You can do it this way but it will return a string:

var double = 3.0var double2 = 3.10func forTrailingZero(temp: Double) -> String {    var tempVar = String(format: "%g", temp)    return tempVar}forTrailingZero(double)   //3forTrailingZero(double2)  //3.1

In Swift 4 you can do it like that:

extension Double {    func removeZerosFromEnd() -> String {        let formatter = NumberFormatter()        let number = NSNumber(value: self)        formatter.minimumFractionDigits = 0        formatter.maximumFractionDigits = 16 //maximum digits in Double after dot (maximum precision)        return String(formatter.string(from: number) ?? "")    }}

example of use: print (Double("128834.567891000").removeZerosFromEnd()) result: 128834.567891

You can also count how many decimal digits has your string:

import Foundationextension Double {    func removeZerosFromEnd() -> String {        let formatter = NumberFormatter()        let number = NSNumber(value: self)        formatter.minimumFractionDigits = 0        formatter.maximumFractionDigits = (self.components(separatedBy: ".").last)!.count        return String(formatter.string(from: number) ?? "")    }}

Removing trailing zeros in output

This scenario is good when the default output precision is desired. We test the value for potential trailing zeros, and we use a different output format depending on it.

extension Double {    var stringWithoutZeroFraction: String {        return truncatingRemainder(dividingBy: 1) == 0 ? String(format: "%.0f", self) : String(self)    }}

(works also with extension Float, but not Float80)

Output:

1.0 → "1"
0.1 → "0.1"
0.01 → "0.01"
0.001 → "0.001"
0.0001 → "0.0001"

Formatting with maximum fraction digits, without trailing zeros

This scenario is good when a custom output precision is desired.This solution seems roughly as fast as NumberFormatter + NSNumber solution from MirekE, but one benefit could be that we're avoiding NSObject here.

extension Double {    func string(maximumFractionDigits: Int = 2) -> String {        let s = String(format: "%.\(maximumFractionDigits)f", self)        for i in stride(from: 0, to: -maximumFractionDigits, by: -1) {            if s[s.index(s.endIndex, offsetBy: i - 1)] != "0" {                return String(s[..<s.index(s.endIndex, offsetBy: i)])            }        }        return String(s[..<s.index(s.endIndex, offsetBy: -maximumFractionDigits - 1)])    }}

(works also with extension Float, but not Float80)

Output for maximumFractionDigits: 2:

1.0 → "1"
0.12 → "0.12"
0.012 → "0.01"
0.0012 → "0"
0.00012 → "0"

Note that it performs a rounding (same as MirekE solution):

0.9950000 → "0.99"
0.9950001 → "1"