Struct JoinFilterRewriter 

struct JoinFilterRewriter<'a> {
    join_side: JoinSide,
    join_side_schema: &'a Schema,
    join_side_projections: Vec<(Arc<dyn PhysicalExpr>, String)>,
    intermediate_column_indices: Vec<ColumnIndex>,
    alias_generator: &'a AliasGenerator,
}

Implements the push-down machinery.

The rewriter starts at the top of the filter expression and traverses the expression tree. For each (sub-)expression, the rewriter checks whether it only refers to one side of the join. If this is never the case, no subexpressions of the filter can be pushed down. If there is a subexpression that can be computed using only one side of the join, the entire subexpression is pushed down to the join side.

Fields

join_side: JoinSide

join_side_schema: &'a Schema

join_side_projections: Vec<(Arc<dyn PhysicalExpr>, String)>

intermediate_column_indices: Vec<ColumnIndex>

alias_generator: &'a AliasGenerator

Implementations

impl<'a> JoinFilterRewriter<'a>

fn new( join_side: JoinSide, join_side_schema: &'a Schema, column_indices: Vec<ColumnIndex>, alias_generator: &'a AliasGenerator, ) -> Self

Creates a new JoinFilterRewriter.

fn rewrite( &mut self, expr: Arc<dyn PhysicalExpr>, ) -> Result<Transformed<Arc<dyn PhysicalExpr>>>

Executes the push-down machinery on expr.

See the JoinFilterRewriter for further information.

fn create_new_column( &mut self, name: String, expr: Arc<dyn PhysicalExpr>, ) -> Result<usize>

Creates a new column in the current join side.

fn depends_on_join_side( &mut self, expr: &Arc<dyn PhysicalExpr>, join_side: JoinSide, ) -> Result<bool>

Checks whether the entire expression depends on the given join_side.