Kristoffer Carlsson
Kristoffer Carlsson
For Pkg, if you update it, I would want the repl to be the updated one. We just have to fix the common load failures from out of date manifests.
👍 CI failure is probably unrelated.
Actually CI on https://github.com/JuliaLang/PackageCompiler.jl/pull/1057 looks better than here. I'll try rebase this.
Actually, can we wait a few days with this? I want to check that some stuff I know interact with the registry is OK with this or if it needs...
It seems this is fine for the use case I wanted to check. So no objection from my pov.
I have no idea what the state of CI is on this repo. Are the test failures something I should look into or are they expected etc?
Comes from https://github.com/tkf/BenchmarkCI.jl/blob/bca191f70d7fe54bb1c6ecc58e4c9b9d0ee8e4c6/src/BenchmarkCI.jl#L127-L151. That doesn't work anymore. Either needs an `@invokelatest` or be written so that the JSON loading is a separate global statement vs the `JSON.print`. I can fix...
> and making that also work would need feedback from the parser state to the lexer (uuugh!) 😢
Right now, there is no way to check if a user has forcefully disabled warn overwrite or if it is just by default so that would need to be updated...
For posterity, this happens because there is a check in `free` that ensures that one doesn't do something like `free Foo#master` but this has a bad interaction with sources which...