Guard comma split against a custom regexes dict missing "commas"

RegexTupleManager falls back to EMPTY_REGEX for any key absent from a
custom regexes dict, and re.compile('').split() matches between every
character rather than not splitting -- so a Constants built with a
custom regexes set that omits "commas" shattered names into single
characters instead of just disabling the comma split. Also clarifies
the suffix_delimiter docstring to mention the comma variants.

Found in review of #281.

Author: derek73

nameparser/config/__init__.py

@@ -590,8 +590,9 @@ class Constants:
     ``None`` (no additional splitting beyond the standard comma split).
 
     Note: setting this to ``","`` or ``", "`` has no additional effect —
-    the full name is already split on bare commas first, and each resulting
-    part is stripped of surrounding whitespace before this step runs.
+    the full name is already split on comma characters first (including the
+    Arabic ``،`` and fullwidth ``，`` variants), and each resulting part is
+    stripped of surrounding whitespace before this step runs.
 
     The delimiter is only applied to parts once they've been identified as
     a suffix group, so it never leaks into a first- or middle-name part. For

nameparser/parser.py

@@ -1195,8 +1195,14 @@ 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.C.regexes.commas.split(self._full_name)]
+        # break up full_name by commas. A missing "commas" key in a custom
+        # regexes dict falls back to RegexTupleManager's EMPTY_REGEX, whose
+        # .split() matches between every character rather than not
+        # splitting at all -- guard against that so a custom regexes dict
+        # that omits "commas" disables the comma split instead of shattering
+        # the name into single characters.
+        commas = self.C.regexes.commas
+        parts = [x.strip() for x in (commas.split(self._full_name) if commas.pattern else [self._full_name])]
         self._had_comma = len(parts) > 1
 
         log.debug("full_name: %s", self._full_name)

tests/test_comma_variants.py

@@ -1,4 +1,6 @@
 from nameparser import HumanName
+from nameparser.config import Constants
+from nameparser.config.regexes import REGEXES
 
 from tests.base import HumanNameTestBase
 
@@ -26,3 +28,20 @@ def test_trailing_arabic_comma_stripped(self) -> None:
         # nothing after the comma, so it's a bare name, not "Last,"
         hn = HumanName("سلمان،")
         self.m(hn.first, "سلمان", hn)
+
+    def test_custom_regexes_without_commas_key_does_not_shatter_name(self) -> None:
+        # A custom regexes dict that omits "commas" entirely must not fall
+        # back to RegexTupleManager's EMPTY_REGEX default for splitting --
+        # re.compile('').split(...) matches between every character, which
+        # explodes any name into single-char pieces instead of leaving it
+        # unsplit (the EMPTY_REGEX convention elsewhere in this codebase
+        # means "feature disabled", not "split on every character").
+        # With comma splitting disabled, "Smith, John" is tokenized like any
+        # other no-comma input (word tokenizing drops the punctuation),
+        # yielding a plain first/last pair -- not the inverted "Last, First"
+        # reading, and definitely not single-character pieces.
+        custom = {k: v for k, v in REGEXES.items() if k != 'commas'}
+        c = Constants(regexes=custom)
+        hn = HumanName("Smith, John", constants=c)
+        self.m(hn.first, "Smith", hn)
+        self.m(hn.last, "John", hn)