It has for a while on 2018. That PR only affects 2015 IIRC
-
-
I coulda sworn I was told by
@ekuber @ rustconf that it used NLL for checking lifetimes but it ran both and in the case of an error it still used the AST borrowck for reporting.1 reply 0 retweets 0 likes -
I believe you if you think I'm mistaken. I'm def not an expert in this domain
1 reply 0 retweets 1 like -
I'm not sure either, lets just wait for esteban to see this.
1 reply 0 retweets 0 likes -
There are still a couple of places where if you use #[feature(nll)] you get different diagnostics. What this PR is changing is enabling the same transitional mode for 2015, IIUC.
1 reply 0 retweets 1 like -
Meaning NLL is used throughout, but the migration is still ongoing.
1 reply 0 retweets 1 like -
Look at the current output: https://github.com/rust-lang/rust/blob/898f36c83cc28d7921a1d7b3605323dc5cfcf533/src/test/ui/associated-types/cache/project-fn-ret-contravariant.transmute.stderr … vs pure NLL: https://github.com/rust-lang/rust/blob/898f36c83cc28d7921a1d7b3605323dc5cfcf533/src/test/ui/associated-types/cache/project-fn-ret-contravariant.transmute.nll.stderr …
1 reply 0 retweets 1 like -
I don't consider the body buried until the .nll.stderr files don't exist anymore
1 reply 0 retweets 3 likes -
TBH I think the nll error is way better in that example
1 reply 0 retweets 4 likes -
*everyone agrees*
2 replies 0 retweets 1 like
Wait did I interpret the last tweet backwards? I thought it was saying we wanted to make nll give the old errors
-
-
Am also curious if the "pure NLL output" is what we can look forward to once the kinks are worked out or if.. not?
1 reply 0 retweets 1 like -
It's what we're looking forward to, definitely
1 reply 0 retweets 1 like - 4 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.