codelucas/newspaper ? reverse-engineered prompt

Build me a Python 3 library for scraping news articles and pulling out the useful stuff in a clean way.

I want to be able to give it a single article URL and get back the title, author names, publish date, main text, top image, embedded videos, keywords, and a short summary. It should also work on whole news sites, so I can point it at a domain and have it discover article links and categories, then let me iterate through the articles it finds. If the page is in another language, it should handle that too, or figure it out automatically when possible.

Please make it feel simple to use, with a small, friendly API, and include the basic documentation and tests so I can see how it works. If anything needs current library behavior or parsing rules checked, look up the latest docs online first.