With the current set of placehoders for the autotype tool, I can use {URL:HOST} to get the host-part of the url. However, for some autotype environments, I need the domainname part of this {URL:HOST}.
For {URL:DOMAINNAME} I have found {T-REPLACE-RX:/{URL:HOST}/^[^.].(.)/$1/} which suits the need: Everything from the {URL:HOST} after the first dot. The problem here is the readability in the autotype settings.
And since I can imagine systems wheren the single hostname is needed, It would be nice to have that too.
For example: With the example url https://www.keepass.info/ :
{URL:HOST} == "www.keepass.info" # already available.
{URL:DOMAIN} == "keepass.info" # my wish: last 2 parts from fqdn
{URL:HOSTNAME} == "www" # my wish: first part.
Detail: In my regexp above, I use all after the first dot for the domainname, that differs from my specification here using the last 2 parts as the domainname, mostly because of the next ideas on subdomain names:
Come to think of it: with https://www.keepass.co.uk/ (wild ideas, just to be complete):
{URL:DOMAIN} == "co.uk" # last 2 parts
{URL:SUBDOMAIN} == "keepass.co.uk" # last 3 parts
{URL:DOMAINNAME} == "keepass.co" # 3rd and 2nd part from end.
{URL:SUBDOMAINNAME} == "keepass" # 3rd part from end
{URL:TLD} == "uk" # top level domain
This scheme can be extended with {URL:SUBSUBDOMAIN}
My point of view:
The {URL:DOMAIN} is to be used in the {USER}@{URL:DOMAIN} accountname on host-specific websites.
The {URL:HOSTNAME} can be used for machine specific logins like msWindows local administrator at {URL:HOSTNAME}\administrator.
The {URL:DOMAINNAME} can frequently safely be used for netbios-domain name as in {URL:DOMAINNAME}{USERNAME}
This is actually a sub-domain, as are the "keepass" and "co" parts of keepass.co.uk. You then get into 3rd subdomain etc and it all gets a bit clumsy.
Maybe have a qualifier with the subdomain placeholder to allow for this.
cheers, Paul