Move language-specific tests to tests/lang

spacy/tests/bn/__init__.py

@@ -1 +0,0 @@
-# coding: utf-8

spacy/tests/fr/__init__.py

@@ -1 +0,0 @@
-# coding: utf-8

spacy/tests/lang/en/test_punct.py

@@ -6,8 +6,8 @@ from __future__ import unicode_literals
 
 import pytest
 
-from ...util import compile_prefix_regex
+from ....util import compile_prefix_regex
-from ...language_data import TOKENIZER_PREFIXES
+from ....lang.punctuation import TOKENIZER_PREFIXES
 
 
 PUNCT_OPEN = ['(', '[', '{', '*']

spacy/tests/lang/fi/test_tokenizer.py

@@ -3,15 +3,14 @@ from __future__ import unicode_literals
 
 import pytest
 
+
 ABBREVIATION_TESTS = [
     ('Hyvää uutta vuotta t. siht. Niemelä!', ['Hyvää', 'uutta', 'vuotta', 't.', 'siht.', 'Niemelä', '!']),
     ('Paino on n. 2.2 kg', ['Paino', 'on', 'n.', '2.2', 'kg'])
 ]
 
-TESTCASES = ABBREVIATION_TESTS
 
-
+@pytest.mark.parametrize('text,expected_tokens', ABBREVIATION_TESTS)
-@pytest.mark.parametrize('text,expected_tokens', TESTCASES)
 def test_tokenizer_handles_testcases(fi_tokenizer, text, expected_tokens):
     tokens = fi_tokenizer(text)
     token_list = [token.text for token in tokens if not token.is_space]

spacy/tests/lang/fr/test_text.py

@@ -1,6 +1,4 @@
 # coding: utf8
-
-
 from __future__ import unicode_literals