On Sunday, 7 August 2022 at 01:22:18 UTC, pascal111 wrote:
Why we use "chain" while we have "~":
'''D
int[] x=[1,2,3];
int[] y=[4,5,6];
auto z=chain(x,y);
auto j=x~y;
'''
They are quite different:
* `chain` gives you "range" (iterator) that starts from the first
element of `x` and ends at the last element of `y` (like e.g.
`zip` in other languages).
* `~` creates a new `int[]` with elements from `x` and the
elements from `y`.
If you use `z[2] = 10;`, you will change `x[2]` as well, since
`z` is just an iterator of which the third element is the value
stored in `x[2]`; if you use `z[4] = 11;`, you will change
`y[1]`; if you change `y[0]`, `z[3]` will also change; etc. (n.b.
if you replace `x` and `y` with other arrays, then this doesn't
apply because then `x` and `y` use different memory.)
`z` is just a range you can use to access the memory of those
`[1,2,3]` and `[4,5,6]` arrays, not a new array.
Here is some example code to show the differences:
```D
void main()
{
import std.range : chain;
import std.stdio : writefln;
int[] x = [ 1, 2, 3 ];
int[] y = [ 4, 5, 6 ];
auto z = chain(x, y);
auto j = x ~ y;
writefln("x: %(%s %)", x);
writefln("y: %(%s %)", y);
writefln("z: %(%s %)", z);
writefln("j: %(%s %)", j);
writefln("----y[1]=10;---");
y[1] = 10;
writefln("x: %(%s %)", x);
writefln("y: %(%s %)", y);
writefln("z: %(%s %)", z);
writefln("j: %(%s %)", j);
writefln("----z[2]=9;----");
z[2] = 9;
writefln("x: %(%s %)", x);
writefln("y: %(%s %)", y);
writefln("z: %(%s %)", z);
writefln("j: %(%s %)", j);
writefln("----j[2]=8;----");
j[2] = 8;
writefln("x: %(%s %)", x);
writefln("y: %(%s %)", y);
writefln("z: %(%s %)", z);
writefln("j: %(%s %)", j);
}
```
output:
```
x: 1 2 3
y: 4 5 6
z: 1 2 3 4 5 6
j: 1 2 3 4 5 6
----y[1]=10;---
x: 1 2 3
y: 4 10 6
z: 1 2 3 4 10 6
j: 1 2 3 4 5 6
----z[2]=9;----
x: 1 2 9
y: 4 10 6
z: 1 2 9 4 10 6
j: 1 2 3 4 5 6
----j[2]=8;----
x: 1 2 9
y: 4 10 6
z: 1 2 9 4 10 6
j: 1 2 8 4 5 6
```
