HTML Scraping

API (Application Programming Interface)

The service explicitly exposes endpoints (usually HTTP/REST or GraphQL) that return structured data, typically JSON or XML. You call a documented URL like GET https://api.example.com/users/123 and get back {"id": 123, "name": "..."}. Pros: stable contract, structured data, authenticated, rate-limited but sanctioned, fast. Cons: only exposes what the provider chose to expose; may require keys, quotas, or payment.

HTML scraping

You fetch the same HTML page a browser would render and parse the markup (with tools like BeautifulSoup, Cheerio, Playwright) to extract data from the DOM. Example: download https://example.com/users/123, then find <h1 class="username"> and read its text. Pros: works on any site, no API needed, can extract anything visible. Cons: fragile (breaks when HTML changes), slower, may violate ToS, often blocked (CAPTCHAs, bot detection), needs a headless browser if the page is JS-rendered.