S singular fix

pattern/text/en/inflect.py

@@ -168,7 +168,6 @@ def referenced(word, article=INDEFINITE):
     ), # 4) Words that do not inflect.
     ((          r"$", ""  , "uninflected", False),
      (          r"$", ""  , "uncountable", False),
-     (         r"s$", "s" , "s-singular" , False),
      (      r"fish$", "fish"       , None, False),
      (r"([- ])bass$", "\\1bass"    , None, False),
      (       r"ois$", "ois"        , None, False),
@@ -240,11 +239,12 @@ def referenced(word, article=INDEFINITE):
      (         r"o$", "i"    ,     "o-i*", True),
      (          r"$", "i"    ,      "-i*", True),
      (          r"$", "im"   ,     "-im*", True)
-    ), # 9) -ch, -sh and -ss take -es in the plural 
+    ), # 9) -ch, -sh and -ss and the s-singular group take -es in the plural
        #    (e.g., churches, classes).
-    ((   r"([cs])h$", "\\1hes"     , None, False),
-     (        r"ss$", "sses"       , None, False),
-     (         r"x$", "xes"        , None, False)
+    ((   r"([cs])h$", "\\1hes"     , None,         False),
+     (        r"ss$", "sses"       , None,         False),
+     (         r"x$", "xes"        , None,         False),
+     (         r"s$", "ses"        , "s-singular", False)
     ), # 10) -f or -fe sometimes take -ves in the plural 
        #     (e.g, lives, wolves).
     (( r"([aeo]l)f$", "\\1ves"     , None, False),
@@ -291,17 +291,19 @@ def referenced(word, article=INDEFINITE):
         "advice"     , "fruit"      , "ketchup"      , "meat"       , "sand"         ,
         "bread"      , "furniture"  , "knowledge"    , "mustard"    , "software"     ,
         "butter"     , "garbage"    , "love"         , "news"       , "understanding",
-        "cheese"     , "gravel"     , "luggage"      , "progress"   , "water"
-        "electricity", "happiness"  , "mathematics"  , "research"   , 
-        "equipment"  , "information", "mayonnaise"   , "rice"
+        "cannabis"   , "gravel"     , "luggage"      , "progress"   , "water"
+        "cheese"     , "happiness"  , "mathematics"  , "research"   ,
+        "electricity", "information", "mayonnaise"   , "rice",
+        "equipment"
         ],
     "s-singular": [
         "acropolis"  , "caddis"     , "dais"         , "glottis"    , "pathos"       ,
-        "aegis"      , "cannabis"   , "digitalis"    , "ibis"       , "pelvis"       ,
-        "alias"      , "canvas"     , "epidermis"    , "lens"       , "polis"        ,
-        "asbestos"   , "chaos"      , "ethos"        , "mantis"     , "rhinoceros"   ,
-        "bathos"     , "cosmos"     , "gas"          , "marquis"    , "sassafras"    ,
-        "bias"       ,                "glottis"      , "metropolis" , "trellis"
+        "aegis"      , "canvas"     , "digitalis"    , "ibis"       , "pelvis"       ,
+        "alias"      , "chaos"      , "epidermis"    , "lens"       , "polis"        ,
+        "asbestos"   , "cosmos"     , "ethos"        , "mantis"     , "rhinoceros"   ,
+        "bathos"     ,                "gas"          , "marquis"    , "sassafras"    ,
+        "bias"       ,                "glottis"      , "metropolis" , "trellis",
+        "bus"
         ],
     "ex-ices": [
         "codex"      , "murex"      , "silex"
@@ -542,6 +544,7 @@ def pluralize(word, pos=NOUN, custom={}, classical=True):
     "collie"     , "hankie"   , "lingerie"    , "reverie"    , "toughie"   , 
     "cookie"     , "hippie"   , "meanie"      , "rookie"     , "valkyrie"  , 
 ))
+singular_s = set(plural_categories['s-singular'])
 singular_irregular = {
        "atlantes": "atlas", 
         "atlases": "atlas", 
@@ -609,7 +612,10 @@ def singularize(word, pos=NOUN, custom={}):
             return word
     for x in singular_ie:
         if w.endswith(x+"s"):
-            return w
+            return word[:-1]
+    for x in singular_s:
+        if w.endswith(x+"es"):
+            return word[:-2]
     for x in singular_irregular:
         if w.endswith(x):
             return re.sub('(?i)'+x+'$', singular_irregular[x], word)

test/test_en.py

@@ -200,6 +200,39 @@ def test_superlative(self):
 
 #---------------------------------------------------------------------------------------------------
 
+class WhiteboxInflectionTestCase(unittest.TestCase):
+
+    def test_s_singular_pluralize(self):
+        self.assertEquals(en.inflect.pluralize('lens'), 'lenses')
+
+    def test_diagnosis_pluralize(self):
+        self.assertEquals(en.inflect.pluralize('diagnosis'), 'diagnoses')
+
+    def test_s_singular_singularize(self):
+        self.assertEquals(en.inflect.singularize('lenses'), 'lens')
+
+    def test_ie_singular_singularize(self):
+        self.assertEquals(en.inflect.singularize('cuties'), 'cutie')
+
+    def test_all_s_singular(self):
+        l = en.inflect.plural_categories['s-singular']
+        singularize = en.inflect.singularize
+        pluralize = en.inflect.pluralize
+        for word in l:
+            self.assertEquals(pluralize(word), word+'es')
+            self.assertEquals(singularize(pluralize(word)), word)
+
+    def test_all_ie_singular(self):
+        l = en.inflect.singular_ie
+        singularize = en.inflect.singularize
+        pluralize = en.inflect.pluralize
+        for word in l:
+            self.assertEquals(pluralize(word), word+'s')
+            self.assertEquals(singularize(pluralize(word)), word)
+
+
+#---------------------------------------------------------------------------------------------------
+
 class TestQuantification(unittest.TestCase):
 
     def setUp(self):