Array literals' default type

[Denis Koroskin]

On Thu, 08 Oct 2009 22:07:32 +0400, Andrei Alexandrescu  
<SeeWebsiteForEmail at erdani.org> wrote:

> Consider:
>
> auto c = [ 2.71, 3.14, 6.023e22 ];
> c ~= 2.21953167;
>
> Should this work? Currently it doesn't because c's type is deduced as  
> double[3].
>
> The literal can initialize either double[3] or double[], so the question  
> is only what the default should be when "auto" is used.
>
>
> Thoughts?
>
> Andrei

I was just about to bump a similar topic.

I strongly believe typeof(c) must be immutable(double)[3].

There are 2 problems in D with current design:

1) The following code:

     auto c = [ 2.71, 3.14, 6.023e22 ];

always allocates memory from a heap. In many cases, a read-only view of  
that array would suffice. In case a mutation is needed, no one stops you  
 from dup'ing that array:

     auto c = [ 2.71, 3.14, 6.023e22 ].dup;

2) There is an inconsistency with strings:

     auto c1 = "Hello"; // immutable
     auto c2 = ['H', 'e', 'l', 'l', 'o']; // mutable

I don't like hidden allocations like this to present in D. I believe this  
is not what most users expect to happen.

Back to your question, I believe it should be fixed-size and immutable.  
Once again, no one stops you from using "[]":

     auto c = [ 2.71, 3.14, 6.023e22 ][]; // a slice is returned

The opposite is impossible - you can't get a fixed-sized array from a  
dynamic one.

Another question is, what type should .dup return? T[new] or T[]? (The  
former one, probably, since the latter one is accessible via .dup[]  
syntax...) Anyway, what is the state of T[new] in future of D?