I wasn’t able to import a PDF document from the URL. When I attempted to, I clicked on import web page, and then type of document PDF. The first time I tried it, I just got code. The second and third times, I got an error message that the URL had failed. So Instead I had to first download the web page and then import it from my hard drive. That work-around was fine, but I thought that importing directly from the URL was supposed to be an option.
Importing web pages from a URL is supported, sure. I don’t really see downloading a file you wish to have available locally on your computer as a “work-around” though. 
That’s how things work. Downloading files straight into your project would represent a security hazard. Browsers have safeguards and malware detection built in for this sort of thing. It would be safer and easier to continue relying upon that safety net.