Explorador Comida di Buteco 2026

A fullstack web application for exploring bars participating in Comida di Buteco 2026 based on the user's proximity.

The project allows users to search for an address, define a search radius in kilometers, and view nearby participating bars on an interactive map. The backend uses a spatial data structure, KDTree, to make coordinate-based searches more efficient.

🔗 Live Demo: https://phcortez.com.br/explorador-comida-di-buteco/
📦 Repository: https://github.com/GL-PC/Explorador-Comida-Di-Buteco

---

Overview

Explorador Comida di Buteco was developed to make it easier to discover participating bars in the event based on a location provided by the user.

The application combines a modern frontend, an interactive map, address geocoding, and spatial processing on the backend to return results sorted by distance.

Main features include:

• Address search;
• Search radius in kilometers;
• Display of nearby bars on an interactive map;
• Results sorted by distance;
• Bar details including address, city, dish image, and link to the bar page;
• Initial loading of participating bars on the map;
• Optimized spatial search using KDTree.

---

Demo

---

Features

Proximity search

The user enters an address and a search radius. The backend geocodes the address, queries the KDTree, and returns only the bars located within the specified distance.

Interactive map

Bars are displayed on a map with markers, allowing users to visually explore participating establishments.

Results sorted by distance

The results list shows the nearest bars first, along with useful information to help the user choose where to go.

Filtering by visible map area

The backend also supports bounding box search, allowing the application to retrieve bars located within the geographical limits of the visible map area.

Geocoded dataset

The project works with a dataset of approximately 1,088 participating bars, including geographic coordinates used for spatial searches.

---

Tech Stack

Frontend

• React 19
• TypeScript
• Vite
• Tailwind CSS
• Leaflet
• React Leaflet
• Lucide React

Backend

• Python
• Flask
• Flask-CORS
• Requests
• Custom KDTree implementation
• Distance calculation using the Haversine formula

---

Author

Developed by Gustavo Ribeiro and Pedro Cortez.