Continue the Cargo Semver Checks goal in 2026 #463
Conversation
|
@obi1kenobi I've opened a draft of for It's mostly updating it to the new template and I've removed sections that seemed to me like they were done. But I found it hard to summarise the status quo section. Would you please take a look at this and help us filling it in? Also, please if I cut too much, let us know! |
obi1kenobi
left a comment
obi1kenobi
left a comment
There was a problem hiding this comment.
Thank you so much for putting this together! I think it looks great, and I added some suggestions for the TODO and a couple of other places in case readers and reviewers want more detail.
Pinging a couple of T-rustdoc folks to see how they want to play things from their end. Whenever they feel this is good to go, I'm onboard too.
Thanks again!
|
|
||
| **Owner:** @obi1kenobi, as maintainer of `cargo-semver-checks` | ||
|
|
||
| I (@obi1kenobi) will be working on this effort. The only other resource request would be occasional discussions and moral support from the [cargo] and [rustdoc] teams, of which I already have the privilege as maintainer of a popular cargo plugin that makes extensive use of rustdoc JSON. |
There was a problem hiding this comment.
@aDotInTheVoid how would you feel about mentioning the cross-crate rustdoc JSON work you've been doing here, to expand slightly from merely "moral support"?
Also @GuillaumeGomez would it be prudent to explicitly request T-rustdoc review capacity for things like rust-lang/rust#148379 or is that fine under "moral support"? I expect that PR is 95% of what we need, and I only expect a couple of tiny follow-up PRs after it (e.g. implied "outlives" bounds for lifetimes, which is fairly simple to do). So I feel a bit silly asking explicitly for PR review capacity when the bulk of the peer-reviewing might be done before this project goal gets merged :) Your call!
| | Implement type-checking in lints | | | | ||
| | Implement linting across crate boundaries | | | |
There was a problem hiding this comment.
I'm fine with this, but just noting that these are both huge tasks and the most likely outcome is that we make significant progress in 2026h1 and yet both boxes remain unchecked and spill over into 2026h2 :)
Hopefully nobody judges this task based on just the boxes in this table, especially when the 2025 project summary is 5000 words and shows the project is literally growing exponentially!
But just flagging it as a concern — and if you think it's worth it, I can try to come up with subtasks so it's easier to demonstrate progress merely with the checkboxes. Your call!
There was a problem hiding this comment.
It's fine as is, imo.
Co-authored-by: Predrag Gruevski <2348618+obi1kenobi@users.noreply.github.com>
| | Implement type-checking in lints | | | | ||
| | Implement linting across crate boundaries | | | |
There was a problem hiding this comment.
It's fine as is, imo.
4 participants
Rendered