Oli Scherer
Oli Scherer
In principle we could use this scheme and make those intrinsics that have no fallback loop, recurse, panic or abort. This is proposed in the issue. The only issue is...
I think by tomorrow I won't be able anymore to replicate whatever model I think I have in my head without rereading the code, but right now I have convinced...
I think we could check this directly in `push_suggestion`, so that no suggestions at all get reported on expansions of derives. Can you try that out and see if it...
Looks like rustc has tests 😆 try running `./x.py test --bless`, that should do it. Though potentially clippy is also affected, so you'd need to bless that, too
I think it's ok to lose this suggestion if it is still reported in another error in the same test. If that is not the case, then we need to...
some tests are still failing
Ah yea, that makes sense :/ Not sure what the best way forward is. I guess ideally we'd make the expansion somehow signal that it is referring to code copied...
> As a different approach, I have an idea to check if the code to which the suggestion is applied is grammatically correct, and suppress it if it is not....
That sounds good to me. We can figure out proc macros later, doesn't have to be in this PR