I have a url:
http://example.com/(S(4txk2wasxh3u0slptzi20qyj))/CWC_Link.aspx
but I only want to extract this portion:
(S(4txk2anwasxh3u0slptzi20qyj))/
Please, can anyone suggest me regex for this
3 Answers
The key point is to notice that the () characters mark the boundaries and that no / character is in the contents:
/(\(S\([^/()]+\)\))/
4 Comments
Seimen
There's a small mistake here, your first "/" in the regex results in
/(S(4txk2anwasxh3u0slptzi20qyj))/.
quetzalcoatl
@Simon: there is no mistake. Those are the bounding characters. Note that there's a capturing group in that expression. After applying the regex you should read the $1 match (first capture,
(S(4txk2anwasxh3u0slptzi20qyj))), not $0 (whole match /(S(4txk2anwasxh3u0slptzi20qyj))/). Without that bounding characters, if you pass an url of http://farmer.gov.in/asdada(S(foo))asdasd/(S(key))/asdasdasd you might catch the 'foo' instead of the 'key'. But of course that's anyways so improbable that you can probably safely remove the extra bounding '/'s.Seimen
Even if the answer is already ticked as solution, there's still a mistake @quetzalcoatl because then your capturing group does not include the forward slash at the end of the string which was described by the OP as desired result. Also you won't catch the foo if you restrict your regex to match for a forward slash after the bracket.
quetzalcoatl
Also you won't catch the foo if you restrict your regex to match for a forward slash - this is exactly why I included a '/' at both sides. Compare that to your regex that is capable of capturing many more false positives. As to the tail, I've completely intentionally left the trailing '/' off the regex, because I take it as a typo on the OP side, because he clearly wanted to catch the 'magic string' from the URL. He didn't complain mind you.Here's your regex. The part in braces will extract needed fragment
/^.+\/([^\/]+)\/.+$/
Basically, the logic is simple:
^ - marks beginning of the string
.+\/ - matches all symbols before the next part. This part of regex is composed taking into account default "greedy" behaviour of regexes, so this part matches http://farmer.gov.in/ in your example
([^\/]+) - matches all symbols between two slashes
\/.+$ - matches all symbols till the end of the string
Example with PHP language:
<?php
$string = "http://farmer.gov.in/(S(4txk2wasxh3u0slptzi20qyj))/CWC_Link.aspx";
$regex = "/^.+\/([^\/]+)\/.+$/";
preg_match($regex, $string, $matches);
var_dump($matches);
?>
In the output $matches[1] will have your needed value (S(4txk2wasxh3u0slptzi20qyj))
Comments
This regex does the job:
\(.*\)\/
Just match an opening bracket, then anything until a closing bracket with a forward slash.
parse_urlfunction. Perl:URImodule. Ruby:URImodule. .NET: 'Uri' class