Trait polars_lazy::physical_plan::expressions::PhysicalExpr

pub trait PhysicalExpr: Send + Sync {
    fn as_expression(&self) -> &Expr;
    fn evaluate(
        &self, 
        df: &DataFrame, 
        _state: &ExecutionState
    ) -> Result<Series>;
    fn evaluate_on_groups<'a>(
        &self, 
        df: &DataFrame, 
        groups: &'a GroupTuples, 
        state: &ExecutionState
    ) -> Result<AggregationContext<'a>>;
    fn to_field(&self, input_schema: &Schema) -> Result<Field>;

    fn as_agg_expr(&self) -> Result<&dyn PhysicalAggregation> { ... }
}

This is supported on crate feature compile only.

Take a DataFrame and evaluate the expressions. Implement this for Column, lt, eq, etc

Required methods

fn as_expression(&self) -> &Expr

fn evaluate(&self, df: &DataFrame, _state: &ExecutionState) -> Result<Series>

Take a DataFrame and evaluate the expression.

fn evaluate_on_groups<'a>(
    &self,
    df: &DataFrame,
    groups: &'a GroupTuples,
    state: &ExecutionState
) -> Result<AggregationContext<'a>>

Some expression that are not aggregations can be done per group Think of sort, slice, filter, shift, etc. defaults to ignoring the group

This method is called by an aggregation function.

In case of a simple expr, like ‘column’, the groups are ignored and the column is returned. In case of an expr where group behavior makes sense, this method is called. For a filter operation for instance, a Series is created per groups and filtered.

An implementation of this method may apply an aggregation on the groups only. For instance on a shift, the groups are first aggregated to a ListChunked and the shift is applied per group. The implementation then has to return the Series exploded (because a later aggregation will use the group tuples to aggregate). The group tuples also have to be updated, because aggregation to a list sorts the exploded Series by group.

This has some gotcha’s. An implementation may also change the group tuples instead of the Series.

fn to_field(&self, input_schema: &Schema) -> Result<Field>

Get the output field of this expr

Provided methods

fn as_agg_expr(&self) -> Result<&dyn PhysicalAggregation>

Convert to a aggregation expression. This can only be done for the final expressions that produce an aggregated result.

The expression sum, min, max etc can be called as evaluate in the standard context, or during a groupby execution, this method is called to convert them to an AggPhysicalExpr

Implementors