Exp¶

Overview¶

The Exp expression computes the exponential function (e^x) of a numeric value. It returns Euler's number (e) raised to the power of the input value, using Java's StrictMath.exp function for precise mathematical calculations.

Syntax¶

EXP(expr)

exp(col("column_name"))

Arguments¶

Argument	Type	Description
child	Expression	The numeric expression to compute the exponential of

Return Type¶

Double - Returns a double-precision floating-point value representing e^x.

Supported Data Types¶

Numeric types (Integer, Long, Float, Double, Decimal)
Input values are automatically cast to Double for computation

Algorithm¶

Inherits from UnaryMathExpression which handles type coercion and null checking
Uses java.lang.StrictMath.exp() for the actual exponential calculation
StrictMath ensures reproducible results across different platforms and JVM implementations
Input is automatically converted to double precision before calculation
Returns IEEE 754 compliant double-precision result

Partitioning Behavior¶

How this expression affects partitioning:

Preserves partitioning - this is a row-level transformation
Does not require shuffle operations
Can be applied within partition boundaries independently

Edge Cases¶

Null handling: Returns null if input is null
Positive infinity: EXP(709.8) and above returns positive infinity
Underflow: Very negative values (below -745) return 0.0
NaN handling: EXP(NaN) returns NaN
Special values: EXP(0) returns exactly 1.0, EXP(1) returns approximately 2.718281828459045

Code Generation¶

This expression supports Tungsten code generation. The doGenCode method generates optimized Java code that directly calls java.lang.StrictMath.exp(), avoiding object creation and method dispatch overhead during execution.

Examples¶

-- Basic exponential calculation
SELECT EXP(0);
-- Result: 1.0

-- Exponential of 1 (Euler's number)
SELECT EXP(1);
-- Result: 2.718281828459045

-- Using with column data
SELECT name, value, EXP(value) as exp_value 
FROM measurements;

// DataFrame API usage
import org.apache.spark.sql.functions.exp

df.select(exp(col("temperature"))).show()

// With column alias
df.select(exp(col("log_value")).alias("original_value")).show()