Given an input with a period, how to get everything trailing it?

Given inputs like the following

bob - invalid
bob. - invalid
bob.edu - valid
bob.com - valid
bob.newco.com - valid

In Ruby, I'd like a way to first determine if there is a period, if there is I want everything after the last period?

I tried doing

domain_ext = search_input.split('.').last

Problem is, if the input is roger开发者_开发知识库. that is returning 5, thinking roger is the ext. I don't know how to tell Ruby only after the last period?

The same regex can be used to filter out domains with a TLD, and return the TLD:

/\.(.+?)$/

Here are some use examples:

domains = %w[ foo foo. foo.com ].select{ |d| d[/\.(.+?)$/, 1] }
=> ["foo.com"]
domains = %w[ foo foo. foo.com ].map{ |d| d[/\.(.+?)$/, 1] }
=> [nil, nil, "com"]

Based on Sawa's comment:

%w[ foo foo. foo.com foo.bar.com ].select{ |d| d[/(?<=\.)([a-z0-9-]+?)$/i, 1] }
=> ["foo.com", "foo.bar.com"]

%w[ foo foo. foo.com foo.bar.com ].map{ |d| d[/(?<=\.)([a-z0-9-]+?)$/i, 1] }
=> [nil, nil, "com", "com"]

 a = %w[
   bob
   bob.
   bob.edu
   bob.com
   bob.newco.com
 ]

 a.each{|str| p str[/\.([^.]+)\z/, 1]}

will give you

nil
nil
"edu"
"com"
"com"

You can just do a substring based on the last index of "."

domain_ext = search_index[search_index.rindex('.')+1..-1]

The above means:

search_index.rindex('.')+1     #Reverse index of "." + 1
search_index.rindex('.')+1..-1 # From that point to the end of the string
search_index[]                 #Simply takes a substring

How about this:

domain_ext = search.split('.').drop(1).last

Which yields, on your examples:

> 'bob'.split('.').drop(1).last
=> nil 

> 'bob.'.split('.').drop(1).last
=> nil 

> 'bob.edu'.split('.').drop(1).last
=> "edu" 

> 'bob.com'.split('.').drop(1).last
=> "com" 

> 'bob.newco.com'.split('.').drop(1).last
=> "com"

> '@gmail.'.split('.').drop(1).last
=> nil

if a trailing period is ok, /(.*)\.(.+)/ will match on the second group, if no trailing is allowed, try /(.*)\.(.*)/