Tracking issue for RFC 2361, "Simpler alternative dbg!() macro"

[Centril]

This is a tracking issue for the RFC "Simpler alternative dbg!() macro" (rust-lang/rfcs#2361).

Steps:

• Implement the RFC: Implement the dbg!(..) macro #54317
• Adjust documentation (see instructions on forge)
• Stabilization PR (see instructions on forge)

Unresolved questions:

There are none.

[Centril]

I am working on this :)

[yoshuawuyts]

I just tried this out, and something that was unexpected for me was that it dbg!() can only take a single value. I think this might be because most other printers I've used to this point support this (e.g. JS console.log(x, y), Rust println!("{:?}, {:?}", x, y), Bash echo $x $y).

I didn't see this mentioned in the RFC anywhere, but it's possible this was brought up before and now falls under "Unbounded bikeshedding". I don't know if it does, but figured it might be worth sharing my experience in the off chance this hadn't been brought up before.

edit (2018-09-25): I just re-read the proposal and found the tuples entry. I guess that answers it, but yeah it still was something unexpected to me. Ah well.

Expected

let x = 1;
let y = 2;
dbg!(x, y);

Current

let x = 1;
let y = 2;
dbg!(x);
dbg!(y);

[SimonSapin]

I suppose one alternative to the tuple discontinuity could be to return nothing at all when more than one argument is provided.

[Centril]

The original RFC, rust-lang/rfcs#2173, considered it and dealt with this by giving you back a tuple; e.g. dbg!(a, b, c) : (typeof(a), typeof(b), typeof(c)). If we're going to support multiple arguments then I think that is the natural approach. Not returning anything would be surprising imo.