Darren Li

@bowenwang1996 Just to get some sanity checks before I attempt to do anything: is there a cache eviction policy in mind that you think would enable best performance in, say,...

Small update: Just realised I forgot to change member role in github org: everyone should be a owner now. Right now we have @nazar-pc and @mvines who are owners of...

I can possibly provide a fix for this by swapping to timedesc, but I am not sure if that is desirable (timedesc does bring in more dependencies).

Currently windows support is still missing, but not exactly what approach is best.

Hm, it does take surprisingly long to resolve, and indeed I think your explanation is correct. Though I vaguely recall having some handling for cases like this, let me check......

> Very nice library! Thank you for your work :) : D --- W.r.t. your proposal: > In any case, may I suggest a different representation for Resolver.result_space than lists...

Right now I suspect there are too many conversions from set to seq - I'll play around with `pattern_resolver.ml` when I have time later. Many thanks for raising the issue!

Hm...now I'm not very sure... Aha...might have found the culprit... Might be missing a call to `slice_result_space_multi` in `aux_inter'`

> Maybe it's because the pattern enumeration only uses the year of the lowerbound? Oh yeah possibly! I'll need to remind myself of how all these work, it's been a...

I think you're spot on. Through `dune utop src/`: ``` let x = Result.get_ok @@ Timere.(resolve (after (Timedesc.make_exn ~year:2022 ~month:01 ~day:01 ~hour:01 ~minute:01 ~second:01 ()) &&& seconds [0])) ``` is...