array-math

##Install
npm install array-math
##Require
var aMath = require('array-math')
##Methods
##aMath.factors(n)
n
is a positive integer
aMath.factors(2) // -> [2]
aMath.factors(96) // -> [2, 2, 2, 2, 2, 3]
aMath.factors(100) // -> [2, 2, 5, 5]
##aMath.divisors(n)
n
is a positive integer
aMath.divisors(2) // -> [2]
aMath.divisors(96) // -> [1, 2, 3, 4, 6, 8, 12, 16, 24, 32, 48, 96]
aMath.divisors(100) // -> [1, 2, 4, 5, 10, 20, 25, 50, 100]
##aMath.isPrime(n)
n
is a positive integer
aMath.isPrime(2) // -> true
aMath.isPrime(3) // -> true
aMath.isPrime(4) // -> false
aMath.isPrime(7) // -> true
aMath.isPrime(96) // -> false
aMath.isPrime(97) // -> true
aMath.isPrime(100) // -> false
aMath.isPrime(113) // -> true
aMath.isPrime(117) // -> false
##aMath.range(n[, m])
n
is a number. If the m
parameter was passed in, then n
is the low number and m is the high number. If the m parameter was not passed in, then n is the high number. (See examples below.)
aMath.range(0) // -> []
aMath.range(1) // -> [0]
aMath.range(2) // -> [0, 1]
aMath.range(2, 2) // -> []
aMath.range(2, 3) // -> [2]
aMath.range(3) // -> [0, 1, 2]
aMath.range(10) // -> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
aMath.range(2, 10) // -> [2, 3, 4, 5, 6, 7, 8, 9]
aMath.range(5, 10) // -> [5, 6, 7, 8, 9]
##aMath.multiply(a)
a
is an array of numbers (integers, floats, negative, whatever)
aMath.multiply([2, 96, 100]) // -> 19200
aMath.multiply([40, 3, 17]) // -> 2040
aMath.multiply([4, 5, 2, 5.2, 3.8]) // -> 790.4
aMath.multiply([520, 0.2, 0.2, 0.8]) // -> 16.64
##aMath.sum(a)
a
is an array of numbers (integers, floats, negative, whatever)
aMath.sum([2, 96, 100]) // -> 198
aMath.sum([2, -96, 100]) // -> 6
aMath.sum([45, 20, 8.3]) // -> 73.3
##aMath.factorial(h[, l])
h
is a number. It is the high number. It defaults to 0.
l
is a number. It is the low number. It defaults to 0.
While multiplying, it will never multiply by 0.
aMath.factorial() // -> 1
aMath.factorial(0) // -> 1
aMath.factorial(1) // -> 1
aMath.factorial(2) // -> 2 (2x1)
aMath.factorial(3) // -> 6 (3x2x1)
aMath.factorial(5) // -> 120 (5x4x3x2x1)
aMath.factorial(5, 0) // -> 120 (5x4x3x2x1)
aMath.factorial(5, 1) // -> 120 (5x4x3x2x1)
aMath.factorial(5, 2) // -> 120 (5x4x3x2)
aMath.factorial(5, 3) // -> 60 (5x4x3)
aMath.factorial(5, 4) // -> 20 (5x4)
aMath.factorial(10) // -> 3628800 (10x9x8x7x6x5x4x3x2x1)
aMath.factorial(10, 3) // -> 1814400 (10x9x8x7x6x5x4x3)
aMath.factorial(10, 5) // -> 151200 (10x9x8x7x6x5)
##License
MIT