macros: hygienize use of `core`/`std` in builtin macros by jseyfried · Pull Request #46550 · rust-lang/rust

jseyfried · 2017-12-07T04:44:21Z

Today, if a builtin macro wants to access an item from core or std (depending #![no_std]), it generates ::core::path::to::item or ::std::path::to::item respectively (c.f. fn std_path() in libsyntax/ext/base.rs).

This PR refactors the builtin macros to instead always emit $crate::path::to::item here. That is, the def site of builtin macros is taken to be in extern crate core; or extern crate std;. Since builtin macros are macros 1.0 (i.e. mostly unhygienic), changing the def site can only effect the resolution of $crate.

r? @nrc

nrc · 2017-12-08T08:16:11Z

@bors: r+

bors · 2017-12-08T08:16:12Z

📌 Commit 6f8df8c has been approved by nrc

bors · 2017-12-08T16:54:51Z

⌛ Testing commit 6f8df8c with merge f40e75d38b2de81b9cf034d4f8c64b003ae79361...

bors · 2017-12-08T18:32:12Z

💔 Test failed - status-travis

jseyfried · 2017-12-09T04:57:45Z

@bors r=nrc

bors · 2017-12-09T04:57:46Z

📌 Commit d4b7c2c has been approved by nrc

bors · 2017-12-09T06:26:01Z

⌛ Testing commit d4b7c2c963266d22f6661fb0febceba20621788b with merge 504235de64dce04ff0d720bec0a16f2dc8bb55f8...

bors · 2017-12-09T08:08:09Z

💔 Test failed - status-travis

kennytm · 2017-12-09T13:59:31Z

The pretty tests produced identifier paths without its crate (::clone::Clone instead of core::clone::Clone), legit.

Details

...
[01:37:23] ---- [pretty] run-pass/uninit-empty-types.rs stdout ----
[01:37:23] [01:37:23] error: pretty-printed source (expanded) does not typecheck
[01:37:23] status: exit code: 101
[01:37:23] command: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2/bin/rustc" "-" "-Zno-trans" "--out-dir" "/checkout/obj/build/x86_64-unknown-linux-gnu/test/run-pass/uninit-empty-types.pretty-out" "--target=x86_64-unknown-linux-gnu" "-L" "/checkout/obj/build/x86_64-unknown-linux-gnu/test/run-pass" "-L" "/checkout/obj/build/x86_64-unknown-linux-gnu/test/run-pass/uninit-empty-types.stage2-x86_64-unknown-linux-gnu.pretty.aux" "-Crpath" "-O" "-Lnative=/checkout/obj/build/x86_64-unknown-linux-gnu/native/rust-test-helpers"
[01:37:23] stdout:
[01:37:23] ------------------------------------------
[01:37:23] [01:37:23] ------------------------------------------
[01:37:23] stderr:
[01:37:23] ------------------------------------------
[01:37:23] error[E0433]: failed to resolve. Did you mean `core::clone`?
[01:37:23] --> <anon>:26:8
[01:37:23] |
[01:37:23] 26 | impl ::clone::Clone for Foo {
[01:37:23] | ^^^^^ Did you mean `core::clone`?
[01:37:23] [01:37:23] warning: unused import: `std::prelude::v1::*`
[01:37:23] --> <anon>:4:5
[01:37:23] |
[01:37:23] 4 | use std::prelude::v1::*;
[01:37:23] | ^^^^^^^^^^^^^^^^^^^
[01:37:23] |
[01:37:23] = note: #[warn(unused_imports)] on by default
[01:37:23] [01:37:23] warning: unused `#[macro_use]` import
[01:37:23] --> <anon>:5:1
[01:37:23] |
[01:37:23] 5 | #[macro_use]
[01:37:23] | ^^^^^^^^^^^^
[01:37:23] [01:37:23] error: cannot continue compilation due to previous error
[01:37:23] [01:37:23] [01:37:23] ------------------------------------------
[01:37:23] [01:37:23] thread '[pretty] run-pass/uninit-empty-types.rs' panicked at 'explicit panic', /checkout/src/tools/compiletest/src/runtest.rs:2770:8
[01:37:23] [01:37:23] [01:37:23] failures:
[01:37:23] [pretty] run-pass/associated-types-binding-in-where-clause.rs
[01:37:23] [pretty] run-pass/deriving-clone-enum.rs
[01:37:23] [pretty] run-pass/deriving-clone-generic-enum.rs
[01:37:23] [pretty] run-pass/deriving-clone-generic-struct.rs
[01:37:23] [pretty] run-pass/deriving-clone-generic-tuple-struct.rs
[01:37:23] [pretty] run-pass/deriving-clone-struct.rs
[01:37:23] [pretty] run-pass/deriving-clone-tuple-struct.rs
[01:37:23] [pretty] run-pass/deriving-enum-single-variant.rs
[01:37:23] [pretty] run-pass/deriving-in-macro.rs
[01:37:23] [pretty] run-pass/deriving-meta-multiple.rs
[01:37:23] [pretty] run-pass/deriving-meta.rs
[01:37:23] [pretty] run-pass/deriving-via-extension-hash-struct.rs
[01:37:23] [pretty] run-pass/issue-14399.rs
[01:37:23] [pretty] run-pass/issue-15689-2.rs
[01:37:23] [pretty] run-pass/issue-19037.rs
[01:37:23] [pretty] run-pass/issue-21402.rs
[01:37:23] [pretty] run-pass/issue-6341.rs
[01:37:23] [pretty] run-pass/uninit-empty-types.rs
[01:37:23] [01:37:23] test result: �[31mFAILED�(B�[m. 2792 passed; 18 failed; 39 ignored; 0 measured; 0 filtered out

…in macros.

jseyfried · 2017-12-11T17:20:13Z

@bors r=nrc

bors · 2017-12-11T17:20:14Z

📌 Commit 3bade7f has been approved by nrc

arielb1 · 2017-12-11T19:10:10Z

This feels like it's ignoring a bit too many tests, which means that pretty might be very broken. If this is not fixed quickly it could remain broken for quite some time.

@bors r-

jseyfried · 2017-12-11T19:20:28Z

Ok, was going to write fix in separate PR -- I'll make sure to avoid regressing these tests then though.

jseyfried · 2017-12-11T19:27:48Z

Pretty is already "broken" for $crate:: paths in the sense that the output doesn't resolve; this PR just makes it worse by using more $crate:: paths.

I can fix the $crate:: case in most cases by using fn eliminate_crate_var (librustc_resolve/macros.rs) as we do for macros 1.1. However, there is still the issue that pretty doesn't play well with hygiene.

arielb1 · 2017-12-12T15:53:56Z

@jseyfried

Sure I know about that, I just don't want to let all the pretty tests bitrot before that gets fixed

jseyfried · 2017-12-13T08:03:02Z

I avoided regressing any pretty tests by printing $crate::.. paths from builtin macros as ::core::.. or ::std::.. where appropriate.

@bors r=nrc

bors · 2017-12-13T08:03:47Z

📌 Commit 85d19b3 has been approved by nrc

kennytm · 2017-12-13T10:02:53Z

@bors p=1 — increase priority to unblock #46551.

bors · 2017-12-13T11:10:02Z

⌛ Testing commit 85d19b3 with merge 3dfbc88...

@nrc

macros: hygienize use of `core`/`std` in builtin macros Today, if a builtin macro wants to access an item from `core` or `std` (depending `#![no_std]`), it generates `::core::path::to::item` or `::std::path::to::item` respectively (c.f. `fn std_path()` in `libsyntax/ext/base.rs`). This PR refactors the builtin macros to instead always emit `$crate::path::to::item` here. That is, the def site of builtin macros is taken to be in `extern crate core;` or `extern crate std;`. Since builtin macros are macros 1.0 (i.e. mostly unhygienic), changing the def site can only effect the resolution of `$crate`. r? @nrc

bors · 2017-12-13T13:43:26Z

☀️ Test successful - status-appveyor, status-travis
Approved by: nrc
Pushing 3dfbc88 to master...

rust-highfive assigned nrc Dec 7, 2017

jseyfried mentioned this pull request Dec 7, 2017
macros: improve 1.0/2.0 interaction #46551
Merged

kennytm added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Dec 7, 2017

jseyfried force-pushed the cleanup_builtin_hygiene branch from 6f8df8c to d4b7c2cCompare December 9, 2017 00:43

kennytm 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-review Status: Awaiting review from the assignee but also interested parties. labels Dec 9, 2017

Use hygiene to access the injected crate (core or std) from built…
9c7969d
…in macros.

jseyfried force-pushed the cleanup_builtin_hygiene branch from d4b7c2c to 3bade7fCompare December 11, 2017 06:09

kennytm 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 11, 2017

kennytm 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-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Dec 11, 2017

Fix fallout in tests.
8dbe79a

Refactor MarkData field modern: bool to kind: MarkKind.
f79c7e4

jseyfried force-pushed the cleanup_builtin_hygiene branch from 3bade7f to 57f373fCompare December 12, 2017 21:15

Improve pretty printing $crate:: paths.
85d19b3

jseyfried force-pushed the cleanup_builtin_hygiene branch from 57f373f to 85d19b3Compare December 13, 2017 06:32

rust-lang deleted a comment from bors Dec 13, 2017

kennytm 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 13, 2017

bors merged commit 85d19b3 into rust-lang:masterDec 13, 2017

bors mentioned this pull request Dec 13, 2017
incr.comp.: Speed up span hashing by caching expansion context hashes. #46562
Merged

jseyfried deleted the cleanup_builtin_hygiene branch December 13, 2017 18:58

xorl mentioned this pull request Dec 14, 2017
New rust core PR breaks rocket_codegen on nightly rwf2/Rocket#503
Closed

Uh oh!

Uh oh!

Conversation

jseyfried commented Dec 7, 2017

Uh oh!

nrc commented Dec 8, 2017

Uh oh!

bors commented Dec 8, 2017

Uh oh!

bors commented Dec 8, 2017

Uh oh!

bors commented Dec 8, 2017

Uh oh!

jseyfried commented Dec 9, 2017

Uh oh!

bors commented Dec 9, 2017

Uh oh!

bors commented Dec 9, 2017

Uh oh!

bors commented Dec 9, 2017

Uh oh!

kennytm commented Dec 9, 2017

Uh oh!

jseyfried commented Dec 11, 2017

Uh oh!

bors commented Dec 11, 2017

Uh oh!

arielb1 commented Dec 11, 2017• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jseyfried commented Dec 11, 2017• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jseyfried commented Dec 11, 2017• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

arielb1 commented Dec 12, 2017

Uh oh!

jseyfried commented Dec 13, 2017• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

bors commented Dec 13, 2017

Uh oh!

kennytm commented Dec 13, 2017

Uh oh!

bors commented Dec 13, 2017

Uh oh!

bors commented Dec 13, 2017

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

arielb1 commented Dec 11, 2017•
edited
Loading

jseyfried commented Dec 11, 2017•
edited
Loading

jseyfried commented Dec 11, 2017•
edited
Loading

jseyfried commented Dec 13, 2017•
edited
Loading