internal: Use upstream exhaustiveness checker! #16420
Conversation
rustbot
added
the
S-waiting-on-review
Nadrieril
force-pushed
the
use-upstream-pattern-analysis
branch
from
821b442
to
2370b70
Compare
| source = "registry+https://github.com/rust-lang/crates.io-index" | ||
| checksum = "6c4085e0c771fd4b883930b599ef42966b855762bbe4052c17673b3253421a6d" | ||
| dependencies = [ | ||
| "derivative", |
There was a problem hiding this comment.
Oof, we went through some pains to get rid of syn 1.0. But you do seem to use it in a couple of places, so doing that by hand would probably be too annoying.
There was a problem hiding this comment.
Oh no 🙈 I mean it's not too hard to impl by hand
There was a problem hiding this comment.
It's not, but I don't want to force you to spend time on that right now. It's okay to merge, then clean it up later (ideally).
There was a problem hiding this comment.
We should just fix up the assist to inline custom proc-macro derive expansions (which currently only supports builtin derives) :)
But to chime in as well, I'd like to see derivative not stick in our dependency tree for long. Iirc it, like syn, has a pretty bad compile time impact.
It's okay to merge, then clean it up later (ideally).
☝️
There was a problem hiding this comment.
Maybe mcarton/rust-derivative#118 can save us! If it ever gets merged
There was a problem hiding this comment.
It seems rustc also wants to get rid of syn 1.0 rust-lang/rust#109302, and they use derivative. That's a good sign
There was a problem hiding this comment.
Hmm, actually rust-derivative does not seem to be maintained anymore (3 years since the last commit). Anyways, since rustc wants to bump to syn 2 as well T-compiler will probably figure something out there at some point.
There was a problem hiding this comment.
Yeah, that PR I linked was opened recently by the person who's working on moving rustc to syn2, hopefully they'll take some kind of action if the maintainer doesn't respond
|
Excited to see what we can do with this! |
|
☀️ Test successful - checks-actions |
lnicola
changed the title
…tive, r=Nilstrieb Stop using derivative in rustc_pattern_analysis CC rust-lang#109302, rust-lang/rust-analyzer#16420 (comment) r? `@Nadrieril`
…tive, r=Nilstrieb Stop using derivative in rustc_pattern_analysis CC rust-lang#109302, rust-lang/rust-analyzer#16420 (comment) r? `@Nadrieril`
…tive, r=Nilstrieb Stop using derivative in rustc_pattern_analysis CC rust-lang#109302, rust-lang/rust-analyzer#16420 (comment) r? ``@Nadrieril``
…tive, r=Nilstrieb Stop using derivative in rustc_pattern_analysis CC rust-lang#109302, rust-lang/rust-analyzer#16420 (comment) r? ```@Nadrieril```
…tive, r=Nilstrieb Stop using derivative in rustc_pattern_analysis CC rust-lang#109302, rust-lang/rust-analyzer#16420 (comment) r? ````@Nadrieril````
Rollup merge of rust-lang#120420 - lnicola:rm-pattern-analysis-derivative, r=Nilstrieb Stop using derivative in rustc_pattern_analysis CC rust-lang#109302, rust-lang/rust-analyzer#16420 (comment) r? ````@Nadrieril````
Because it has been librarified!
The extra
Apache-2.0 WITH LLVM-exceptionlicense is forrustc_apfloat. Also this duplicatesrustc_indexbecause the other upstream deps are still on an earlier version. They should be bumpable now though. Good thing is that we don't need this new crate to be synchronized with the others, which will make our lives easier.