Damn! Part 1 was difficult to follow but once you read it and programmed against what was written, it was straightforward. This part didn't take long.
Part 2 was a whole other beast. The ranges got stupid big and so there was just way too much seeds to evaluate and find locations for. I tried leaving my computer to run overnight but it was still nowhere close to the answer. Some people did succeed in bruteforcing it but it looks like I won't be able to.
I'm not sure what to do for part 2 but I might just skip it.
This year looks to be harder than the previous years!
I did have the idea of creating a table of locations mapped to soils and then checking the table but that also sounds like it's going to take forever.
I feel like there is a trick but I'm not sure what, the reddit threads don't have any hints that blew the entire thing wide open so hopefully something comes up later today.
Just Part 1 today.