NAs not handled

[davidski]

Passing NAs to tldextract generates warnings/errors. Opening this as a note to myself to come back and contribute a PR. :grin:

[Horopter]

Doesn't exactly implement the python counterpart. For ex: http://developer.twitter.com gives an error. Passing a longer web-url crashes the function with a timeout. The documentation is for me, insufficient to map my program in python to that of R. In python get_tld("https://www.google.co.in/") returns google.co.in which is cleaner and better. In R, this version gives me the following output. host subdomain domain tld 1 http://www.google.com/ <NA> <NA> <NA> It doesn't make any sense to use this in my situation. Thank you for listening to me. Hope to hear from you soon.