Improve handling of rustdoc lints when used with raw doc fragments. #136400
Conversation
rustbot
added
S-waiting-on-review
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
lolbinarycat
force-pushed
the
rustdoc-link-lint-135851
branch
from
a9a435b to
c6d7696
Compare
| sp = sp.with_lo(sp.lo() + BytePos(match_start as u32)); | ||
| sp = sp.with_hi(sp.lo() + BytePos((md_range.end - md_range.start) as u32)); | ||
| return Some(sp); | ||
| } |
There was a problem hiding this comment.
This logic is run on a compiler's happy path, when rustdoc and its lints are not involved.
Could you do a perf run before merging this?
There was a problem hiding this comment.
source_span_to_markdown_range is? for what purpose?
There was a problem hiding this comment.
i just checked every usage of this function in this repo, and they're all in rustdoc lints
There was a problem hiding this comment.
Huh, that's odd. source_span_to_markdown_range is indeed not used in the compiler. It should be part of rustdoc instead (it doesn't rely on any compiler-internal APIs, so that would technically work) — except that this situation is slightly suspicious. Maybe it used to be used by the compiler in the past, we need to check the history. Also, it might indicate that whoever uses that function doesn't handle cross-crate intra-doc links correctly.
There was a problem hiding this comment.
i feel like i'm missing something, i don't see how intra-doc lints are relevant here.
There was a problem hiding this comment.
i just checked every usage of this function in this repo, and they're all in rustdoc lints
If something is not used in the compiler, then it should not live in compiler/rustc_resolve.
There was a problem hiding this comment.
do you want me to move it in this PR, or can it be done in a followup?
|
@fmease what module do you want me to move |
|
@fmease reminder to look at this. |
|
@fmease it's been a week, are you going to get to this, or should I reroll reviewers? |
rustdoc::bare_urlsno longer outputs incoherent suggestions ifsource_span_for_markdown_rangereturns None, instead outputting no suggestionsource_span_for_markdown_rangehas one more heuristic, so it will returnNoneless often.fixes #135851