Wildberries product parser with a simple web interface and Excel export.
Enter a search query and the number of items: the service collects product name, brand, price and rating through the marketplace search API and returns an .xlsx file.
- FastAPI web server with Jinja2 templates
- httpx (async) for requests to the marketplace search API
- pandas for the Excel export
- Poetry, Docker / docker compose
poetry install
poetry run uvicorn main:app --reloadOpen http://127.0.0.1:8000, enter a query (e.g. iphone 15), set the number of items and start parsing. When it finishes, download the Excel file; files are also saved to exports/.
docker compose up -d --buildServer deployment steps are described in DEPLOY.md; a detailed walkthrough is in walkthrough.md.
| File | Purpose |
|---|---|
main.py |
FastAPI app: web UI, parsing endpoint, file download |
scraper.py |
Async client for the marketplace search API |
exporter.py |
Export of results to Excel |
templates/ |
Web interface |
test_export.py, verify_parsing.py |
Checks for export and parsing |
The marketplace rate-limits requests from data-center IPs. For server use, run with a reasonable request rate or configure a proxy in scraper.py.