std/bigfloat

BigFloat provides arbitrary-precision decimal floating-point arithmetic for Zen-C. It is implemented as a scaled BigInt, allowing for high-precision calculations without binary rounding errors.

Overview

• Arbitrary Precision: Supports decimal numbers of any size, limited only by memory.
• Scaled Representation: Uses a BigInt magnitude and an integer scale to represent decimal values.
• Precision Control: Easily align scales for precise addition and subtraction.
• RAII: Memory for the underlying magnitude is automatically managed via the Drop trait.

Usage

import "std/bigfloat.zc"

fn main() {
    let a = BigFloat::from_int(123);
    a.scale = 2; // Represents 1.23
    
    let b = BigFloat::from_int(4567);
    b.scale = 3; // Represents 4.567
    
    let sum = a.add(b);
    
    let s = sum.to_string();
    println "Sum: {s}"; // Sum: 5.797
    free(s);
} // a, b, and sum are freed automatically here

Struct Definition

struct BigFloat {
    magnitude: BigInt;
    scale: int;
}

Methods

Construction

Method	Signature	Description
new	BigFloat::new() -> BigFloat	Creates a new BigFloat initialized to 0.0.
from_int	BigFloat::from_int(val: u64) -> BigFloat	Creates a BigFloat from an integer with scale 0.

Modification

Method	Signature	Description
add	add(self, other: BigFloat) -> BigFloat	Adds two BigFloat values, automatically aligning their scales. Returns a new BigFloat.
align_scale	align_scale(self, target_scale: int)	Increases the scale of the BigFloat to target_scale by shifting the magnitude.

Utility

Method	Signature	Description
clone	clone(self) -> BigFloat	Returns a deep copy of the BigFloat.
to_string	to_string(self) -> char*	Returns a heap-allocated string representation with the decimal point.

Memory Management

Method	Signature	Description
free_mem	free_mem(self)	Manually frees the underlying BigInt memory.
Trait	impl Drop for BigFloat	Automatically calls free_mem() when out of scope.