CBQN

We implemented something like this in CBQN last year (mainly for modulus, as floor division isn't a primitive). Commit is https://github.com/dzaima/CBQN/commit/d333902, some proofs of when and why it works at https://mlochbaum.github.io/BQN/implementation/primitive/ari....

For most things (or, well, that depends on what one commits I guess), the filtered form will likely be the more preferable one. If you want to keep around some of the largely-useless stuff just in case, you can `git tag backup/sso-garbage` or similar before rebasing. For fun, here's a git log pre-rebase of some backup tag I have (final commits: https://github.com/dzaima/CBQN/commits/eccbac37ab15bd68320d9...): f6bc866f (tag: backup/pre-rebase-si-bitwiden) ..aarch64 0fd19c39 Singeli n→8 bitwiden ed2a0655 more Singeli utils 556aa17b use q_fbit more a28adcbb minor src/README.md cleanup d6d40fc7 .warning comment 9dcf2f75 .don't need customizeShape for explicitly-created bitarr 958a04a1 update Singeli submodule 159ee16e .. ea43cb2d . b067d7a8 !! c588a381 fix ⟨1‿2⟩⊸⊏˘ mat 0e251720 . 8f049ed

I was able to make a variant of the higher-base version that runs in a single pass, by stopping when one partition fills up and using a different method for the remaining (asymptotically few) elements. I described the idea, which is based on another effort called MergeShuffle, here: https://mlochbaum.github.io/BQN/implementation/primitive/ran... And it is better when N gets large. My implementation set the cutoff at 2^19 elements, although the effect isn't too big for a few more powers of two. Here's the main radix loop: https://github.com/dzaima/CBQN/blob/v0.7.0/src/builtins/sysf...