Skip to content

Arbitrary Precision Library for Delphi with Support for Integer and Floating Point Computations.

License

Notifications You must be signed in to change notification settings

Xor-el/DelphiBigNumberXLib

Repository files navigation

BigNumberXLib

BigNumberXLib is a Delphi Big Number Library with support for Integer and Floating Point Computations. It consists of three custom types namely

  1. TIntegerX which provides support for Integer Computations and is based on ClojureCLR BigInteger Class

  2. TDecimalX which provides support for Floating Point Computations and is based on ClojureCLR BigDecimal Class.

  3. TContext which defines the precision and rounding mode for Floating Point Computations.

Hints about the code:

1.  Multi-condition "for" loops and loops where iterator gets changed inside 
  the loop were converted to while loops. 
    
2.  This Library is written with (Delphi XE7 Update 1). 
    This Library have been tested to work properly with Delphi (XE7 Update 1). 
    This Library might work with other Unicode versions of Delphi (at least XE3 
    upwards) with little or no modifications but have not been tested by me. 
   
3. Finally, Mobile Compilers are Supported.

4. This Library is Locale-Aware but it's Unit Tests are not.

Usage Hints

1. This Library supports Operator Overloading as well as "Chaining Mode".
 
2. When Performing Floating Point Arithmetics, especially Division, it 
  is advisable to use the "Chaining Mode" so we can specify the "TContext" 
  in order to avoid issues with "Non-terminating decimal expansion" like (1/3) 
  since this Library is an "Unlimited" Precision Library.
  When using the "Chaining Mode", do note that it uses Left to Right precedence 
  rather than "PEMDAS" or "BODMAS".
  In order to use "PEMDAS" or "BODMAS" precedence in Chaining Mode, you have
  to group the preferred operations in Parenthesis.

3. Note that when using the Double Constructor of "TDecimalX", 
   TDecimalX.Create(0.1) <> TDecimalX.Create('0.1');
   This is as a result of Computers being unable to store floating point values 
   exactly Internally.

4. Note that the "=" operator or ".Equals" of TDecimalX compares both precision, not
   just coefficients and exponents, so assertions 
   like 1.0 = 1.00 will return false.
   for such comparisons, use the "CompareTo" method.
   Examples
   var
   a, b: TDecimalX;
   result: Boolean;
   begin
     a:= TDecimalX.Create('1.0');
     b:= TDecimalX.Create('1.00');
     result := a = b;  // returns false
     result := a.CompareTo(b) = 0; // returns true
   end; 

Code Examples

To Perform Operations with TIntegerX and TDecimalX:

Using Operator Overloading Method;

uses
SysUtils, IntegerX, DecimalX;

procedure Calc();
var
  int1, int2, intans: TIntegerX;
  dec1, dec2, decans: TDecimalX;

begin
  int1 := TIntegerX.Create(4);
  int2 := TIntegerX.Create('8');
  intans := int1 * int2 + int1 - int2 div int1;
  ShowMessage(intans.ToString());

  dec1 := TDecimalX.Create(8.97);
  dec2 := TDecimalX.Create('6.2');
  decans := dec1 * dec2 + dec1 - dec2 / dec1;
  ShowMessage(decans.ToString());

end;

Using Chaining Method;

uses
SysUtils, IntegerX, DecimalX;

procedure Calc();
var
  int1, int2, intans: TIntegerX;
  dec1, dec2, decans: TDecimalX;
  c: TContext;

begin
  int1 := TIntegerX.Create(4);
  int2 := TIntegerX.Create('8');
  // left to right
  intans := int1.Multiply(int2).Add(int1).Subtract(int2).Divide(int1);
  ShowMessage(intans.ToString());

  c:= TContext.Create(5, TRoundingMode.HalfUp);
  dec1 := TDecimalX.Create(8.97);
  dec2 := TDecimalX.Create('6.2');
  // left to right
  decans := dec1.Multiply(dec2, c).Add(dec1, c).Subtract(dec2, c).Divide(dec1, c);
  ShowMessage(decans.ToString());

end;

Unit Tests

  Unit Tests can be found in BigNumberXLib.Test Folder.
The Unit tests makes use of DUnitX and TestInsight.

License

This Program is Licensed Under Mozilla Public License v. 2.0

Conclusion

Special Thanks to the makers of ClojureCLR and Leo Aso.

About

Arbitrary Precision Library for Delphi with Support for Integer and Floating Point Computations.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages