Stabilize `underscore_imports` #56303

petrochenkov · 2018-11-28T00:00:30Z

petrochenkov · 2018-11-28T00:00:40Z

petrochenkov · 2018-11-28T00:40:38Z

Stabilization report

Feature name: #![feature(underscore_imports)]
Version target: 1.32 (2019-01-18)

Implemented in #48922 and landed 8.5 months ago.
Tracking issue is #48216.
Documentation issue is rust-lang/reference#470.
Tests for the feature can be found in src/test/ui/rfc-2166-underscore-imports.

What is being stabilized

use path as _; imports the path path and renames it into some unique identifier that can't be written explicitly.
For traits this means Trait's methods are now in scope in module/block into which use Trait as _; is placed.
For non-traits the import has no effect and is reported as unused by the unused_imports lint.

Additional effect on 2018 edition is that use my_crate::a::b as _; can refer to an extern crate my_crate and link it.
However, the import is still reported as unused in this case, unless it's of the form use my_crate as _; (~~add test~~ tested in #56392).

extern crate name as _; imports the crate name and renames it into some unique identifier that can't be written explicitly.
The crate imported this way is linked, so the feature can be used for link-only crates, and the crate is not reported as unused in this case (~~add test~~ tested in #56392).
If #[macro_use] is attached to such crate it works as usual and imports macros from name into prelude (~~add test?~~ tested in #56392).

Some details and corner cases

Despite the name introduced with _ being unnameable explicitly, it still can be fetched by glob imports.
So, something like use super::*; reimports traits brought in scope with use Trait as _; in the parent module, and their methods become available in the child module.

Underscore imports can be public - pub use Trait as _;, this also can make sense in combination with glob imports.

Open questions

If procedural macro clones the _ token from its input, will it refer to the same unique identifier?
Right now the answer is yes, the "unique identifier" is created during parsing, so procedural macros (and declarative macros as well, if you try hard enough (~~add test~~ tested in #56392)) can clone it while keeping its identity.
It's possible to implement it differently, so unique identifiers are only created during name resolution, after macro expansion.
UPDATE: Creation of gensyms is delayed until name resolution in #56392, so macros cannot clone them anymore.

Known issues

"Unique identifiers" (gensyms) are lost in crate metadata, so all _s become same after being imported from another crate.
This is a hard issue to solve, but it doesn't prevent the common use case of _ imports for importing traits, which doesn't normally assume reexporting _ imports to other crates.
UPDATE: Mostly fixed in #56392.

_ imports and other imports in general, are known to be buggy when used together with macros 2.0 (macro items or Span::def_site), but this only affects unstable code.

petrochenkov · 2018-11-28T00:43:04Z

I'm especially interested in opinions regarding interactions with proc macros ("Open questions" in #56303 (comment)).
cc @dtolnay

Centril

Looks good pending tests passing and pending some of the additional tests discussed above.

Centril · 2018-11-28T01:09:09Z

Thanks for the thoroughly detailed report <3.

I propose that we stabilize #![feature(underscore_imports)] with a version target of 1.32 (or later if necessary...) for all editions (2015 and 2018).

As for the open questions re. macros, I propose that we create unique identifiers during name resolution and after macro expansion such that (#56303 (comment)):

#[my_macro]
use path::to::Trait as _;

// expands to

use path::to::Trait1 as _;
use path::to::Trait2 as _; // the same underscore token

is not an error (same for the declarative macro case in #56303 (comment)).

@rfcbot merge

rfcbot · 2018-11-28T01:09:10Z

dtolnay · 2018-11-28T01:09:30Z

To make sure I understand the procedural macro consideration: would the following be an example of the edge case you have in mind?

#[my_macro]
use path::to::Trait as _;

// expands to

use path::to::Trait1 as _;
use path::to::Trait2 as _; // the same underscore token

and these would be conflicting imports because they reuse the same gensym?

petrochenkov · 2018-11-28T01:21:51Z

@dtolnay
Yes, exactly.

The declarative macro equivalent is

#![feature(underscore_imports)]
#![allow(unused)]

macro_rules! m {
    ($item: item) => { $item $item }
}

m!(use std as _;); // ERROR the name `_` is defined multiple times

fn main() {}

https://play.rust-lang.org/?version=nightly&mode=debug&edition=2015&gist=a56fe430e06fcb300afcb01a64bcfe8b

dtolnay · 2018-11-28T01:26:54Z

Thanks @petrochenkov. If you were to ask me which of the two behaviors I would want for the code in my previous comment, I would mildly prefer for those not to be conflicting imports -- like how if a macro expands to let x: Result<_, _> = ... with the two underscores being the same token, they still operate independently and wouldn't somehow end up being constrained to be the same inferred type.

But I would be comfortable stabilizing this as currently implemented. I don't foresee the behavior being at all problematic for procedural macros in actual usage.

Centril · 2018-11-28T01:40:16Z

@dtolnay

like how if a macro expands to let x: Result<_, _> = ... with the two underscores being the same token, they still operate independently and wouldn't somehow end up being constrained to be the same inferred type.

I agree with this rationale and have amended the proposal accordingly.

Nemo157 · 2018-11-28T08:42:24Z

"Unique identifiers" (gensyms) are lost in crate metadata, so all _s become same after being imported from another crate.
This is a hard issue to solve, but it doesn't prevent the common use case of _ imports for importing tratis, which doesn't normally assume reexporting _ imports to other crates.

Re-exporting _ traits from a prelude is one of the usecases I was looking forward to. embedded_hal::prelude uses manually renamed traits to try and avoid conflicts, trying to replace this with underscore_imports compiles the crate fine, but gives an error related to this when glob imported:

error[E0428]: the name `_` is defined multiple times
  |
  = note: `_` must be defined only once in the type namespace of this module

It would be nice if there were a specific diagnostic for this usecase (and an issue to track implementing a fix?).

petrochenkov · 2018-11-28T14:16:25Z

One more example showing that gensym creation at parse-time is at least suspicious:

duplicate! {
    // The macro argument is parsed as an unstructured token stream, `_` is parsed as a plain `_` token, gensym is not created.
    // Duplication results in two plain `_` tokens.
    // Two plain `_`s are parsed after the macro is expanded, two different gensyms are created, no name conflict happens during resolution.
    use foo as _;
}

// The macro argument is parsed as an item, gensym is created when parsing `_`.
// Duplication results in two identical gensyms.
// Two gensyms are parsed after the macro is expanded, name conflict happens during resolution.
#[duplicate]
use foo as _;

petrochenkov · 2018-11-28T14:26:03Z

@Nemo157
FWIW, we can easily invert the bug, especially if gensym creation is delayed until name resolution.

Now we always consider gensyms from _ same cross-crate, while they can actually be different.
But instead we could always consider them different cross-crate, while they can actually be same.

Actually, I have hard time imagining situation in which the reverted bug would be observable, if we delay the gensym creation until resolve.
We can create a "name conflict":

// Crate 1
pub mod m {
    pub use foo as _;
}

pub use m::*;

// Crate 2
use crate1::m::*;
use crate1::*;

but it won't result in an error, because glob vs glob conflict is only an error if definitions of conflicting names are different, but in this case it's the same foo.
And apparently we can't create non-glob conflicts because gensyms from _ are unnameable.

bors · 2018-11-29T09:45:44Z

☔ The latest upstream changes (presumably #56245) made this pull request unmergeable. Please resolve the merge conflicts.

petrochenkov · 2018-11-30T23:07:24Z

All the suggestions and additional tests are implemented in #56392.

scottmcm · 2018-11-30T23:40:10Z

@rfcbot concern waiting-on-56392-fix

It looks like there's currently one test in https://github.com/rust-lang/rust/tree/master/src/test/ui/rfc-2166-underscore-imports. Is that sufficient? Are there success stories of people trying this out and it working, so people wanting it stabilized to be able to check in? (I didn't see any in #48216, and only a failure story here.)

petrochenkov · 2018-12-01T00:03:16Z

It looks like there's currently one test in https://github.com/rust-lang/rust/tree/master/src/test/ui/rfc-2166-underscore-imports. Is that sufficient?

Well, this is a small feature. Most of intra-crate cases are basically covered in that test. (And #56392 adds some more.)

Nemo157 · 2018-12-01T00:11:54Z

Oh, I can give a success story where I've been using this for a while as well 😄 here I've been using it for bringing in link-only crates (and by the sounds of #56392 that will fix the allow(unused_imports) I had to use).

Delay gensym creation for "underscore items" (`use foo as _`/`const _`) until name resolution So they cannot be cloned by macros. See #56303 for the discussion. Mostly fix cross-crate use of underscore items by inverting the "gensyms are lost in metadata" bug as described in #56303 (comment). Fix unused import warnings for single-segment imports (first commit) and `use crate_name as _` imports (as specified in #56303 (comment)). Prohibit accidentally implemented `static _: TYPE = EXPR;` (cc #55983). Add more tests for `use foo as _` imports.

rfcbot · 2018-12-07T00:42:38Z

🔔 This is now entering its final comment period, as per the review above. 🔔

src/libsyntax/feature_gate.rs

torkleyy · 2018-12-13T11:06:36Z

Awesome! When exporting traits from a prelude like this, what is rustdoc going to generate?

petrochenkov · 2018-12-13T11:34:40Z

@torkleyy
Some quick testing shows this:

rfcbot · 2018-12-17T00:51:12Z

The final comment period, with a disposition to merge, as per the review above, is now complete.

cramertj · 2018-12-17T18:44:24Z

r=me with @Centril's above fix to the feature gate stabilization version.

petrochenkov · 2018-12-17T19:43:42Z

@bors r=cramertj

bors · 2018-12-17T19:43:43Z

📌 Commit 7c901ba has been approved by cramertj

bors · 2018-12-17T22:18:33Z

⌛ Testing commit 7c901ba with merge c9bb68f...

Stabilize `underscore_imports` Closes #48216

bors · 2018-12-18T00:39:33Z

☀️ Test successful - status-appveyor, status-travis
Approved by: cramertj
Pushing c9bb68f to master...

Pkgsrc changes: * Bump required rust version to build to 1.32.0. * Adapt patches to changed file locations. * Since we now patch some more vendor/ modules, doctor the corresponding .cargo-checksum.json files accordingly Upstream changes: Version 1.33.0 (2019-02-28) ========================== Language -------- - [You can now use the `cfg(target_vendor)` attribute.][57465] E.g. `#[cfg(target_vendor="apple")] fn main() { println!("Hello Apple!"); }` - [Integer patterns such as in a match expression can now be exhaustive.][56362] E.g. You can have match statement on a `u8` that covers `0..=255` and you would no longer be required to have a `_ => unreachable!()` case. - [You can now have multiple patterns in `if let` and `while let` expressions.][57532] You can do this with the same syntax as a `match` expression. E.g. ```rust enum Creature { Crab(String), Lobster(String), Person(String), } fn main() { let state = Creature::Crab("Ferris"); if let Creature::Crab(name) | Creature::Person(name) = state { println!("This creature's name is: {}", name); } } ``` - [You can now have irrefutable `if let` and `while let` patterns.][57535] Using this feature will by default produce a warning as this behaviour can be unintuitive. E.g. `if let _ = 5 {}` - [You can now use `let` bindings, assignments, expression statements, and irrefutable pattern destructuring in const functions.][57175] - [You can now call unsafe const functions.][57067] E.g. ```rust const unsafe fn foo() -> i32 { 5 } const fn bar() -> i32 { unsafe { foo() } } ``` - [You can now specify multiple attributes in a `cfg_attr` attribute.][57332] E.g. `#[cfg_attr(all(), must_use, optimize)]` - [You can now specify a specific alignment with the `#[repr(packed)]` attribute.][57049] E.g. `#[repr(packed(2))] struct Foo(i16, i32);` is a struct with an alignment of 2 bytes and a size of 6 bytes. - [You can now import an item from a module as an `_`.][56303] This allows you to import a trait's impls, and not have the name in the namespace. E.g. ```rust use std::io::Read as _; // Allowed as there is only one `Read` in the module. pub trait Read {} ``` - [You may now use `Rc`, `Arc`, and `Pin` as method receivers][56805]. Compiler -------- - [You can now set a linker flavor for `rustc` with the `-Clinker-flavor` command line argument.][56351] - [The mininum required LLVM version has been bumped to 6.0.][56642] - [Added support for the PowerPC64 architecture on FreeBSD.][57615] - [The `x86_64-fortanix-unknown-sgx` target support has been upgraded to tier 2 support.][57130] Visit the [platform support][platform-support] page for information on Rust's platform support. - [Added support for the `thumbv7neon-linux-androideabi` and `thumbv7neon-unknown-linux-gnueabihf` targets.][56947] - [Added support for the `x86_64-unknown-uefi` target.][56769] Libraries --------- - [The methods `overflowing_{add, sub, mul, shl, shr}` are now `const` functions for all numeric types.][57566] - [The methods `rotate_left`, `rotate_right`, and `wrapping_{add, sub, mul, shl, shr}` are now `const` functions for all numeric types.][57105] - [The methods `is_positive` and `is_negative` are now `const` functions for all signed numeric types.][57105] - [The `get` method for all `NonZero` types is now `const`.][57167] - [The methods `count_ones`, `count_zeros`, `leading_zeros`, `trailing_zeros`, `swap_bytes`, `from_be`, `from_le`, `to_be`, `to_le` are now `const` for all numeric types.][57234] - [`Ipv4Addr::new` is now a `const` function][57234] Stabilized APIs --------------- - [`unix::FileExt::read_exact_at`] - [`unix::FileExt::write_all_at`] - [`Option::transpose`] - [`Result::transpose`] - [`convert::identity`] - [`pin::Pin`] - [`marker::Unpin`] - [`marker::PhantomPinned`] - [`Vec::resize_with`] - [`VecDeque::resize_with`] - [`Duration::as_millis`] - [`Duration::as_micros`] - [`Duration::as_nanos`] Cargo ----- - [Cargo should now rebuild a crate if a file was modified during the initial build.][cargo/6484] Compatibility Notes ------------------- - The methods `str::{trim_left, trim_right, trim_left_matches, trim_right_matches}` are now deprecated in the standard library, and their usage will now produce a warning. Please use the `str::{trim_start, trim_end, trim_start_matches, trim_end_matches}` methods instead. - The `Error::cause` method has been deprecated in favor of `Error::source` which supports downcasting. [55982]: rust-lang/rust#55982 [56303]: rust-lang/rust#56303 [56351]: rust-lang/rust#56351 [56362]: rust-lang/rust#56362 [56642]: rust-lang/rust#56642 [56769]: rust-lang/rust#56769 [56805]: rust-lang/rust#56805 [56947]: rust-lang/rust#56947 [57049]: rust-lang/rust#57049 [57067]: rust-lang/rust#57067 [57105]: rust-lang/rust#57105 [57130]: rust-lang/rust#57130 [57167]: rust-lang/rust#57167 [57175]: rust-lang/rust#57175 [57234]: rust-lang/rust#57234 [57332]: rust-lang/rust#57332 [57465]: rust-lang/rust#57465 [57532]: rust-lang/rust#57532 [57535]: rust-lang/rust#57535 [57566]: rust-lang/rust#57566 [57615]: rust-lang/rust#57615 [cargo/6484]: rust-lang/cargo#6484 [`unix::FileExt::read_exact_at`]: https://doc.rust-lang.org/std/os/unix/fs/trait.FileExt.html#method.read_exact_at [`unix::FileExt::write_all_at`]: https://doc.rust-lang.org/std/os/unix/fs/trait.FileExt.html#method.write_all_at [`Option::transpose`]: https://doc.rust-lang.org/std/option/enum.Option.html#method.transpose [`Result::transpose`]: https://doc.rust-lang.org/std/result/enum.Result.html#method.transpose [`convert::identity`]: https://doc.rust-lang.org/std/convert/fn.identity.html [`pin::Pin`]: https://doc.rust-lang.org/std/pin/struct.Pin.html [`marker::Unpin`]: https://doc.rust-lang.org/stable/std/marker/trait.Unpin.html [`marker::PhantomPinned`]: https://doc.rust-lang.org/nightly/std/marker/struct.PhantomPinned.html [`Vec::resize_with`]: https://doc.rust-lang.org/std/vec/struct.Vec.html#method.resize_with [`VecDeque::resize_with`]: https://doc.rust-lang.org/std/collections/struct.VecDeque.html#method.resize_with [`Duration::as_millis`]: https://doc.rust-lang.org/std/time/struct.Duration.html#method.as_millis [`Duration::as_micros`]: https://doc.rust-lang.org/std/time/struct.Duration.html#method.as_micros [`Duration::as_nanos`]: https://doc.rust-lang.org/std/time/struct.Duration.html#method.as_nanos [platform-support]: https://forge.rust-lang.org/platform-support.html

rust-highfive assigned Centril Nov 28, 2018

petrochenkov mentioned this pull request Nov 28, 2018

Document use path as _; and extern crate name as _; rust-lang/reference#470

Closed

petrochenkov mentioned this pull request Nov 28, 2018

Tracking issue for RFC #2166: impl-only-use #48216

Closed

3 tasks

Centril approved these changes Nov 28, 2018

View reviewed changes

Centril added T-lang Relevant to the language team, which will review and decide on the PR/issue. relnotes Marks issues that should be documented in the release notes of the next release. labels Nov 28, 2018

rfcbot added proposed-final-comment-period Proposed to merge/close by relevant subteam, see T-<team> label. Will enter FCP once signed off. disposition-merge This issue / PR is in PFCP or FCP with a disposition to merge it. labels Nov 28, 2018

bors added the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Nov 29, 2018

petrochenkov mentioned this pull request Nov 30, 2018

Delay gensym creation for "underscore items" (use foo as _/const _) until name resolution #56392

Merged

petrochenkov added S-blocked Status: Blocked on something else such as an RFC or other implementation work. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Nov 30, 2018

rfcbot removed the proposed-final-comment-period Proposed to merge/close by relevant subteam, see T-<team> label. Will enter FCP once signed off. label Dec 7, 2018

petrochenkov added S-waiting-on-team Status: Awaiting decision from the relevant subteam (see the T-<team> label). and removed S-blocked Status: Blocked on something else such as an RFC or other implementation work. labels Dec 7, 2018

Centril reviewed Dec 8, 2018

View reviewed changes

src/libsyntax/feature_gate.rs Outdated Show resolved Hide resolved

rfcbot added finished-final-comment-period The final comment period is finished for this PR / Issue. and removed final-comment-period In the final comment period and will be merged soon unless new substantive objections are raised. labels Dec 17, 2018

petrochenkov added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-team Status: Awaiting decision from the relevant subteam (see the T-<team> label). labels Dec 17, 2018

Stabilize underscore_imports

7c901ba

petrochenkov force-pushed the stabuseas branch from 9b8ec84 to 7c901ba Compare December 17, 2018 19:43

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Dec 17, 2018

bors added a commit that referenced this pull request Dec 17, 2018

Auto merge of #56303 - petrochenkov:stabuseas, r=cramertj

c9bb68f

Stabilize `underscore_imports` Closes #48216

bors merged commit 7c901ba into rust-lang:master Dec 18, 2018

This was referenced Dec 18, 2018

Make the getter for NonZero types into a const fn #56739

Closed

Stabilize uniform_paths #56759

Merged

Centril mentioned this pull request Dec 30, 2018

Document underscore imports. rust-lang/reference#492

Merged

Centril added this to the 1.33 milestone Apr 26, 2019

petrochenkov deleted the stabuseas branch June 5, 2019 16:24

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Stabilize `underscore_imports` #56303

Stabilize `underscore_imports` #56303

petrochenkov commented Nov 28, 2018 •

edited

Loading

petrochenkov commented Nov 28, 2018

petrochenkov commented Nov 28, 2018 •

edited

Loading

petrochenkov commented Nov 28, 2018

Centril left a comment •

edited

Loading

Centril commented Nov 28, 2018 •

edited

Loading

rfcbot commented Nov 28, 2018 •

edited by eddyb

Loading

dtolnay commented Nov 28, 2018

petrochenkov commented Nov 28, 2018

dtolnay commented Nov 28, 2018

Centril commented Nov 28, 2018

Nemo157 commented Nov 28, 2018 •

edited

Loading

petrochenkov commented Nov 28, 2018

petrochenkov commented Nov 28, 2018 •

edited

Loading

bors commented Nov 29, 2018

petrochenkov commented Nov 30, 2018

scottmcm commented Nov 30, 2018

petrochenkov commented Dec 1, 2018

Nemo157 commented Dec 1, 2018

rfcbot commented Dec 7, 2018

torkleyy commented Dec 13, 2018

petrochenkov commented Dec 13, 2018

rfcbot commented Dec 17, 2018

cramertj commented Dec 17, 2018

petrochenkov commented Dec 17, 2018

bors commented Dec 17, 2018

bors commented Dec 17, 2018

bors commented Dec 18, 2018

Stabilize underscore_imports #56303

Stabilize underscore_imports #56303

Conversation

petrochenkov commented Nov 28, 2018 • edited Loading

petrochenkov commented Nov 28, 2018

petrochenkov commented Nov 28, 2018 • edited Loading

Stabilization report

What is being stabilized

Some details and corner cases

Open questions

Known issues

petrochenkov commented Nov 28, 2018

Centril left a comment • edited Loading

Choose a reason for hiding this comment

Centril commented Nov 28, 2018 • edited Loading

rfcbot commented Nov 28, 2018 • edited by eddyb Loading

dtolnay commented Nov 28, 2018

petrochenkov commented Nov 28, 2018

dtolnay commented Nov 28, 2018

Centril commented Nov 28, 2018

Nemo157 commented Nov 28, 2018 • edited Loading

petrochenkov commented Nov 28, 2018

petrochenkov commented Nov 28, 2018 • edited Loading

bors commented Nov 29, 2018

petrochenkov commented Nov 30, 2018

scottmcm commented Nov 30, 2018

petrochenkov commented Dec 1, 2018

Nemo157 commented Dec 1, 2018

rfcbot commented Dec 7, 2018

torkleyy commented Dec 13, 2018

petrochenkov commented Dec 13, 2018

rfcbot commented Dec 17, 2018

cramertj commented Dec 17, 2018

petrochenkov commented Dec 17, 2018

bors commented Dec 17, 2018

bors commented Dec 17, 2018

bors commented Dec 18, 2018

Stabilize `underscore_imports` #56303

Stabilize `underscore_imports` #56303

petrochenkov commented Nov 28, 2018 •

edited

Loading

petrochenkov commented Nov 28, 2018 •

edited

Loading

Centril left a comment •

edited

Loading

Centril commented Nov 28, 2018 •

edited

Loading

rfcbot commented Nov 28, 2018 •

edited by eddyb

Loading

Nemo157 commented Nov 28, 2018 •

edited

Loading

petrochenkov commented Nov 28, 2018 •

edited

Loading