Skip to content

Wave

Wave

Wave(year: str, wave_folder: str, country: Country)

A single survey wave within a country.

Typically obtained via bracket notation on a :class:Country::

wave = uga['2019-20']
df = wave.household_roster()

Parameters:

Name Type Description Default
year str

Wave label (e.g. '2019-20').

required
wave_folder str

Subdirectory name on disk (may differ from year for multi-round surveys).

required
country Country

Parent country instance.

required

data_scheme property

data_scheme: list[str]

resources property

resources: dict[str, Any]

Load the data_info.yml that describes table structure, merges, etc.

formatting_functions property

formatting_functions: dict[str, Callable[..., Any]]

column_mapping

column_mapping(request: str, data_info: dict[str, Any] | None = None) -> dict[str, Any]

Retrieve column mappings for a given dataset request and map into a dictionary to be ready for df_data_grabber.

Input: request: str, the request data name in data_scheme (e.g. 'cluster_features', 'household_roster', 'food_acquired', 'interview_date')

Output: final_mapping: dict, {file_name: {'idxvars': idxvar_dic, 'myvars': myvars_dic}}

Example: {'data_file.dta': {'idxvars': {'cluster': ('cluster', )}), 'myvars': {'region': ('region', ), 'urban': ('urban', )}}}

grab_data

grab_data(request: str, *, _with_origin: bool = False) -> pd.DataFrame | tuple[pd.DataFrame, str]

get data from the data file Input: request: str, the request data name (e.g. 'cluster_features', 'household_roster', 'food_acquired', 'interview_date') Output: df: pd.DataFrame, the data requested

_with_origin is private producer plumbing for the actual getters. It returns (frame, origin) without adding public DataFrame attributes. Origin describes the route that supplied this result, not whether its cache path still exists (GH #1078; task ledger sections 2 and 4).