How bboatman/magicdeckgenerator works
Main Components
• Backend Application (Python): This is encapsulated in the magicDeckGenerator directory. Key modules include:
– Core Engine (magicDeckGenerator/main.py) launching workflows.
– Libraries under magicDeckGenerator/lib handling:
▪ Deck and DeckMember classes.
▪ Model definitions (e.g., Card).
▪ Data scraping and scraping abstractions (e.g., AbstractDataScraper, MtgTop8Scraper, scraper.py, deckScraper.py).
▪ Card vectorization components (cardVectorizor, vectorize module's main.py) which also includes the cosine_sim module.
▪ Utility modules (log.py, service.py, timer.py).
• Frontend Visualizer (Angular): Found under magicVisualizer, the Angular project is responsible for visualizing vectorized card relationships and deck structures.
– It includes assets (CSV files of graph points), Angular routing, and Angular Material-based views.
• Database (PostgreSQL via Docker): Managed by docker-compose.yml and init scripts (in postgres/migrations), the DB stores card and deck data, and is updated through migrations on container startup.
Diagraming Relationships and Interactions
– Data Scraping: Components within lib/dataScrapers and deckScraper.py.
– Card Vectorization: Modules in cardVectorizor and lib/vectorize (including cosine_sim).
– Core Logic: Deck/DeckMember modules, Models (Card), logging, and service utilities.
Architectural Patterns and Design Principles
• Modularity in the Python backend: modules are separated by responsibility (data scraping vs. vectorization vs. core deck modeling).
• Mention the loosely coupled integration between the generation engine and visualization tool – one processes data while the other visualizes it.
• Layered approach: data is scraped, then vectorized (using cosine similarity and other mathematical measures), and finally utilized for deck generation and visualization.
• Use of containerization (Docker) to standardize the database environment.
Specific Technologies and Frameworks
• Python for the backend.
• PostgreSQL as the database (managed via docker-compose).
• Angular/Angular Material for the frontend application.
• Docker for containerization.
• Additionally, list any notable libraries such as those used in vectorization (cosine_sim) or data scraping.
Important Elements
– “Python Backend Service (magicDeckGenerator)” subdivided as needed (Scraper, Vectorizer, Logic).
– “Angular Visualizer (magicVisualizer)”.
– “PostgreSQL Database” (inside a Docker container).
– Denote the workflow from scraping → vectorization → deck generation → visualizer.
– Indicate read/write flows between the backend and the database.
• Annotations for significant external tools (Docker, Node package management for Angular).