clean-housenumbers'}
+    for mark in request.node.iter_markers(name="sanitizer_params"):
+        sanitizer_args.update({k.replace('_', '-') : v for k,v in mark.kwargs.items()})
+
+    def _run(**kwargs):
+        place = PlaceInfo({'address': kwargs})
+        _, address = PlaceSanitizer([sanitizer_args]).process_names(place)
+
+        return sorted([(p.kind, p.name) for p in address])
+
+    return _run
+
+
+def test_simple_number(sanitize):
+    assert sanitize(housenumber='34') == [('housenumber', '34')]
+
+
+@pytest.mark.parametrize('number', ['1;2;3', '1,2,3', '1; 3 ,2',
+                                    '2,,3,1', '1;2;3;;', ';3;2;1'])
+def test_housenumber_lists(sanitize, number):
+    assert sanitize(housenumber=number) == \
+           [('housenumber', '1'), ('housenumber', '2'), ('housenumber', '3')]
+
+
+@pytest.mark.sanitizer_params(filter_kind=('number', 'streetnumber'))
+def test_filter_kind(sanitize):
+    assert sanitize(housenumber='34', number='4', badnumber='65') == \
+            [('badnumber', '65'), ('housenumber', '34'), ('housenumber', '4')]

test/python/tokenizer/sanitizers/test_helpers.py

@@ -0,0 +1,43 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# This file is part of Nominatim. (https://nominatim.org)
+#
+# Copyright (C) 2022 by the Nominatim developer community.
+# For a full list of authors see the git log.
+"""
+Tests for sanitizer helper functions.
+"""
+import pytest
+
+from nominatim.errors import UsageError
+import nominatim.tokenizer.sanitizers.helpers as helpers
+
+@pytest.mark.parametrize('inp', ('fg34', 'f\\f', 'morning [glory]', '56.78'))
+def test_create_split_regex_no_params_unsplit(inp):
+    regex = helpers.create_split_regex({})
+
+    assert list(regex.split(inp)) == [inp]
+
+
+@pytest.mark.parametrize('inp,outp', [('here,there', ['here', 'there']),
+                                      ('ying;;yang', ['ying', 'yang']),
+                                      (';a; ;c;d,', ['', 'a', '', 'c', 'd', '']),
+                                      ('1,  3  ,5', ['1', '3', '5'])
+                                     ])
+def test_create_split_regex_no_params_split(inp, outp):
+    regex = helpers.create_split_regex({})
+
+    assert list(regex.split(inp)) == outp
+
+
+@pytest.mark.parametrize('delimiter', ['.', '\\', '[]', '   ', '/.*+'])
+def test_create_split_regex_custom(delimiter):
+    regex = helpers.create_split_regex({'delimiters': delimiter})
+
+    assert list(regex.split(f'out{delimiter}house')) == ['out', 'house']
+    assert list(regex.split('out,house')) == ['out,house']
+
+
+def test_create_split_regex_empty_delimiter():
+    with pytest.raises(UsageError):
+        regex = helpers.create_split_regex({'delimiters': ''})

test/python/tokenizer/sanitizers/test_split_name_list.py

@@ -5,7 +5,7 @@
 # Copyright (C) 2022 by the Nominatim developer community.
 # For a full list of authors see the git log.
 """
-Tests for the sanitizer that splitts multivalue lists.
+Tests for the sanitizer that splits multivalue lists.
 """
 import pytest