Regex URL Path from URL

Question

I am having a little bit of regex trouble.

I am trying to get the path in this url videoplay.

http://video.google.co.uk:80/videoplay?docid=-7246927612831078230&hl=en#hello

If I use this regex /.+ it matches /video as well.

I would need some kind of anti / negative match to not include //

When I have to use regexes on urls fast and dirty, I usually include // at the beginning, before the capture group. Note you can't do http://, because they might be accessing it using a different protocol, or even ://, because they might specify the port number. — jwrush, Aug 19 '12 at 01:06
possible duplicate of [Getting parts of a URL (Regex)](http://stackoverflow.com/questions/27745/getting-parts-of-a-url-regex) — Raniz, Jun 04 '15 at 02:07

score 45 · Answer 1 · answered Jul 03 '15 at 05:22

In case if you need this for your JavaScript web-app: the best answer I ever found on this topic is here. Basic (and also original) version of the code looks like this:

var parser = document.createElement('a');
parser.href = "http://example.com:3000/pathname/?search=test#hash";

parser.protocol; // => "http:"
parser.hostname; // => "example.com"
parser.port;     // => "3000"
parser.pathname; // => "/pathname/"
parser.search;   // => "?search=test"
parser.hash;     // => "#hash"
parser.host;     // => "example.com:3000"

Thank you John Long, you made by day!

score 16 · Answer 2 · answered Aug 19 '14 at 08:54

16

(http[s]?:\/\/)?([^\/\s]+\/)(.*) group 3
Demo: http://regex101.com/r/vK4rV7/1

answered Aug 19 '14 at 08:54

M G

1,240
14
26

2

It wouldn't work if there for a path such as `www.abc.com?param=xyz`. I slightly modified it like this to make it work (I also use non-matching group for the first two groups). `(?:https?:\/\/)?(?:[^?\/\s]+[?\/])(.*)` Demo: https://regex101.com/r/eNUBb9 – nbeuchat May 24 '18 at 16:52

score 10 · Accepted Answer · answered Aug 25 '15 at 11:06

This expression gets everything after videoplay, aka the url path.

/\/(videoplay.+)/

This expression gets everything after the port. Also consisting of the path.

/\:\d./(.+)/

However If using Node.js I recommend the native url module.

var url = require('url')
var youtubeUrl = "http://video.google.co.uk:80/videoplay?docid=-7246927612831078230&hl=en#hello"
url.parse(youtubeUrl)

Which does all of the regex work for you.

{
  protocol: 'http:',
  slashes: true,
  auth: null,
  host: 'video.google.co.uk:80',
  port: '80',
  hostname: 'video.google.co.uk',
  hash: '#hello',
  search: '?docid=-7246927612831078230&hl=en',
  query: 'docid=-7246927612831078230&hl=en',
  pathname: '/videoplay',
  path: '/videoplay?docid=-7246927612831078230&hl=en',
  href: 'http://video.google.co.uk:80/videoplay?docid=-7246927612831078230&hl=en#hello' 
}

The `url` node module is in legacy mode. The docs recommend using the `URL` class instead. See here: https://nodejs.org/dist/latest-v14.x/docs/api/url.html#url_legacy_url_api — darksinge, Jul 28 '21 at 16:29

score 5 · Answer 4 · edited Jan 23 '20 at 13:42

function getPath(url, defaults){
    var reUrlPath = /(?:\w+:)?\/\/[^\/]+([^?#]+)/;
    var urlParts = url.match(reUrlPath) || [url, defaults];
    return urlParts.pop();
}
alert( getPath('http://stackoverflow.com/q/123/regex-url', 'unknown') );
alert( getPath('https://stackoverflow.com/q/123/regex-url', 'unknown') );
alert( getPath('//stackoverflow.com/q/123/regex-url', 'unknown') );
alert( getPath('http://stackoverflow.com/q/123/regex-url?foo', 'unknown') );
alert( getPath('http://stackoverflow.com/q/123/regex-url#foo', 'unknown') );
alert( getPath('http://stackoverflow.com/q/123/regex-url/', 'unknown') );
alert( getPath('http://stackoverflow.com/q/123/regex-url/?foo', 'unknown') );
alert( getPath('http://stackoverflow.com/q/123/regex-url/#foo', 'unknown') );
alert( getPath('http://stackoverflow.com/', 'unknown') );

score 4 · Answer 5 · edited May 23 '17 at 11:55

You can try this:

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

([^?]+) above is the capturing group which returns your path.

Please note that this is not an all-URL regex. It just solves your problem of matching all the text between the first "/" occurring after "//" and the following "?" character.

If you need an all-matching regex, you can check this StackOverflow link where they have discussed and dissected all possibilities of an URI into its constituent parts including your "path".
If you consider that an overkill AND if you know that your input URL will always follow a pattern of having your path between the first "/" and following "?", then the above regex should be sufficient.

Try this url: http://video.google.co.uk:80?docid=-7246927612831078230&hl=en#hello, this regex returns group1 = o — FiftiN, Apr 04 '19 at 12:13

score 3 · Answer 6 · answered Aug 11 '21 at 09:06

3

for new Googlers, use JavaScript web api URL at any environment:

new URL('your url string').pathname

https://developer.mozilla.org/en-US/docs/Web/API/URL/URL

answered Aug 11 '21 at 09:06

Mohammad Javad Khademian

508
1
4
16

1

This is beautiful. – suchislife Nov 22 '21 at 01:15
Regex URL Path from URL? – Max Barrass Jan 25 '22 at 16:39
He is asking about Regex not existing functions – Alin Mar 14 '22 at 15:29

score 2 · Answer 7 · answered Nov 10 '21 at 08:01

Even though the answers using language features are good, here is one more way to split URL to components using REGEXP:

^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\\?([^#]*))?(#(.*))?
     ||            |  |          |       |   |        | |
     12 - scheme   |  |          |       |   |        | |
                   3  4 - authority, includes hostname/ip and port number.
                                 5 - path|   |        | |
                                         6   7 - query| |
                                                      8 9 - fragment

Alin · Answer 8 · 2022-03-14T15:33:33.313

I have worked on it extensively and here is the result:

(?i)(?<scheme>http|https|ftp|sftp|sip|sips|file):\/\/(?:(?<username>[^`!@#$^&*()+=,:;'"{}\|\[\]\s\/\\]+)(?::(?<password>[^`!@#$^&*()+=,:;'"{}\|\[\]\s\/\\]+))?@)?(?:(?<ipv4>((?:(?:25[0-5]|2[0-4]\d|1?\d\d?)\.){3}(?:25[0-5]|2[0-4]\d|1?\d\d?)))|\[(?<ipv6>(?i)(?:[\da-f]{0,4}:){1,7}(?:(?<ipv4_in_ipv6>(?:(?:25[0-5]|2[0-4]\d|1?\d\d?)\.){3}(?:25[0-5]|2[0-4]\d|1?\d\d?))|[\da-f]{0,4}))\]|(?:(?<sub_domain>[^\s~`!@#$%^&*()_+=,.?:;'"{}\|\[\]\/\\]+\.)*(?<domain>[^\s~`!@#$%^&*()_+=,.?:;'"{}\|\[\]\/\\]+)(?<tld>\.[^\s~`!@#$%^&*()\-_+=,.?:;'"{}\|\[\]\/\\0-9]{2,})))+(?<port>:\d+)?(?:\/(?<path>\/?[^\s`@#$^&=.?"{}\\]+\/)*(?<file>[^\s`@#$^&=?"{}\/\\]+)?(?<query>\?[^\s`#$^"{}\\]+)*(?<fragment>#[^\s`$^&=?"{}\/\\]+)?)?

Demo | Git Repository

So, in your case, there is just a need to get the group contains the path and add the word you like, i.e. videoplay. To be more specific, I am talking about this:

(?:\/videoplay(?<path>\/?[^\s`@#$^&=.?"{}\\]+\/)*(?<file>[^\s`@#$^&=?"{}\/\\]+)?(?<query>\?[^\s`#$^"{}\\]+)*(?<fragment>#[^\s`$^&=?"{}\/\\]+)?)?

score 1 · Answer 9 · answered Aug 19 '12 at 01:06

1

You mean a negative lookbehind? (?<!/)

answered Aug 19 '12 at 01:06

Niet the Dark Absol

320,036
81
464
592

Peter · Answer 10 · 2020-05-05T18:55:55.160

var subject =
'<link rel="shortcut icon" href="https://cdn.sstatic.net/Sites/stackoverflow/img/favicon.ico?v=ec617d715196"><link rel="apple-touch-icon" href="https://cdn.sstatic.net/Sites/stackoverflow/img/apple-touch-icon.png?v=c78bd457575a"><link rel="image_src" href="https://cdn.sstatic.net/Sites/stackoverflow/img/apple-touch-icon.png?v=c78bd457575a">';
var re=/\"[a-z]+:\/\/[^ ]+"/m;
document.write(subject.match(re));

You can try this

/\"[a-z]+:\/\/[^ ]+/

Usage

if (/\"[a-z]+:\/\/[^ ]+/m.test(subject)) {  // Successful match } else {    // Match attempt failed }

score 0 · Answer 11 · answered Aug 19 '12 at 18:33

0

Its not a regex solution, but most languages have a URL library that will parse any URL into its constituent parts. This may be a better solution for what you are doing.

answered Aug 19 '12 at 18:33

Toby Allen

10,997
11
73
124

score -1 · Answer 12 · answered Nov 10 '21 at 07:33

-1

Please try this:

^http[s]?:\/\/(www\.)?(.*)?\/?(.)*

answered Nov 10 '21 at 07:33

Mohammad Hosseini

575
2
7

score -2 · Answer 13 · answered Aug 19 '12 at 11:29

-2

I think this is what you're after: [^/]+$

Demo: http://regex101.com/r/rG8gB9

answered Aug 19 '12 at 11:29

Firas Dib

2,743
19
38

4

This doesn't match the path of a URL, just the very last part of the path. With "http://google.com/foo/bar" it matches "bar" – justderb May 30 '14 at 19:24

Regex URL Path from URL

13 Answers13

Linked