* Add is_oov property, and fix up handling of attributes

spacy/attrs.pxd

@@ -12,19 +12,8 @@ cpdef enum attr_id_t:
     LIKE_NUM
     LIKE_EMAIL
     IS_STOP
-    FLAG0
-    FLAG1
-    FLAG2
-    FLAG3
-    FLAG4
-    FLAG5
-    FLAG6
-    FLAG7
-    FLAG8
-    FLAG9
-    FLAG10
-    FLAG11
-    FLAG12
+    IS_OOV
+    
     FLAG13
     FLAG14
     FLAG15

spacy/en/__init__.py

@@ -25,9 +25,9 @@ from ..util import read_lang_data
 from ..attrs import TAG, HEAD, DEP, ENT_TYPE, ENT_IOB
 
 
-def get_lex_props(string, oov_prob=-30):
+def get_lex_props(string, oov_prob=-30, is_oov=False):
     return {
-        'flags': get_flags(string),
+        'flags': get_flags(string, is_oov=is_oov),
         'length': len(string),
         'orth': string,
         'lower': string.lower(),

spacy/en/attrs.pxd

@@ -1,4 +1,4 @@
-from ..attrs cimport FLAG11, FLAG12, FLAG13, FLAG14
+from ..attrs cimport FLAG13, FLAG14
 from ..attrs cimport FLAG15, FLAG16, FLAG17, FLAG18, FLAG19, FLAG20, FLAG21
 from ..attrs cimport FLAG22, FLAG23, FLAG24, FLAG25, FLAG26, FLAG27, FLAG28
 from ..attrs cimport FLAG29,  FLAG30, FLAG31, FLAG32
@@ -10,6 +10,7 @@ from ..attrs cimport IS_PUNCT as _IS_PUNCT
 from ..attrs cimport IS_SPACE as _IS_SPACE
 from ..attrs cimport IS_TITLE as _IS_TITLE
 from ..attrs cimport IS_UPPER as _IS_UPPER
+from ..attrs cimport IS_OOV as _IS_OOV
 from ..attrs cimport LIKE_EMAIL as _LIKE_EMAIL
 from ..attrs cimport LIKE_URL as _LIKE_URL
 from ..attrs cimport LIKE_NUM as _LIKE_NUM
@@ -43,6 +44,7 @@ cpdef enum:
     LIKE_NUM = _LIKE_NUM
     LIKE_EMAIL = _LIKE_EMAIL
     IS_STOP = _IS_STOP
+    IS_OOV = _IS_OOV
  
     ORTH = _ORTH
     SHAPE = _SHAPE

spacy/lexeme.pyx

@@ -11,6 +11,7 @@ import numpy
 
 from .attrs cimport IS_ALPHA, IS_ASCII, IS_DIGIT, IS_LOWER, IS_PUNCT, IS_SPACE
 from .attrs cimport IS_TITLE, IS_UPPER, LIKE_URL, LIKE_NUM, LIKE_EMAIL, IS_STOP
+from .attrs cimport IS_OOV
 
 
 memset(&EMPTY_LEXEME, 0, sizeof(LexemeC))
@@ -51,6 +52,9 @@ cdef class Lexeme:
         cdef flags_t one = 1
         return self.flags & (one << flag_id)
 
+    property is_oov:
+        def __get__(self): return self.check_flag(IS_OOV)
+
     property is_alpha:
         def __get__(self): return self.check_flag(IS_ALPHA)
     

spacy/tokens/token.pyx

@@ -18,8 +18,7 @@ from ..parts_of_speech cimport CONJ, PUNCT
 
 from ..attrs cimport IS_ALPHA, IS_ASCII, IS_DIGIT, IS_LOWER, IS_PUNCT, IS_SPACE
 from ..attrs cimport IS_TITLE, IS_UPPER, LIKE_URL, LIKE_NUM, LIKE_EMAIL, IS_STOP
-
-
+from ..attrs cimport IS_OOV
 
 
 cdef class Token:
@@ -286,7 +285,9 @@ cdef class Token:
         def __get__(self):
             return self.vocab.strings[self.c.dep]
 
-    
+    property is_oov:
+        def __get__(self): return check_flag(self.c.lex, IS_OOV)
+
     property is_alpha:
         def __get__(self): return check_flag(self.c.lex, IS_ALPHA)
     

spacy/vocab.pyx

@@ -95,7 +95,7 @@ cdef class Vocab:
         if len(string) < 3:
             mem = self.mem
         lex = <LexemeC*>mem.alloc(sizeof(LexemeC), 1)
-        props = self.lexeme_props_getter(string, self.oov_prob)
+        props = self.lexeme_props_getter(string, self.oov_prob, is_oov=is_oov)
         set_lex_struct_props(lex, props, self.strings, EMPTY_VEC)
         if is_oov:
             lex.id = 0
@@ -119,7 +119,7 @@ cdef class Vocab:
         if len(string) < 3:
             mem = self.mem
         lex = <LexemeC*>mem.alloc(sizeof(LexemeC), 1)
-        props = self.lexeme_props_getter(string)
+        props = self.lexeme_props_getter(string, self.oov_prob, is_oov=is_oov)
         set_lex_struct_props(lex, props, self.strings, EMPTY_VEC)
         if is_oov:
             lex.id = 0