Rewrite `inline` attribute parser to use new infrastructure and improve diagnostics for all parsed attributes #138165

jdonszelmann · 2025-03-07T16:07:27Z

This PR:

creates a new parser for inline attributes
creates consistent error messages and error codes between attribute parsers; inline and others
as such changes a few error messages for other attributes to be (in my eyes) much more consistent
tests ast-lowering lints introduced by Infrastructure for lints during attribute parsing, specifically duplicate usages of attributes #138164 since this is now useful for the first time
Coalesce some useless error codes

Builds on top of #138164

jdonszelmann · 2025-03-08T23:37:17Z

compiler/rustc_codegen_ssa/src/codegen_attrs.rs

@@ -584,7 +546,6 @@ fn codegen_fn_attrs(tcx: TyCtxt<'_>, did: LocalDefId) -> CodegenFnAttrs {
            return OptimizeAttr::Default;
        };

-        inline_span = Some(attr.span());


This fixes #137950

bors · 2025-03-11T05:03:38Z

☔ The latest upstream changes (presumably #138302) made this pull request unmergeable. Please resolve the merge conflicts.

jdonszelmann · 2025-06-09T22:54:26Z

@rustbot review

jdonszelmann · 2025-06-09T23:21:16Z

ah shit, broke clippy 😭

bors · 2025-06-13T08:40:56Z

☔ The latest upstream changes (presumably #142432) made this pull request unmergeable. Please resolve the merge conflicts.

compiler/rustc_attr_data_structures/src/attributes.rs

compiler/rustc_attr_parsing/src/context.rs

oli-obk · 2025-06-16T10:23:05Z

tests/rustdoc-json/attrs/inline.rs

-//@ is "$.index[?(@.name=='just_inline')].attrs" '["#[inline]"]'
+//@ is "$.index[?(@.name=='just_inline')].attrs" '["#[attr = Inline(Hint)]"]'


So to understand https://github.com/rust-lang/rust/pull/138165/files?diff=unified&w=1#r2143839007 right, we could represent this attrs list in json not as a list of strings, but as a list of either strings (unstable format, only print or view it) or json maps that contain all the information in a structured way and doesn't require consumers to re-parse it.

If we change the map representation, we bump the format. If we change the string representation, we do not.

Either way, we either break csc or we bump the version so csc can gate its parser on it. The only question we're asking right now is whether we bump the version twice (once now, and once for when we move to json maps).

is that about right @aDotInTheVoid @obi1kenobi ?

From a c-s-c perspective I don't have a strong opinion, we don't check #[inline] and the change shouldn't break us.

From a rustdoc JSON format perspective, I think it's kinder to other rustdoc JSON users if we bump.

@oli-obk if we bump, we should also decide if we bump from now on on every reworked attribute parser.

@obi1kenobi and I talked synchronously for a bit, and for now we both think it's best to do version bumps for a bit until we can implement a more structured approach. Pushed a commit that does the bump.

rustbot · 2025-06-16T17:16:46Z

rustdoc-json-types is a public (although nightly-only) API. If possible, consider changing src/librustdoc/json/conversions.rs; otherwise, make sure you bump the FORMAT_VERSION constant.

cc @CraftSpider, @aDotInTheVoid, @Enselic, @obi1kenobi

jdonszelmann · 2025-06-16T17:19:37Z

I think that addresses all the review comments by others @oli-obk :)

bors · 2025-06-16T21:55:45Z

☔ The latest upstream changes (presumably #142589) made this pull request unmergeable. Please resolve the merge conflicts.

aDotInTheVoid · 2025-06-16T23:30:17Z

src/rustdoc-json-types/lib.rs

@@ -30,7 +30,7 @@ pub type FxHashMap<K, V> = HashMap<K, V>; // re-export for use in src/librustdoc
 /// This integer is incremented with every breaking change to the API,
 /// and is returned along with the JSON blob as [`Crate::format_version`].
 /// Consuming code should assert that this value matches the format version(s) that it supports.
-pub const FORMAT_VERSION: u32 = 46;
+pub const FORMAT_VERSION: u32 = 47;


If this PR isn't bitrotty or urgent, it'd be really nice if could land after #142601 (and a rebase), so it's guaranteed not to race with #142502 to do the same bump. (#142601 fixes this by making this a conflict, so we'll never need to manually worry about this again 🤞)

this one seems to have few conflicts, will preemptively bump to 48

rustbot assigned oli-obk Mar 7, 2025

rustbot added A-attributes Area: Attributes (`#[…]`, `#![…]`) S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Mar 7, 2025

This comment has been minimized.

Sign in to view

jdonszelmann force-pushed the inline branch from f2f2286 to 93a2322 Compare March 8, 2025 19:05

This comment has been minimized.

Sign in to view

jdonszelmann commented Mar 8, 2025

View reviewed changes

jdonszelmann force-pushed the inline branch 2 times, most recently from e6dad01 to 2a707bf Compare March 9, 2025 15:34

This comment has been minimized.

Sign in to view

jdonszelmann force-pushed the inline branch from 2a707bf to 7787c54 Compare March 9, 2025 16:09

This comment has been minimized.

Sign in to view

jdonszelmann force-pushed the inline branch from 7787c54 to 251ad20 Compare March 9, 2025 20:10

jdonszelmann mentioned this pull request Mar 9, 2025

rewrite optimize attribute to use new attribute parsing infrastructure #138291

Draft

oli-obk 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 Apr 28, 2025

jdonszelmann force-pushed the inline branch 2 times, most recently from b7dda57 to 88a713a Compare June 9, 2025 22:40

This comment has been minimized.

Sign in to view

jdonszelmann mentioned this pull request Jun 9, 2025

Infrastructure for lints during attribute parsing, specifically duplicate usages of attributes #138164

Merged

jdonszelmann force-pushed the inline branch from 88a713a to e44a373 Compare June 9, 2025 22:50

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jun 9, 2025

This comment has been minimized.

Sign in to view

jdonszelmann mentioned this pull request Jun 9, 2025

Exhaustively handle parsed attributes in CheckAttr #140372

Merged

jdonszelmann force-pushed the inline branch from e44a373 to 2751f40 Compare June 10, 2025 06:53

jdonszelmann force-pushed the inline branch from 36c5c90 to 6745f0a Compare June 12, 2025 23:49

jdonszelmann force-pushed the inline branch from 6745f0a to f50c732 Compare June 13, 2025 09:49

This was referenced Jun 13, 2025

Tracking issue: Attribute refactor #131229

Open

Rework #[cold] attribute parser #142491

Draft

rework #[naked] attribute parser #142493

Draft

Fix missing docs in rustc_attr_parsing #142494

Draft

Better template for #[repr] attributes #142495

Draft

mejrs approved these changes Jun 14, 2025

View reviewed changes

jdonszelmann mentioned this pull request Jun 15, 2025

use #[align] attribute for fn_align #142507

Open

oli-obk reviewed Jun 16, 2025

View reviewed changes

jdonszelmann force-pushed the inline branch from f50c732 to 1e8aac1 Compare June 16, 2025 15:14

This comment has been minimized.

Sign in to view

jdonszelmann force-pushed the inline branch 2 times, most recently from 220615f to 43dd618 Compare June 16, 2025 17:16

aDotInTheVoid reviewed Jun 16, 2025

View reviewed changes

jdonszelmann and others added 5 commits June 17, 2025 01:33

implement inline parser

c6e0b18

implement rustc_force_inline parser

50b1e53

convert entire codebase to parsed inline attrs

e58beaa

fix bugs in inline/force_inline and diagnostics of all attr parsers

71cf44f

use consistent attr errors in all attribute parsers

14d0c1d

jdonszelmann force-pushed the inline branch from 43dd618 to aa6facc Compare June 16, 2025 23:35

This comment has been minimized.

Sign in to view

jdonszelmann added 3 commits June 17, 2025 01:54

make error codes reflect reality better

16d9c6b

fix clippy

8e6771d

bump rustdoc json version for inline attribute representation

c7ad3d7

jdonszelmann force-pushed the inline branch from aa6facc to c7ad3d7 Compare June 16, 2025 23:54

		//@ is "$.index[?(@.name=='just_inline')].attrs" '["#[inline]"]'
		//@ is "$.index[?(@.name=='just_inline')].attrs" '["#[attr = Inline(Hint)]"]'

Rewrite inline attribute parser to use new infrastructure and improve diagnostics for all parsed attributes #138165

Are you sure you want to change the base?

Rewrite inline attribute parser to use new infrastructure and improve diagnostics for all parsed attributes #138165

Conversation

jdonszelmann commented Mar 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment has been minimized.

This comment has been minimized.

jdonszelmann Mar 8, 2025

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

This comment has been minimized.

bors commented Mar 11, 2025

Uh oh!

This comment has been minimized.

jdonszelmann commented Jun 9, 2025

Uh oh!

This comment has been minimized.

jdonszelmann commented Jun 9, 2025

Uh oh!

bors commented Jun 13, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

oli-obk Jun 16, 2025

Choose a reason for hiding this comment

Uh oh!

obi1kenobi Jun 16, 2025

Choose a reason for hiding this comment

Uh oh!

jdonszelmann Jun 16, 2025

Choose a reason for hiding this comment

Uh oh!

jdonszelmann Jun 16, 2025

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

rustbot commented Jun 16, 2025

Uh oh!

jdonszelmann commented Jun 16, 2025

Uh oh!

bors commented Jun 16, 2025

Uh oh!

aDotInTheVoid Jun 16, 2025

Choose a reason for hiding this comment

Uh oh!

jdonszelmann Jun 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

Uh oh!

Rewrite `inline` attribute parser to use new infrastructure and improve diagnostics for all parsed attributes #138165

Rewrite `inline` attribute parser to use new infrastructure and improve diagnostics for all parsed attributes #138165

jdonszelmann commented Mar 7, 2025 •

edited

Loading

jdonszelmann Jun 16, 2025 •

edited

Loading