UnaryPositive¶

Overview¶

The UnaryPositive expression represents the unary plus operator (+) applied to a numeric or interval expression. It is a runtime-replaceable expression that effectively returns the input value unchanged, serving as a no-op operation that preserves the original expression's value and data type.

Syntax¶

SELECT +expression

Arguments¶

Argument	Type	Description
child	Expression	The numeric or interval expression to apply the unary plus operator to

Return Type¶

Returns the same data type as the input expression. The output data type is determined by child.dataType.

Supported Data Types¶

All numeric types (IntegerType, LongType, FloatType, DoubleType, DecimalType, etc.)
All interval types (CalendarIntervalType, YearMonthIntervalType, DayTimeIntervalType)

Algorithm¶

The expression implements the RuntimeReplaceable trait, meaning it is replaced at runtime
During optimization, the entire UnaryPositive expression is replaced with its child expression
No actual computation is performed - it's essentially a pass-through operation
The replacement happens through the lazy val replacement: Expression = child implementation
Supports context-independent constant folding when the child expression is foldable

Partitioning Behavior¶

Preserves partitioning since it's a no-op that doesn't modify data distribution
Does not require shuffle operations
Maintains the same partitioning properties as the child expression

Edge Cases¶

Null handling: Preserves null values unchanged (returns null if input is null)
Empty input: Not applicable as this operates on individual expressions, not collections
Overflow behavior: No overflow concerns since no arithmetic computation is performed
Type preservation: Exactly preserves the input data type and precision for decimal types

Code Generation¶

This expression supports code generation through the RuntimeReplaceable mechanism. Since it gets replaced by its child expression during query planning, the code generation behavior depends entirely on the child expression's code generation capabilities.

Examples¶

-- Example SQL usage
SELECT +1;
-- Returns: 1

SELECT +(-5);
-- Returns: -5

SELECT +salary FROM employees;
-- Returns the salary values unchanged

// Example DataFrame API usage
import org.apache.spark.sql.functions._

// Using unary plus (though rarely needed in practice)
df.select(col("amount").unary_+)

// More commonly seen in SQL expressions
spark.sql("SELECT +price FROM products")