pyflink.table.udf.udtf#

udtf(f: Union[Callable, pyflink.table.udf.TableFunction, Type], input_types: Optional[Union[List[pyflink.table.types.DataType], pyflink.table.types.DataType, str, List[str]]] = None, result_types: Optional[Union[List[pyflink.table.types.DataType], pyflink.table.types.DataType, str, List[str]]] = None, deterministic: Optional[bool] = None, name: Optional[str] = None, concurrency: Optional[int] = None, num_gpus: Optional[float] = None, gpu_type: Optional[str] = None) → pyflink.table.udf.UserDefinedTableFunctionWrapper[source]#

udtf(f: None = None, input_types: Optional[Union[List[pyflink.table.types.DataType], pyflink.table.types.DataType, str, List[str]]] = None, result_types: Optional[Union[List[pyflink.table.types.DataType], pyflink.table.types.DataType, str, List[str]]] = None, deterministic: Optional[bool] = None, name: Optional[str] = None, concurrency: Optional[int] = None, num_gpus: Optional[float] = None, gpu_type: Optional[str] = None) → Callable[[Union[Callable, pyflink.table.udf.TableFunction, Type]], pyflink.table.udf.UserDefinedTableFunctionWrapper]

Helper method for creating a user-defined table function.

Example

>>> # The input_types is optional.
>>> @udtf(result_types=[DataTypes.BIGINT(), DataTypes.BIGINT()])
... def range_emit(s, e):
...     for i in range(e):
...         yield s, i

>>> # Specify result_types via string
>>> @udtf(result_types=['BIGINT', 'BIGINT'])
... def range_emit(s, e):
...     for i in range(e):
...         yield s, i

>>> # Specify result_types via row string
>>> @udtf(result_types='Row<a BIGINT, b BIGINT>')
... def range_emit(s, e):
...     for i in range(e):
...         yield s, i

>>> class MultiEmit(TableFunction):
...     def eval(self, i):
...         return range(i)
>>> multi_emit = udtf(MultiEmit(), DataTypes.BIGINT(), DataTypes.BIGINT())

Parameters

f – user-defined table function.
input_types – optional, the input data types.
result_types – the result data types.
name – the function name.
deterministic – the determinism of the function’s results. True if and only if a call to this function is guaranteed to always return the same result given the same parameters. (default True)
concurrency – optional concurrency (parallelism) for this UDTF operator.
num_gpus – optional number of GPUs requested for this UDTF.
gpu_type – optional GPU type. Required when num_gpus is set.

Returns

UserDefinedTableFunctionWrapper or function.

New in version 1.11.0.