allow ref on locals, globals, and statics

[WalterBright]

Implementation of the proposal discussed in the n.g.:

https://www.digitalmars.com/d/archives/digitalmars/dip/development/First_Draft_ref_For_Variable_Declarations_74.html

[dlang-bot]

Thanks for your pull request, @WalterBright!

Bugzilla references

Your PR doesn't reference any Bugzilla issue.

If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog.

Testing this PR locally

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

dub run digger -- build "master + dmd#16428"

[thewilsonator]

are these such refs implicitly scope? if not should they be? Also this needs a changelog entry

[WalterBright]

are these such refs implicitly scope`?

Yes.

[thewilsonator]

Good to hear. Please add a fail_compilation (e.g. to diag9679.d) that tests that.

[WalterBright]

Please add a fail_compilation (e.g. to diag9679.d) that tests that.

done.

[WalterBright]

Cybershadow/DAutoTest gives no indication why it failed.

[thewilsonator]

As always , scroll to the bottom and work your way back up:

object.Exception@../tools/changed.d(256): Changelog entries should consist of one title line, a blank separator line, and a description.

[WalterBright]

@thewilsonator I did scroll through it, but did not see that message, lost in the thousands of lines of output. Thanks for finding it for me!

[ntrel]

What about (ref i) { ref r = i; ... } - is the declaration of r an error?

[WalterBright]

@ntrel it's the same as the way ref is used in any other context. Same semantics.

[WalterBright]

@ntrel it's the same behavior as:

void foo(ref int x);
void bar(ref int y)
{
    foo(y);
}

[WalterBright]

This is ready to go now.

[dkorpel]

The changelog makes it sound like the DIP was already accepted. Should we wait with merging or word is as a preview feature?

[thewilsonator]

We should wait, the DIP has not even been submitted, let alone reviewed by the wider community.

[WalterBright]

the DIP has not even been submitted, let alone reviewed by the wider community

That's what https://www.digitalmars.com/d/archives/digitalmars/dip/development/First_Draft_ref_For_Variable_Declarations_74.html is for.

[ntrel]

Suggested change

	See the [DIP](https://github.com/WalterBright/documents/blob/master/varRef.md).

[ntrel]

The DIP says:

This DIP is for declaring local variables as ref. Not globals, statics, externs, __gshareds or fields.