Fix Issue 10535 - Add initialize function to allocate an empty AA

[ntrel]

See changelog/docs for details. I'll make a dlang.org pull for this soon.

This is adapted from a patch by Ketmar Dark: https://issues.dlang.org/show_bug.cgi?id=16269#c4.
CC'ing @jmdavis @schveiguy.

[dlang-bot]

Thanks for your pull request and interest in making D better, @ntrel! We are looking forward to reviewing it, and you should be hearing from a maintainer soon.
Please verify that your PR follows this checklist:

• My PR is fully covered with tests (you can see the annotated coverage diff directly on GitHub with CodeCov's browser extension
• My PR is as minimal as possible (smaller, focused PRs are easier to review than big ones)
• I have provided a detailed rationale explaining my changes
• New or modified functions have Ddoc comments (with Params: and Returns:)

Please see CONTRIBUTING.md for more information.

---

If you have addressed all reviews or aren't sure how to proceed, don't hesitate to ping us with a simple comment.

Bugzilla references

Auto-close	Bugzilla	Severity	Description
✓	10535	normal	[AA] Add a function to druntime which returns an empty AA

Testing this PR locally

If you don't have a local development environment setup, you can use Digger to test this PR:

dub fetch digger
dub run digger -- build "master + druntime#2504"

[wilzbach]

You might want to have a look at a very similar attempt: reserve -> #1929

[jacob-carlborg]

I don't like filling up the object module with a lot of symbols. This is a pretty common name as well.

[rainers]

I guess it has been suggested before, but maybe it would be better to make

    int[int] aa = [];

allocate the internal data struct. It currently yields a nonsense Error: cannot have associative array key of void.

[ntrel]

@wilzbach reserve(0) could work for this, but it looks like we don't have an implementation for non-zero sizes ATM - would that be a blocker? (I think reserve is less intuitive).

@jacob-carlborg Andrei suggested adding a function to initialize an empty AA, instead of adding built-in syntax, in 2013:
https://forum.dlang.org/post/[email protected]

@rainers [] could work, but it then means a silent allocation in an AA context, and no allocation in a slice context, and which of these contexts it is is sometimes not visible at the use-site. I think users would also do things like aa == [], causing unnecessary allocations and possible confusion. I suggested aa = new AA as being clearer syntax in the issue tracker.

[rainers]

new AA sounds good to me too. With alias AA = int[string] it currently errors out with new can only create structs, dynamic arrays or class objects, not int[string]'s and new int[string] yields Error: cannot pass type string as a function argument. So code breakage is unlikely.

[jacob-carlborg]

jacob-carlborg Andrei suggested adding a function to initialize an empty AA, instead of adding built-in syntax, in 2013:
https://forum.dlang.org/post/[email protected]

I still don't like adding a new function for it. Using new AA works for me too. It's not really like we're adding new built-in syntax. It's already there, we're just adding a meaning to it.