Volodymyr Agafonkin

@hborrelli1 currently the only 100% reliable workaround is to upload and use it as a vector tile dataset instead of GeoJSON — Mapbox vector tile server has advanced processing that...

I tried looking into what might be the problem in `binary-split`, but it seems to work fine with Node v21.3. The logs for the failure above also says `ConnectTimeoutError: Connect...

Good point. Hat is only used for random ids to assign to drawn features so this should be easy to replace.

Those are two different `filter`s. `filter` on a layer filters features within a layer. `filter` on a GeoJSON source filters features within that source. > So in the clustering case,...

I'd love to see this too, since this is how a human would write a JSON file to be readable. For example, I often work with GeoJSON and TopoJSON formats,...

@dzhibas looks great, thanks for looking into this! The next step would be to handle nested arrays, e.g. stuff like this: ``` json { "geometry": [ [ [3415, -64], [3323,...

This led me to a simpler idea that got us 3x perf improvement! https://github.com/mourner/icomesh/commit/4cc07dcf001d145b8ea82b68980d0d68de332268

> * [robust](https://docs.rs/robust/1.1.0/robust/index.html)::[incircle](https://docs.rs/robust/1.1.0/robust/fn.incircle.html#) for your `Point::in_circle`; This was omitted for consistency with the JavaScript version, which didn't add it because it was not that important for robustness but increased the...

Sorry for getting around to this so late! I took a different route in https://github.com/mourner/kdbush/commit/d1d856fb12f52de642c3ac721bf7e35cd63e3478

Hmm, interesting. Thanks for the report! I'll think about this. The algorithm itself probably wasn't designed for cases like this, so at this point I'm not sure how to do...