Amando
Amando
Just a heads up, I have unpacked the `ml` sub-directories, as it is not a very big module yet so the extra folders are not necessary imo.
Have made `polars` crate part of an opt-in feature.
The way I have it currently reduces the number of dependencies compiled from 400+ to around 270-280. Are you saying we should remove `optional = true` and change `polars` to...
But `polars` is already not included unless `data` is specified as a feature. Why do you want to have `data = []` instead of `data = ["dep:polars"]` ?
@kelvincesar thanks for the comment :) I have considered doing this but am worried about running into issues with the orphan rules.
Hi :) I'm also a little confused about using it within a package. For example: ```R ## foo.R foo
I'll take a closer look over the weekend, but I don't see an issue with hashing a pointer.
Hi everyone, just revisiting this after quite a long time. Can we just hash the inner SEXP instead of deriving ?
I also realise that hashing the pointer and not the actual contents would essentially allow duplicate keys. Roughly (abusing R + Rust syntax): ```rust HashMap.from([ (c(1,2), "Bob"), (c(1,2), "Jane"), ])...
cdylib crate-type fails to build with linking issue 'ld: symbol(s) not found for architecture arm64'
I am getting a similar error ``` dyld[39406]: symbol not found in flat namespace '_PyExc_BaseException' ```