Rename stop_words.py to word_sets.py and include more sets

NUM_WORDS and ORDINAL_WORDS are currently not used, but the hard-coded
list should be removed from orth.pyx and replaced to use
language-specific functions. This will later allow other languages to
use their own functions to set those flags. (In English, this is easier
because it only needs to be checked against a set – in German for
example, this requires a more complex function, as most number words
are one word.)

spacy/en/language_data.py

@@ -6,7 +6,7 @@ from ..language_data import update_exc, strings_to_exc, expand_exc
 from ..symbols import ORTH, LEMMA
 
 from .tag_map import TAG_MAP
-from .stop_words import STOP_WORDS
+from .word_sets import STOP_WORDS, NUM_WORDS
 from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS, ORTH_ONLY
 from .morph_rules import MORPH_RULES
 

spacy/en/word_sets.py

@@ -1,7 +1,9 @@
-# encoding: utf8
+# coding: utf8
 from __future__ import unicode_literals
 
 
+# Stop words
+
 STOP_WORDS = set("""
 a about above across after afterwards again against all almost alone along
 already also although always am among amongst amount an and another any anyhow
@@ -65,3 +67,24 @@ whither who whoever whole whom whose why will with within without would
 
 yet you your yours yourself yourselves
 """.split())
+
+
+# Number words
+
+NUM_WORDS = set("""
+zero one two three four five six seven eight nine ten eleven twelve thirteen
+fourteen fifteen sixteen seventeen eighteen nineteen twenty thirty forty fifty
+sixty seventy eighty ninety hundred thousand million billion trillion
+quadrillion gajillion bazillion
+""".split())
+
+
+# Ordinal words
+
+ORDINAL_WORDS = set("""
+first second third fourth fifth sixth seventh eigth ninth tenth eleventh twelveth
+thirteenth fourteenth fifteenth sixteenth sventeenth eighteenth nineteenth
+twentieth  thirtieth fortieth fiftieth sixtieth seventieth eightieth ninetieth
+hundreth thousandth millionth billionth trillionth quadrillionth gajillionth
+bazillionth
+""".split())