Question

I’m looking specifically for a regular expression that will grab the last term of a URL. This is not always a file name, it may not end in .html or .php, so I’ll need to make sure that the regular expression is grabbing the last term from the URL.

---

Example:

I need to grab http://www.mydomain.com/anything_can_be_here/thankyoupage

I need to extract “thankyoupage” even when there can be any term preceding it in the URL.

Also note, there is no file extension on the thankyoupage URL segment.

Answer 1

This should do it:

/^(?:http:\/\/)?(?:[^\/]+)\/.*?\/([^\/]+)(?:\?.*)?$/

For example, the result of this:

m = 'http://example.com/where/is?the=pancakes/house'.match(/^(?:http:\/\/)?(?:[^\/]+)\/.*?\/([^\/]+)(?:\?.*)?$/);

is this array:

["http://example.com/where/is?the=pancakes/house", "is"]

And this:

m = 'http://example.com/where/is'.match(/^(?:http:\/\/)?(?:[^\/]+)\/.*?\/([^\/]+)(?:\?.*)?$/)

Results in:

["http://example.com/where/is", "is"]

And this:

m = 'http://example.com/'.match(/^(?:http:\/\/)?(?:[^\/]+)\/.*?\/([^\/]+)(?:\?.*)?$/)

Results in null.

And your component is in m[1] and that comes from ([^\/]+). The (?:[^\/]+) will take care of the hostname (and the userinfo if it happens to be present), the (?:\?.*)?$ part will take care of any trailing CGI arguments.

Depending on your URLs, you could replace ^(?:http:\/\/)? with ^http:\/\/.