Micha Reiser
Micha Reiser
The pydantic speedup is interesting I added a CI job in https://github.com/astral-sh/ruff/pull/21864 that should make it easier to verify a fix.
Claude claims that ty should exit with 130 after an interrupt. I don't feel like I know enough whether it's worth the complexity or not
I feel pretty good about the approach taken in this change so I'll go ahead and merge this PR tomorrow unless someone objects
The speed-ups in the ty-benchmark are between 0-6%
Not sure how to tell whether there are any new mypy primer hits or if it's just the usual non-deterministic cases
Yeah, they look suspicious.
I'll go ahead and merge this, given that @ibraheemdev is out for the rest of the week.
I'm very confused why the benchmark job fails here when it succeeded on https://github.com/astral-sh/ruff/actions/runs/20426087014/job/58686630346
Need to update cargo fuzz too. I'll do that once the upstream PR is merged
There are a few more salsa changes that we want to get in, where pinning is very convenient, but we do plan to move to releases eventually.