def test__get_ts__asof_latest():
""" I can get the latest known value for each sample date
"""
df = groupby_asof(get_bitemporal_test_data())
assert len(df) == 4
assert all(df['OPEN'] == [1.1, 2.1, 3.1, 4.1])
assert all(df['CLOSE'] == [10.1, 20.1, 30.1, 40.1])
def test__get_ts__asof_datetime():
""" I can get a timeseries as-of a particular point in time
"""
df = groupby_asof(get_bitemporal_test_data(), as_of=dt('2015-01-05'))
assert len(df) == 3
assert all(df['OPEN'] == [1.1, 2.1, 3.0])
assert all(df['CLOSE'] == [10.1, 20.1, 30.0])
def test__get_ts__asof_datetime():
""" I can get a timeseries as-of a particular point in time
"""
df = groupby_asof(get_bitemporal_test_data(), as_of=dt('2015-01-05'))
assert len(df) == 3
assert all(df['OPEN'] == [1.1, 2.1, 3.0])
assert all(df['CLOSE'] == [10.1, 20.1, 30.0])
def test__get_ts__asof_latest():
""" I can get the latest known value for each sample date
"""
df = groupby_asof(get_bitemporal_test_data())
assert len(df) == 4
assert all(df['OPEN'] == [1.1, 2.1, 3.1, 4.1])
assert all(df['CLOSE'] == [10.1, 20.1, 30.1, 40.1])
def read(self, symbol, as_of=None, raw=False, **kwargs):
# TODO: shall we block from_version from getting into super.read?
"""Read data for the named symbol. Returns a BitemporalItem object with
a data and metdata element (as passed into write).
Parameters
----------
symbol : `str`
symbol name for the item
as_of : `datetime.datetime`
Return the data as it was as_of the point in time.
raw : `bool`
If True, will return the full bitemporal dataframe (i.e. all versions of the data). This also means as_of is
ignored.
Returns
-------
BitemporalItem namedtuple which contains a .data and .metadata element
"""
item = self._store.read(symbol, **kwargs)
last_updated = max(item.data.index.get_level_values(self.observe_column))
if raw:
return BitemporalItem(symbol=symbol, library=self._store._arctic_lib.get_name(), data=item.data,
metadata=item.metadata,
last_updated=last_updated)
else:
index_names = list(item.data.index.names)
index_names.remove(self.observe_column)
return BitemporalItem(symbol=symbol, library=self._store._arctic_lib.get_name(),
data=groupby_asof(item.data, as_of=as_of, dt_col=index_names,
asof_col=self.observe_column),
metadata=item.metadata, last_updated=last_updated)
def test_fancy_group_by_multi_index():
ts = multi_index_df_from_arrs(
index_headers=('index 1', 'index 2', 'observed_dt'),
index_arrs=[[
'2012-09-08 17:06:11.040',
'2012-09-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-11-08 17:06:11.040',
], ['SPAM Index', 'EGG Index', 'SPAM Index', 'SPAM Index'] +
['EGG Index', 'SPAM Index'] * 2,
['2015-01-01'] * 3 + ['2015-01-05'] + ['2015-01-01'] * 4],
data_dict={'near': [1.0, 1.6, 2.0, 4.2, 2.1, 2.5, 2.6, 3.0]})
expected_ts = multi_index_df_from_arrs(
index_headers=('index 1', 'index 2'),
index_arrs=[[
'2012-09-08 17:06:11.040',
'2012-09-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-11-08 17:06:11.040',
], ['EGG Index', 'SPAM Index'] * 3 + ['SPAM Index']],
data_dict={'near': [1.6, 1.0, 2.1, 4.2, 2.6, 2.5, 3.0]})
assert_frame_equal(
expected_ts,
groupby_asof(ts, dt_col=['index 1', 'index 2'],
asof_col='observed_dt'))
def test__can_insert_row():
""" I can insert a new row into a bitemp ts and it comes back when selecting the latest data
"""
df = get_bitemporal_test_data()
df = insert_at(df, dt('2014-01-03'), [[9, 90]])
assert len(df) == 9
df = groupby_asof(df)
assert len(df) == 4
assert df.loc[dt('2014-01-03')]['OPEN'] == 9
assert df.loc[dt('2014-01-03')]['CLOSE'] == 90
def test__can_insert_row():
""" I can insert a new row into a bitemp ts and it comes back when selecting the latest data
"""
df = get_bitemporal_test_data()
df = insert_at(df, dt('2014-01-03'), [[9, 90]])
assert len(df) == 9
df = groupby_asof(df)
assert len(df) == 4
assert df.loc[dt('2014-01-03')]['OPEN'] == 9
assert df.loc[dt('2014-01-03')]['CLOSE'] == 90
def test__get_ts__unsorted_index():
""" I can get a timeseries as-of a date when the index isn't sorted properly
"""
df = get_bitemporal_test_data()
# Swap the third and fourth rows around. This would break the group-by if we didn't check
# for sortedness
df = df.reindex(df.index[[0, 1, 3, 2, 4, 5, 6, 7]])
df = groupby_asof(df)
assert len(df) == 4
assert all(df['OPEN'] == [1.1, 2.1, 3.1, 4.1])
assert all(df['CLOSE'] == [10.1, 20.1, 30.1, 40.1])
def test__get_ts__unsorted_index():
""" I can get a timeseries as-of a date when the index isn't sorted properly
"""
df = get_bitemporal_test_data()
# Swap the third and fourth rows around. This would break the group-by if we didn't check
# for sortedness
df = df.reindex(df.index[[0, 1, 3, 2, 4, 5, 6, 7]])
df = groupby_asof(df)
assert len(df) == 4
assert all(df['OPEN'] == [1.1, 2.1, 3.1, 4.1])
assert all(df['CLOSE'] == [10.1, 20.1, 30.1, 40.1])
def test_fancy_group_by_multi_index():
ts = read_str_as_pandas(""" index 1 | index 2 | observed_dt | near
2012-09-08 17:06:11.040 | SPAM Index | 2015-01-01 | 1.0
2012-09-08 17:06:11.040 | EGG Index | 2015-01-01 | 1.6
2012-10-08 17:06:11.040 | SPAM Index | 2015-01-01 | 2.0
2012-10-08 17:06:11.040 | SPAM Index | 2015-01-05 | 4.2
2012-10-08 17:06:11.040 | EGG Index | 2015-01-01 | 2.1
2012-10-09 17:06:11.040 | SPAM Index | 2015-01-01 | 2.5
2012-10-09 17:06:11.040 | EGG Index | 2015-01-01 | 2.6
2012-11-08 17:06:11.040 | SPAM Index | 2015-01-01 | 3.0""", num_index=3)
expected_ts = read_str_as_pandas(""" index 1 | index 2 | near
2012-09-08 17:06:11.040 | EGG Index | 1.6
2012-09-08 17:06:11.040 | SPAM Index | 1.0
2012-10-08 17:06:11.040 | EGG Index | 2.1
2012-10-08 17:06:11.040 | SPAM Index | 4.2
2012-10-09 17:06:11.040 | EGG Index | 2.6
2012-10-09 17:06:11.040 | SPAM Index | 2.5
2012-11-08 17:06:11.040 | SPAM Index | 3.0""", num_index=2)
assert_frame_equal(expected_ts, groupby_asof(ts, dt_col=['index 1', 'index 2'], asof_col='observed_dt'))
def test_fancy_group_by_multi_index():
ts = multi_index_df_from_arrs(
index_headers=('index 1', 'index 2', 'observed_dt'),
index_arrs=[
[
'2012-09-08 17:06:11.040',
'2012-09-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-11-08 17:06:11.040',
],
['SPAM Index', 'EGG Index', 'SPAM Index', 'SPAM Index'] + ['EGG Index', 'SPAM Index'] * 2,
['2015-01-01'] * 3 + ['2015-01-05'] + ['2015-01-01'] * 4
],
data_dict={'near': [1.0, 1.6, 2.0, 4.2, 2.1, 2.5, 2.6, 3.0]}
)
expected_ts= multi_index_df_from_arrs(
index_headers=('index 1', 'index 2'),
index_arrs=[
[
'2012-09-08 17:06:11.040',
'2012-09-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-08 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-10-09 17:06:11.040',
'2012-11-08 17:06:11.040',
],
['EGG Index', 'SPAM Index'] * 3 + ['SPAM Index']
],
data_dict={'near': [1.6, 1.0, 2.1, 4.2, 2.6, 2.5, 3.0]}
)
assert_frame_equal(expected_ts, groupby_asof(ts, dt_col=['index 1', 'index 2'], asof_col='observed_dt'))
def test_fancy_group_by_multi_index():
ts = read_str_as_pandas(""" index 1 | index 2 | observed_dt | near
2012-09-08 17:06:11.040 | SPAM Index | 2015-01-01 | 1.0
2012-09-08 17:06:11.040 | EGG Index | 2015-01-01 | 1.6
2012-10-08 17:06:11.040 | SPAM Index | 2015-01-01 | 2.0
2012-10-08 17:06:11.040 | SPAM Index | 2015-01-05 | 4.2
2012-10-08 17:06:11.040 | EGG Index | 2015-01-01 | 2.1
2012-10-09 17:06:11.040 | SPAM Index | 2015-01-01 | 2.5
2012-10-09 17:06:11.040 | EGG Index | 2015-01-01 | 2.6
2012-11-08 17:06:11.040 | SPAM Index | 2015-01-01 | 3.0""",
num_index=3)
expected_ts = read_str_as_pandas(""" index 1 | index 2 | near
2012-09-08 17:06:11.040 | EGG Index | 1.6
2012-09-08 17:06:11.040 | SPAM Index | 1.0
2012-10-08 17:06:11.040 | EGG Index | 2.1
2012-10-08 17:06:11.040 | SPAM Index | 4.2
2012-10-09 17:06:11.040 | EGG Index | 2.6
2012-10-09 17:06:11.040 | SPAM Index | 2.5
2012-11-08 17:06:11.040 | SPAM Index | 3.0""",
num_index=2)
assert_frame_equal(
expected_ts,
groupby_asof(ts, dt_col=['index 1', 'index 2'],
asof_col='observed_dt'))
