I want to get ASIA from following sample.
/scripts/skw.asp?term=&department=ASIA
Do you know how can I extract department value from whole text.
-
Possible duplicates: here, and here.user4235730– user42357302014-11-30 21:59:13 +00:00Commented Nov 30, 2014 at 21:59
Add a comment
|
2 Answers
string = "/scripts/skw.asp?term=&department=ASIA&a=b"
puts string[/department=(\w+)/, 1] # => "ASIA"
or you can parse this as query(which is in my opinion more appropriate):
require 'cgi'
string = "/scripts/skw.asp?term=&department=ASIA&a=b"
query = string.split('?')[1] # => "term=&department=ASIA&a=b"
parsed_query = CGI::parse(query) # => {"term"=>[""], "department"=>["ASIA"], "a"=>["b"]}
puts parsed_query['department'][0] # => "ASIA"
2 Comments
Mark Thomas
This won't work if the department contains a non-word character such as a URL-encoded space, i.e.
&department=Customer%20Service.
Tom Gasaner
/department=(.+)&?/str = '/scripts/skw.asp?term=&department=ASIA'
You could use a capture group:
str[/\bdepartment=(.+$)/, 1]
#=> "ASIA"
or a positive lookbehind:
str[/(?<=\bdepartment=).+$/]
#=> "ASIA"
6 Comments
Tom Gasaner
Though it is an interesting approach, both solutions will return incorrect result if there would be another parameter in the end or if you just switch positions of
term and department in existing string like this /scripts/skw.asp?department=ASIA&term=
Cary Swoveland
I assumed from the question that the string ends with the substring to be returned. If that's not the case, one would need to know how it would be determined where the substring to be returned ends. That is not given in the question and I've made no assumptions about where the string came from or how it will be used.
pguardiario
I don't think that
\W does what you think it does. Maybe you had \b in mind.Cary Swoveland
@pguardiario, sorry for the late reply. I don't want
\b because "department" could be preceded by a non-alphameric character, as in the example. \W is not quite that, but I thought it was close enough.pguardiario
It's the difference between checking for the existence of a non-word char and checking for the non-existence of a word char.
\b is what you want there. |