MakeDecimal¶

Overview¶

MakeDecimal is an internal Spark Catalyst expression that creates a Decimal value from an unscaled Long integer. This expression is generated exclusively by the optimizer during query planning and is not directly accessible to users. It handles the conversion of Long values to Decimal types with specified precision and scale parameters.

Syntax¶

This expression is internal only and not directly accessible via SQL or DataFrame API. It is created automatically by the Catalyst optimizer during decimal operations.

Arguments¶

Argument	Type	Description
child	Expression	The input expression that evaluates to a Long value
precision	Int	The total number of digits in the decimal number
scale	Int	The number of digits after the decimal point
nullOnOverflow	Boolean	Whether to return null on overflow (true) or throw exception (false)

Return Type¶

Returns DecimalType(precision, scale) where precision and scale are the specified parameters.

Supported Data Types¶

This expression specifically requires:

Input: Long integers (unscaled decimal values)
Output: Decimal type with specified precision and scale

Algorithm¶

The expression evaluation follows these steps:

Casts the input value to Long type
Creates a new Decimal instance
Uses either setOrNull or set method based on overflow handling preference
Returns the constructed Decimal value with the specified precision and scale
Handles overflow according to the nullOnOverflow flag

Partitioning Behavior¶

This expression preserves partitioning characteristics:

Does not require data shuffle as it operates on individual values
Maintains the same partitioning scheme as the input expression
Can be evaluated locally on each partition

Edge Cases¶

Null handling: Expression is null-intolerant, meaning null inputs produce null outputs
Overflow behavior: Controlled by nullOnOverflow parameter - either returns null or throws exception
ANSI mode: When ANSI SQL compliance is enabled, overflow throws exceptions instead of returning null
Precision limits: Subject to Spark's decimal precision and scale constraints
Default constructor: Uses !SQLConf.get.ansiEnabled as default for nullOnOverflow parameter

Code Generation¶

This expression supports Tungsten code generation through the doGenCode method. It generates optimized bytecode that:

Uses nullSafeCodeGen for efficient null handling
Dynamically selects between setOrNull and set methods
Includes conditional null checking only when the result is nullable

Examples¶

-- This expression is internal only and not directly accessible in SQL
-- It may be generated during decimal arithmetic operations
SELECT CAST(123 AS DECIMAL(10,2)) + CAST(456 AS DECIMAL(10,2));

// This expression is internal only and not directly accessible in DataFrame API
// It may be generated during decimal operations
import org.apache.spark.sql.functions._
df.select(col("long_column").cast(DecimalType(10, 2)))