I also fixed the rustdoc bug upstream!
github.com/rust-lang/ru...
@predr.ag
Author of cargo-semver-checks & Trustfall // https://github.com/sponsors/obi1kenobi // https://predr.ag/blog // ex Principal Eng @Kensho // MIT alum // https://hachyderm.io/@predrag // not from around here π²π° // he-him
I also fixed the rustdoc bug upstream!
github.com/rust-lang/ru...
cargo-semver-checks v0.47 is live now β¨
Inside, you'll find:
- a workaround for a Rust 1.94 rustdoc bug that caused crashes in some projects
- 7 new lints
Happy SemVer Sunday!
github.com/obi1kenobi/c...
When someone says βScientists do not want you to knowβ you can dismiss everything from there on. Scientists want you to know. They are desperate that you know. They canβt shut up about what they found out and want you to know.
Here's a tiny bit of hopium:
In 2024 we were planning on adding 160 GW of renewable energy to the grid by 2030.
As of the start of this year, we had added 100 GW and are planning on adding 220 GW more by 2030.
(In 2022 we planned for 60 GW in 2030. We hit that by the end of 2023)
This is wild, nice find! I see why you describe it as a rustc bug.
All good! Thank you!
It would still be helpful to know the "what did the upstream crate have, and what changed to trigger the bug" though.
You've thought 1000x more about this than any of the cargo-semver-checks contributors. So things that are obvious to you are quite opaque to us, even with the playground link :/
I'd love some help from @radbuglet.bsky.social to help me understand this better. We definitely don't catch this now, but I'd like to fix that!
It'd be helpful to frame this as "here's code that existed initially, then X got added, and behold how that caused breakage."
Important news!
I am delighted to announce the launch of my new podcast: The Untitled History Podcast.
It's me, and guests, falling down all the history rabbitholes you can imagine.
Please, please spread the word, subscribe and listen. Because... SHIPS CATS!! cms.megaphone.fm/channel/the-...
β°οΈ 1 day left to submit a talk!
The #rustconf program is shaped by the Rust community β and the CFP closes tomorrow @ 11:59pm PST.
If youβve been building, experimenting, debating, or refining how Rust is used in practice, this is your chance to put that work!
ngl when I saw this, I thought espresso must be a new library for working with gRPC over HTTP2 via tonic...
Ugh, you're telling me that "great landing" and "good landing" are not `Ord`.
In retrospect this makes sense. This plane is clearly a float.
A rare case of a great landing (you can reuse the airframe after landing) that wasn't necessarily also a good landing (you can walk away from the airframe after landing).
Hey chat is it good when your security audit recommends this?
Image from the Trail of Bits report showing the offending code associated with the second high-severity vulnerability. It includes an empty "else" branch with a "to-do" comment that reads: "TODO (eventually) reject if signature is missing"
Why was this code ever shipped?!
This is from the second vuln, where keys' signatures aren't checked before they're stored in the trusted key store.
Why would you ever ship a "TODO, actually validate signatures lol" in your secure messenger?!
Folks, there's an impersonation scam on Discord (and possibly elsewhere) that uses tech people's names and photos to slide into your DMs.
If you get a friend request / DM from "me" out of the blue:
(1) it's almost certainly not me
(2) feel free to ping me here to check!
Thereβs so much demand for a T thatβs so fundamentally disappointing.
Neat idea! Is there a way to make `cargo update` inside the project work nicely though, without updating the trip wire?
Asking because I have a weekly job to update deps, and I want to figure out how I can adopt this reasonably. Maybe have it `cargo update` then explicitly downgrade the trip wire?
And in case you missed it, here's our annual recap and our plans for 2026, which should look somewhat different than the last few years!
bsky.app/profile/pred...
cargo-semver-checks v0.46 was released earlier this week β¨
It comes with 45 *new* lints. Almost 20% of *all lints* are new in this release β this is wild!! Exponential growth, manifested.
github.com/obi1kenobi/c...
Yep. Only way to:
- have a dependency on Rust
- support multiple Rust versions
- retain any semblance of sanity for me
Very soon, just probably not the way you had in mind π
It'll be a runtime dependency!
Funny, that's also how to build cargo-semver-checks too.
Mastering that one technique seems to cover a lot of cases.
We are excited to announce that we can successfully use Rust's standard library from the GPU. This has never been done before.
www.vectorware.com/blog/rust-st...
Supporting Rust's standard library enables existing Rust code to work on the GPU and makes GPU programming feel normal.
I just woke up from a nap and somehow while I was asleep, everyone on the bus has figured out we are not going to the right place
This is part of how cargo-semver-checks has been growing so quickly.
It's a lot of work to discover and formulate SemVer rules. It's fairly trivial for an LLM to take those detailed notes and write *deterministic* queries in the Trustfall query language, and back them up with tests.
I'm thrilled it's still helpful!
You might also like this other post which has more of those same shenanigans:
predr.ag/blog/when-is...
I built but never open-sourced a Python equivalent of cargo-semver-checks too. What held me back was feeling like there was no funding interest for it, and that I wouldn't have cycles or willpower to maintain it in such an environment.
What you describe doesn't seem like a big stretch from there
This is the same tech that backs cargo-semver-checks. It isn't specific to Rust or Python or any other fixed set of data sources or languages.
Curious about your use case!
I've built something not that dissimilar to what you're describing, and would be happy to chat about it more if useful:
predr.ag/blog/how-to-...