isURL www.google is valid

[thisbejim]

www.google is valid with the default isURL settings.

[thisbejim]

Additionally, with the following setup:

{
    protocols: ['https'],
    require_protocol: true,
    require_tld: true,
    require_host: true,
 }

The following is valid: https://www.google. You would expect the validator to require the tld.

[chriso]

https://www.google is a valid URL, and .google might be a valid TLD. See #429 and #428. The only way for the library to determine whether or not .google is a valid TLD would be to include a big list of them, which would bloat the size of the library. Another alternative would be to make a request to the URL to see if it resolves and responds to the request, but this is outside the scope of the library.

[thisbejim]

Ah cool thanks for the explanation and apologies for not searching the issues thoroughly enough.

Given the confusion and the generally expected behaviour from a few different users, it might be worth adding an option for a mode that blacklists www as the start of the hostname (e.g. www.google) unless there is another . in the host name www.google.com. This gets around having to track tlds and bloating the library. Or if you'd prefer not to add anything extra to the library you could maybe document how to blacklist these domains in the readme.

e.g.

host_whitelist: [/(?:\w+\.)+\w+\./]

^ not sure how good that is

Anyway, just a thought! Cheers 👍