64

Imagine you were given a string and you had to sort that string alphabetically using a function. Example:

sortAlphabets( 'drpoklj' ); //=> returns 'djklopr'

What would be the best way to do this?

1
  • 5
    I tried myself, but I couldn't find what I was looking for. This answer just adds to my experience.
    – Marco V
    Commented Jun 18, 2015 at 10:26

3 Answers 3

115

You can use array sort function:

var sortAlphabets = function(text) {
    return text.split('').sort().join('');
};

STEPS

  1. Convert string to array
  2. Sort array
  3. Convert back array to string

Demo

3
  • 10
    What's the space and time complexity of this solution?
    – kdizzle
    Commented May 18, 2019 at 7:46
  • 1
    @kdizzle Each of these methods is O(N) on its own... Commented Apr 29, 2021 at 19:05
  • 7
    Time Complexity: sort is O(N logN), split and join is O(N), so time complexity is O(N logN). Space Complexity is O(N).
    – Xitang
    Commented Jul 9, 2021 at 16:55
38

Newer browsers support String.prototype.localeCompare() which makes sorting utf8 encoded strings really simple. Note that different languages may have a different order of characters. More information on MDN about localCompare.

function sortAlphabet(str) {
  return [...str].sort((a, b) => a.localeCompare(b)).join("");
}

console.log(sortAlphabet("drpoklj")); // Logs: "djklopr"

If you only have to support ascii strings then the default sorting implementation will do.

function sortAlphabet(str) {
  return [...str].sort().join("");
}
2
  • Note that the returned value is an array while it should be a string, so need to add .join("")
    – yl2015
    Commented Nov 1, 2020 at 10:36
  • 1
    @yl2015 Good catch! I've updated the answer accordingly Commented Nov 9, 2020 at 21:13
21

As previous answers have shown, you convert the string to an array of single-character strings, sort that, and then recombine it into a string. But, using split isn't the best practice way to do that first step, because a JavaScript string is a series of UTF-16 code units with invalid surrogate pairs tolerated, and split("") splits up surrogate pairs into their individual code units, potentially separating them, thus breaking the code point (loosely: character) they're supposed to form as a pair. So if you have an emoji in the string (for instance) or any of hundreds of thousands of characters in non-Western scripts, those can get broken.

In ES5 and earlier, correctly splitting the string required that you detect and handle surrogate pairs to ensure they stayed together, which was a bit of a pain and involved checking charCodeAt for specific ranges of values.

As of ES2015+, it's really easy: You just use the string's iterator, which is defined to provide each code point in the string, whether that's a single code unit or two. To get an array of the code points, you can use the iterator via spread notation ([...str]) or Array.from (Array.from(str)).

So using that, we get:

function sortAlphabets(str) {
    return [...str].sort((a, b) => a.localeCompare(b)).join("");
}

Live Example:

// Using the iterator
function sortAlphabets(str) {
    return [...str].sort((a, b) => a.localeCompare(b)).join("");
}

// Using split("")
function sortAlphabetsUsingSplit(str) {
    return str.split("").sort((a, b) => a.localeCompare(b)).join("");
}

const str = "😀देवनागरी😃";
console.log("Original string    : " + str);
console.log("Using the iterator : " + sortAlphabets(str));
console.log("Using split('')    : " + sortAlphabetsUsingSplit(str));

Note how using split, some of the characters have gotten mangled.

0

Not the answer you're looking for? Browse other questions tagged or ask your own question.