Fractions

What is Fractions?

Fractions is a unit for doing calculations with fractions.
The unit was inspired by a discussion on the Lazarus forum.

Subversion

You can get a copy of fractions on

svn https://lazarus-ccr.svn.sourceforge.net/svnroot/lazarus-ccr/components/fractions

Or you can use viewcv to preview the files:

http://sourceforge.net/p/lazarus-ccr/svn/HEAD/tree/components/fractions/

Features

The TFraction type

TFraction is an "advanced record": a record with methods and properties.

Porperties of TFraction

• property Numerator: Int64;
  • Gets and sets the Numerator part of the fraction
• property Denominator: Int64;
  • Gets and sets the Denominator part of the fraction.
  • Trying to assign 0 (zero) to TFraction.Denomiator will raise an EZeroDivide exception

Methods of TFraction

• procedure Normalize;
  • Normalizes the fraction by dividing the Numerator and Dominatr by their greatest common divisor, e.g. 10/5 becomes 2/1
• function ToString: String;
  • returns the string representation of the fraction (no normalizaltion is done), e.g. when Numerator = 10 and Denominator = 5, the function returns '10/5'
• function Resolve: String;
  • retruns the string representation of the fraction after normalization, e.g. when Numerator = 4 and Denominator = 3, the function returns '1 1/3'
• function ToFloat: Double;
  • returns the floating point representation of the fraction, e.g. when Numerator = 1 and Denominator = 3, the function returns 0.3333333333333

Assigning and creating Fractions

• function Fraction(ANumerator, ADenominator: Int64): TFraction;
  • creates the fraction ANumerator/ADenominator, e.g. Fraction(1,2) gives 1/2 (one half).
• function Fraction(AIntPart, ANumerator, ADenominator: Int64): TFraction;
  • creates the fraction ANumerator/ADenominator, e.g. Fraction(1,1,2) gives 1 1/2 (one and one half).

Supported calculations

Overloaded operators

• = (equals)
• < (less than)
• > (greater than)
• <= (less than or equal)
• >= (greater than or equal)
• := (assign): this operator allows an Int64 or a String on the right hand side
• + (add)
• - (subtract)
• * (multiply): overloaded for fractions and Int64
• / (divide): overloaded for fractions and Int64
• ** (power): only allowed for Integers (on right hand side)

Mathematical functions

• function Min(a, b: TFraction): TFraction; inline; overload;
  • returns a if (a <= b), otherwise retruns b
• function Max(a, b: TFraction): TFraction; inline; overload;
  • returns a if (a >= b), otherwise returns b
• function InRange(const AValue, AMin, AMax: TFraction): Boolean; inline; overload;
  • returns True if (AValue >= AMin) and (AValue <= AMax), otherwise returns False
• function EnsureRange(const AValue, AMin, AMax: TFraction): TFraction; inline; overload;
  • returns AMin if (AValue < AMin), otherwise returns AMax if (AValue > AMax), otherwise returns AValue
• function Sign(const AValue: TFraction): TValueSign; inline; overload;
  • retruns NegativeValue if (AValue < 0), otherwise returns PositiveValue if (AValue >0), otherwise returns ZeroValue
• function IsZero(const AValue: TFraction): Boolean; overload;
  • returns True if (AValue = 0), otherwise returns False
• function Abs(const AValue: TFraction): TFraction; overload;
  • return AValue if (AValue >= 0), otherwise returns -AValue;

General purpose functions

• function GreatestCommonDivisor(a, b: Int64): Int64;
  • returns the greates common divisor of a and b
• function Floor(D: Double): Int64; overload;
  • returns the integer part of D (subtracted with 1, if (D < 0))

Conversion routines

• function FloatToFraction(Value, Precision: Double): TFraction;
  • returns an approximation of Value in the form of a fraction, e.g. FloatTofraction(0.5,0.01) returns 1/2 (one half)
  • Precision defines the acceptable value for which (Abs(Result) - Abs(Value) <= Precision.
  • Precision is bound to Value by a magnitude of 15 at most
  • FloatToFraction in fact is a function varable which is initialized to the MF_FloatTofraction() function. You can assign your own function to it if wanted.
• function TryFloatToFraction(Value, Precision: Double; out F: TFraction; AcceptPrecisionError: Boolean): Boolean;
  • return value depends on AcceptPrecisionError value:
    • if (AcceptPrecisionError = True) then the function retruns True if an approxiamtion can be found
    • if (AcceptPrecisionError = False) the functions returns True if an approximation can be found AND Abs(Abs(Result)-Abs(Value)) <= Precision
  • the value of F only makes sense if the function returns True
• function FloatToFractionDef(Value, Precision: Double; Def: TFraction; AcceptPrecisionError: Boolean): TFraction;
  • returns the found approximation (F) if TryFloatToFraction(Value, Precision, F, AcceptPrecisionError: Boolean) succeeds, otherwise returns Def
• function StrToFraction(const S: String): TFraction;
  • returns fraction that is represented in S, raises an EConvertError upon failure
• function TryStrToFraction(const S: String; out F: TFraction): Boolean;
  • retruns True if S can be converted to a fraction, otherwise returns False
  • the value of F only makes sense f the function returns True
• function StrToFractionDef(const S: String; Def: TFraction): TFraction;
  • returns the fraction represented in S if conversion succeeds, otherwise returns Def

Example program

program example;

{$mode objfpc}{$H+}
{$apptype console}

uses
  Classes, sysutils, fractions, math;

var
  F1, F2: TFraction;
  D, Prec: Double;
  i: Integer;
begin
  F1 := Fraction(1,1,3); // 1 1/3
  F2 := Fraction(4,3);   // 4/3
  writeln('F1.ToString = ',F1.ToString); // '4/3'
  writeln('F1.Resolve  = ',F1.Resolve);  // '1 1/3'
  writeln('F1.ToFloat  = ',F1.ToFloat:16:16); // 1.3333333333333333
  writeln('F2.ToString = ',F2.ToString); // '4/3'
  writeln('(F1 = F2)   = ',F1=F2);         //True

  F1 := Fraction(1,2);
  F2 := Fraction(1,3);
  writeln(F1.ToString,' * ',F2.ToString,'   = ',(F1*F2).Resolve);  // '1/6'
  writeln(F1.ToString,' / ',F2.ToString,'   = ',(F1/F2).Resolve);  // '1 1/2'
  writeln(F1.ToString,' + ',F2.ToString,'   = ',(F1+F2).Resolve);  // '5/6'
  writeln(F1.ToString,' - ',F2.ToString,'   = ',(F1-F2).Resolve);  // '1/6'
  writeln(F1.ToString,' ** 2    = ',(F1**2).Resolve);  // '1/6'

  D := 0.25;
  F1 := FloatToFraction(D, 0.000001);
  writeln('FloatTofraction(0.25) -> ',F1.ToString);   // '1/4'
  writeln;
  writeln('Approximations of Pi:');
  writeln('                                              [Pi = ',Pi:16:16,']');
  Prec := 1.0;
  for i := 1 to 10 do
  begin
    Prec := Prec / 10;
    F2 := FloatTofraction(Pi, Prec);
    writeln('FloatTofraction(Pi,',Prec:10:10,') = ',Format('%-13s',[F2.Resolve]),'   [',F2.ToFloat:16:16,']');
  end;
end.

This outputs:

F1.ToString = 4/3
F1.Resolve  = 1 1/3
F1.ToFloat  = 1.3333333333333333
F2.ToString = 4/3
(F1 = F2)   = TRUE
1/2 * 1/3   = 1/6
1/2 / 1/3   = 1 1/2
1/2 + 1/3   = 5/6
1/2 - 1/3   = 1/6
1/2 ** 2    = 1/4
FloatTofraction(0.25) -> 1/4

Approximations of Pi:
                                              [Pi = 3.1415926535897932]
FloatTofraction(Pi,0.1000000000) = 3 1/7           [3.1428571428571428]
FloatTofraction(Pi,0.0100000000) = 3 1/7           [3.1428571428571428]
FloatTofraction(Pi,0.0010000000) = 3 16/113        [3.1415929203539825]
FloatTofraction(Pi,0.0001000000) = 3 16/113        [3.1415929203539825]
FloatTofraction(Pi,0.0000100000) = 3 16/113        [3.1415929203539825]
FloatTofraction(Pi,0.0000010000) = 3 16/113        [3.1415929203539825]
FloatTofraction(Pi,0.0000001000) = 3 4703/33215    [3.1415926539214212]
FloatTofraction(Pi,0.0000000100) = 3 4703/33215    [3.1415926539214212]
FloatTofraction(Pi,0.0000000010) = 3 4703/33215    [3.1415926539214212]
FloatTofraction(Pi,0.0000000001) = 3 14093/99532   [3.1415926536189365]

From the Free Pascal wiki, page "Fractions", as it stood on 2022-08-06 after a change by Kai Burghardt, out of the wiki's dump of 2024-05-26. Copyright stays with the author who wrote it, according to the old wiki's Licensing page. If you have concerns about this material, message the admin.