to! converting 1D to 2D array

[ed]

On Wednesday, 12 March 2014 at 02:14:45 UTC, bearophile wrote:
> ed:
>
>> I am trying to convert a 1D array to a 2D array
>
> If you have a dynamic array (1D), you can convert it to a 
> dynamic array of dynamic arrays (2D) using chunks:
>
>
> void main() {
>     import std.stdio, std.range, std.algorithm;
>
>     int[] a = [1, 2, 3, 4, 5, 6];
>     int[][] b = a.chunks(2).array;
>     b.writeln;
> }
>
>
> Output:
>
> [[1, 2], [3, 4], [5, 6]]
>
> Your problems are caused by mixing fixed-size arrays (that are 
> values, allocated in-place), with dynamic arrays (that are 
> little length+pointer structs that often point to 
> heap-allocated memory).
>
> Bye,
> bearophile

Thanks for explaining this, it makes sense what you said. But I'm 
still not sure why my original Example 1 worked.

~~~
// This works OK and converts long[4] to int[] then implicitly to 
int[2][2]
long[4] a=[1,2,3,4];
int[2][2] b = to!(int[])(a);

// Why does this not work the same way?
long[4] a=[1,2,3,4];
int[2][2] b;
b = to!(int[])(a);
~~~

My understanding of your explanation is that it shouldn't work.


Cheers,
ed