def get_data_last(self,
as_of: Optional[Union[dt.date, dt.datetime]],
start: Optional[Union[dt.date, dt.datetime]] = None,
fields: Optional[Iterable[str]] = None,
**kwargs) -> pd.DataFrame:
"""
Get the last point for this DataSet, at or before as_of
:param as_of: The date or time as of which to query
:param start: The start of the range to query
:param fields: The fields for which to query
:param kwargs: Additional query parameters, e.g., city='Boston'
:return: A Dataframe of values
**Examples**
>>> from gs_quant.data import Dataset
>>> import datetime as dt
>>>
>>> weather = Dataset('WEATHER')
>>> last = weather.get_data_last(dt.datetime.now())
"""
query = self.provider.build_query(start=start,
end=as_of,
fields=fields,
**kwargs)
data = self.provider.last_data(query, self.id)
return construct_dataframe_with_types(self.id, data)
def test_construct_dataframe_with_types(get_types):
get_types.return_value = test_types
df = construct_dataframe_with_types(str(Dataset.TR.TREOD), test_data)
assert np.issubdtype(df.index.dtype, datetime64)
assert df['adjustedAskPrice'].dtype == int64
assert df['adjustedBidPrice'].dtype == float64
assert df['assetId'].dtype == object # https://pbpython.com/pandas_dtypes.html python str == dtype object
assert np.issubdtype(df['updateTime'].dtype, datetime64)
def get_data_series(self,
field: Union[str, Fields],
start: Optional[Union[dt.date, dt.datetime]] = None,
end: Optional[Union[dt.date, dt.datetime]] = None,
as_of: Optional[dt.datetime] = None,
since: Optional[dt.datetime] = None,
**kwargs) -> pd.Series:
"""
Get a time series of data for a field of a dataset
:param field: The DataSet field to use
:param start: Requested start date/datetime for data
:param end: Requested end date/datetime for data
:param as_of: Request data as_of
:param since: Request data since
:param kwargs: Extra query arguments, e.g. ticker='EDZ19'
:return: A Series of the requested data, indexed by date or time, depending on the DataSet
**Examples**
>>> from gs_quant.data import Dataset
>>> import datetime as dt
>>>
>>> weather = Dataset('WEATHER')
>>> dew_point = weather
>>>> .get_data_series('dewPoint', dt.date(2016, 1, 15), dt.date(2016, 1, 16), city=('Boston', 'Austin'))
"""
field_value = field if isinstance(field, str) else field.value
query = self.provider.build_query(start=start,
end=end,
as_of=as_of,
since=since,
fields=(field_value, ),
**kwargs)
symbol_dimensions = self.provider.symbol_dimensions(self.id)
if len(symbol_dimensions) != 1:
raise MqValueError(
'get_data_series only valid for symbol_dimensions of length 1')
symbol_dimension = symbol_dimensions[0]
data = self.provider.query_data(query, self.id)
df = construct_dataframe_with_types(self.id, data)
gb = df.groupby(symbol_dimension)
if len(gb.groups) > 1:
raise MqValueError(
'Not a series for a single {}'.format(symbol_dimension))
return pd.Series(index=df.index, data=df.loc[:, field_value].values)
def get_data(self,
start: Optional[Union[dt.date, dt.datetime]] = None,
end: Optional[Union[dt.date, dt.datetime]] = None,
as_of: Optional[dt.datetime] = None,
since: Optional[dt.datetime] = None,
fields: Optional[Iterable[Union[str, Fields]]] = None,
asset_id_type: str = None,
**kwargs) -> pd.DataFrame:
"""
Get data for the given range and parameters
:param start: Requested start date/datetime for data
:param end: Requested end date/datetime for data
:param as_of: Request data as_of
:param since: Request data since
:param fields: DataSet fields to include
:param kwargs: Extra query arguments, e.g. ticker='EDZ19'
:return: A Dataframe of the requested data
**Examples**
>>> from gs_quant.data import Dataset
>>> import datetime as dt
>>>
>>> weather = Dataset('WEATHER')
>>> weather_data = weather.get_data(dt.date(2016, 1, 15), dt.date(2016, 1, 16), city=('Boston', 'Austin'))
"""
field_names = None if fields is None else list(
map(lambda f: f if isinstance(f, str) else f.value, fields))
query = self.provider.build_query(start=start,
end=end,
as_of=as_of,
since=since,
fields=field_names,
**kwargs)
data = self.provider.query_data(query,
self.id,
asset_id_type=asset_id_type)
return construct_dataframe_with_types(self.id, data)
def test_get_coverage(get_coverage, get_types):
get_coverage.return_value = test_coverage_data
get_types.return_value = {'gsid': 'string'}
data = Dataset(Dataset.TR.TREOD).get_coverage()
assert data.equals(construct_dataframe_with_types(str(Dataset.TR.TREOD), test_coverage_data))
def test_last_data(query_data, get_types):
query_data.return_value = [test_data[-1]]
get_types.return_value = test_types
dataset = Dataset(Dataset.TR.TREOD)
data = dataset.get_data_last(dt.date(2019, 1, 9), assetId='MA4B66MW5E27U8P32SB')
assert data.equals(construct_dataframe_with_types(str(Dataset.TR.TREOD), ([test_data[-1]])))
