def test_dataframe_backed_repr_complex():
df = pd.DataFrame([(1, 'Alice', 100), (2, 'Bob', -200),
(3, 'Charlie', 300), (4, 'Denis', 400),
(5, 'Edith', -500)],
columns=['id', 'name', 'balance'])
t = Data(df)
repr(t[t['balance'] < 0])
def test_table_resource():
with tmpfile('csv') as filename:
csv = CSV(filename, 'w', schema='{x: int, y: int}')
csv.extend([[1, 2], [10, 20]])
t = Data(filename)
assert isinstance(t.data, CSV)
assert list(compute(t)) == list(csv)
def test_repr():
result = expr_repr(t['name'])
print(result)
assert isinstance(result, str)
assert 'Alice' in result
assert 'Bob' in result
assert '...' not in result
result = expr_repr(t['amount'] + 1)
print(result)
assert '101' in result
t2 = Data(tuple((i, i**2) for i in range(100)), fields=['x', 'y'])
result = expr_repr(t2)
print(result)
assert len(result.split('\n')) < 20
assert '...' in result
def test_create_with_schema():
t = Data(data, schema='{name: string, amount: float32}')
assert t.schema == dshape('{name: string, amount: float32}')
def test_compute_on_Data_gives_back_data():
assert compute(Data([1, 2, 3])) == [1, 2, 3]
def test_table_raises_on_inconsistent_inputs():
with pytest.raises(ValueError):
t = Data(data,
schema='{name: string, amount: float32}',
dshape=dshape("{ name : string, amount : float32 }"))
def test_data_passes_kwargs_to_resource():
assert Data(example('iris.csv'), encoding='ascii').data.encoding == 'ascii'
def test_can_trivially_create_pytables():
Data(example('accounts.h5') + '::/accounts')
def test_mutable_backed_repr():
mutable_backed_table = Data([[0]], fields=['col1'])
repr(mutable_backed_table)
def test_can_trivially_create_csv_Data():
Data(example('iris.csv'))
def test_Data_attribute_repr():
path = os.path.join(os.path.dirname(__file__), 'accounts.csv')
t = Data(CSV(path))
result = t.timestamp.day
expected = pd.DataFrame({'timestamp_day': [25] * 3})
assert repr(result) == repr(expected)
def test_into_nd_array_column_failure():
tble = Data(L, fields=['id', 'name', 'balance'])
expr = tble[tble['balance'] < 0]
colarray = into(nd.array, expr)
assert len(list(compute(expr))) == len(colarray)
def test_into_nd_array_selection():
t = Data(L, fields=['id', 'name', 'balance'])
expr = t[t['balance'] < 0]
selarray = into(nd.array, expr)
assert len(list(compute(expr))) == len(selarray)
def test_into_np_ndarray_column():
t = Data(L, fields=['id', 'name', 'balance'])
expr = t[t.balance < 0].name
colarray = into(np.ndarray, expr)
assert len(list(compute(expr))) == len(colarray)
def test_create_with_raw_data():
t = Data(data, columns=['name', 'amount'])
assert t.schema == dshape('{name: string, amount: int64}')
assert t.name
assert t.data == data
def test_can_trivially_create_csv_Data_with_unicode():
if sys.version[0] == '2':
assert isinstance(Data(example(u'iris.csv')).data, CSV)
def test_can_trivially_create_sqlite_table():
Data('sqlite:///' + example('iris.db') + '::iris')
def test_dataframe_backed_repr():
df = pd.DataFrame(data=[0], columns=['col1'])
dataframe_backed_table = Data(df)
repr(dataframe_backed_table)
from blaze.expr.table import Symbol
from datashape import dshape
from blaze.utils import tmpfile, example
import pytest
import sys
import pandas as pd
import numpy as np
from dynd import nd
data = (('Alice', 100), ('Bob', 200))
L = [[1, 'Alice', 100], [2, 'Bob', -200], [3, 'Charlie', 300],
[4, 'Denis', 400], [5, 'Edith', -500]]
t = Data(data, columns=['name', 'amount'])
def test_table_raises_on_inconsistent_inputs():
with pytest.raises(ValueError):
t = Data(data,
schema='{name: string, amount: float32}',
dshape=dshape("{ name : string, amount : float32 }"))
def test_resources():
assert t._resources() == {t: t.data}
def test_resources_fail():
t = Symbol('t', 'var * {x: int, y: int}')