Minor issues
Hello,
I've found some issues
gov.vcis a website, not tldipv6.yethere we getparts: [ 'ipv6', 'ye' ] --- tld_level: 2gov.gealso the website, not tld
From "The list" https://publicsuffix.org/list/public_suffix_list.dat. gov.vc is declared as a tld. I do not know how to handle thoses cases (where it is both used as a tld AND a domain name) If you suggest me a behavior, i'll gladly implement it
From "The list" https://publicsuffix.org/list/public_suffix_list.dat. gov.vc is declared as a tld. I do not know how to handle thoses cases (where it is both used as a tld AND a domain name) If you suggest me a behavior, i'll gladly implement it
Hello, just consider gov.vc as a site and agriculture.gov.vc as a subdomain. Maybe the database above is outdated.
Create an additional layer, where handle such specific cases like gov.vc
Like, first of all, check these cases ['gov.vc', 'gov.ge', etc], then go to the rest in the list public_suffix_list.dat
There are several so-called TLDs in the Public Suffix List that are websites, but there is no rule that a TLD cannot be a website.
The traditional definition says that the TLD is the right-most part of a hostname, basically the string before the first period going from right to left. Based on this, a TLD cannot be a website, but since this rule is not respected by the Public Suffix List, then you have TLDs like gov.uk or gov.ge which are also websites.
@vladtreny sounds like your use case requires a TLD to not be a website. I suggest you handle this in your source code because the change you are requesting would break the expected behaviour of this library given it's based on the Public Suffix List. You can loop a curl or fetch request for all TLDs in the list and see which ones respond and thus handle these ones however you need based on your requirements.
Out of 50m websites, I found only a dozen of such cases. So do not insist to apply these changes.
The allowDotlessTLD will allow this behavior