def test_apply_adj_raise_error():
with pytest.raises(ValueError):
df = pd.read_csv('BTC.csv', parse_dates=['date'])
adj_df = apply_adjustment(df,
adj_date='2018-08-01',
adj_value=-100,
adj_type='div')
def test_apply_adj_cols():
df = pd.read_csv('BTC.csv', parse_dates=['date'])
adj_df = apply_adjustment(df,
adj_date='2018-07-21',
adj_value=1 / 2,
cols=['open', 'high'])
adj_df = adj_df.set_index('date').sort_index()
cols = ['open', 'high', 'low', 'close']
assert adj_df.loc['2018-07-11', 'open'] == 3151.24
assert adj_df.loc['2018-07-11', 'close'] == 6381.87
def test_apply_adj_sub_negative():
df = pd.read_csv('BTC.csv', parse_dates=['date'])
adj_df = apply_adjustment(df,
adj_date='2018-08-01',
adj_value=-100,
adj_type='sub')
adj_df = adj_df.set_index('date').sort_index()
assert adj_df.loc['2018-07-16', 'close'] == 6826.4
assert adj_df.loc['2018-08-01', 'high'] == 15500.16
assert adj_df.loc['2018-08-10', 'close'] == 12286.6
def test_apply_adj_date_col():
df = pd.read_csv('BTC.csv', parse_dates=['date'])
df['timestamp'] = df['date']
del df['date']
adj_df = apply_adjustment(df,
adj_date='2018-07-21',
adj_value=1 / 2,
date_col='timestamp')
adj_df = adj_df.set_index('timestamp').sort_index()
assert adj_df.loc['2018-07-11', 'open'] == 3151.24
assert adj_df.loc['2018-07-21', 'close'] == 3702.15
def test_apply_adj_mul():
df = pd.read_csv('BTC.csv', parse_dates=['date'])
adj_df = apply_adjustment(df, adj_date='2018-07-21', adj_value=1 / 2)
adj_df = adj_df.set_index('date').sort_index()
cols = ['open', 'high', 'low', 'close']
assert adj_df.loc['2018-07-11', 'open'] == 3151.24
assert adj_df.loc['2018-07-21', 'close'] == 3702.15
for a, b in zip(adj_df.loc['2018-07-21', cols],
(3665.27, 3721.2, 3608.47, 3702.15)):
assert a == b
assert adj_df.loc['2018-08-01', 'high'] == 15500.16
assert adj_df.loc['2018-07-11', 'volume'] == 2481016