Nathan Zimmerberg

I'm closing this as non-reproducible, but please let me know if the issue persists.

Yes, this would be a nice feature to add. Also check out https://geographiclib.sourceforge.io/, which, from what I can tell, supports WMMHR2025 and has a nice algorithm for high-degree spherical harmonics....

Maybe it is https://gitlab.gnome.org/GNOME/librsvg/-/issues/1023 I recently updated Librsvg from v2.54.7 to v2.58.5 in https://github.com/JuliaPackaging/Yggdrasil/pull/11801, which is likely the problem.

This also updates the rust version, and there might be some issues related to https://github.com/rust-lang/rust/pull/121337#issuecomment-2049298524

I created https://github.com/JuliaPackaging/Yggdrasil/issues/11804 to track the new problem. The easy fix is to use 2.54.7. Maybe we will be stuck on that version until upstream/rust can be fixed.

Should we switch to using JSONX instead of JSON? https://github.com/JuliaIO/JSON.jl/tree/master/vendor

Just FYI, when running Downstream tests in Julia 1.12 in https://github.com/JuliaIO/CodecZstd.jl/actions/runs/19185940798/job/54852442426, the tests for Arrow.jl seem to stall, but do eventually pass after about 2 hours.

~~This implementation seems to cause extra allocations. I'll try something else.~~

Here are some basic benchmarks: ```julia @btime $(1//1)/$(complex(1,2)) ``` PR: 109.592 ns (0 allocations: 0 bytes) master: 41.488 ns (0 allocations: 0 bytes) ```julia @btime $(big(1//1))/$(big(complex(1//1, 2//1))) ``` PR: 1.272...

> Are you sure this is causing allocations? That would be pretty surprising. My previous attempt to fix this was, this version doesn't have that issue.