Oh dear, looks like that's a bug in cargo itself.

rust-lang/rust#137640 for the lint failure.

As for the failed test, I can't look too closely now but I suspect it's due to the recent changes in tree-sitter-rust: tree-sitter/tree-sitter-rust#254

Edit: I guess we'll have to wait until cargo's nightly channel is fixed.

Again, I'm wondering why you're YOLOing nightly Rust in CI...

Again, I'm wondering why you're YOLOing nightly Rust in CI...

¯_(ツ)_/¯

I think amaanq likes to stay on top of the nightly lints. Maybe it would be better to run the nightly lints as a separate job and use stable for everything else? That way any new lint failures are visible, but issues in the nightly toolchain won't be a blocker like here.

Yes, that would be reasonable (and not mark it as required).

Something something first try?

Changes now include fixes for two new lints, test updates to reflect changes in tree-sitter-rust, as well as moving the nightly clippy invocation to a separate job that doesn't cause the action to fail. I'm not sure why +stable is necessary for the regular clippy invocation. Without it, clippy nightly 1.86 is used, leading to the crash mentioned above.

rust-lang/rust#137640 is closed as completed, but I'm still seeing the crash with nightly. It might be worth opening a separate issue.

Successfully created backport PR for release-0.25:

• adapt to new rust lints and tweak ci #4252