ScribblyBirb

Cannot replicate this with `helix 22.05 (27609f50)`, can replicate with master; currently git bisecting, but my computer builds very slowly... :laughing:

Okay, bisect done, this was introduced by bumping `tree-sitter` in fd644ccfa24b8141c1b8ea32349a0844fe7c884a So maybe an infinite loop in https://github.com/elm-tooling/tree-sitter-elm that showed up with the new tree-sitter version?

Building with ```toml [[package]] name = "tree-sitter" version = "0.20.7" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "549a9faf45679ad50b7f603253635598cf5e007d8ceb806a23f95355938f76a0" ``` solves the problem, so it was introduced in v0.20.8.

Poking around upstream, another bisect shows that tree-sitter/tree-sitter@d223a81b5064587af3a5f61f52d519670ba8995f (between v0.20.7 and v0.20.8) breaks 3 tests in https://github.com/elm-tooling/tree-sitter-elm so it's possible that that is the offending commit, though I haven't been...

I think that the ability to opt-out of the exhaustiveness check is good, as long as it's explicitly opt-**out**. I would like the ability to write both functions that I...

> What I don't like so much about baz as _ is that _ is just a regular old variable name in other contexts where it doesn't mean "ignore me"....

> _ is in fact the only variable name you'll find in α-normalized Dhall, which ultimately is just plain old Dhall. Okay, fair. XD > But this is not very...

> What about { bar = renamed_bar }, so that deconstruction mirrors construction ? Personally `{ renamed_bar = bar }` makes more sense to me, since assignment normally works in...

> Can you explain the desugaring in some more detail, with examples? This obviously would have to occur after resolving the import (if it's one), since before, we aren't sure...

> @SiriusStarr there's a problem with a simple desugar. > > ```haskell > expose { a = 1, b = 2, c = 3 } > expose { d =...