rustc: factor out most of hir::def::Def's variants into DefKind, and rename to Res. #60462

eddyb · 2019-05-01T23:31:40Z

The first two commits are about introducing DefKind, both to simplify/orthogonalize hir::def::Def, and to allow reasoning about the kind of a definition without dealing with the redundant DefId.
(There are likely more changes to be made, such as adding enough DefKind variants for tcx.def_kind(def_id) to return just DefKind, not Option<DefKind>, but this is pretty big as-is)

The third commit frees up the Def name (which we may want to use in the future for "definitions", in the sense of "entities with a DefId") by renaming hir::def::Def to Res ("resolution").
IMO this fits, as it represents all the possible name resolution results, not just "definitions (with a DefId)".

Quick examples:

// Before:
if tcx.describe_def(def_id) == Some(Def::Struct(def_id)) {...}
if let Def::Struct(def_id) = path.def {...}

// After:
if tcx.def_kind(def_id) == Some(DefKind::Struct) {...}
if let Res::Def(DefKind::Struct, def_id) = path.res {...}

r? @petrochenkov cc @rust-lang/compiler

src/librustc/hir/def.rs

src/librustc/hir/mod.rs

bors · 2019-05-02T05:02:16Z

☔ The latest upstream changes (presumably #60460) made this pull request unmergeable. Please resolve the merge conflicts.

oli-obk · 2019-05-02T06:56:59Z

src/librustc_resolve/lib.rs

@@ -1104,7 +1124,7 @@ impl ModuleOrUniformRoot<'_> {
    fn same_def(lhs: Self, rhs: Self) -> bool {
        match (lhs, rhs) {
            (ModuleOrUniformRoot::Module(lhs),
-             ModuleOrUniformRoot::Module(rhs)) => lhs.def() == rhs.def(),
+             ModuleOrUniformRoot::Module(rhs)) => lhs.def_id() == rhs.def_id(),


just to confirm, this change is ok, because the DefKind would be the same if the DefId is the same?

Yes, DefKinds must be the same for same DefIds.

Especially since this is a Module, so a mod/enum/trait, which don't even have the potential ambiguity that constructors can.

petrochenkov · 2019-05-02T17:30:33Z

src/librustc/hir/map/mod.rs

            Node::Ctor(variant_data) => {
+                // FIXME(eddyb) is this even possible, if we have a `Node::Ctor`?
+                if variant_data.ctor_hir_id().is_none() {


Insert an assert to find out?

petrochenkov · 2019-05-02T17:49:11Z

src/librustc/hir/map/mod.rs

@@ -310,14 +310,14 @@ impl<'hir> Map<'hir> {
        self.definitions.as_local_node_id(def_id.to_def_id()).unwrap()
    }

-    pub fn describe_def(&self, node_id: NodeId) -> Option<Def> {
+    fn def_kind(&self, node_id: NodeId) -> Option<DefKind> {


I was surprised that this worked and users of describe_def never needed the DefId-less variants, but if it's called only from a DefId-based query, then it's understandable.

petrochenkov · 2019-05-02T18:19:32Z

r=me after rebase
@bors p=1

eddyb · 2019-05-03T16:05:38Z

@bors r=petrochenkov

bors · 2019-05-03T16:05:39Z

📌 Commit 393c8bc24885671ebde87a74ca8d651a79f9bd23 has been approved by petrochenkov

bors · 2019-05-03T18:19:00Z

☔ The latest upstream changes (presumably #60510) made this pull request unmergeable. Please resolve the merge conflicts.

…ind.

eddyb · 2019-05-03T19:51:52Z

@bors r=petrochenkov

bors · 2019-05-03T19:51:53Z

📌 Commit ff174fe has been approved by petrochenkov

eddyb · 2019-05-03T19:53:44Z

@bors p=100 (this bitrots almost instantly, should be merged before the next rollup)

bors · 2019-05-03T23:11:24Z

⌛ Testing commit ff174fe with merge 13fde05...

@petrochenkov

rustc: factor out most of hir::def::Def's variants into DefKind, and rename to Res. The first two commits are about introducing `DefKind`, both to simplify/orthogonalize `hir::def::Def`, and to allow reasoning about the kind of a definition without dealing with the redundant `DefId`. (There are likely more changes to be made, such as adding enough `DefKind` variants for `tcx.def_kind(def_id)` to return just `DefKind`, not `Option<DefKind>`, but this is pretty big as-is) The third commit frees up the `Def` name (which we may want to use in the future for "definitions", in the sense of "entities with a `DefId`") by renaming `hir::def::Def` to `Res` ("resolution"). IMO this fits, as it represents all the possible name resolution results, not just "definitions (with a `DefId`)". Quick examples: ```rust // Before: if tcx.describe_def(def_id) == Some(Def::Struct(def_id)) {...} if let Def::Struct(def_id) = path.def {...} ``` ```rust // After: if tcx.def_kind(def_id) == Some(DefKind::Struct) {...} if let Res::Def(DefKind::Struct, def_id) = path.res {...} ``` r? @petrochenkov cc @rust-lang/compiler

bors · 2019-05-04T01:59:08Z

☀️ Test successful - checks-travis, status-appveyor
Approved by: petrochenkov
Pushing 13fde05 to master...

rust-highfive · 2019-05-04T02:00:18Z

📣 Toolstate changed by #60462!

Tested on commit 13fde05.
Direct link to PR: #60462

💔 clippy-driver on windows: test-pass → build-fail (cc @Manishearth @llogiq @mcarton @oli-obk @phansch, @rust-lang/infra).
💔 clippy-driver on linux: test-pass → build-fail (cc @Manishearth @llogiq @mcarton @oli-obk @phansch, @rust-lang/infra).
💔 miri on windows: test-pass → build-fail (cc @oli-obk @RalfJung @eddyb, @rust-lang/infra).
💔 miri on linux: test-pass → build-fail (cc @oli-obk @RalfJung @eddyb, @rust-lang/infra).
💔 rls on windows: test-fail → build-fail (cc @Xanewok, @rust-lang/infra).
💔 rls on linux: test-fail → build-fail (cc @Xanewok, @rust-lang/infra).

@Manishearth

Tested on commit rust-lang/rust@13fde05. Direct link to PR: <rust-lang/rust#60462> 💔 clippy-driver on windows: test-pass → build-fail (cc @Manishearth @llogiq @mcarton @oli-obk @phansch, @rust-lang/infra). 💔 clippy-driver on linux: test-pass → build-fail (cc @Manishearth @llogiq @mcarton @oli-obk @phansch, @rust-lang/infra). 💔 miri on windows: test-pass → build-fail (cc @oli-obk @RalfJung @eddyb, @rust-lang/infra). 💔 miri on linux: test-pass → build-fail (cc @oli-obk @RalfJung @eddyb, @rust-lang/infra). 💔 rls on windows: test-fail → build-fail (cc @Xanewok, @rust-lang/infra). 💔 rls on linux: test-fail → build-fail (cc @Xanewok, @rust-lang/infra).

Rustup to rustc 1.36.0-nightly (13fde05b1 2019-05-03) Trying to deal with changes from rust-lang/rust#60462 Moved from #4060 so everyone can collab on the rustup branch.

@petrochenkov

rustc: collapse relevant DefPathData variants into {Type,Value,Macro,Lifetime}Ns. `DefPathData` was meant to disambiguate within each namespace, but over the years, that purpose was overlooked, and it started to serve a double-role as a sort of `DefKind` (which #60462 properly adds). Now, we can go back to *only* categorizing namespaces (at least for the variants with names in them). r? @petrochenkov or @nikomatsakis cc @michaelwoerister

@eddyb

Rename `PathResolution` to `PartialRes` Don't use `PartialRes` when `Res` is enough. Rename `Res::kind_name` to `Res::descr` for consistency. Remove `Res::Label`, paths can never resolve to labels. Some further cleanup after #60462 r? @eddyb

rust-highfive assigned petrochenkov May 1, 2019

rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label May 1, 2019

Centril reviewed May 2, 2019

View reviewed changes

src/librustc/hir/def.rs Show resolved Hide resolved

src/librustc/hir/mod.rs Show resolved Hide resolved

eddyb mentioned this pull request May 2, 2019

RFC: Rust Symbol Mangling (v0) rust-lang/rfcs#2603

Merged

oli-obk reviewed May 2, 2019

View reviewed changes

oli-obk approved these changes May 2, 2019

View reviewed changes

petrochenkov reviewed May 2, 2019

View reviewed changes

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-review Status: Awaiting review from the assignee but also interested parties. labels May 2, 2019

eddyb force-pushed the def-1-a-mere-resolution branch from c4c23c9 to 393c8bc Compare May 3, 2019 16:02

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 May 3, 2019

bors 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 May 3, 2019

eddyb added 3 commits May 3, 2019 22:48

rustc: factor most DefId-containing variants out of Def and into DefK…

a3fcab3

…ind.

rustc: use DefKind instead of Def, where possible.

b92b1a7

rustc: rename hir::def::Def to Res (short for "resolution").

ff174fe

eddyb force-pushed the def-1-a-mere-resolution branch from 393c8bc to ff174fe Compare May 3, 2019 19:50

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 May 3, 2019

eddyb mentioned this pull request May 3, 2019

rustc: collapse relevant DefPathData variants into {Type,Value,Macro,Lifetime}Ns. #60525

Merged

matthiaskrgr added a commit to matthiaskrgr/rust-clippy that referenced this pull request May 4, 2019

start attempt to rustup to rust-lang/rust#60462

bb4440c

matthiaskrgr mentioned this pull request May 4, 2019

start attempt to rustup to https://github.com/rust-lang/rust/pull/60462 rust-lang/rust-clippy#4060

Closed

bors added the merged-by-bors This PR was explicitly merged by bors. label May 4, 2019

bors merged commit ff174fe into rust-lang:master May 4, 2019

This was referenced May 4, 2019

miri no longer builds after rust-lang/rust#60462 #60533

Closed

rls no longer builds after rust-lang/rust#60462 #60534

Closed

eddyb deleted the def-1-a-mere-resolution branch May 4, 2019 02:09

This was referenced May 4, 2019

Ban multi-trait objects via trait aliases #59445

Merged

[let_chains, 1/6] Remove hir::ExprKind::If #59288

Merged

Add outlives suggestions for some lifetime errors #58281

Merged

Implementation of RFC 2289 (associated_type_bounds) #57428

Merged

Manishearth mentioned this pull request May 4, 2019

Rustup to rustc 1.36.0-nightly (13fde05b1 2019-05-03) rust-lang/rust-clippy#4061

Merged

petrochenkov mentioned this pull request May 4, 2019

Rename PathResolution to PartialRes #60544

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

rustc: factor out most of hir::def::Def's variants into DefKind, and rename to Res. #60462

rustc: factor out most of hir::def::Def's variants into DefKind, and rename to Res. #60462

eddyb commented May 1, 2019

bors commented May 2, 2019

oli-obk May 2, 2019

petrochenkov May 2, 2019

eddyb May 3, 2019

petrochenkov May 2, 2019

petrochenkov May 2, 2019

petrochenkov commented May 2, 2019

eddyb commented May 3, 2019

bors commented May 3, 2019

bors commented May 3, 2019

eddyb commented May 3, 2019

bors commented May 3, 2019

eddyb commented May 3, 2019

bors commented May 3, 2019

bors commented May 4, 2019

rust-highfive commented May 4, 2019

rustc: factor out most of hir::def::Def's variants into DefKind, and rename to Res. #60462

rustc: factor out most of hir::def::Def's variants into DefKind, and rename to Res. #60462

Conversation

eddyb commented May 1, 2019

bors commented May 2, 2019

oli-obk May 2, 2019

Choose a reason for hiding this comment

petrochenkov May 2, 2019

Choose a reason for hiding this comment

eddyb May 3, 2019

Choose a reason for hiding this comment

petrochenkov May 2, 2019

Choose a reason for hiding this comment

petrochenkov May 2, 2019

Choose a reason for hiding this comment

petrochenkov commented May 2, 2019

eddyb commented May 3, 2019

bors commented May 3, 2019

bors commented May 3, 2019

eddyb commented May 3, 2019

bors commented May 3, 2019

eddyb commented May 3, 2019

bors commented May 3, 2019

bors commented May 4, 2019

rust-highfive commented May 4, 2019