rustdoc: inline all the impls by mitaa · Pull Request #33133 · rust-lang/rust

mitaa · 2016-04-21T16:21:30Z

This used to be done to avoid inlining impls referencing private items, but is now unnecessary since we actually check that impls do not reference non-doc-reachable items.

fixes#32881
fixes#33025
fixes#33113

r? @alexcrichton

petrochenkov · 2016-04-21T16:35:05Z

we actually check that impls do not reference non-doc-reachable items

I'm curious, is this done with a separate check somewhere in rustdoc?
Because exactly this information is gathered in rustc_privacy and returned by access_levels.is_doc_reachable(impl_id)

mitaa · 2016-04-21T16:59:28Z

That was introduced in #33002. AFAIU the access_levels gathered in rustc_privacy are only for the local crate - so rustdoc tries to also probe the crate metadata in a similar way (i.e. also takes doc(hidden) into account) to librustc_privacy::EmbargoVisitor and adds that to access_levels.

edit: What I meant with

we actually check that impls do not reference non-doc-reachable items

was that the impl trait and type are checked, nothing more.

alexcrichton · 2016-04-22T20:30:23Z

@bors: r+ 8c302a00223abb46a2f42e5331fc27d1b1d35e1a

Nice!

bors · 2016-04-24T00:03:10Z

⌛ Testing commit 8c302a0 with merge 326dd17...

bors · 2016-04-24T01:02:05Z

💔 Test failed - auto-win-gnu-32-opt-rustbuild

This used to be done to avoid inlining impls referencing private items, but is now unnecessary since we actually check that impls do not reference non-doc-reachable items.

mitaa · 2016-04-24T12:38:38Z

(updated)

alexcrichton · 2016-04-24T17:16:55Z

Looks like there's still an error on Travis?

failures:
---- [rustdoc] rustdoc/inline_local/issue-32343.rs stdout ----
error: htmldocck failed!
status: exit code: 1
command: "/usr/bin/python2.7" "/home/travis/build/rust-lang/rust/src/etc/htmldocck.py" "x86_64-unknown-linux-gnu/test/rustdoc/inline_local/issue-32343.stage2-x86_64-unknown-linux-gnu" "/home/travis/build/rust-lang/rust/src/test/rustdoc/inline_local/issue-32343.rs"
stdout:
------------------------------------------
------------------------------------------
stderr:
------------------------------------------
21: @has check failed
`XPATH PATTERN` did not match
// @has - '//code/a' 'Bar'
Encountered 1 errors
------------------------------------------
thread '[rustdoc] rustdoc/inline_local/issue-32343.rs' panicked at 'explicit panic', /home/travis/build/rust-lang/rust/src/tools/compiletest/src/runtest.rs:1597
note: Run with `RUST_BACKTRACE=1` for a backtrace.
failures:
[rustdoc] rustdoc/inline_local/issue-32343.rs

An item is inlined and recorded as inlined even if it is `doc(hidden)`, leading to unchecked external links.

mitaa · 2016-04-24T18:41:08Z

Sorry, I guess checking local item links is a bit of an issue, and unnecessary. Because of that I've changed to only checking extern items.

alexcrichton · 2016-04-25T04:51:00Z

@bors: r+ 6603c95

bors · 2016-04-25T10:20:13Z

⌛ Testing commit 6603c95 with merge a5b5cab...

bors · 2016-04-25T11:32:37Z

💔 Test failed - auto-win-msvc-32-opt

alexcrichton · 2016-04-25T18:08:35Z

@bors: retry

On Mon, Apr 25, 2016 at 4:32 AM, bors notifications@github.com wrote:

[image: 💔] Test failed - auto-win-msvc-32-opt
http://buildbot.rust-lang.org/builders/auto-win-msvc-32-opt/builds/3158
—
You are receiving this because you were assigned.
Reply to this email directly or view it on GitHub
#33133 (comment)

@alexcrichton

…ichton rustdoc: inline all the impls This used to be done to avoid inlining impls referencing private items, but is now unnecessary since we actually check that impls do not reference non-doc-reachable items. fixesrust-lang#32881fixesrust-lang#33025fixesrust-lang#33113 r? @alexcrichton

Rollup of 7 pull requests - Successful merges: #33107, #33133, #33160, #33167, #33194, #33196, #33200 - Failed merges:

jethrogb · 2016-05-10T03:56:39Z

This PR should be backported into beta, to fix this regression:
https://doc.rust-lang.org/stable/std/any/trait.Any.html
https://doc.rust-lang.org/beta/std/any/trait.Any.html

There was an issue (rust-lang#33025) which caused these tests to not work. The issue has since been fixed in rust-lang#33133, and so we can now include them.

rust-highfive assigned alexcrichton Apr 21, 2016

mitaa added 2 commits April 24, 2016 08:17

Inline impls on traits too
cc615b2

Remove, now unnecessary, workaround
dca7f01
This used to be done to avoid inlining impls referencing private items, but is now unnecessary since we actually check that impls do not reference non-doc-reachable items.

mitaa force-pushed the rdoc-smth-smth-impl branch from 8c302a0 to 01f7c74Compare April 24, 2016 12:37

mitaa force-pushed the rdoc-smth-smth-impl branch from 01f7c74 to 3c3ee35Compare April 24, 2016 18:12

Check reachability for inlined extern links too
6603c95
An item is inlined and recorded as inlined even if it is `doc(hidden)`, leading to unchecked external links.

mitaa force-pushed the rdoc-smth-smth-impl branch from 3c3ee35 to 6603c95Compare April 24, 2016 18:33

Manishearth mentioned this pull request Apr 25, 2016
Rollup of 7 pull requests #33204
Merged

bors added a commit that referenced this pull request Apr 25, 2016
Auto merge of#33204- Manishearth:rollup, r=Manishearth
149cf77
Rollup of 7 pull requests - Successful merges: #33107, #33133, #33160, #33167, #33194, #33196, #33200 - Failed merges:

bors added a commit that referenced this pull request Apr 26, 2016
Auto merge of#33204- Manishearth:rollup, r=Manishearth
41028de
Rollup of 7 pull requests - Successful merges: #33107, #33133, #33160, #33167, #33194, #33196, #33200 - Failed merges:

bors merged commit 6603c95 into rust-lang:masterApr 26, 2016

mitaa deleted the rdoc-smth-smth-impl branch April 26, 2016 08:59

jethrogb mentioned this pull request May 10, 2016
rustdoc: do not strip blanket impls in crate of origin #33514
Merged

jethrogb mentioned this pull request May 10, 2016
Blanket trait impls are not listed as implementations in rustdoc #29503
Closed

Uh oh!

Uh oh!

Conversation

mitaa commented Apr 21, 2016

Uh oh!

petrochenkov commented Apr 21, 2016

Uh oh!

mitaa commented Apr 21, 2016• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

alexcrichton commented Apr 22, 2016

Uh oh!

bors commented Apr 24, 2016

Uh oh!

bors commented Apr 24, 2016

Uh oh!

mitaa commented Apr 24, 2016

Uh oh!

alexcrichton commented Apr 24, 2016

Uh oh!

mitaa commented Apr 24, 2016

Uh oh!

alexcrichton commented Apr 25, 2016

Uh oh!

bors commented Apr 25, 2016

Uh oh!

bors commented Apr 25, 2016

Uh oh!

alexcrichton commented Apr 25, 2016

Uh oh!

jethrogb commented May 10, 2016

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

mitaa commented Apr 21, 2016•
edited
Loading