Medicine autocomplete with Node.js, React and Python

#Run the starter

Download the starter files, set MEDIFONTIS_API_KEY privately in your environment, then run node server.mjs with Node.js 24 or later. Open http://127.0.0.1:3000. No dependencies are needed.

• server.mjs
• index.html
• MedicineSearch.jsx
• README.txt

#Use it in React

Mount MedicineSearch.jsx in your existing React app and route /api/suggestions to the Node server on the same origin. The component waits 300 milliseconds after typing, cancels stale requests, and waits for text composition to finish.

import MedicineSearch from './MedicineSearch.jsx'

export default function App() {
  return <MedicineSearch />
}

#Call suggestions from Python

import json, os
from urllib.request import Request, urlopen
from urllib.parse import urlencode

params = urlencode({'country': 'fr', 'q': 'dolip', 'limit': 10})
request = Request('https://api.medifontis.com/v1/drugs/suggestions?' + params,
    headers={'Authorization': 'Bearer ' + os.environ['MEDIFONTIS_API_KEY']})
with urlopen(request, timeout=10) as response:
    print(json.load(response)['suggestions'])

#Connect it to your application

Keep keys on your server. The starter binds to loopback and limits requests; before public deployment, add your application authentication, shared rate limiting and HTTPS. Suggestions are names, not product identities: use the search and record endpoints to select a specific source record.