Dart Decimal

Features

Dart Library that solve the floating point issue by using BigInt.

Note: DartDecimal supports precision up to 18 digits because it uses int instead of BigInt. The purpose is to avoid unnecessarily large numbers or high precision, as they weren't needed for my use case. Additionally, int uses a fixed amount of memory and may be more efficient than BigInt.

Getting started

# pubspec.yaml
dependencies:
  dart_decimal:
    git:
      url: [email protected]:zdirnecamlcs96/dart-decimal.git
      ref: main

$ flutter pub upgrade dart_decimal

Usage

// 1.005 * 1000 = 1004.9999999999999
final a = DartDecimal(amount: 1005, precision: 3);
final b = DartDecimal.parse(1000);
final result = a - b; // 1.005

Edge Cases

Intermediate rounding with limited precision

When performing calculations with limited precision (e.g., rounding to 10 decimal places), discrepancies can arise if intermediate results are rounded too early. Rounding early may cause small errors when those rounded values are used in subsequent calculations, leading to minor discrepancies in the final result.

Example 1: Continuous Calculation (No Intermediate Rounding)

// No rounding until the final result (a / b * c)
final a = DartDecimal.parse(114.4);
final b = DartDecimal.parse(15.2);
final c = a / b; // 7.526315789473684 (full precision)
final result = c * b; // 114.4 (exact result)

Example 2: Intermediate Rounding (Round After Division)

// Intermediate rounding (a / b).roundTo(10) * c
final a = DartDecimal.parse(114.4);
final b = DartDecimal.parse(15.2);
final c = (a / b)roundToPrecision(10); // 7.5263157895 (rounded to 10 decimal places)
final result = c * b; // 114.4000000004 (slight discrepancy due to rounding)

How to handle it

1. Avoid Intermediate Rounding to prevent such discrepancies.
2. For the use case above, check if b is the same as the c and return a directly without further manipulation to ensure the expected outcome.

Disclaimer

This project is intended for personal use only. Use at your own risk. No warranty or support is provided. However, you are welcome to raise issues or submit pull requests.

License

This project is licensed under the MIT License.

Acknowledgements

• Inspired by dart-decimal.
• Learned about scientific notation from MathsIsFun.