Source code for dagster_pandas.validation

from dagster import (
    DagsterInvariantViolationError,
    _check as check,
)
from pandas import DataFrame, Timestamp
from pandas.core.dtypes.common import (
    is_bool_dtype,
    is_float_dtype,
    is_integer_dtype,
    is_numeric_dtype,
    is_string_dtype,
)

from dagster_pandas.constraints import (
    CategoricalColumnConstraint,
    ColumnDTypeFnConstraint,
    ColumnDTypeInSetConstraint,
    Constraint,
    ConstraintViolationException,
    DataFrameConstraint,
    InRangeColumnConstraint,
    NonNullableColumnConstraint,
    UniqueColumnConstraint,
)

PANDAS_NUMERIC_TYPES = {"int64", "float"}


def _construct_keyword_constraints(non_nullable, unique, ignore_missing_vals):
    non_nullable = check.bool_param(non_nullable, "exists")
    unique = check.bool_param(unique, "unique")
    ignore_missing_vals = check.bool_param(ignore_missing_vals, "ignore_missing_vals")
    if non_nullable and ignore_missing_vals:
        raise DagsterInvariantViolationError(
            "PandasColumn cannot have a non-null constraint while also ignore missing values"
        )
    constraints = []
    if non_nullable:
        constraints.append(NonNullableColumnConstraint())
    if unique:
        constraints.append(UniqueColumnConstraint(ignore_missing_vals=ignore_missing_vals))
    return constraints


[docs]class PandasColumn: """The main API for expressing column level schemas and constraints for your custom dataframe types. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If th column exists, the validate function will validate the column. Defaults to True. constraints (Optional[List[Constraint]]): List of constraint objects that indicate the validation rules for the pandas column. """ def __init__(self, name, constraints=None, is_required=None): self.name = check.str_param(name, "name") self.is_required = check.opt_bool_param(is_required, "is_required", default=True) self.constraints = check.opt_list_param(constraints, "constraints", of_type=Constraint) def validate(self, dataframe): if self.name not in dataframe.columns: # Ignore validation if column is missing from dataframe and is not required if self.is_required: raise ConstraintViolationException( "Required column {column_name} not in dataframe with columns" " {dataframe_columns}".format( column_name=self.name, dataframe_columns=dataframe.columns ) ) else: for constraint in self.constraints: constraint.validate(dataframe, self.name) @staticmethod def exists(name, non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None): """Simple constructor for PandasColumns that expresses existence constraints. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ return PandasColumn( name=check.str_param(name, "name"), constraints=_construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def boolean_column( name, non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None ): """Simple constructor for PandasColumns that expresses boolean constraints on boolean dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ return PandasColumn( name=check.str_param(name, "name"), constraints=[ColumnDTypeFnConstraint(is_bool_dtype)] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def numeric_column( name, min_value=-float("inf"), max_value=float("inf"), non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None, ): """Simple constructor for PandasColumns that expresses numeric constraints numeric dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. min_value (Optional[Union[int,float]]): The lower bound for values you expect in this column. Defaults to -float('inf') max_value (Optional[Union[int,float]]): The upper bound for values you expect in this column. Defaults to float('inf') non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ return PandasColumn( name=check.str_param(name, "name"), constraints=[ ColumnDTypeFnConstraint(is_numeric_dtype), InRangeColumnConstraint( check.numeric_param(min_value, "min_value"), check.numeric_param(max_value, "max_value"), ignore_missing_vals=ignore_missing_vals, ), ] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def integer_column( name, min_value=-float("inf"), max_value=float("inf"), non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None, ): """Simple constructor for PandasColumns that expresses numeric constraints on integer dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. min_value (Optional[Union[int,float]]): The lower bound for values you expect in this column. Defaults to -float('inf') max_value (Optional[Union[int,float]]): The upper bound for values you expect in this column. Defaults to float('inf') non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ return PandasColumn( name=check.str_param(name, "name"), constraints=[ ColumnDTypeFnConstraint(is_integer_dtype), InRangeColumnConstraint( check.numeric_param(min_value, "min_value"), check.numeric_param(max_value, "max_value"), ignore_missing_vals=ignore_missing_vals, ), ] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def float_column( name, min_value=-float("inf"), max_value=float("inf"), non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None, ): """Simple constructor for PandasColumns that expresses numeric constraints on float dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. min_value (Optional[Union[int,float]]): The lower bound for values you expect in this column. Defaults to -float('inf') max_value (Optional[Union[int,float]]): The upper bound for values you expect in this column. Defaults to float('inf') non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ return PandasColumn( name=check.str_param(name, "name"), constraints=[ ColumnDTypeFnConstraint(is_float_dtype), InRangeColumnConstraint( check.numeric_param(min_value, "min_value"), check.numeric_param(max_value, "max_value"), ignore_missing_vals=ignore_missing_vals, ), ] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def datetime_column( name, min_datetime=Timestamp.min, max_datetime=Timestamp.max, non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None, tz=None, ): """Simple constructor for PandasColumns that expresses datetime constraints on 'datetime64[ns]' dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. min_datetime (Optional[Union[int,float]]): The lower bound for values you expect in this column. Defaults to pandas.Timestamp.min. max_datetime (Optional[Union[int,float]]): The upper bound for values you expect in this column. Defaults to pandas.Timestamp.max. non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. tz (Optional[str]): Required timezone for values eg: tz='UTC', tz='Europe/Dublin', tz='US/Eastern'. Defaults to None, meaning naive datetime values. """ if tz is None: datetime_constraint = ColumnDTypeInSetConstraint({"datetime64[ns]"}) else: datetime_constraint = ColumnDTypeInSetConstraint({f"datetime64[ns, {tz}]"}) # One day more/less than absolute min/max to prevent OutOfBoundsDatetime errors when converting min/max to be tz aware if min_datetime.tz_localize(None) == Timestamp.min: min_datetime = Timestamp("1677-09-22 00:12:43.145225Z") if max_datetime.tz_localize(None) == Timestamp.max: max_datetime = Timestamp("2262-04-10 23:47:16.854775807Z") # Convert bounds to same tz if Timestamp(min_datetime).tz is None: min_datetime = Timestamp(min_datetime).tz_localize(tz) if Timestamp(max_datetime).tz is None: max_datetime = Timestamp(max_datetime).tz_localize(tz) return PandasColumn( name=check.str_param(name, "name"), constraints=[ datetime_constraint, InRangeColumnConstraint( min_datetime, max_datetime, ignore_missing_vals=ignore_missing_vals ), ] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def string_column( name, non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None ): """Simple constructor for PandasColumns that expresses constraints on string dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ return PandasColumn( name=check.str_param(name, "name"), constraints=[ColumnDTypeFnConstraint(is_string_dtype)] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, ) @staticmethod def categorical_column( name, categories, of_types=frozenset({"category", "object"}), non_nullable=False, unique=False, ignore_missing_vals=False, is_required=None, ): """Simple constructor for PandasColumns that expresses categorical constraints on specified dtypes. Args: name (str): Name of the column. This must match up with the column name in the dataframe you expect to receive. categories (List[Any]): The valid set of buckets that all values in the column must match. of_types (Optional[Union[str, Set[str]]]): The expected dtype[s] that your categories and values must abide by. non_nullable (Optional[bool]): If true, this column will enforce a constraint that all values in the column ought to be non null values. unique (Optional[bool]): If true, this column will enforce a uniqueness constraint on the column values. ignore_missing_vals (Optional[bool]): A flag that is passed into most constraints. If true, the constraint will only evaluate non-null data. Ignore_missing_vals and non_nullable cannot both be True. is_required (Optional[bool]): Flag indicating the optional/required presence of the column. If the column exists the validate function will validate the column. Default to True. """ of_types = {of_types} if isinstance(of_types, str) else of_types return PandasColumn( name=check.str_param(name, "name"), constraints=[ ColumnDTypeInSetConstraint(of_types), CategoricalColumnConstraint(categories, ignore_missing_vals=ignore_missing_vals), ] + _construct_keyword_constraints( non_nullable=non_nullable, unique=unique, ignore_missing_vals=ignore_missing_vals ), is_required=is_required, )
def validate_constraints(dataframe, pandas_columns=None, dataframe_constraints=None): dataframe = check.inst_param(dataframe, "dataframe", DataFrame) pandas_columns = check.opt_list_param( pandas_columns, "column_constraints", of_type=PandasColumn ) dataframe_constraints = check.opt_list_param( dataframe_constraints, "dataframe_constraints", of_type=DataFrameConstraint ) if pandas_columns: for column in pandas_columns: column.validate(dataframe) if dataframe_constraints: for dataframe_constraint in dataframe_constraints: dataframe_constraint.validate(dataframe)