mario.parse_figaro(
path: str | None = None,
table: str = 'SUT',
year: int | None = None,
iot_mode: str = 'auto',
unit: str = 'MIO_EUR',
countries: str | list[str] | tuple[str, ...] | None = None,
timeout: int = 60,
model: str = 'Database',
name: str = None,
calc_all: bool = False,
**kwargs,
) → object#

Parse FIGARO tables from the Eurostat API.

The parser uses the dedicated Eurostat FIGARO dataflows: naio_10_fcp_s* and naio_10_fcp_u* for SUTs, and naio_10_fcp_ip* or naio_10_fcp_ii* for IOTs. The suffix is selected automatically from year.

Parameters:

  • path (str, optional) – Optional local cache directory for normalized FIGARO API payloads. When provided, MARIO saves downloaded CSV payloads there and reuses matching files on later runs.

  • table (str, optional) – either SUT or IOT.

  • year (int) – FIGARO reference year. This is required.

  • iot_mode (str, optional) – FIGARO IOT variant. Supported values are auto, product and industry. auto defaults to the product-by-product dataflow.

  • unit (str, optional) – Eurostat unit. The current FIGARO parser supports MIO_EUR.

  • countries (str or list-like, optional) – Optional country subset using Eurostat FIGARO country codes. Use it to parse smaller slices without downloading the full multiregional table.

  • timeout (int, optional) – HTTP timeout in seconds for each Eurostat request; defaults to 60.

  • model (str, optional) – Public MARIO model class to instantiate; defaults to Database.

  • name (str, optional) – Dataset name stored in metadata; defaults to the detected source name.

  • calc_all (bool, optional) – Materialize derived matrices immediately after parsing; defaults to False.

Notes

Supply dataflows encode domestic production with one geo dimension. MARIO assigns that region to both axes of S. Use and IOT dataflows retain their separate origin and destination dimensions.