How igor-ferreira-bezerra/average-price-ir-calculator works
Project Type and Purpose
• This is a monolithic, full-stack Python web application built with Streamlit.
• Its purpose is to calculate the average price of securities (ações and FIIs) from an uploaded Excel “Movimentação” sheet, handle special operations (desdobro, bonificação, transferência, atualização) and enrich results by merging static CNPJ CSV files.
File Structure
• Top-level entrypoint: calculator.py (Streamlit app containing UI + business logic).
• Data folder: data/CNPJs_FII.csv, data/CNPJs_ACOES.csv (external metadata).
• Config/development: .devcontainer/devcontainer.json (defines a reproducible dev environment).
• Dependency manifest: requirements.txt (Pandas, OpenPyXL, Streamlit).
Main Components
a. User (browser)
b. Streamlit Front-End (calculator.py UI components)
c. Calculation Engine (data ingestion, preprocessing, average price logic, special operation handlers)
d. Data Sources
• Uploaded Excel (runtime input)
• Static CSVs (CNPJs_FII.csv, CNPJs_ACOES.csv)
e. Python Runtime / DevContainer (environment boundary)
Relationships and Data Flow
• User → Streamlit UI: uploads Excel, sets year-base, confirms updates
• Streamlit UI → Calculation Engine: invokes preprocessing & calculation functions
• Calculation Engine → Excel Reader (OpenPyXL/Pandas): reads “Movimentação” sheet
• Calculation Engine → CSV Reader (Pandas): loads and merges CNPJ data
• Calculation Engine → Streamlit UI: returns two result tables (FIIs / Ações)
• (Optional) DevContainer wraps the Python runtime and dependencies
Architectural Patterns and Principles
• Monolith with clearly separated concerns: UI vs. processing vs. data
• MVC-ish: Streamlit handles view + controller, Pandas DataFrames are the model
• Pipeline-style data processing: ingest → transform → merge → output
• File-based integration instead of networked microservices
Diagramming Conventions
• Blue rectangle for UI
• Green rectangle for Processing/Calculation Engine
• UI → Calculation Engine (“trigger preprocessing & calc”)
• Calculation Engine → UI (“render result tables”)
Detail Without Over-Modularizing
• Every major step—file ingestion, special-case handling (desdobro, bonificação, atualização), data enrichment, result rendering.
• Avoid splitting into dozens of microservices; keep it monolithic but clearly partitioned by responsibility.