I finally wrote a little bit about verification and why Rust is interesting https://t.co/RTaqnisZXR. If it interest people I'm planning on writing more on design of specification languages for Rust and the implementation of verifiers in and about Rust
The PR was adding the `never_patterns` feature gate, which is an important step in my little project of fixing the support of empty types in pattern matching https://t.co/6f52nQUoxX
This is your regularly scheduled remainder that what the Rust project could use the most right now is non-coding organisational help. Ppl to run meetings, to document what the project is doing, to help triage, track projects, get the right people to look at the right things, ...
My goals for the exhaustiveness algorithm:
- linear pass, because it's currently quadratic. won't save as much time as it sounds but cleaner
- explain it clearly for everyone
- librarification, so rust-analyzer can do cool things
- never patterns, to stabilize exhaustive_patterns