A few suggestions

[WilliamDenniss]

Hey @eselkin, nice work on your Swift implementation of Open Location Code! As it happens, I had a similar idea around the same time. In my case I picked the existing Python code to port to Swift. Took me a while to go through my company's open source process, but you can now check it out (pod 'OpenLocationCodeFramework'). Pity about the duplicated work, when I started your port wasn't yet published, just how it goes I guess.

A few suggestions for your port:

1/ I don't think it works with Objective-C, as your classes don't descend from NSObject. Two options to do that (if you want to support ObjC): either modify your main classes, or provide new ones as a bridge. For my library, I did the latter, as objects that descend from NSObject can't be used in pure Swift on Linux – so the Objective-C stuff has to be separable to maintain Linux support. That approach also means you can continue to use pure swift constructs (like structs) in the swift API which is nice.

2/ I highly recommend implementing the core SDK tests driven by the CSV data. I know they caught a few bugs in my port when I ran them! You can probably take my implementation directly, and adapt it for your API.

3/ I also included ports of the recover nearest implementation. I like the short-code feature of OLC, it's useful to be able to give someone in the same city as you a shorter 6 digit code for locations within the city ☺️. Some apps like Google Maps already highlight the short code characters to speed up communication time (e.g. when you're telling someone where you are on the phone).

[eselkin]

For some reason, I didn't see your comment until now. I will probably descend from NSObject since I don't think it requires anything outside that scope. I'll check in the future. I'll check the testing. I hadn't needed the short codes for my implementation but I had planned on adding it, just hadn't gotten around to testing my code yet.