DynamicPruningExpression¶

Overview¶

DynamicPruningExpression is a marker expression used during query planning to enable dynamic partition pruning optimization. It wraps an underlying predicate expression and defers all evaluation logic to its child expression while providing metadata for the Catalyst optimizer to identify dynamic pruning opportunities.

Syntax¶

This expression is created internally during query planning and is not directly accessible through SQL or DataFrame API syntax. It wraps predicates that can benefit from dynamic pruning optimization.

Arguments¶

Argument	Type	Description
child	Expression	The underlying predicate expression that performs the actual evaluation

Return Type¶

Returns the same data type as its child expression, as it delegates all evaluation to the child.

Supported Data Types¶

Supports all data types that the underlying child expression supports, as this is purely a wrapper expression that does not modify the evaluation logic.

Algorithm¶

Acts as a transparent wrapper around the child expression during evaluation
Delegates eval() calls directly to the child expression without modification
Provides tree pattern matching capability through DYNAMIC_PRUNING_EXPRESSION pattern
Maintains expression tree structure while adding pruning metadata
Supports both interpreted and code-generated evaluation modes

Partitioning Behavior¶

How this expression affects partitioning:

Does not directly affect partitioning as it's evaluation-transparent
Enables partition pruning optimizations that can reduce the number of partitions scanned
May allow the optimizer to eliminate entire partitions before execution

Edge Cases¶

Null handling: Inherits null handling behavior from the child expression
Empty input: Behavior depends entirely on the child expression's implementation
Error propagation: Any exceptions from child evaluation are propagated unchanged
Planning phase: Expression must be resolved during planning or query execution may fail

Code Generation¶

Supports Tungsten code generation through the doGenCode() method, which delegates code generation to the child expression. The generated code is identical to what the child expression would produce.

Examples¶

-- This expression is not directly accessible in SQL
-- It is created internally during optimization of queries like:
SELECT * FROM table1 t1 
JOIN table2 t2 ON t1.id = t2.id 
WHERE t2.category = 'active'

// Not directly accessible through DataFrame API
// Created internally during catalyst planning phase
// when dynamic pruning optimization is applicable