Ranges and backward iteration

[Timon Gehr]

On 07/30/2012 01:26 AM, Andrew wrote:
> I have a use case where I would like to be able to pass both a
> forward and backward iteration of an array to a function:
>
> void foo(InputR, OutputR)(InputR i, OutputR j)
> if (isInputRange!InputR && isOutputRange!(OutputR, InputR))
> {
> ...
> }
>
> main()
> {
> //forward:
> foo(a[1..n], b[1..n]);
> //backward:
> foo(retro(a[n..$]), retro(b[n..$]));
>
> //do stuff with b as it has been constructed now...
> }
>
> This doesn't work

Works for me.

This is how I test:

void foo(InputR, OutputR)(InputR i, OutputR j) if (isInputRange!InputR 
&& isOutputRange!(OutputR, InputR)){
     j.put(i);
}

void main() {
     int[] a=[2,0,0,3], b=[1,2,3,4];
     int n=2;
     foo(a[0..n], b[0..n]);
     assert(b==[2,0,3,4]);
     foo(retro(a[n..$]), retro(b[n..$]));
     assert(b==[2,0,0,3]);
}

> (retro doesn't appear to return a Range, but
> rather an object of type "Result", which I don't understand)

'Result' implements the range interface and is a local struct of the 
'retro' function.

>, but the intent should be clear.
>
> How do I do this? What am I missing? There doesn't seem to be a
> "backward iterator" equivalent in std.range.

retro should do the job.