Modernise and merge lexeme vocab tests

spacy/tests/vocab/test_lexeme.py

@@ -1,28 +1,58 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
+from ...attrs import *
+
 import pytest
 
-from spacy.attrs import *
+@pytest.mark.parametrize('text1,prob1,text2,prob2', [("NOUN", -1, "opera", -2)])
+def test_vocab_lexeme_lt(en_vocab, text1, text2, prob1, prob2):
+    """More frequent is l.t. less frequent"""
+    lex1 = en_vocab[text1]
+    lex1.prob = prob1
+    lex2 = en_vocab[text2]
+    lex2.prob = prob2
+
+    assert lex1 < lex2
+    assert lex2 > lex1
 
 
+@pytest.mark.parametrize('text1,text2', [("phantom", "opera")])
+def test_vocab_lexeme_hash(en_vocab, text1, text2):
+    """Test that lexemes are hashable."""
+    lex1 = en_vocab[text1]
+    lex2 = en_vocab[text2]
+    lexes = {lex1: lex1, lex2: lex2}
+    assert lexes[lex1].orth_ == text1
+    assert lexes[lex2].orth_ == text2
 
 
-def test_lexeme_lt(en_vocab):
+def test_vocab_lexeme_is_alpha(en_vocab):
-    '''More frequent is l.t. less frequent'''
+    assert en_vocab['the'].flags & (1 << IS_ALPHA)
-    noun = en_vocab['NOUN']
+    assert not en_vocab['1999'].flags & (1 << IS_ALPHA)
-
+    assert not en_vocab['hello1'].flags & (1 << IS_ALPHA)
-    opera = en_vocab['opera']
-
-    assert noun < opera
-    assert opera > noun
 
 
-def test_lexeme_hash(en_vocab):
+def test_vocab_lexeme_is_digit(en_vocab):
-    '''Test that lexemes are hashable.'''
+    assert not en_vocab['the'].flags & (1 << IS_DIGIT)
-    phantom = en_vocab['phantom']
+    assert en_vocab['1999'].flags & (1 << IS_DIGIT)
+    assert not en_vocab['hello1'].flags & (1 << IS_DIGIT)
 
-    opera = en_vocab['opera']
 
-    lexes = {phantom: phantom, opera: opera}
+def test_vocab_lexeme_add_flag_auto_id(en_vocab):
-    assert lexes[phantom].orth_ == 'phantom'
+    is_len4 = en_vocab.add_flag(lambda string: len(string) == 4)
-    assert lexes[opera].orth_ == 'opera'
+    assert en_vocab['1999'].check_flag(is_len4) == True
+    assert en_vocab['1999'].check_flag(IS_DIGIT) == True
+    assert en_vocab['199'].check_flag(is_len4) == False
+    assert en_vocab['199'].check_flag(IS_DIGIT) == True
+    assert en_vocab['the'].check_flag(is_len4) == False
+    assert en_vocab['dogs'].check_flag(is_len4) == True
+
+
+def test_vocab_lexeme_add_flag_provided_id(en_vocab):
+    is_len4 = en_vocab.add_flag(lambda string: len(string) == 4, flag_id=IS_DIGIT)
+    assert en_vocab['1999'].check_flag(is_len4) == True
+    assert en_vocab['199'].check_flag(is_len4) == False
+    assert en_vocab['199'].check_flag(IS_DIGIT) == False
+    assert en_vocab['the'].check_flag(is_len4) == False
+    assert en_vocab['dogs'].check_flag(is_len4) == True

spacy/tests/vocab/test_lexeme_flags.py

@@ -1,42 +0,0 @@
-from __future__ import unicode_literals
-
-import pytest
-
-from spacy.attrs import *
-
-
-def test_is_alpha(en_vocab):
-    the = en_vocab['the']
-    assert the.flags & (1 << IS_ALPHA)
-    year = en_vocab['1999']
-    assert not year.flags & (1 << IS_ALPHA)
-    mixed = en_vocab['hello1']
-    assert not mixed.flags & (1 << IS_ALPHA)
-
-
-def test_is_digit(en_vocab):
-    the = en_vocab['the']
-    assert not the.flags & (1 << IS_DIGIT)
-    year = en_vocab['1999']
-    assert year.flags & (1 << IS_DIGIT)
-    mixed = en_vocab['hello1']
-    assert not mixed.flags & (1 << IS_DIGIT)
-
-
-def test_add_flag_auto_id(en_vocab):
-    is_len4 = en_vocab.add_flag(lambda string: len(string) == 4)
-    assert en_vocab['1999'].check_flag(is_len4) == True
-    assert en_vocab['1999'].check_flag(IS_DIGIT) == True
-    assert en_vocab['199'].check_flag(is_len4) == False
-    assert en_vocab['199'].check_flag(IS_DIGIT) == True
-    assert en_vocab['the'].check_flag(is_len4) == False
-    assert en_vocab['dogs'].check_flag(is_len4) == True
-
-
-def test_add_flag_provided_id(en_vocab):
-    is_len4 = en_vocab.add_flag(lambda string: len(string) == 4, flag_id=IS_DIGIT)
-    assert en_vocab['1999'].check_flag(is_len4) == True
-    assert en_vocab['199'].check_flag(is_len4) == False
-    assert en_vocab['199'].check_flag(IS_DIGIT) == False
-    assert en_vocab['the'].check_flag(is_len4) == False
-    assert en_vocab['dogs'].check_flag(is_len4) == True