Sentimentify, the app who analyze your tweets :)

The architecture was built using the concepts of Clean Architecture and the SOLID principles. The modularization was composed by:

• SentimentifyEngine: Domain and presentation layers of Search and Analyze features plus abstractions of data layer, like HTTP client.
• SentimentifyiOS: User interface implementation of Search and Analyze features using UIKit.
• TwitterSearchInfrastructure: Infrastructure details of how to do a twitter request, how authenticate and how map the data. Implements the SearchLoader protocol and add behaviour of tweets search.
• GoogleAnalyzeInfrastructure: Infrastructure details of how to do a natural language request, how authenticate and how map the data. Implements both AnalyzeLoader and AnalyzeClassifier protocol, and add behaviour of analyze sentiment and classify the content.
• SentimentifyTestExtensions: Add extra behaviours specific for unit tests.
• SentimentifyApp: The main module, that create features and inject the dependencies, control the navigation flow, add new behaviours trough decorators and adapters.

demo.mov

Features

• Search and list the latest tweets
• See the emotion result based on selected tweet

Non-functional features

• Clean Architecture
• SOLID Principles
• Unit Tests using TDD
• Modular projects using Frameworks
• Design Patterns (Strategy, Proxy, Decorator and Adapter)
• Twitter and Google Natural Language as infrastrucure details, adding the possibility to extend new behaviours
• UITableViewDiffableDataSource
• Child View Controllers (Plugin like)
• Dark Mode
• HTTP requests
• Error Handling
• CI using GitHub Actions
• Swift Package Manager

Third-party dependencies

• Kingfisher

Architecture Diagram