fix: recognize single-letter roman numeral suffixes in suffix-comma format (closes #136) (#174)

* fix: recognize single-letter suffix_not_acronyms (e.g. 'V') in suffix-comma format (closes #136)

Roman numeral suffixes like 'V' are single uppercase letters, which the
initial regex also matches. In suffix-comma format ("John Ingram, V") the
post-comma position is unambiguous, so the initial guard should not block
recognition. Add are_suffixes_after_comma() that bypasses the initial check
for suffix_not_acronyms members and use it at the suffix-comma detection site,
leaving is_suffix() unchanged for the no-comma path where 'V' mid-name should
still be treated as an initial.

* test: add coverage for suffix_not_acronyms members in suffix-comma format

Add tests for 'I' (same single-letter ambiguity as 'V'), a mixed
suffix-comma case (V MD), and two suffix_not_acronyms members together
(V Jr.). Also tighten the are_suffixes_after_comma docstring to describe
the short-circuit mechanism accurately and enumerate the full scope of
affected suffix_not_acronyms entries.

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: derek73

nameparser/parser.py

@@ -487,6 +487,23 @@ def are_suffixes(self, pieces: Iterable[str]) -> bool:
                 return False
         return True
 
+    def are_suffixes_after_comma(self, pieces: Iterable[str]) -> bool:
+        """Like are_suffixes, but pieces found in suffix_not_acronyms are
+        accepted unconditionally without passing through is_suffix().
+
+        Used when detecting suffix-comma format (e.g. "John Ingram, V") where
+        the post-comma position is unambiguous. This covers all
+        suffix_not_acronyms members (i, ii, iii, iv, v, jr, sr, etc.),
+        case-insensitively, including single-letter entries that is_suffix()
+        would otherwise reject via is_an_initial().
+        """
+        for piece in pieces:
+            if lc(piece) in self.C.suffix_not_acronyms:
+                continue
+            if not self.is_suffix(piece):
+                return False
+        return True
+
     def is_rootname(self, piece: str) -> bool:
         """
         Is not a known title, suffix or prefix. Just first, middle, last names.
@@ -686,7 +703,7 @@ def parse_full_name(self) -> None:
 
             post_comma_pieces = self.parse_pieces(parts[1].split(' '), 1)
 
-            if self.are_suffixes(parts[1].split(' ')) \
+            if self.are_suffixes_after_comma(parts[1].split(' ')) \
                     and len(parts[0].split(' ')) > 1:
 
                 # suffix comma:

tests/test_suffixes.py

@@ -34,6 +34,35 @@ def test_two_suffixes_lastname_comma_format(self) -> None:
         # NOTE: this adds a comma when the original format did not have one.
         self.m(hn.suffix, "Jr., MD", hn)
 
+    def test_roman_numeral_v_suffix_comma_format(self) -> None:
+        # suffix-comma position is unambiguous: 'V' must be a suffix, not a single-letter initial
+        hn = HumanName("John W. Ingram, V")
+        self.m(hn.first, "John", hn)
+        self.m(hn.middle, "W.", hn)
+        self.m(hn.last, "Ingram", hn)
+        self.m(hn.suffix, "V", hn)
+
+    def test_roman_numeral_i_suffix_comma_format(self) -> None:
+        # 'I' has the same single-letter ambiguity as 'V'
+        hn = HumanName("John W. Smith, I")
+        self.m(hn.first, "John", hn)
+        self.m(hn.middle, "W.", hn)
+        self.m(hn.last, "Smith", hn)
+        self.m(hn.suffix, "I", hn)
+
+    def test_suffix_not_acronym_then_acronym_suffix_comma_format(self) -> None:
+        # single-letter suffix_not_acronyms entry followed by an acronym suffix
+        hn = HumanName("John Smith, V MD")
+        self.m(hn.first, "John", hn)
+        self.m(hn.last, "Smith", hn)
+        self.m(hn.suffix, "V MD", hn)
+
+    def test_two_suffix_not_acronyms_suffix_comma_format(self) -> None:
+        hn = HumanName("John Smith, V Jr.")
+        self.m(hn.first, "John", hn)
+        self.m(hn.last, "Smith", hn)
+        self.m(hn.suffix, "V Jr.", hn)
+
     def test_two_suffixes_suffix_comma_format(self) -> None:
         hn = HumanName("Franklin Washington, Jr. MD")
         self.m(hn.first, "Franklin", hn)