Chandler Carruth

And merging (confirmed with the author).

Title says "BindExport -> BindDecl", I think it should be "... -> ExportDecl"

To be clear -- I think the leads need to decide both on a name, but also on whether they want a keyword marking this at all. While there has...

@zygoloid and I discussed this and I think we have a thought of what might be a reasonable direction here, unless someone finds a problem with it. The key is...

> In the case where you have an extern declaration in the API file of library "a", a forward declaration in the API file of library "b", and a definition...

FWIW, the `extern library "..."` approach is growing on me. That said, I definitely see the point about a risk of verbosity. I think some of this depends on whether...

After checking with leads, let's call this decided with the `extern library "..."` direction, which should unblock the proposal in this space.

From @zygoloid's comment above: > The hope is that this will give us better diagnostics, better human readability and understandability of the code, a single meaning for extern, and only...

> Sorry, one point of clarification... Clarifications are good! > Is this allowed: > > ``` > library "a"; > > // The existence of `extern library` is not required....

Thanks for creating the issue, seems good to have something to track this. Also gives me a good place to mention a high-level concern I have in this direction --...