Pinterest Scraper Web App

This project provides a modern web interface and API for scraping public Pinterest board URLs. The frontend is built with React and Vite, and the backend uses FastAPI to consume the public RSS feed that Pinterest exposes for boards.

Features

• Animated single-page UI with dedicated tabs for scraping, viewing results, and configuring preferences.
• Backend endpoint (POST /api/scrape) that reads the RSS feed for any public Pinterest board and extracts pin metadata.
• Keyword, width, and height filters to narrow the results that are displayed.
• Copy helpers for quickly exporting pin data as TSV or copying all pin links.

Getting Started

Prerequisites

• Node.js 18+
• Python 3.10+

Install frontend dependencies

npm install

Install backend dependencies

pip install -r requirements.txt

Run the development servers

Start the FastAPI backend first:

uvicorn backend.app:app --reload

In a new terminal, start the Vite development server:

npm run dev

The React app will be available on http://localhost:5173. Requests to /api/* are proxied to the FastAPI server at http://localhost:8000.

Scraping limitations

• The API relies on Pinterest's public RSS feed, so only public boards are supported. Private boards still require the original desktop automation script.
• Some pins may not expose width or height metadata; these entries are still returned but may not pass size filters if Pinterest omits the dimensions.

Production build

To build the frontend for production, run:

npm run build

The compiled assets will be located in the dist/ directory.