Struct datafusion_expr::function::AccumulatorArgs

pub struct AccumulatorArgs<'a> {
    pub data_type: &'a DataType,
    pub schema: &'a Schema,
    pub dfschema: &'a DFSchema,
    pub ignore_nulls: bool,
    pub sort_exprs: &'a [Expr],
    pub is_reversed: bool,
    pub name: &'a str,
    pub is_distinct: bool,
    pub input_types: &'a [DataType],
    pub input_exprs: &'a [Expr],
}

AccumulatorArgs contains information about how an aggregate function was called, including the types of its arguments and any optional ordering expressions.

Fields

data_type: &'a DataType

The return type of the aggregate function.

schema: &'a Schema

The schema of the input arguments

dfschema: &'a DFSchema

The schema of the input arguments

ignore_nulls: bool

Whether to ignore nulls.

SQL allows the user to specify IGNORE NULLS, for example:

SELECT FIRST_VALUE(column1) IGNORE NULLS FROM t;

sort_exprs: &'a [Expr]

The expressions in the ORDER BY clause passed to this aggregator.

SQL allows the user to specify the ordering of arguments to the aggregate using an ORDER BY. For example:

SELECT FIRST_VALUE(column1 ORDER BY column2) FROM t;

If no ORDER BY is specified, `sort_exprs`` will be empty.

is_reversed: bool

Whether the aggregation is running in reverse order

name: &'a str

The name of the aggregate expression

is_distinct: bool

Whether the aggregate function is distinct.

SELECT COUNT(DISTINCT column1) FROM t;

input_types: &'a [DataType]

The input types of the aggregate function.

input_exprs: &'a [Expr]

The logical expression of arguments the aggregate function takes.

Trait Implementations

impl<'a> Debug for AccumulatorArgs<'a>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.