Rust 1.30 #281

steveklabnik · 2018-10-24T20:19:03Z

r? @rust-lang/core

@ashleygwilliams will be making me a GIF to include in the post tonight or tomorrow, but we can't forget that before a merge.

petrochenkov · 2018-10-24T20:52:00Z

_posts/2018-10-25-Rust-1.30.md

+its rules felt awkward in practice. These changes are the first steps we're
+taking to make the module system feel more straightforward.
+
+[`mod.rs` files are now optional][optionalmod]. Imagine we had a `foo`


This one is un-stabilized again in rust-lang/rust#55315 due to issues.

I think you can mention tool attributes instead (like #[rustfmt::skip]), they are well hidden in https://github.com/rust-lang/rust/blob/master/RELEASES.md#misc in the release notes.

FYI, I've just opened up rust-lang/rust#55331. The tool_lints feature doesn't actually land until 1.31, ~~even though the compiler warns and recommends the new syntax~~.

Edit: I've confirmed that 1.30 doesn't actually generate a warning on this when running cargo clippy. It looks like this was backed out of beta some time ago rust-lang/rust@27daef9.

@jtgeibel
The release notes talk about tool attributes (#[rustfmt::skip]) rather than tool lints (#[allow(clippy::something)]).
Tool attributes were indeed stabilized on 1.30.

@petrochenkov thanks a lot for the clarification, I wasn't aware those were separate. If it is added to the blog post it may be worth clarifying the distinction because I could see others being confused as well.

Mark-Simulacrum · 2018-10-24T20:51:26Z

_posts/2018-10-25-Rust-1.30.md

+its rules felt awkward in practice. These changes are the first steps we're
+taking to make the module system feel more straightforward.
+
+[`mod.rs` files are now optional][optionalmod]. Imagine we had a `foo`


This is being dropped from the release (last minute bug found)

Mark-Simulacrum · 2018-10-24T20:52:34Z

_posts/2018-10-25-Rust-1.30.md

+
+// new
+let json = serde_json::from_str("...");
+```


This doesn't demonstrate use but the text indicates it should?

petrochenkov · 2018-10-24T20:58:25Z

_posts/2018-10-25-Rust-1.30.md

+
+There's two changes to `use` as well. Well, three changes: we already
+mentioned that you can use `use` to bring macros into scope in the macros
+section. There's two more. The first is that you can [`use` an extern crate


This not about use, this is about anything but use actually.
What happens is that extern crate names are put into prelude, so in relative paths they can be used from the whole crate without importing, similarly to e.g. Vec or Copy.
(Paths in use are absolute on 2015 edition, so they are not affected by this change.)

CryZe · 2018-10-24T20:58:35Z

I feel like the fact that libcore can be used on stable now deserves its own section (panic handling being unstable made it impossible to use stable for any actual libcore only binaries or libraries before). This is huge for embedded.

pietroalbini · 2018-10-24T21:36:14Z

cc @rust-lang/release

pietroalbini · 2018-10-24T21:49:46Z

_posts/2018-10-25-Rust-1.30.md

+The largest feature of Cargo in this release is that we now [have a progress
+bar!](https://github.com/rust-lang/cargo/pull/5995/)
+
+> GIF goes here


Just a reminder to put the gif here (or maybe asciinema?).

I think a GIF is nicer as it starts automatically, so it is more eye-catching.

carols10cents · 2018-10-24T23:53:22Z

I just pushed some changes without seeing the other comments-- I was confused by the use section as well and I tried to clarify but I'm not sure if I succeeded.

PLEASE check my work and back out anything that's wrong ❤️

dtolnay · 2018-10-25T01:41:40Z

As part of the procedural macro stabilizations, it would be good to call out that span information on errors in generated code are transparently getting way better. For example in Serde derives (notice where the span points to):

#[derive(Serialize)]
struct Demo {
    ok: String,
    bad: std::thread::Thread,
}

Before

error[E0277]: the trait bound `std::thread::Thread: _IMPL_SERIALIZE_FOR_Demo::_serde::Serialize` is not satisfied
 --> src/main.rs:3:10
  |
3 | #[derive(Serialize)]
  |          ^^^^^^^^^ the trait `_IMPL_SERIALIZE_FOR_Demo::_serde::Serialize` is not implemented for `std::thread::Thread`

After

error[E0277]: the trait bound `std::thread::Thread: serde::Serialize` is not satisfied
 --> src/main.rs:7:5
  |
7 |     bad: std::thread::Thread,
  |     ^^^ the trait `serde::Serialize` is not implemented for `std::thread::Thread`

The improvement comes automatically with no code changes for macros that are based on syn / quote / proc-macro2.

The libproc_macro stabilizations to enable this actually landed in 1.29 but @alexcrichton had wanted to delay drawing attention to it until the big bang stabilization of procedural macros in 1.30.

petrochenkov · 2018-10-25T09:28:04Z

_posts/2018-10-25-Rust-1.30.md

+
+See the [detailed release notes][notes] for more.
+
+### Library stabilizations


Perhaps it makes sense to mention proc macro API here (https://doc.rust-lang.org/stable/proc_macro/index.html).
It was secretly stabilized in 1.29 and then never announced officially.

petrochenkov · 2018-10-25T09:33:15Z

_posts/2018-10-25-Rust-1.30.md

-structs and enums; attributes can go on other places as well, like functions.
-As an example of using an attribute-like macro, you might have something like
-this when using a web application framework:
+Attribute-like macros are similar to custom derive macros, but instead of generating code


Wording nit: attribute macros are not "-like" attributes, they are attributes.

(While function-like macros are obviously not functions, so the "-like" suffix is justified there.)

petrochenkov · 2018-10-25T09:40:33Z

_posts/2018-10-25-Rust-1.30.md

-imports much less when moving code around.
+The old behavior of `use` paths always starting from the crate root still applies. But
+after making a one-time switch to the new style, these rules will apply uniformly to
+paths everywhere, and you'll need to tweak your imports much less when moving code around.


I don't know what this paragraph is talking about, the changes stabilized in 1.30 neither reduce "import tweaking when moving code around" in any way, nor "lead to a more straightforward understanding of how paths resolve".

All of those are supposed benefits of the 2018 edition module system, that's not stabilized yet and require explicit switching to 2018 edition.
Right now (1.30 / 2015 edition) crate:: in paths is supported to simplify migration to 2018 rather than anything else.

A bit more constructively:

Non-imports starting with a crate name require less tweaking when moving code around due to the "::serde_json::from_str -> serde_json::from_str" change.

If crate:: is used consistently in imports (the "still on 2015, but ready for 2018 migration" mode), then we have more straightforward import logic - they always start with a crate name or a special keyword.

i was purely talking about moving code within submodules, not moving to or from 2015/2018, like your first bullet point

steveklabnik · 2018-10-25T15:56:06Z

Thanks folks, I'm pretty sure I got everything here.

petrochenkov · 2018-10-25T16:22:26Z

_posts/2018-10-25-Rust-1.30.md

+```
+
+Moving a function to a submodule and having your imports break was not a great
+experience. Now, `use` will check the first part of the path and see if it's an `extern


"use will check" -> "relative paths will check"
"imports break" -> "non-imports break" (well, there's probably some better wording for "non-imports")

i tried a better wording, thank you :)

Rust 1.30

cfe470c

petrochenkov reviewed Oct 24, 2018

View reviewed changes

Mark-Simulacrum requested changes Oct 24, 2018

View reviewed changes

petrochenkov reviewed Oct 24, 2018

View reviewed changes

pietroalbini reviewed Oct 24, 2018

View reviewed changes

Clarifications, grammar, punctuation

c57ade2

petrochenkov reviewed Oct 25, 2018

View reviewed changes

ashleygwilliams and others added 2 commits October 25, 2018 11:00

feat(post): add cargo demo gif

2906893

review comments

7fdf631

petrochenkov reviewed Oct 25, 2018

View reviewed changes

Mark-Simulacrum approved these changes Oct 25, 2018

View reviewed changes

steveklabnik added 2 commits October 25, 2018 12:25

one more

f2b7528

use the full version

2680b32

Mark-Simulacrum merged commit c784c7d into gh-pages Oct 25, 2018

Mark-Simulacrum deleted the 1.30-announcement branch October 25, 2018 16:36

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Rust 1.30 #281

Rust 1.30 #281

steveklabnik commented Oct 24, 2018

petrochenkov Oct 24, 2018

petrochenkov Oct 24, 2018

jtgeibel Oct 25, 2018 •

edited

Loading

petrochenkov Oct 25, 2018

jtgeibel Oct 25, 2018

Mark-Simulacrum Oct 24, 2018

Mark-Simulacrum Oct 24, 2018

petrochenkov Oct 24, 2018 •

edited

Loading

CryZe commented Oct 24, 2018 •

edited

Loading

pietroalbini commented Oct 24, 2018

pietroalbini Oct 24, 2018

otavio Oct 25, 2018

carols10cents commented Oct 24, 2018

dtolnay commented Oct 25, 2018 •

edited

Loading

petrochenkov Oct 25, 2018

petrochenkov Oct 25, 2018

petrochenkov Oct 25, 2018 •

edited

Loading

petrochenkov Oct 25, 2018

steveklabnik Oct 25, 2018

steveklabnik commented Oct 25, 2018

petrochenkov Oct 25, 2018

steveklabnik Oct 25, 2018


		See the [detailed release notes][notes] for more.

		### Library stabilizations

Rust 1.30 #281

Rust 1.30 #281

Conversation

steveklabnik commented Oct 24, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jtgeibel Oct 25, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

petrochenkov Oct 24, 2018 • edited Loading

Choose a reason for hiding this comment

CryZe commented Oct 24, 2018 • edited Loading

pietroalbini commented Oct 24, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

carols10cents commented Oct 24, 2018

dtolnay commented Oct 25, 2018 • edited Loading

Before

After

Choose a reason for hiding this comment

Choose a reason for hiding this comment

petrochenkov Oct 25, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

steveklabnik commented Oct 25, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jtgeibel Oct 25, 2018 •

edited

Loading

petrochenkov Oct 24, 2018 •

edited

Loading

CryZe commented Oct 24, 2018 •

edited

Loading

dtolnay commented Oct 25, 2018 •

edited

Loading

petrochenkov Oct 25, 2018 •

edited

Loading