associative array: unexpected results after static initialization

[kdevel]

On Friday, 1 December 2017 at 17:02:48 UTC, H. S. Teoh wrote:
> [...]
> There is at least 1 use case in the bugzilla issue that 
> justifies AA literals with (possibly) duplicated keys:
>
> 	https://issues.dlang.org/show_bug.cgi?id=15290#c1
>
> Code snippet:
> -------
> foreach (i; 0..10)
>     foreach (j; 0..10) if (pred2 (i, j))
>         foreach (k; 0..10) if (pred3 (i, j, k))
>             ... and maybe more ...
>         {
>             auto set = [i: true, j: true; k: true];
>             if (set.length == 3)
>             {
>                  ... we found distinct i, j, k, ...
>             }
>         }
> -------

Sure. this looks "idiomatic" but the keys are not constants. If 
the compiler only complained on compile-time duplicate keys it 
would not do so in this case.

BTW: I was under the impression (never checked or profiled such 
code) that in the following code trans_AA and trans_switch are 
technically equivalent:

```
string trans_AA (string s)
{
    immutable string[string] aa = [
       "src1" : "repl1",
       "src2" : "repl2",
       "src3" : "repl3",
    ];
    return aa[s];
}

string trans_switch (string s)
{
    switch (s) {
       case "src1": return "repl1";
       case "src2": return "repl2";
       case "src3": return "repl3";
//      case "src3": return "repl3";
       default:
          throw new Exception ("no key " ~ s);
    }
}

void main ()
{
    import std.stdio;
    import std.string;
    auto a = stdin.readln.chomp;
    a.trans_switch.writeln;
    a.trans_AA.writeln;
}
```

If one uncomments the "src3" line in trans_switch the compiler 
complains with

    aa.d(17): Error: duplicate case "src3" in switch statement

while a duplication in trans_AA goes unnoticed.