Julian Kenwood

Hi there, I contributed a bit to `im` on the Vector front and I have my own experimental implementation of RRB trees including the `Focus`/`FocusMut`. Although my implementation has substantial...

Greetings, just checking in. I haven't forgotten about this, I didn't have time to much work on this this past weekend. I have gone through the code and just made...

Yes, that sounds correct. One thing I've realised `unmut` is weird, I'm not sure why you'd ever need it. You'd already need to have gotten mutable access to the Vector...

Ah yes, sorry in my experimental repo I have worked around the need for having a lock. When split is called, I do a `(A)rc::make_mut` call, also I have a...

I have been unhappy with the RRB documentation since I started implementing it nearly 2 years ago. I only really understood the intention of the concatenation balancing algorithm relatively recently....

Sorry, I've been pretty busy recently. I meant to add that what after your comment is that what L'orange (and the original paper) calls left-packed is a regular M-1 -...

I believe the copying and shifting of is handled by `createConcatenationPlan` and `executeConcatenationPlan` so I was looking mainly at the those functions for insight there. I figured they were using...

I've had another read through of imbl's concatenation and the algorithm in Nicolas Stucki's thesis. I'm fairly certain the two correspond pretty much exactly and there no major differences. The...

Its kind of silly but there is another parameter `RRB_INVARIANT` which is 1. I think it should actually just actually just be `RRB_EXTRAS/2` (`e/2`). I think it would be great...

Okay, so I went through some of the literature again (I found that L'orange's website also links to a youtube video by Phil Bagwell on RRB's trees). They don't seem...