How rvish-glitch/smart_trip_planner works
Project type and purpose
• This is a cross-platform Flutter application (“Smart Trip Planner”) for conversational, AI-powered travel itinerary generation.
• It uses Google Gemini AI for NLP and itinerary creation, Isar for local persistence, and integrates Google Maps for geospatial visualization.
File structure
• Top-level directories for each platform (android, ios, linux, macos, windows, web) contain the native runners and configuration.
• The core Flutter code lives under lib/:
• core/utils & core/services – shared utilities and foundational services (e.g., reading API key from .env)
• data/datasources – REST/HTTP clients for Google Gemini AI (gemini_api.dart / gemini_api_fixed.dart)
• data/models – data objects (Itinerary, generated code)
• data/repositories – abstraction over local (Isar, SharedPreferences) and remote sources
• presentation/screens & widgets – UI pages (chat, memory, trip editor, token manager) and reusable UI components
• presentation/services – view-model-like services managing state and bridging UI to data layer
• Configuration/build scripts: pubspec.yaml, Android Gradle scripts, iOS Xcode projects, CMakeLists for desktop.
README for additional insights
• Confirms Clean Architecture: separation into core, data, presentation layers.
• Specifies repository pattern, dependency injection, and use cases (CRUD itinerary, chat messaging, map visualization).
• Shows high-level tech stack (Flutter, Dart, Gemini AI, Isar).