From: Andrew Timberlake <andrew@...>
Date: 2009-04-22T22:25:34+09:00
Subject: Re: Needed only the Domain name from an url

On Wed, Apr 22, 2009 at 2:28 PM, Srikanth Jeeva <sri.jjhero@gmail.com> wrote:
> hi,
> Suppose if i give,
> url="www.google.com/picasa/images/1"
>
> I need only the string "google" which is the domain name.
>
> right now i have tried this,
>
> require 'rubygems'
> require 'uri'
> url = "http://www.google.com/picasa/images/1"
> puts url = URI.parse(url).host
> -----------
> output: "google.com"
> -----------
> but i want only the word "google". please help.
>
> Pts:
> the domain name can also be, "google.co.in"
> ie) for example, "http://www.google.co.in/picasa/images/1"
>
> Thanks in advance,
> srikanth


Technically google.com is the domain
If I run your code, I get "www.google.com" as the output, not "google.com"
This poses an interesting problem because a domain can be:
one.two.three.four.google.co.uk
What is the domain for you in that?

Otherwise you could do something like
uri.host.sub(/.*?\.?([^.]+)(?:\.\w{3}|\.\w{2,3}\.\w{2,3})$/, '\1')

It may not be the prettiest regex but it gets the job done for the following:
  * one.two.google.com
  * one.two.google.co.in
  * www.somesite.com.au
  * www.one.co.uk
but it won't work for
  * www.one.com (but it will work for one.com)

Andrew Timberlake
http://ramblingsonrails.com
http://www.linkedin.com/in/andrewtimberlake

"I have never let my schooling interfere with my education" - Mark Twain