GEO_PROJECT/ │ ├── data/ │ ├── properties_geocoded.csv # Master dataset (source of truth) │ └── properties.db # SQLite database used by app │ ├── app.py # Main application ├── reload_db.py # Reload DB from CSV ├── requirements.txt └── README.md

🔁 Data Update Workflow

When the geocoded CSV is updated, you must reload the database before running the app.

Step 1 — Replace the CSV

Place the updated file at:

data/properties_geocoded.csv

This file is the master dataset.

Step 2 — Reload the Database

From project root:

py reload_db.py

This will:

Overwrite the properties table

Sync the database with the CSV

Ensure the app uses updated coordinates

Expected output:

Database successfully reloaded. Step 3 — Run the Application py app.py

(or however the app is normally launched)

⚠ Important Rules

The CSV is the source of truth.

The SQLite DB is derived from the CSV.

Don't manually edit the SQLite database.

Always reload using reload_db.py.

Keep only one database file: data/properties.db.

🧠 Why This Architecture

CSV = portable, human-readable, version-controlled

SQLite = fast query layer for runtime

reload_db.py = deterministic, repeatable synchronization

This keeps the system clean, modular, and reproducible.

🔎 Quick Database Check

To verify DB integrity:

py -c "import sqlite3; c=sqlite3.connect('data/properties.db'); print(c.execute('select count(*) from properties').fetchone()); c.close()"

Future Improvements

Add timestamp validation, add option to upload a new csv rather than do it from the backend to update DB

Change UI saved route area

Create app by switching app structure entirely

Migrate to Postgres for production scale