[rustdoc] Fix how `Deref` items is handled. by GuillaumeGomez · Pull Request #160915 · rust-lang/rust

GuillaumeGomez · 2026-08-11T14:37:27Z

This PR fixes a few things around how we handle Deref:

Since nothing except for methods is callable through a Deref, only methods should be kept
If the Deref::Target is a type implementing Copy, then methods taking self work and should be displayed.

During the clean pass, I store the item DefId on which Deref is implemented in a DefIdSet if the Deref::Target implements Copy. Then during rendering, we check if the "parent item" is in in the DefIdSet, and if so, we keep methods with self.

Now you might wonder why the item and not the derefed item. It's because the DefId we get from the computed type doesn't match the DefId of the actual type (that's where I spent most of my time, finding an ID (DefId/ItemId) I can use as key T_T).

It computes in html/render if the Deref::Target item is copy before rendering it.

So to resume:

&self is always kept.
&mut self is kept if DerefMut is implemented (nothing changed there).
self is kept only if Deref::Target is Copy.
Everything else disappears as they're not callable through Deref.

r? @camelid

rustbot · 2026-08-11T14:37:34Z

camelid is currently at their maximum review capacity.
They may take a while to respond.

GuillaumeGomez · 2026-08-11T14:38:24Z

Oof, that's a big diff. It's mostly because of reindent, not much that we can do about. ^^'

GuillaumeGomez · 2026-08-11T16:00:25Z

Fixed fmt. ^^'

camelid

Btw there's some missing wording in this sentence of the PR description, so I'm not quite sure what it says:

Since since except for methods is callable through a Deref, only methods should be kept

View changes since this review

GuillaumeGomez · 2026-08-11T19:16:27Z

Btw there's some missing wording in this sentence of the PR description, so I'm not quite sure what it says:

Since since except for methods is callable through a Deref, only methods should be kept

Fixed the typo. Correct sentence is:

Since nothing except for methods is callable through a Deref, only methods should be kept

GuillaumeGomez · 2026-08-11T23:15:04Z

Thanks to @camelid's suggestion, code is now much simpler. Thanks!

@rustbot ready

camelid · 2026-08-12T15:32:07Z

+    let for_ = clean_ty(impl_.self_ty, cx);
+


nit: Moving this let is now unnecessary. Better to move it back to where it's used.

View changes since the review

camelid · 2026-08-12T15:32:41Z

@@ -295,6 +295,8 @@ pub(crate) fn build_deref_target_impls(
                    inline::build_impls(cx, did, None, ret);
                });
            }
+        } else {
+            break;


Why was this break added?

View changes since the review

Because it's actually looking for the Deref::Target item and doesn't do anything. So at that point, we can break as soon as we found it. Can remove it though, considering there are only 2 items, doesn't matter much.

Also just realized that it shouldn't be in a else condition. Anyway, removing it.

camelid · 2026-08-12T15:34:26Z

@@ -3150,3 +3174,30 @@ fn repr_attribute<'tcx>(

    (!result.is_empty()).then(|| format!("#[repr({})]", result.join(", ")).into())
 }
+
+pub(crate) fn compute_if_deref_target_implements_copy(
+    cx: &Context<'_>,


Let's have this take TyCtxt instead of Context and change the callers as well. It's better to take as little input as possible.

View changes since the review

camelid · 2026-08-12T15:35:00Z

    AlsoCollectAssocFns { assoc_fns: &'r mut Vec<Link<'l>> },
 }

 fn get_methods<'a>(
    i: &'a clean::Impl,
    mut mode: GetMethodsMode<'_, 'a>,
    used_links: &mut FxHashSet<String>,
-    tcx: TyCtxt<'_>,
+    cx: &Context<'_>,


Ditto here.

View changes since the review

GuillaumeGomez · 2026-08-12T20:44:39Z

Applied suggestions.

@rustbot ready

GuillaumeGomez · 2026-08-24T15:13:30Z

ping @camelid ;)

camelid · 2026-09-02T16:41:00Z

-    } else {
-        false
-    }
+    (deref_mut_ || !by_mut_ref) && !by_box && (!by_value || target_is_copy)


Any idea why Box is special-cased here? Seems weird to special-case at all, and also I would expect needing special cases for things like Rc or Pin as well if Box truly needs it.

View changes since the review

I think it's because Box is a known type implementing Deref for the end type, common enough to make sense as an optimization? And should likely add the other types you mentioned. However in another PR might make more sense.

This causes a behavior difference though, right? It's not just a matter of optimization/performance.

Could you see what behavior changes if you remove the !by_box part of this expression?

Without by_box, when running the rustdoc-html tests:

failures: ---- [rustdoc-html] tests/rustdoc-html/deref-mut-35169-2.rs stdout ---- ------python3 stdout------------------------------ ------python3 stderr------------------------------ 38: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]//h4[@class="code-header"]' 'fn by_explicit_box(self: Box<Foo>)' 39: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]' 'fn by_explicit_box(self: Box<Foo>)' 40: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]//h4[@class="code-header"]' 'fn by_explicit_self_box(self: Box<Self>)' 41: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]' 'fn by_explicit_self_box(self: Box<Self>)' Encountered 4 errors ------------------------------------------ error: htmldocck failed! status: exit status: 1 command: "/usr/bin/python3" "/home/imperio/rust/rust/src/etc/htmldocck.py" "/home/imperio/rust/rust/build/x86_64-unknown-linux-gnu/test/rustdoc-html/deref-mut-35169-2" "/home/imperio/rust/rust/tests/rustdoc-html/deref-mut-35169-2.rs" stdout: none --- stderr ------------------------------- 38: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]//h4[@class="code-header"]' 'fn by_explicit_box(self: Box<Foo>)' 39: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]' 'fn by_explicit_box(self: Box<Foo>)' 40: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]//h4[@class="code-header"]' 'fn by_explicit_self_box(self: Box<Self>)' 41: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]' 'fn by_explicit_self_box(self: Box<Self>)' Encountered 4 errors ------------------------------------------ ---- [rustdoc-html] tests/rustdoc-html/deref-mut-35169-2.rs stdout end ---- ---- [rustdoc-html] tests/rustdoc-html/deref-mut-35169.rs stdout ---- ------python3 stdout------------------------------ ------python3 stderr------------------------------ 33: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]//h4[@class="code-header"]' 'fn by_explicit_box(self: Box<Foo>)' 34: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]' 'fn by_explicit_box(self: Box<Foo>)' 35: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]//h4[@class="code-header"]' 'fn by_explicit_self_box(self: Box<Self>)' 36: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]' 'fn by_explicit_self_box(self: Box<Self>)' Encountered 4 errors ------------------------------------------ error: htmldocck failed! status: exit status: 1 command: "/usr/bin/python3" "/home/imperio/rust/rust/src/etc/htmldocck.py" "/home/imperio/rust/rust/build/x86_64-unknown-linux-gnu/test/rustdoc-html/deref-mut-35169" "/home/imperio/rust/rust/tests/rustdoc-html/deref-mut-35169.rs" stdout: none --- stderr ------------------------------- 33: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]//h4[@class="code-header"]' 'fn by_explicit_box(self: Box<Foo>)' 34: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_box"]' 'fn by_explicit_box(self: Box<Foo>)' 35: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]//h4[@class="code-header"]' 'fn by_explicit_self_box(self: Box<Self>)' 36: !has check failed `XPATH PATTERN` unexpectedly matched //@ !has - '//*[@id="method.by_explicit_self_box"]' 'fn by_explicit_self_box(self: Box<Self>)' Encountered 4 errors ------------------------------------------ ---- [rustdoc-html] tests/rustdoc-html/deref-mut-35169.rs stdout end ---- failures: [rustdoc-html] tests/rustdoc-html/deref-mut-35169-2.rs [rustdoc-html] tests/rustdoc-html/deref-mut-35169.rs

I think we should leave it as is. ;)

Hmm, I still think this special-casing is a little sketchy but it's pre-existing so we should deal with it in another PR.

…uwer Rollup of 12 pull requests Successful merges: - #163531 (rustc_ast_lowering: track implicit Self via explicit flag instead of name) - #160915 ([rustdoc] Fix how `Deref` items is handled.) - #163364 (make semicolon_in_expressions_from_non_local_macros not report-in-deps) - #163770 (document the rustc_comptime attribute) - #163780 (Skip optional asserts in SsaRangePropagation) - #163785 (Bump Windows CI LLVM to 22.1.8) - #163796 (continue crashes tests `-Znext-solver` work) - #163646 (Get `inputs_hir` directly from `decl`) - #163733 (Fix extra spaces in integer format_into docs) - #163773 (Update GitHub Actions to v26) - #163787 (moves rustc_legacy_const_generics checks into attribute parsing) - #163810 (Fix an issue for clippy's `search_is_some` with the next-solver)

JonathanBrouwer · 2026-10-05T17:23:07Z

💔 I suspect this PR failed tests as part of a rollup
@bors r-

Failed during a local run

error[E0433]: cannot find module or crate `hir` in this scope
    --> src/librustdoc/html/render/mod.rs:3188:16
     |
3188 |         && let hir::ItemKind::Impl(impl_item) = item.kind
     |                ^^^ use of unresolved module or unlinked crate `hir`
     |
     = help: if you wanted to use a crate named `hir`, use `cargo add hir` to add it to your `Cargo.toml`
help: consider importing one of these enums
     |
  42 + use crate::ast::ItemKind;
     |
  42 + use crate::clean::ItemKind;
     |
  42 + use rustc_ast::ItemKind;
     |
  42 + use rustc_hir::ItemKind;
     |
     = and 1 other candidate
help: if you import `ItemKind`, refer to it directly
     |
3188 -         && let hir::ItemKind::Impl(impl_item) = item.kind
3188 +         && let ItemKind::Impl(impl_item) = item.kind
     |

I think you need to rebase

rust-bors · 2026-10-05T17:23:17Z

This pull request was unapproved.

This PR was contained in a rollup (#163827), which was unapproved.

View changes since this unapproval

…opy`

…t in `Cache`

rustbot · 2026-10-05T17:41:10Z

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

GuillaumeGomez · 2026-10-05T17:42:04Z

@JonathanBrouwer Yeah it needed a rebase.

@bors r=camelid

rust-bors · 2026-10-05T17:42:07Z

📌 Commit 32561a0 has been tentatively approved by camelid

It will be put into the queue for this repository once PR CI succeeds.

@camelid

[rustdoc] Fix how `Deref` items is handled. Fixes rust-lang#160236. This PR fixes a few things around how we handle `Deref`: * Since nothing except for methods is callable through a `Deref`, only methods should be kept * If the `Deref::Target` is a type implementing `Copy`, then methods taking `self` work and should be displayed. <del>During the `clean` pass, I store the item `DefId` on which `Deref` is implemented in a `DefIdSet` if the `Deref::Target` implements `Copy`. Then during rendering, we check if the "parent item" is in in the `DefIdSet`, and if so, we keep methods with `self`.</del> <del>Now you might wonder why the item and not the derefed item. It's because the `DefId` we get from the computed type doesn't match the `DefId` of the actual type (that's where I spent most of my time, finding an ID (`DefId`/`ItemId`) I can use as key T_T).</del> It computes in `html/render` if the `Deref::Target` item is copy before rendering it. So to resume: * `&self` is always kept. * `&mut self` is kept if `DerefMut` is implemented (nothing changed there). * `self` is kept only if `Deref::Target` is `Copy`. * Everything else disappears as they're not callable through `Deref`. r? @camelid

…uwer Rollup of 4 pull requests Successful merges: - #160915 ([rustdoc] Fix how `Deref` items is handled.) - #163267 (fix `VisibleForLeakCheck` in `RegionOutlives` fast path) - #163673 (Move #[doc(inline)] and #[doc(no_inline)] validation into rustc_attr_parsing) - #163814 (Add some comments and docs related to `AllocatorNightly`)

…uwer Rollup of 5 pull requests Successful merges: - #160915 ([rustdoc] Fix how `Deref` items is handled.) - #163267 (fix `VisibleForLeakCheck` in `RegionOutlives` fast path) - #163673 (Move #[doc(inline)] and #[doc(no_inline)] validation into rustc_attr_parsing) - #163795 (When documenting without a specified `--edition`, emit a message) - #163814 (Add some comments and docs related to `AllocatorNightly`)

@camelid

Rollup merge of #160915 - GuillaumeGomez:deref-items, r=camelid [rustdoc] Fix how `Deref` items is handled. Fixes #160236. This PR fixes a few things around how we handle `Deref`: * Since nothing except for methods is callable through a `Deref`, only methods should be kept * If the `Deref::Target` is a type implementing `Copy`, then methods taking `self` work and should be displayed. <del>During the `clean` pass, I store the item `DefId` on which `Deref` is implemented in a `DefIdSet` if the `Deref::Target` implements `Copy`. Then during rendering, we check if the "parent item" is in in the `DefIdSet`, and if so, we keep methods with `self`.</del> <del>Now you might wonder why the item and not the derefed item. It's because the `DefId` we get from the computed type doesn't match the `DefId` of the actual type (that's where I spent most of my time, finding an ID (`DefId`/`ItemId`) I can use as key T_T).</del> It computes in `html/render` if the `Deref::Target` item is copy before rendering it. So to resume: * `&self` is always kept. * `&mut self` is kept if `DerefMut` is implemented (nothing changed there). * `self` is kept only if `Deref::Target` is `Copy`. * Everything else disappears as they're not callable through `Deref`. r? @camelid

rust-timer · 2026-10-06T06:17:14Z

Note

This PR was benchmarked as part of triage of its containing rollup: triage URL.

Finished benchmarking commit (9d02933): comparison URL.

Overall result: ❌ regressions - please read:

Our benchmarks found a performance regression caused by this PR.
This might be an actual regression, but it can also be just noise.

Next Steps:

If the regression was expected or you think it can be justified,
please write a comment with sufficient written justification, and add
@rustbot label: +perf-regression-triaged to it, to mark the regression as triaged.
If you think that you know of a way to resolve the regression, try to create
a new PR with a fix for the regression.
If you do not understand the regression or you think that it is just noise,
you can ask the @rust-lang/wg-compiler-performance working group for help (members of this group
were already notified of this PR).

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	0.3%	[0.3%, 0.3%]	1
Regressions ❌ (secondary)	1.2%	[1.2%, 1.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.3%	[0.3%, 0.3%]	1

Max RSS (memory usage)

This perf run didn't have relevant results for this metric.

Cycles

Results (secondary 6.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	6.5%	[6.5%, 6.5%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

This perf run didn't have relevant results for this metric.

Artifact size: 408.58 MiB -> 408.75 MiB (0.04%)

rustbot assigned camelid Aug 11, 2026

This comment has been minimized.

Sign in to view

GuillaumeGomez force-pushed the deref-items branch from 32c5654 to e3fb18c Compare August 11, 2026 14:58

camelid requested changes Aug 11, 2026

View reviewed changes

Comment thread src/librustdoc/clean/inline.rs Outdated

Comment thread src/librustdoc/clean/utils.rs Outdated

Comment thread tests/rustdoc-html/deref/deref-to-primitive.rs

rustbot 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 Aug 11, 2026

GuillaumeGomez removed the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Aug 11, 2026

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Aug 11, 2026

camelid reviewed Aug 12, 2026

View reviewed changes

camelid 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 Aug 12, 2026

GuillaumeGomez force-pushed the deref-items branch from ca399ba to 7a71fdf Compare August 12, 2026 20:43

This comment has been minimized.

Sign in to view

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 Aug 12, 2026

camelid reviewed Sep 2, 2026

View reviewed changes

camelid 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 Sep 2, 2026

GuillaumeGomez force-pushed the deref-items branch from 7a71fdf to 6eaa385 Compare September 11, 2026 13:02

This comment has been minimized.

Sign in to view

JonathanBrouwer mentioned this pull request Oct 5, 2026

Rollup of 12 pull requests #163827

Closed

rust-bors Bot 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 Oct 5, 2026

GuillaumeGomez added 5 commits October 5, 2026 19:25

rustdoc: Only show useable impl items through Deref

8333607

Correctly show methods taking self if the target type implements `C…

d4e1f42

…opy`

Compute "is copy" information directly in render instead of storing i…

bc54245

…t in `Cache`

Improve code

b0d1567

Fix import errors

32561a0

GuillaumeGomez force-pushed the deref-items branch from 5463f66 to 32561a0 Compare October 5, 2026 17:41

rust-bors Bot 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 Oct 5, 2026

JonathanBrouwer mentioned this pull request Oct 5, 2026

Rollup of 4 pull requests #163838

Closed

JonathanBrouwer mentioned this pull request Oct 5, 2026

Rollup of 5 pull requests #163842

Merged

rust-bors Bot merged commit 3c67d2d into rust-lang:main Oct 6, 2026
14 checks passed

rustbot added this to the 1.101.0 milestone Oct 6, 2026

rustbot added the perf-regression Performance regression. label Oct 6, 2026

GuillaumeGomez deleted the deref-items branch October 6, 2026 09:24

Conversation

GuillaumeGomez commented Aug 11, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Aug 11, 2026

Uh oh!

GuillaumeGomez commented Aug 11, 2026

Uh oh!

This comment has been minimized.

GuillaumeGomez commented Aug 11, 2026

Uh oh!

camelid left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

GuillaumeGomez commented Aug 11, 2026

Uh oh!

GuillaumeGomez commented Aug 11, 2026

Uh oh!

camelid Aug 12, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

camelid Aug 12, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

camelid Aug 12, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

camelid Aug 12, 2026 • edited by rustbot 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.

GuillaumeGomez commented Aug 12, 2026

Uh oh!

GuillaumeGomez commented Aug 24, 2026

Uh oh!

camelid Sep 2, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

This comment has been minimized.

JonathanBrouwer commented Oct 5, 2026

Uh oh!

rust-bors Bot commented Oct 5, 2026

Uh oh!

rustbot commented Oct 5, 2026

Uh oh!

GuillaumeGomez commented Oct 5, 2026

Uh oh!

rust-bors Bot commented Oct 5, 2026

Uh oh!

Uh oh!

rust-timer commented Oct 6, 2026

Overall result: ❌ regressions - please read:

Uh oh!

Reviewers

Assignees

GuillaumeGomez commented Aug 11, 2026 •

edited by rustbot

Loading

camelid left a comment •

edited by rustbot

Loading

camelid Aug 12, 2026 •

edited by rustbot

Loading

camelid Aug 12, 2026 •

edited by rustbot

Loading

camelid Aug 12, 2026 •

edited by rustbot

Loading

camelid Aug 12, 2026 •

edited by rustbot

Loading

camelid Sep 2, 2026 •

edited by rustbot

Loading