Discover the nations of the world with the Countries iOS application. Built with cutting-edge AI from ChatGPT 4.o3-mini and Grok 3 xAI, this app presents countries organized by region (Europe, Asia, Latin America, etc.). For each country, access key information: official name, flag, population figures, GDP data, and a detailed map view showcasing its location and borders.
The iOS application leverages World Bank data, with the explicit challenge given to the AI: to autonomously discover the necessary websites and data structures for development.
For visualizing country borders, we tasked Grok 3 xAI with creating polygon models using geo-coordinates in SwiftUI and MapKit for iOS 17. Our resource assessment for border data led us to recommend martynafford/natural-earth-geojson. We selected this source for its up-to-date nature (last updated January 24, 2024), its direct provision of GEOJSON files, and its consistency with the most recent Natural Earth data.
- MVVM design pattern
- SwiftUI
- async / await
- Swift 6 strict concurrency using @MainActor and marking selection functions as nonisolated or using Task.detached.
- API with Map (position: $position) and Maker
- CLGeocoder() to get more accurate geographic coordinates of the country's capital and Task.detached to run in the background.
- Coped with decoding JSON data without any problems
- Used a modern async / await system for working with multithreading. Suggested several options for switching to Swift 6 strict concurrency using @MainActor and marking selection functions as nonisolated or using Task.detached.
- At first, I suggested the old Map API with Map (coordinateRegion: $region, annotationItems: [country]) and MapMaker, but after receiving the corresponding warnings, I switched to the new API with Map (position: $position) and Maker.
- Used CLGeocoder() to get more accurate geographic coordinates of the country's capital and Task.detached to run in the background.
- The reasoning is short, to the point and lasts from 1 to 25 seconds, the average time is 8 seconds.
- I can recommend it as a good training material for iOS developers.
- After evaluating several sources of country border data, recommended the martynafford/natural-earth-geojson repository due to its newness (January 24, 2024), direct availability of GeoJSON, and its compliance with the latest Natural Earth data.
- Provided code for two ways to decode JSON data: MKGeoJSONDecoder() in MapKit and the standard JSONDecoder()
- Used modern async / await multithreaded fetching of JSON data from the web
- Integrated country border data into SwiftUI Map View
- Can handle non-standard JSON data perfectly
- Proved to be an effective and reliable assistant in iOS programming