How pt-indonesia-comnets-plus/database-management-web works
Project type and purpose
• This is a containerized, data‐driven Streamlit web application with two principal user interfaces (“Home” and “Admin”) plus an ETL pipeline and AI‐powered services (RAG, agent graph, Airflow triggers, etc.).
• It supports user management, asset data ingestion, dashboard views, chat/search features, and email notifications.
File structure
• Top level:
• Dockerfile/.dockerignore – container definition
• Main_Page.py & pages/ – Streamlit entry points
• etl_proces.py & ETL_proces.ipynb – batch ETL jobs ingesting data.xlsx, PDF → database
• core/ – business-logic package
· configs/ (YAML data & tools configurations)
· models/ (ORM-style classes: User, Asset)
· services/ (AssetDataService, UserService, EmailService, RAG, plus an agent_graph subpackage with tool connectors)
· utils/ (DB connection, Firebase, cookies, CSS loader)
• features/ – MVC-style controllers & views for admin and home flows
• static/ – CSS/image assets
• data/ – raw data files
• init.sql & schema.rb – database schema
• Patterns: layered MVC (features/controllers → services → models → utils/database), plus a separate ETL module.