Anshul Singhvi
Anshul Singhvi
A generic tiled iterator would be a bit tough. I think you'd indicate an approximate size, and the iterator could partition into tiles that fit the chunk pattern most optimally....
~Ah crap, converting cellarea to Proj means we have to implement some form of GFT conversion in Proj as well...~ Edit: Proj has both GFT conversions and an inbuilt geographic...
https://github.com/meggart/DiskArrayEngine.jl/tree/main kind of implements tiled iteration, but it's also huge. I don't think it offers multithreading by default either, and it pulls in a crazy number of dependencies. We could...
I just realized that Proj allows you to return data in radians (which is what it uses internally). We could potentially switch to that and avoid cosd and sind. Might...
Ah, I guess I missed reading that :D - will see if there are any actually missing cases and add those.
JuliaGeo./GeoJSON.jl will also have to be completely rewritten.
It doesn't yet but it depends on JSON3.jl, which it should switch from to JSON.jl especially since that will allow completely lazy parsing
Feel free to ignore the cosd/sind stuff if there's no time, I just discovered that they use an extended-precision deg2rad that might be useful.
Ideally we would say `cellarea(Linear(), raster)` or `cellarea(Spherical(), raster)` I guess, that will have to wait for a bunch of reactors in GeometryOps so that Rasters can depend on a...
It would take about 3 days to get out, but we could technically define a GeometryOpsCore.jl package today and get it registered, Rasters could then depend on that (it should...