v0.3.11: Fix #39, capitalization of initials with periods that are also conjunctions

Author: derek73

docs/release_log.rst

@@ -1,5 +1,7 @@
 Release Log
 ===========
+* 0.3.11 - October 17, 2015
+    - Fix bug capitalization exceptions (#39)
 * 0.3.10 - September 19, 2015
     - Fix encoding of byte strings on python 2.x (#37)
 * 0.3.9 - September 5, 2015

nameparser/__init__.py

@@ -1,4 +1,4 @@
-VERSION = (0, 3, 10)
+VERSION = (0, 3, 11)
 __version__ = '.'.join(map(str, VERSION))
 __author__ = "Derek Gulbranson"
 __author_email__ = '[email protected]'

nameparser/config/regexes.py

@@ -4,7 +4,7 @@
 
 REGEXES = set([
     ("spaces", re.compile(r"\s+", re.U)),
-    ("word", re.compile(r"\w+", re.U)),
+    ("word", re.compile(r"(\w|\.)+", re.U)),
     ("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)),

nameparser/parser.py

@@ -265,11 +265,11 @@ def is_title(self, value):
 
     def is_conjunction(self, piece):
         """Is in the conjuctions set and not :py:func:`is_an_initial()`."""
-        return lc(piece) in self.C.conjunctions and not self.is_an_initial(piece)
+        return piece.lower() in self.C.conjunctions and not self.is_an_initial(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)
+        return piece.lower() in self.C.prefixes and not self.is_an_initial(piece)
 
     def is_roman_numeral(self, value):
         """
@@ -657,7 +657,7 @@ def join_on_conjunctions(self, pieces, additional_parts_count=0):
 
     def cap_word(self, word):
         if self.is_prefix(word) or self.is_conjunction(word):
-            return lc(word)
+            return word.lower()
         exceptions = self.C.capitalization_exceptions
         if lc(word) in exceptions:
             return exceptions[lc(word)]

tests.py

@@ -1708,16 +1708,21 @@ def test_capitalization_with_Mac_as_hyphenated_names(self):
         hn.capitalize()
         self.m(str(hn), 'Donovan McNabb-Smith', hn)
 
+    def test_capitization_middle_initial_is_also_a_conjunction(self):
+        hn = HumanName('scott e. werner')
+        hn.capitalize()
+        self.m(str(hn), 'Scott E. Werner', hn)
+
     # Leaving already-capitalized names alone
-    def test123(self):
+    def test_no_change_to_mixed_chase(self):
         hn = HumanName('Shirley Maclaine')
         hn.capitalize()
         self.m(str(hn), 'Shirley Maclaine', hn)
 
     def test_capitalize_diacritics(self):
-        hn = HumanName('matth\xe4us schmidt')
+        hn = HumanName('matthëus schmidt')
         hn.capitalize()
-        self.m(u(hn), 'Matth\xe4us Schmidt', hn)
+        self.m(u(hn), 'Matthëus Schmidt', hn)
 
     # http://code.google.com/p/python-nameparser/issues/detail?id=15
     def test_downcasing_mac(self):
@@ -1966,7 +1971,8 @@ def test_variations_of_TEST_NAMES(self):
         name = sys.argv[1]
         hn = HumanName(name, encoding=sys.stdout.encoding)
         print((repr(hn)))
-        print((hn.capitalize()))
+        hn.capitalize()
+        print((repr(hn)))
     else:
         # if log.level > 0:
         #     for name in TEST_NAMES: