Recognize Arabic and fullwidth CJK comma as name-part delimiters

The "Lastname, Firstname" split only matched the ASCII comma, so Arabic
(U+060C) and fullwidth (U+FF0C) commas failed to trigger the comma format
and leaked into the parsed output. Adds a shared "commas" regex used for
both the split and the trailing-comma strip.

Fixes #265.

Author: derek73

nameparser/config/regexes.py

@@ -29,6 +29,10 @@
     "bidi": re_bidi,
     "phd": re.compile(r'\s(ph\.?\s+d\.?)', re.I),
     "space_before_comma": re.compile(r'\s+,'),
+    # ASCII comma plus its Arabic (U+060C) and fullwidth CJK (U+FF0C)
+    # counterparts, used to split "Last, First" format and to strip a
+    # trailing comma before parsing (#265).
+    "commas": re.compile(r'[,،，]'),
     "east_slavic_patronymic": re.compile(
         r'(ovich|ovna|evich|evna|ichna|ilyich|kuzmich|lukich|fomich|fokich)$',
         re.I,

nameparser/parser.py

@@ -923,7 +923,7 @@ def _apply_full_name(self, value: str | bytes, *, stacklevel: int) -> None:
     def collapse_whitespace(self, string: str) -> str:
         # collapse multiple spaces into single space
         string = self.C.regexes.spaces.sub(" ", string.strip())
-        if string.endswith(","):
+        if string and self.C.regexes.commas.fullmatch(string[-1]):
             string = string[:-1]
         return string
 
@@ -1196,7 +1196,7 @@ def parse_full_name(self) -> None:
         self._full_name = self.collapse_whitespace(self._full_name)
 
         # break up full_name by commas
-        parts = [x.strip() for x in self._full_name.split(",")]
+        parts = [x.strip() for x in self.C.regexes.commas.split(self._full_name)]
         self._had_comma = len(parts) > 1
 
         log.debug("full_name: %s", self._full_name)

tests/test_comma_variants.py

@@ -0,0 +1,28 @@
+from nameparser import HumanName
+
+from tests.base import HumanNameTestBase
+
+
+class HumanNameCommaVariantsTests(HumanNameTestBase):
+    """Non-ASCII comma characters should split "Last, First" the same as ',' (#265)."""
+
+    def test_arabic_comma_splits_lastname_format(self) -> None:
+        hn = HumanName("سلمان، محمد")
+        self.m(hn.first, "محمد", hn)
+        self.m(hn.last, "سلمان", hn)
+
+    def test_fullwidth_comma_splits_lastname_format(self) -> None:
+        hn = HumanName("Smith，John")
+        self.m(hn.first, "John", hn)
+        self.m(hn.last, "Smith", hn)
+
+    def test_arabic_comma_does_not_pollute_output(self) -> None:
+        hn = HumanName("سلمان، محمد")
+        self.assertNotIn("،", hn.last)
+        self.assertNotIn("،", str(hn))
+
+    def test_trailing_arabic_comma_stripped(self) -> None:
+        # matches ASCII behavior: a single word with a trailing comma has
+        # nothing after the comma, so it's a bare name, not "Last,"
+        hn = HumanName("سلمان،")
+        self.m(hn.first, "سلمان", hn)