Simplified (and slow) implementation:

T[] splitLength(T)(T arr, size_t count) if (isArray!T)
{
    T[] result;

    while (arr.length)
    {
        result ~= arr.take(count);
        arr.popFrontN(count);
    }

    return result;
}

Strings are an exception again, that code won't work for strings. Damn..

On Monday, August 15, 2011 04:49:59 Andrej Mitrovic wrote:
> Is there something in Phobos with which I could do:
> 
> auto arr = [1, 2, 3, 4, 5, 6];
> 
> int[][] newarr = arr.splitLength(2);
> assert(newarr.length == 3);

I could have sworn that there was a function that did this, but I can't find it at the moment.

- Jonathan M Davis

On Sun, 14 Aug 2011 23:00:26 -0400, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:

> Simplified (and slow) implementation:
>
> T[] splitLength(T)(T arr, size_t count) if (isArray!T)
> {
>     T[] result;
>
>     while (arr.length)
>     {
>         result ~= arr.take(count);
>         arr.popFrontN(count);
>     }
>
>     return result;
> }

Ouch!

Slicing is your friend :)  Also, threw in a reserve for good measure, and you really need an assert to ensure count is not 0:

assert(count > 0);

result.reserve((arr.length + count - 1) / count)

while(arr.length > count)
{
   result ~= arr[0..count];
   arr = arr[count..$];
}

if(arr.length)
  result ~= arr;

-Steve

On 8/15/11, Steven Schveighoffer <schveiguy@yahoo.com> wrote:
> Ouch!

It's not that big of an ouch actually: http://codepad.org/m1zKlP0e

Am 15.08.2011, 18:06 Uhr, schrieb Andrej Mitrovic <andrej.mitrovich@gmail.com>:

> On 8/15/11, Steven Schveighoffer <schveiguy@yahoo.com> wrote:
>> Ouch!
>
> It's not that big of an ouch actually: http://codepad.org/m1zKlP0e

Compiled in optimized 64-bit with dmd it gives me:
old: 393 ms
new: 310 ms
which is ~27% overhead. It is an average ouch.