CohenArthur
CohenArthur
> Also I see the build-and-check with glibcxx assertions turned on ICE's. I'll try to figure out what's going on there. (gcc is giving me trouble with compiling :/ )...
Closing in favor of a new attempt
yeah we should probably do that in the attributes checker I think? would probably make sense, as for example `repr(...)` cannot use type aliases or anything like that: https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=eb6470d9855df67c57f53619392970b8
I don't know if this is the best way to go about this, it's definitely hacky. maybe we could do all the desugars as part of the fixed point and...
> It looks like doing exactly that would screw up recursion_limit tracking how do you mean? desugaring shouldn't count for the expansion limit, as we control the expansion (as opposed...
I think this is not really an issue, if we hit the expansion limit we'll have an error anyway and will stop the pipeline. so the desugaring does not matter....
hi @varun-r-mallya, sorry about this another contributor already asked to work on the issue beforehand on Zulip. apologies. if you want, you can do the same thing in our `HIR`...
@varun-r-mallya there isn't much more to say haha. have a look in the code for the class, find it, and then just delete it. this will cause a bunch of...
@dylngg no worries!! glad you got it, thanks for your contribution :D
> @CohenArthur would like in the long run to switch to a system with less side effects, every visit function should return it's own modifications, errors would be reported as...