v0.3.8, use regex to check for roman numerals at end of name (#36)

Author: derek73

docs/release_log.rst

@@ -1,8 +1,11 @@
 Release Log
 ===========
+* 0.3.8 - September 2, 2015
+    - Use regex to check for roman numerals at end of name (#36)
+    - Add DVM to suffixes
 * 0.3.7 - August 30, 2015
-    - Make HumanName instances pickleable
     - Speed improvement, 3x faster
+    - Make HumanName instances pickleable
 * 0.3.6 - August 6, 2015
     - Fix strings that start with conjunctions (#20)
     - handle assigning lists of names to a name attribute

nameparser/__init__.py

@@ -1,4 +1,4 @@
-VERSION = (0, 3, 7)
+VERSION = (0, 3, 8)
 __version__ = '.'.join(map(str, VERSION))
 __author__ = "Derek Gulbranson"
 __author_email__ = '[email protected]'

nameparser/config/regexes.py

@@ -8,7 +8,7 @@
     ("mac", re.compile(r'^(ma?c)(\w+)', re.I | re.U)),
     ("initial", re.compile(r'^(\w\.|[A-Z])?$', re.U)),
     ("nickname", re.compile(r'\s*?[\("](.+?)[\)"]', re.U)),
-    ("roman_numeral", re.compile(r'^(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$', re.I | re.U)),
+    ("roman_numeral", re.compile(r'^(X|IX|IV|V?I{0,3})$', re.I | re.U)),
     ("no_vowels",re.compile(r'^[^aeyiuo]+$', re.I | re.U))
 ])
 """

nameparser/parser.py

@@ -270,6 +270,13 @@ def is_conjunction(self, piece):
     def is_prefix(self, piece):
         """Is in the prefixes set and not :py:func:`is_an_initial()`."""
         return lc(piece) in self.C.prefixes and not self.is_an_initial(piece)
+
+    def is_roman_numeral(self, value):
+        """
+        Matches the ``roman_numeral`` regular expression in 
+        :py:data:`~nameparser.config.regexes.REGEXES`.
+        """
+        return bool(self.C.regexes.roman_numeral.match(value))
 
     def is_suffix(self, piece):
         """Is in the suffixes set and not :py:func:`is_an_initial()`."""
@@ -295,7 +302,7 @@ def is_an_initial(self, value):
         Matches the ``initial`` regular expression in 
         :py:data:`~nameparser.config.regexes.REGEXES`.
         """
-        return self.C.regexes.initial.match(value) or False
+        return bool(self.C.regexes.initial.match(value))
 
     # def is_a_roman_numeral(value):
     #     return re_roman_numeral.match(value) or False
@@ -397,21 +404,25 @@ def parse_full_name(self):
             #            part[0]
 
             pieces = self.parse_pieces(parts)
-            
+            p_len = len(pieces)
             for i, piece in enumerate(pieces):
                 try:
                     nxt = pieces[i + 1]
                 except IndexError:
                     nxt = None
 
                 # title must have a next piece, unless it's just a title
-                if self.is_title(piece) and (nxt or len(pieces) == 1):
+                if self.is_title(piece) and (nxt or p_len == 1):
                     self.title_list.append(piece)
                     continue
                 if not self.first:
                     self.first_list.append(piece)
                     continue
-                if self.are_suffixes(pieces[i+1:]):
+                if self.are_suffixes(pieces[i+1:]) or \
+                        (self.is_roman_numeral(nxt) and i == p_len - 2 \
+                         and not self.is_an_initial(piece)):
+                        # if the next piece is the last piece and a roman numeral 
+                        # but this piece is not an initial
                     self.last_list.append(piece)
                     self.suffix_list += pieces[i+1:]
                     break

tests.py

@@ -1135,7 +1135,8 @@ def test_title_with_three_part_name_last_initial_is_suffix_uppercase_no_period(s
         hn = HumanName("King John Alexander V")
         self.m(hn.title, "King", hn)
         self.m(hn.first, "John", hn)
-        self.m(hn.last, "V", hn)
+        self.m(hn.last, "Alexander", hn)
+        self.m(hn.suffix, "V", hn)
 
     def test_four_name_parts_with_suffix_that_could_be_initial_lowercase_no_period(self):
         hn = HumanName("larry james edward johnson v")
@@ -1144,19 +1145,20 @@ def test_four_name_parts_with_suffix_that_could_be_initial_lowercase_no_period(s
         self.m(hn.last, "johnson", hn)
         self.m(hn.suffix, "v", hn)
 
-    @unittest.expectedFailure
     def test_four_name_parts_with_suffix_that_could_be_initial_uppercase_no_period(self):
         hn = HumanName("Larry James Johnson I")
         self.m(hn.first, "Larry", hn)
         self.m(hn.middle, "James", hn)
         self.m(hn.last, "Johnson", hn)
-        # if it's in upper case, we currently assume it's an initial
-        # it's not really clear if we can assume it's one or the other.
-        # If they really are the "first", they are probably used to using a 
-        # comma to avoid confusion. Humans know that "Johnson" is a last name,
-        # but that wouldn't really be a "simple" nameparser. 
         self.m(hn.suffix, "I", hn)
 
+    def test_roman_numeral_initials(self):
+        hn = HumanName("Larry V I")
+        self.m(hn.first, "Larry", hn)
+        self.m(hn.middle, "V", hn)
+        self.m(hn.last, "I", hn)
+        self.m(hn.suffix, "", hn)
+
     # tests for Rev. title (Reverend)
     def test124(self):
         hn = HumanName("Rev. John A. Kenneth Doe")