Utility library for numerics. Float versions of ceil(), floor() and rand() with precision. An open minded numeric parser, formatter, plus some additional functions.

Current version supports PHP ^8.0.

Installation

Install with Composer;

composer require phrity/util-numerics

The Numerics class

Class synopsis

Phrity\Util\Numerics {

    /* Methods */
    public __construct(int|null $precision = null, string|null $locale = null);
    public ceil(float $number, int|null $precision = null): float;
    public floor(float $number, int|null $precision = null): float;
    public round(float $number, int|null $precision = null): float;
    public mceil(float $number, float $multipleOf): float;
    public mfloor(float $number, float $multipleOf): float;
    public mround(float $number, float $multipleOf): float;
    public parse(int|float|string $numeric): float|null;
    public format(float $number, int|null $precision = null): string;
    public rand(float $min = 0, float[null $max = null, int|null $precision = null): float|null;
    public precision(float $number, bool $wide = false): int;
    public setLocale(string $locale): void;
}

Constructor

The constructor accepts default precision and/or locale. Precision can also be negative to achive integer rounding.

$numerics = new Numerics(); // Default precision is 0
$numerics = new Numerics(3, 'en_US'); // Default precision is 3, default locale en-US

Ceil method

Round fractions up, according to precision specifier. A precision of 0 corresponds to PHP ceil() function. Precision can also be negative.

// Precison specified on each call
$numerics = new Numerics();
$numerics->ceil(1234.5678,  2); // 1234.57
$numerics->ceil(1234.5678,  1); // 1234.60
$numerics->ceil(1234.5678,  0); // 1235.00
$numerics->ceil(1234.5678, -1); // 1240.00
$numerics->ceil(1234.5678, -2); // 1300.00

// Precison specified as default, override is possible
$numerics = new Numerics(2);
$numerics->ceil(1234.5678); // 1234.57
$numerics->ceil(1234.5678, 0); // 1235.00

Floor method

Round fractions down, according to precision specifier. A precision of 0 corresponds to PHP floor() function. Precision can also be negative.

// Precison specified on each call
$numerics = new Numerics();
$numerics->floor(1234.5678,  2); // 1234.56
$numerics->floor(1234.5678,  1); // 1234.50
$numerics->floor(1234.5678,  0); // 1234.00
$numerics->floor(1234.5678, -1); // 1230.00
$numerics->floor(1234.5678, -2); // 1200.00

// Precison specified as default, override is possible
$numerics = new Numerics(2);
$numerics->floor(1234.5678); // 1234.56
$numerics->floor(1234.5678, 0); // 1234.00

Round method

Standard round, according to precision specifier. Precision can also be negative.

// Precison specified on each call
$numerics = new Numerics();
$numerics->round(1234.5678,  2); // 1234.57
$numerics->round(1234.5678,  1); // 1234.60
$numerics->round(1234.5678,  0); // 1235.00
$numerics->round(1234.5678, -1); // 1230.00
$numerics->round(1234.5678, -2); // 1200.00

// Precison specified as default, override is possible
$numerics = new Numerics(2);
$numerics->round(1234.5678); // 1234.57
$numerics->round(1234.5678, 0); // 1235.00

Ceil multiple-of method

Round up, according to multiple-of specifier.

$numerics = new Numerics();
$numerics->mceil(123.123, 0.25); // 123.25
$numerics->mceil(123.123, 2.5); // 125
$numerics->mceil(123.123, 25); // 125
$numerics->mceil(123.123, 250); // 250

Floor multiple-of method

Round down, according to multiple-of specifier.

$numerics = new Numerics();
$numerics->mfloor(456.789, 0.25); // 456.75
$numerics->mfloor(456.789, 2.5); // 455
$numerics->mfloor(456.789, 25); // 450
$numerics->mfloor(456.789, 250); // 250

Round multiple-of method

Round, according to multiple-of specifier.

$numerics = new Numerics();
$numerics->mround(456.789, 0.25); // 456.75
$numerics->mround(456.789, 2.5); // 457.5
$numerics->mround(456.789, 25); // 450
$numerics->mround(456.789, 250); // 500

Parse method

Numeric parser. Parses number by evaluating input rather than using locale or making explicit assumtions. Returns float, or null if provided input can not be parsed.

$numerics = new Numerics();

// Integer and float input
$numerics->parse(1234.56); // 1234.56
$numerics->parse(1234); // 1234.00

// String input
$numerics->parse('1234.56'); // 1234.56
$numerics->parse('1234,56'); // 1234.56
$numerics->parse('1 234.56'); // 1234.56
$numerics->parse('1 234,56'); // 1234.56
$numerics->parse('1,234.56'); // 1234.56
$numerics->parse('1.234,56'); // 1234.56

// Evaluated string input
$numerics->parse(' 1 234.56 '); // 1234.56
$numerics->parse('-1,234.56'); // -1234.56
$numerics->parse('+1.234,56'); // 1234.56
$numerics->parse('.56'); // 0.56
$numerics->parse(',56'); // 0.56
$numerics->parse('12.'); // 12.0
$numerics->parse('12,'); // 12.0

// Locale support for fringe cases
$numerics->setLocale('sv_SE');
$numerics->parse('123,456'); // 123.456
$numerics->setLocale('en_US');
$numerics->parse('123,456'); // 123456.0

Format method

Numeric formatter. Formats numbers according to precision (rounding and padding) and locale. Precision can also be negative.

$numerics = new Numerics();

$numerics->setLocale('sv_SE');
$numerics->format(1234.5678, 2); // "1 234,56"
$numerics->format(1234, 2); // "1 234,00"
$numerics->format(1234.5678, 0); // "1 234"
$numerics->format(1234, -2); // "1 200"

$numerics->setLocale('en_US');
$numerics->format(1234.5678, 2); // "1,234.56"
$numerics->format(1234, 2); // "1,234.00"
$numerics->format(1234.5678, 0); // "1,234"
$numerics->format(1234, -2); // "1,200"

Using format with many decimals.

$big = 0.123456789123456789;
$numerics->format($big); // "0.123457" - same as below
$numerics->format($big, $numerics->precision($big)); // "0.123457" - php decimal precision
$numerics->format($big, $numerics->precision($big, true)); // "0.123456789123457" - without precision loss
$numerics->format($big, 50); // "0.12345678912345678379658409085095627233386039733887" - with precision loss

Rand method

Float random number with precision. Precision can also be negative. Returns float, or null if impossible to generate result.

// Precison specified on each call
$numerics = new Numerics();
$numerics->rand(0, 10); // 0.0 … 10.0
$numerics->rand(0, 100, 2); // 0.00 … 100.00
$numerics->rand(-100, 100, 4); // -100.0000 … 100.0000
$numerics->rand(0.01, 0.97, 4); // 0.0100 … 0.9700
$numerics->rand(9, 11, -1); // 10.0
$numerics->rand(90, 110, -2); // 100.0

// Precison specified as default, override is possible
$numerics = new Numerics(2);
$numerics->rand(0, 100); // 0.00 … 100.00

Precision method

Count number of relevant decimals in a number. By default the number of serializable decimals (as of php evaluation) is returned.

$numerics = new Numerics();
$numerics->precision(12); // 0
$numerics->precision(12.0); // 0
$numerics->precision(12.34); // 2
$numerics->precision(0.123456789123456789); // 6

By setting the wide option, number of decimals usable without precision loss is returned.

$numerics = new Numerics();
$numerics->precision(0.123456789123456789, true); // 15

setLocale method

Affects the format() method and fringe case in parse() method. If not set, current locale will be used in these methods.

$numerics = new Numerics();
$numerics->setLocale('sv_SE'); // Set to Swedish

Versions

Version	PHP
2.4	^8.0	Multiple-of rounding: mround() mfloor() mceil() methods
2.3	^7.4|^8.0	Precision improvements, negative precision in format()
2.2	^7.4|^8.0	Default locale
2.1	^7.1|^8.0
2.0	^7.1	Instanceable, format() method, ability to specify locale
1.2	>=5.6	rand() and precision() methods
1.1	>=5.6	parse() method
1.0	>=5.6	ceil() and floor() methods