various fixes; use pytest

Author: alimanfoo

numcodecs/blosc.c

@@ -3,7 +3,7 @@
 
 
 from .abc import Codec
-from .compat import ndarray_from_buffer, buffer_copy, ensure_text, ensure_bytes
+from .compat import ndarray_from_buffer, buffer_copy, ensure_text
 
 
 import numpy as np
@@ -25,18 +25,18 @@ class Categorize(Codec):
     --------
     >>> import numcodecs
     >>> import numpy as np
-    >>> x = np.array([b'male', b'female', b'female', b'male', b'unexpected'])
+    >>> x = np.array(['male', 'female', 'female', 'male', 'unexpected'], dtype=object)
     >>> x
-    array([b'male', b'female', b'female', b'male', b'unexpected'],
-          dtype='|S10')
-    >>> codec = numcodecs.Categorize(labels=[b'female', b'male'], dtype=x.dtype)
+    array(['male', 'female', 'female', 'male', 'unexpected'],
+          dtype=object)
+    >>> codec = numcodecs.Categorize(labels=['female', 'male'], dtype=object)
     >>> y = codec.encode(x)
     >>> y
     array([2, 1, 1, 2, 0], dtype=uint8)
     >>> z = codec.decode(y)
     >>> z
-    array([b'male', b'female', b'female', b'male', b''],
-          dtype='|S10')
+    array(['male', 'female', 'female', 'male', ''],
+          dtype=object)
 
     """
 

numcodecs/categorize.py

@@ -85,17 +85,6 @@ def buffer_copy(buf, out=None):
     return out
 
 
-# def buffer_size(v):
-#     from array import array as _stdlib_array
-#     if PY2 and isinstance(v, _stdlib_array):  # pragma: no cover
-#         # special case array.array because does not support buffer
-#         # interface in PY2
-#         return v.buffer_info()[1] * v.itemsize
-#     else:
-#         v = memoryview(v)
-#         return reduce(operator.mul, v.shape) * v.itemsize
-
-
 def ndarray_from_buffer(buf, dtype):
     if isinstance(buf, np.ndarray):
         arr = buf.reshape(-1, order='A').view(dtype)
@@ -104,17 +93,10 @@ def ndarray_from_buffer(buf, dtype):
     return arr
 
 
-def ensure_bytes(l, encoding='utf-8'):
-    if isinstance(l, binary_type):
-        return l
-    else:
-        return l.encode(encoding=encoding)
-
-
 def ensure_text(l, encoding='utf-8'):
     if isinstance(l, text_type):
         return l
-    else:
+    else:  # pragma: py3 no cover
         return text_type(l, encoding=encoding)
 
 

numcodecs/compat.py

@@ -4,7 +4,7 @@
 
 import numpy as np
 from numpy.testing import assert_array_equal
-from nose.tools import eq_ as eq
+from nose.tools import assert_raises
 
 
 from numcodecs.categorize import Categorize
@@ -51,14 +51,14 @@ def test_encode():
         expect = np.array([1, 2, 1, 3, 0], dtype='u1')
         enc = codec.encode(arr)
         assert_array_equal(expect, enc)
-        eq(expect.dtype, enc.dtype)
+        assert expect.dtype == enc.dtype
 
         # test decoding with unexpected value
         dec = codec.decode(enc)
         expect = arr.copy()
         expect[expect == u'ƪùüx'] = u''
         assert_array_equal(expect, dec)
-        eq(arr.dtype, dec.dtype)
+        assert arr.dtype == dec.dtype
 
 
 def test_config():
@@ -77,7 +77,7 @@ def test_repr():
     else:  # pragma: py2 no cover
         expect += "labels=['foo', 'bar', 'baz', ...])"
     actual = repr(codec)
-    eq(expect, actual)
+    assert expect == actual
 
     if not PY2:  # pragma: py2 no cover
 
@@ -87,7 +87,7 @@ def test_repr():
         expect = "Categorize(dtype='<U4', astype='|u1', " \
                  "labels=['ƒöõ', 'ßàř', 'ßāẑ', ...])"
         actual = repr(codec)
-        eq(expect, actual)
+        assert expect == actual
 
 
 def test_backwards_compatibility():
@@ -97,3 +97,8 @@ def test_backwards_compatibility():
     codec = Categorize(labels=labels, dtype=object, astype='u1')
     check_backwards_compatibility(Categorize.codec_id, arrays_object,
                                   [codec], prefix='O')
+
+
+def test_errors():
+    with assert_raises(ValueError):
+        Categorize(labels=['foo', 'bar'], dtype='S6')