def test_into_ctable_list_datetimes():
from datetime import datetime
L = [datetime(2012, 1, 1), datetime(2013, 2, 2)]
b = into(bcolz.carray, L)
assert np.issubdtype(b.dtype, np.datetime64)
assert list(into(Iterator, b)) == L
def test_into_ctable_list_datetimes():
from datetime import datetime
L = [datetime(2012, 1, 1), datetime(2013, 2, 2)]
b = into(bcolz.carray, L)
assert np.issubdtype(b.dtype, np.datetime64)
assert list(into(Iterator, b)) == L
def test_into_bcolz_from_many_numpy_arrays():
x = np.array([(1, 'Hello'), (2, 'abc')],
dtype=[('num', 'i4'), ('name', 'S5')])
b = into(bcolz.ctable, x)
assert len(b) == len(x)
b = into(b, x)
assert len(b) == 2 * len(x)
def test_into_chunks():
from blaze.compute.numpy import chunks, compute_up
from blaze.compute.chunks import chunks, compute_up, ChunkIterator
from blaze import into
x = np.array([(int(i), float(i)) for i in range(100)],
dtype=[('a', np.int32), ('b', np.float32)])
cs = chunks(x, chunksize=10)
b1 = into(bcolz.ctable, ChunkIterator(cs))
b2 = into(bcolz.ctable, x)
assert str(b1) == str(b2)
def test_into_chunks():
from blaze.compute.numpy import chunks, compute_up
from blaze.compute.chunks import chunks, compute_up, ChunkIterator
from blaze import into
x = np.array([(int(i), float(i)) for i in range(100)],
dtype=[('a', np.int32), ('b', np.float32)])
cs = chunks(x, chunksize=10)
b1 = into(bcolz.ctable, ChunkIterator(cs))
b2 = into(bcolz.ctable, x)
assert str(b1) == str(b2)
def test_selection_head():
b = into(bcolz.ctable,
((i, i + 1, float(i)**2) for i in range(10000)),
names=['a', 'b', 'c'])
t = Symbol('t', 'var * {a: int32, b: int32, c: float64}')
assert compute((t.a < t.b).all(), b) == True
assert list(compute(t[t.a < t.b].a.head(10), b)) == list(range(10))
assert list(compute(t[t.a > t.b].a.head(10), b)) == []
assert into([], compute(t[t.a + t.b > t.c], b)) == [(0, 1, 0),
(1, 2, 1),
(2, 3, 4)]
assert len(compute(t[t.a + t.b > t.c].head(10), b)) # non-empty
assert len(compute(t[t.a + t.b < t.c].head(10), b)) # non-empty
def test_selection_head():
b = into(bcolz.ctable,
((i, i + 1, float(i)**2) for i in range(10000)),
names=['a', 'b', 'c'])
t = TableSymbol('t', '{a: int32, b: int32, c: float64}')
assert compute((t.a < t.b).all(), b) == True
assert list(compute(t[t.a < t.b].a.head(10), b)) == list(range(10))
assert list(compute(t[t.a > t.b].a.head(10), b)) == []
assert into([], compute(t[t.a + t.b > t.c], b)) == [(0, 1, 0),
(1, 2, 1),
(2, 3, 4)]
assert len(compute(t[t.a + t.b > t.c].head(10), b)) # non-empty
assert len(compute(t[t.a + t.b < t.c].head(10), b)) # non-empty
def test_into_ctable_numpy():
assert str(
into(bcolz.ctable, np.array([(1, 1.0), (2, 2.0), (3, 3.0)], dtype=[("a", np.int32), ("b", np.float32)]))
) == str(
bcolz.ctable(
[np.array([1, 2, 3], dtype=np.int32), np.array([1.0, 2.0, 3.0], dtype=np.float32)], names=["a", "b"]
)
)
def test_into_ctable_DataFrame():
df = DataFrame([[1, "Alice"], [2, "Bob"], [3, "Charlie"]], columns=["id", "name"])
b = into(bcolz.ctable, df)
assert list(b.names) == list(df.columns)
assert list(b["id"]) == [1, 2, 3]
print(b["name"])
print(b["name"][0])
print(type(b["name"][0]))
assert list(b["name"]) == ["Alice", "Bob", "Charlie"]
def test_into_ctable_DataFrame():
df = DataFrame([[1, 'Alice'], [2, 'Bob'], [3, 'Charlie']],
columns=['id', 'name'])
b = into(bcolz.ctable, df)
assert list(b.names) == list(df.columns)
assert list(b['id']) == [1, 2, 3]
print(b['name'])
print(b['name'][0])
print(type(b['name'][0]))
assert list(b['name']) == ['Alice', 'Bob', 'Charlie']
def test_into_ctable_DataFrame():
df = DataFrame([[1, 'Alice'],
[2, 'Bob'],
[3, 'Charlie']], columns=['id', 'name'])
ds = dshape('var * {id: int32, name: string[7, "U32"]}')
b = into(bcolz.ctable, df, dshape=ds)
assert list(b.names) == list(df.columns)
assert list(b['id']) == [1, 2, 3]
assert list(b['name']) == ['Alice', 'Bob', 'Charlie']
assert discover(b).measure == ds.measure
def test_into_ctable_DataFrame():
df = DataFrame([[1, 'Alice'],
[2, 'Bob'],
[3, 'Charlie']], columns=['id', 'name'])
b = into(bcolz.ctable, df)
assert list(b.names) == list(df.columns)
assert list(b['id']) == [1, 2, 3]
print(b['name'])
print(b['name'][0])
print(type(b['name'][0]))
assert list(b['name']) == ['Alice', 'Bob', 'Charlie']
def test_into_ndarray_ctable():
assert str(into(np.ndarray, b)) == \
str(np.array([(1, 1.), (2, 2.), (3, 3.)],
dtype=[('a', int), ('b', float)]))
def test_into_ctable_numpy():
assert str(into(bcolz.ctable, np.array([(1, 1.), (2, 2.), (3, 3.)],
dtype=[('a', np.int32), ('b', np.float32)]))) == \
str(bcolz.ctable([np.array([1, 2, 3], dtype=np.int32),
np.array([1., 2., 3.], dtype=np.float32)],
names=['a', 'b']))
def test_into_ctable_list():
b = into(bcolz.ctable, [(1, 1.), (2, 2.), (3, 3.)], names=['a', 'b'])
assert list(b['a']) == [1, 2, 3]
assert b.names == ['a', 'b']
def test_into_ctable_list_datetimes():
from datetime import datetime
b = into(bcolz.carray, [datetime(2012, 1, 1), datetime(2013, 2, 2)])
assert np.issubdtype(b.dtype, np.datetime64)
def test_into_ctable_iterator():
b = into(bcolz.ctable, iter([(1, 1.), (2, 2.), (3, 3.)]), names=['a', 'b'])
assert list(b['a']) == [1, 2, 3]
assert b.names == ['a', 'b']
def test_into_DataFrame_ctable():
result = into(DataFrame(), b)
expected = DataFrame([[1, 1.], [2, 2.], [3, 3.]], columns=['a', 'b'])
assert str(result) == str(expected)
def test_into_list_carray():
assert into([], b['a']) == [1, 2, 3]
def test_into_ctable_numpy():
assert str(into(bcolz.ctable, np.array([(1, 1.), (2, 2.), (3, 3.)],
dtype=[('a', np.int32), ('b', np.float32)]))) == \
str(bcolz.ctable([np.array([1, 2, 3], dtype=np.int32),
np.array([1., 2., 3.], dtype=np.float32)],
names=['a', 'b']))
def test_into_ctable_iterator():
b = into(bcolz.ctable, iter([(1, 1.), (2, 2.), (3, 3.)]), names=['a', 'b'])
assert list(b['a']) == [1, 2, 3]
assert b.names == ['a', 'b']
def test_into_list_ctable():
assert into([], b) == [(1, 1.), (2, 2.), (3, 3.)]
def test_into_ndarray_ctable():
assert str(into(np.ndarray, b)) == str(np.array([(1, 1.0), (2, 2.0), (3, 3.0)], dtype=[("a", int), ("b", float)]))
def test_into_ndarray_carray():
assert str(into(np.ndarray, b['a'])) == \
str(np.array([1, 2, 3]))
def test_into_list_carray():
assert into([], b['a']) == [1, 2, 3]
def test_into_ctable_list():
b = into(bcolz.ctable, [(1, 1.), (2, 2.), (3, 3.)], names=['a', 'b'])
assert list(b['a']) == [1, 2, 3]
assert b.names == ['a', 'b']
def test_into_DataFrame_ctable():
result = into(DataFrame(), b)
expected = DataFrame([[1, 1.0], [2, 2.0], [3, 3.0]], columns=["a", "b"])
assert str(result) == str(expected)
def test_chunks():
assert len(list(chunks(b, chunksize=2))) == 2
assert (next(chunks(b, chunksize=2)) == into(np.array(0), b)[:2]).all()
def test_chunks():
assert len(list(chunks(b, chunksize=2))) == 2
assert (next(chunks(b, chunksize=2)) == into(np.array(0), b)[:2]).all()
def test_into_ndarray_carray():
assert str(into(np.ndarray, b['a'])) == \
str(np.array([1, 2, 3]))
def test_into_list_ctable():
assert into([], b) == [(1, 1.), (2, 2.), (3, 3.)]
def test_into_DataFrame_ctable():
result = into(DataFrame(), b)
expected = DataFrame([[1, 1.], [2, 2.], [3, 3.]], columns=['a', 'b'])
assert str(result) == str(expected)
def test_into_ctable_list():
b = into(bcolz.ctable, [(1, 1.0), (2, 2.0), (3, 3.0)], names=["a", "b"])
assert list(b["a"]) == [1, 2, 3]
assert b.names == ["a", "b"]
def test_into_ctable_iterator():
b = into(bcolz.ctable, iter([(1, 1.0), (2, 2.0), (3, 3.0)]), names=["a", "b"])
assert list(b["a"]) == [1, 2, 3]
assert b.names == ["a", "b"]
def test_into_ndarray_ctable():
assert str(into(np.ndarray, b)) == \
str(np.array([(1, 1.), (2, 2.), (3, 3.)],
dtype=[('a', int), ('b', float)]))
