Allow alternative space characters as group separator when parsing numbers (#1007)

The French group separator is `"\u202f"` (narrow non-breaking space),
but when parsing numbers in the real world, you will most often encounter
either a regular space character (`" "`) or a non-breaking space character
(`"\xa0"`).

The issue was partially adressed earlier in #637,
but only to allow regular spaces instead of non-breaking spaces `"\xa0"` in
`parse_decimal`.

This commit goes further by changing both `parse_number` and `parse_decimal`
to allow certain other space characters when the group character is itself a space character,
but is not present in the string to parse.

Unit tests are included.

Author: ronnix

babel/numbers.py

@@ -998,6 +998,15 @@ def __init__(self, message: str, suggestions: list[str] | None = None) -> None:
         self.suggestions = suggestions
 
 
+SPACE_CHARS = {
+    ' ',  # space
+    '\xa0',  # no-break space
+    '\u202f',  # narrow no-break space
+}
+
+SPACE_CHARS_RE = re.compile('|'.join(SPACE_CHARS))
+
+
 def parse_number(
     string: str,
     locale: Locale | str | None = LC_NUMERIC,
@@ -1026,8 +1035,18 @@ def parse_number(
     :raise `NumberFormatError`: if the string can not be converted to a number
     :raise `UnsupportedNumberingSystemError`: if the numbering system is not supported by the locale.
     """
+    group_symbol = get_group_symbol(locale, numbering_system=numbering_system)
+
+    if (
+        group_symbol in SPACE_CHARS and  # if the grouping symbol is a kind of space,
+        group_symbol not in string and  # and the string to be parsed does not contain it,
+        SPACE_CHARS_RE.search(string)  # but it does contain any other kind of space instead,
+    ):
+        # ... it's reasonable to assume it is taking the place of the grouping symbol.
+        string = SPACE_CHARS_RE.sub(group_symbol, string)
+
     try:
-        return int(string.replace(get_group_symbol(locale, numbering_system=numbering_system), ''))
+        return int(string.replace(group_symbol, ''))
     except ValueError as ve:
         raise NumberFormatError(f"{string!r} is not a valid number") from ve
 
@@ -1085,12 +1104,12 @@ def parse_decimal(
     decimal_symbol = get_decimal_symbol(locale, numbering_system=numbering_system)
 
     if not strict and (
-        group_symbol == '\xa0' and  # if the grouping symbol is U+00A0 NO-BREAK SPACE,
+        group_symbol in SPACE_CHARS and  # if the grouping symbol is a kind of space,
         group_symbol not in string and  # and the string to be parsed does not contain it,
-        ' ' in string  # but it does contain a space instead,
+        SPACE_CHARS_RE.search(string)  # but it does contain any other kind of space instead,
     ):
         # ... it's reasonable to assume it is taking the place of the grouping symbol.
-        string = string.replace(' ', group_symbol)
+        string = SPACE_CHARS_RE.sub(group_symbol, string)
 
     try:
         parsed = decimal.Decimal(string.replace(group_symbol, '')

tests/test_numbers.py

@@ -751,6 +751,15 @@ def test_parse_number():
     with pytest.raises(numbers.UnsupportedNumberingSystemError):
         numbers.parse_number('1.099,98', locale='en', numbering_system="unsupported")
 
+@pytest.mark.parametrize('string', [
+    '1 099',
+    '1\xa0099',
+    '1\u202f099',
+])
+def test_parse_number_group_separator_can_be_any_space(string):
+    assert numbers.parse_number(string, locale='fr') == 1099
+
+
 def test_parse_decimal():
     assert (numbers.parse_decimal('1,099.98', locale='en_US')
             == decimal.Decimal('1099.98'))
@@ -761,6 +770,15 @@ def test_parse_decimal():
     assert excinfo.value.args[0] == "'2,109,998' is not a valid decimal number"
 
 
+@pytest.mark.parametrize('string', [
+    '1 099,98',
+    '1\xa0099,98',
+    '1\u202f099,98',
+])
+def test_parse_decimal_group_separator_can_be_any_space(string):
+    assert decimal.Decimal('1099.98') == numbers.parse_decimal(string, locale='fr')
+
+
 def test_parse_grouping():
     assert numbers.parse_grouping('##') == (1000, 1000)
     assert numbers.parse_grouping('#,###') == (3, 3)