fix #8, add support for chained suffixes. remove support for titles with periods in them, e.g "Lt.Gen."

current parser logic won't support both, seems fixable though

Author: derek73

docs/index.rst

@@ -30,7 +30,7 @@ Supports 3 different comma placement variations in the input string.
 
 * Title Firstname "Nickname" Middle Middle Lastname Suffix
 * Lastname [Suffix], Title Firstname (Nickname) Middle Middle[,] Suffix [, Suffix]
-* Title Firstname M Lastname [Suffix], Suffix [, Suffix]
+* Title Firstname M Lastname [Suffix], Suffix [Suffix] [, Suffix]
 
 When there is ambiguity that cannot be resolved by a rule-based approach,
 HumanName prefers to handle the most common cases correctly. For example,

nameparser/config/suffixes.py

@@ -18,6 +18,8 @@
     'chfc',
     'cfp',
     'md',
+    'mba',
+    'ma',
     'phd',
     'mp',
     'qc',

nameparser/parser.py

@@ -2,11 +2,11 @@
 from __future__ import unicode_literals
 
 import logging
-from .util import u
-from .util import text_type
-from .util import lc
-from .config import CONSTANTS
-from .config import Constants
+from nameparser.util import u
+from nameparser.util import text_type
+from nameparser.util import lc
+from nameparser.config import CONSTANTS
+from nameparser.config import Constants
 
 # http://code.google.com/p/python-nameparser/issues/detail?id=10
 log = logging.getLogger('HumanName')
@@ -264,7 +264,15 @@ def is_prefix(self, piece):
 
     def is_suffix(self, piece):
         """Is in the suffixes set or :py:func:`is_an_initial()`."""
-        return lc(piece) in self.C.suffixes and not self.is_an_initial(piece)
+        # suffixes may have periods inside them like "M.D."
+        return lc(piece).replace('.','') in self.C.suffixes and not self.is_an_initial(piece)
+    
+    def are_suffixes(self, pieces):
+        """Return True if all pieces are suffixes."""
+        for piece in pieces:
+            if not self.is_suffix(piece):
+                return False
+        return True
 
     def is_rootname(self, piece):
         '''Is not a known title, suffix or prefix. Just first, middle, last names.'''
@@ -391,26 +399,24 @@ def parse_full_name(self):
                 if not self.first:
                     self.first_list.append(piece)
                     continue
-                if (i == len(pieces) - 2) and self.is_suffix(nxt):
+                if self.are_suffixes(pieces[i+1:]):
                     self.last_list.append(piece)
-                    self.suffix_list.append(nxt)
+                    self.suffix_list += pieces[i+1:]
                     break
                 if not nxt:
                     self.last_list.append(piece)
                     continue
 
                 self.middle_list.append(piece)
         else:
-            if self.is_suffix(parts[1]):
+            if self.are_suffixes(parts[1].split(' ')):
 
-                # suffix comma: title first middle last [suffix], suffix [, suffix]
+                # suffix comma: title first middle last [suffix], suffix [suffix] [, suffix]
                 #               parts[0],                         parts[1:...]
 
                 self.suffix_list += parts[1:]
-                
                 pieces = self.parse_pieces(parts[0].split(' '))
                 log.debug("pieces: {0}".format(u(pieces)))
-                
                 for i, piece in enumerate(pieces):
                     try:
                         nxt = pieces[i + 1]
@@ -423,9 +429,9 @@ def parse_full_name(self):
                     if not self.first:
                         self.first_list.append(piece)
                         continue
-                    if (i == len(pieces) - 2) and self.is_suffix(nxt):
+                    if self.are_suffixes(pieces[i+1:]):
                         self.last_list.append(piece)
-                        self.suffix_list.insert(0,nxt)
+                        self.suffix_list = pieces[i+1:] + self.suffix_list
                         break
                     if not nxt:
                         self.last_list.append(piece)
@@ -475,6 +481,20 @@ def parse_full_name(self):
             self.unparsable = False
             self.post_process()
 
+
+    # def split_periods(self, pieces):
+    #     """
+    #     If there is a period that is not at the end of a piece, split it on periods.
+    #     """
+    #     tmp = []
+    #     for piece in pieces:
+    #         if piece[:-1].find('.') >= 0:
+    #             p = [_f for _f in piece.split('.') if _f]
+    #             tmp += [x+'.' for x in p]
+    #         else:
+    #             tmp += [piece]
+    #     return tmp
+
     def parse_pieces(self, parts, additional_parts_count=0):
         """
         Split parts on spaces and remove commas, join on conjunctions and
@@ -489,21 +509,11 @@ def parse_pieces(self, parts, additional_parts_count=0):
         :rtype: list
         """
 
-        ps = []
+        tmp = []
         for part in parts:
-            ps += [x.strip(' ,') for x in part.split(' ')]
-        
-        # if there is a period that is not at the end of a piece, split it on periods
-        pieces = []
-        for piece in ps:
-            if piece[:-1].find('.') >= 0:
-                p = [_f for _f in piece.split('.') if _f]
-                pieces += [x+'.' for x in p]
-            else:
-                pieces += [piece]
-        
+            tmp += [x.strip(' ,') for x in part.split(' ')]
 
-        return self.join_on_conjunctions(pieces, additional_parts_count)
+        return self.join_on_conjunctions(tmp, additional_parts_count)
 
     def join_on_conjunctions(self, pieces, additional_parts_count=0):
         """

nameparser/util.py

@@ -22,4 +22,4 @@ def lc(value):
     """Lower case and remove any periods to normalize for comparison."""
     if not value:
         return ''
-    return value.lower().replace('.','')
+    return value.lower().strip('.')

tests.py

@@ -1044,7 +1044,9 @@ def test_lowercase_middle_initial_comma_lastname_and_suffix_conflict_with_conjun
         self.m(hn.last, "Smith", hn)
         self.m(hn.suffix, "III, Jr", hn)
 
+    @unittest.expectedFailure
     def test_two_initials_conflict_with_conjunction(self):
+        # Supporting this seems to screw up titles with periods in them like M.B.A.
         hn = HumanName('E.T. Smith')
         self.m(hn.first, "E.", hn)
         self.m(hn.middle, "T.", hn)
@@ -1270,7 +1272,85 @@ def test_parenthesis_are_removed2(self):
         self.m(hn.suffix, "Jr.", hn)
 
 
+class SuffixesTestCase(HumanNameTestBase):
+    
+    def test_suffix(self):
+        hn = HumanName("Joe Franklin Jr")
+        self.m(hn.first, "Joe", hn)
+        self.m(hn.last, "Franklin", hn)
+        self.m(hn.suffix, "Jr", hn)
+
+    def test_suffix_with_periods(self):
+        hn = HumanName("Joe Dentist D.D.S.")
+        self.m(hn.first, "Joe", hn)
+        self.m(hn.last, "Dentist", hn)
+        self.m(hn.suffix, "D.D.S.", hn)
+
+    def test_two_suffixes(self):
+        hn = HumanName("Kenneth Clarke QC MP")
+        self.m(hn.first, "Kenneth", hn)
+        self.m(hn.last, "Clarke", hn)
+        # NOTE: this adds a comma when the orginal format did not have one. 
+        # not ideal but at least its in the right bucket
+        self.m(hn.suffix, "QC, MP", hn)
+
+    def test_two_suffixes_lastname_comma_format(self):
+        hn = HumanName("Washington Jr. MD, Franklin")
+        self.m(hn.first, "Franklin", hn)
+        self.m(hn.last, "Washington", hn)
+        # NOTE: this adds a comma when the orginal format did not have one. 
+        self.m(hn.suffix, "Jr., MD", hn)
+
+    def test_two_suffixes_suffix_comma_format(self):
+        hn = HumanName("Franklin Washington, Jr. MD")
+        self.m(hn.first, "Franklin", hn)
+        self.m(hn.last, "Washington", hn)
+        self.m(hn.suffix, "Jr. MD", hn)
+
+    def test_suffix_containing_periods(self):
+        hn = HumanName("Kenneth Clarke Q.C.")
+        self.m(hn.first, "Kenneth", hn)
+        self.m(hn.last, "Clarke", hn)
+        self.m(hn.suffix, "Q.C.", hn)
+
+    def test_suffix_containing_periods_lastname_comma_format(self):
+        hn = HumanName("Clarke, Kenneth, Q.C. M.P.")
+        self.m(hn.first, "Kenneth", hn)
+        self.m(hn.last, "Clarke", hn)
+        self.m(hn.suffix, "Q.C. M.P.", hn)
+
+    def test_suffix_containing_periods_suffix_comma_format(self):
+        hn = HumanName("Kenneth Clarke Q.C., M.P.")
+        self.m(hn.first, "Kenneth", hn)
+        self.m(hn.last, "Clarke", hn)
+        self.m(hn.suffix, "Q.C., M.P.", hn)
+
+    def test_suffix_with_single_comma_format(self):
+        hn = HumanName("John Doe jr., MD")
+        self.m(hn.first, "John", hn)
+        self.m(hn.last, "Doe", hn)
+        self.m(hn.suffix, "jr., MD", hn)
+
+    def test_suffix_with_double_comma_format(self):
+        hn = HumanName("Doe, John jr., MD")
+        self.m(hn.first, "John", hn)
+        self.m(hn.last, "Doe", hn)
+        self.m(hn.suffix, "jr., MD", hn)
+
+    #http://en.wikipedia.org/wiki/Ma_(surname)
+    def test_potential_suffix_that_is_also_last_name(self):
+        hn = HumanName("Jack Ma")
+        self.m(hn.first, "Jack", hn)
+        self.m(hn.last, "Ma", hn)
+    
+    def test_potential_suffix_that_is_also_last_name_with_suffix(self):
+        hn = HumanName("Jack Ma Jr")
+        self.m(hn.first, "Jack", hn)
+        self.m(hn.last, "Ma", hn)
+        self.m(hn.suffix, "Jr", hn)
+
 class HumanNameTitleTestCase(HumanNameTestBase):
+
     def test_last_name_is_also_title(self):
         hn = HumanName("Amy E Maid")
         self.m(hn.first, "Amy", hn)
@@ -1306,18 +1386,6 @@ def test_title_is_title(self):
         hn = HumanName("Coach")
         self.m(hn.title, "Coach", hn)
 
-    def test_suffix_with_single_comma_format(self):
-        hn = HumanName("John Doe jr., MD")
-        self.m(hn.first, "John", hn)
-        self.m(hn.last, "Doe", hn)
-        self.m(hn.suffix, "jr., MD", hn)
-   
-    def test_suffix_with_double_comma_format(self):
-        hn = HumanName("Doe, John jr., MD")
-        self.m(hn.first, "John", hn)
-        self.m(hn.last, "Doe", hn)
-        self.m(hn.suffix, "jr., MD", hn)
-   
     # TODO: fix handling of U.S.
     @unittest.expectedFailure
     def test_chained_title_first_name_initial(self):
@@ -1354,17 +1422,17 @@ def test_chained_hyphenated_title_with_comma_suffix(self):
         self.m(hn.middle, "G", hn)
         self.m(hn.last, "Davis", hn)
         self.m(hn.suffix, "III", hn)
-    
+
     @unittest.expectedFailure
     def test_title_multiple_titles_with_conjunctions(self):
-        # I think it finds the index of the wrong 'the'. I get confused because it
+        # FIXME: I think it finds the index of the wrong 'the'. I get confused because it
         # loops in reverse order.
         hn = HumanName("The Right Hon. the President of the Queen's Bench Division")
         self.m(hn.title, "The Right Hon. the President of the Queen's Bench Division", hn)
 
     @unittest.expectedFailure
     def test_conjunction_before_title(self):
-        # TODO: seems fixable
+        # FIXME: seems fixable
         hn = HumanName('The Lord of the Universe')
         self.m(hn.title, "The Lord of the Universe", hn)
 
@@ -1388,7 +1456,9 @@ def test_title_with_last_initial_is_suffix(self):
         self.m(hn.first, "John", hn)
         self.m(hn.last, "V.", hn)
 
-    def test_lc_comparison_of_title(self):
+    @unittest.expectedFailure
+    def test_two_title_parts_separated_by_commas(self):
+        # supporting this currently messes up supporting suffixes like M.B.A.
         hn = HumanName("Lt.Gen. John A. Kenneth Doe IV")
         self.m(hn.title, "Lt. Gen.", hn)
         self.m(hn.first, "John", hn)
@@ -1438,10 +1508,11 @@ def test_possible_conflict_with_suffix_that_could_be_initial(self):
 
     # 'ben' is removed from PREFIXES in v0.2.5
     # this test could re-enable this test if we decide to support 'ben' as a prefix
-    # def test_ben_as_conjunction(self):
-    #     hn = HumanName("Ahmad ben Husain")
-    #     self.m(hn.first,"Ahmad", hn)
-    #     self.m(hn.last,"ben Husain", hn)
+    @unittest.expectedFailure
+    def test_ben_as_conjunction(self):
+        hn = HumanName("Ahmad ben Husain")
+        self.m(hn.first,"Ahmad", hn)
+        self.m(hn.last,"ben Husain", hn)
 
     def test_ben_as_first_name(self):
         hn = HumanName("Ben Johnson")