More intra-doc links, add explicit exception list to linkchecker #74485
Conversation
|
(rust_highfive has picked a reviewer for you, use r? to override) |
rust-highfive
added
the
S-waiting-on-review
This won't actually fix #32553 because it is about the missing |
|
@ollie27 oh heh I just copied the issue number from the source when I deleted the linkcheck line |
Manishearth
force-pushed
the
more-intra-doc
branch
from
7ee0f2a
to
a968093
Compare
jyn514
added
the
A-intra-doc-links
| /// # Examples | ||
| /// | ||
| /// You can create a `String` from a literal string with [`String::from`]: | ||
| /// | ||
| /// [`String::from`]: From::from |
There was a problem hiding this comment.
Let's link to the relevant impl, not just the trait (this is &'_ str)
| /// [`String::from`]: From::from | |
| /// [`String::from`]: #impl-From<%26%27_%20str> |
There was a problem hiding this comment.
the exact impl doesn't have special docs and the point is to reduce the direct links :)
There was a problem hiding this comment.
I think linking to the type is helpful though. It's not clear that 'literal string' means &'_ str, especially to beginners.
Also, same comment as below about fragments loading faster than different pages.
There was a problem hiding this comment.
We can just fix that with an &str link :)
| /// [`with_capacity_and_hasher`]: #method.with_capacity_and_hasher | ||
| /// [`RefCell`]: crate::cell::RefCell | ||
| /// [`Cell`]: crate::cell::Cell | ||
| /// [`default`]: Default::default |
There was a problem hiding this comment.
This now goes to the trait page, instead of the list of implementations. I know HashMap::default doesn't currently work, can we keep #method.default until then? We don't have the ambiguity of String::from because there can only be one Default impl.
There was a problem hiding this comment.
IMO the trait page is fine but I'd rather get rid of the #s and have links work over having them link to the trait impl instead of the trait. I don't see a difference in utility there
There was a problem hiding this comment.
The difference is that going to the trait takes you to a new page and away from the docs you were originally reading. #method.default loads basically instantaneously.
| const LINKCHECK_EXCEPTIONS: &[(&str, &[&str])] = &[ | ||
| // These are methods on slice, and `Self` does not work on primitive impls | ||
| // in intra-doc links (intra-doc links are weird) | ||
| // https://github.com/rust-lang/rust/issues/62834 is necessary to be |
There was a problem hiding this comment.
I think you meant the issue on primitive types.
| // https://github.com/rust-lang/rust/issues/62834 is necessary to be | |
| // https://github.com/rust-lang/rust/issues/63351 is necessary to be |
There was a problem hiding this comment.
oh, in my mind the generic issue is also what blocks slice, but your thing works
src/tools/linkchecker/main.rs
Outdated
| // are cases where that does not work | ||
| const LINKCHECK_EXCEPTIONS: &[(&str, &[&str])] = &[ | ||
| // These are methods on slice, and `Self` does not work on primitive impls | ||
| // in intra-doc links (intra-doc links are weird) |
There was a problem hiding this comment.
I actually meant to say "primitive impls are weird" lol
Co-authored-by: Joshua Nelson <joshua@yottadb.com>
Co-authored-by: Joshua Nelson <joshua@yottadb.com>
Manishearth
force-pushed
the
more-intra-doc
branch
from
e9424e8
to
ec966ae
Compare
|
@bors r+ rollup=iffy The changes to linkchecker will mean any PR with broken links would break during rollup but not in a check build. |
|
📌 Commit ec966ae has been approved by |
bors
added
S-waiting-on-bors
…arth Rollup of 4 pull requests Successful merges: - rust-lang#74333 (Deny unsafe operations in unsafe functions in libstd/alloc.rs) - rust-lang#74356 (Remove combine function) - rust-lang#74419 (Add a thumbv4t-none-eabi target) - rust-lang#74485 (More intra-doc links, add explicit exception list to linkchecker) Failed merges: - rust-lang#74486 (Improve Read::read_exact documentation) r? @ghost
Fixes the broken links behind #32553
Progress on #32130 and #32129 except for a small number of links. Instead of whitelisting entire files, I've changed the code to whitelist specific links in specific files, and added a comment requesting people explain the reasons they add exceptions. I'm not sure if we should close those issues in favor of the already filed intra-doc link issues.