one line transpose for 2-dimensional Javascript arrays

transpose.js

function transpose(a)
{
  return a[0].map(function (_, c) { return a.map(function (r) { return r[c]; }); });
  // or in more modern dialect
  // return a[0].map((_, c) => a.map(r => r[c]));
}

Nice gist!
Thank you!

Your gist turned up as my first google hit so I thought I would add more coffee: found this on rosetta code (http://rosettacode.org/wiki/Matrix_transposition#CoffeeScript). Note that each map() call costs a function call. Not sure when eliminating a couple of function calls would be significant...

transpose = (matrix) ->
    (t[i] for t in matrix) for i in [0...matrix[0].length]

ES6 version:

transpose = a => a[0].map((_, c) => a.map(r => r[c]));

(edited, was missing the [0] before the first map)

👍

thanks! and ES6 version looks neat!

Hmm... a[0].map? What about empty arrays?
lubien's ES6 version works correctly only for square matrix.
The solution I suggest does not rely on any specific part of the matrix (which could crash the code) nor contains itself within (making use of dot notation easier):

transpose = matrix => matrix.reduce(($, row) => row.map((_, i) => [...($[i] || []), row[i]]), [])

transpose([[0, 1], [2, 3], [4, 5]]) // [[0, 2, 4], [1, 3, 5]]

@tatomyr empty arrays (or non-2-dimensional arrays) a legit concern for a library method, not sure if it's important for a one liner. Probably best dealt with something like this

function transpose(a)
{
  return a && a.length && a[0].map && a[0].map(function (_, c) { return a.map(function (r) { return r[c]; }); }) || [];
}