How to accept port number in this regex?

Question

This is url validation regex

/^(https?:\/\/)?([\da-z\.-]+)\.([a-z\.]{2,6})([\/\w \.-]*)*\/?$/

This works Okay. However if I type url with port number such as http://localhost:3000, it gives an error.

How can I change this expression to accept a port number as well?

I don't think there's anything in the original regexp that's specific to PHP. It should work as is. — Barmar, Apr 17 '23 at 05:06
Why do you think `http` by itself should match? If you want to match a URL, you need more than just the protocol name. — Barmar, Apr 17 '23 at 05:08
Can you post the exact PHP code that prints true for `"http"`? From a quick glance, the `.` in this regex looks to be a required character. — Dogbert, Apr 17 '23 at 05:08
You don't need to escape `.` inside `[]` in either language. — Barmar, Apr 17 '23 at 05:10
@GoonGamja Yes. But nothing makes the rest of the URL optional. It needs to be something like `http://foo.bar.com/x/y/z` — Barmar, Apr 17 '23 at 05:26
Your PHP regexp works with JS: https://regex101.com/r/a4RtLl/1 — Barmar, Apr 17 '23 at 05:27
@GoonGamja Don't use regex to validate URL. There are better ways. See https://stackoverflow.com/questions/5717093/check-if-a-javascript-string-is-a-url — nice_dev, Apr 17 '23 at 06:43
Why is this tagged with JS and PHP, but does not contain any such code? — Nico Haase, Apr 17 '23 at 06:57
@NicoHaase when I edit my question, forgot to remove the tag — GoonGamja, Apr 17 '23 at 07:08

Shubham Nikam · Accepted Answer · 2023-04-17T07:13:36.707

Please can try this one

https?:\/\/(?:w{1,3}\.)?[^\s.]+(?:\.[a-z]+)*(?::\d+)?(?![^<]*(?:<\/\w+>|\/?>))

https?: This part matches the protocol of the URL. The s is optional, so it matches both http and https.

:\/\/: This matches the colon and two forward slashes that are part of the protocol.

(?:w{1,3}\.)?: This matches the optional www subdomain, if it exists. The (?: ) creates a non-capturing group, and the {1,3} quantifier matches between 1 and 3 instances of the letter "w".

[^\s.]+: This matches the domain name, which consists of one or more characters that are not whitespace or a dot.

(?:\.[a-z]+)*: This matches any optional subdomains that come before the domain, separated by dots. The (?: ) creates another non-capturing group, and the * quantifier matches zero or more instances of this group. The [a-z]+ matches one or more lowercase letters.

(?::\d+)?: This matches an optional port number that might come after the domain, preceded by a colon. The (?: ) creates another non-capturing group, and the ? quantifier makes it optional. The : matches the colon, and \d+ matches one or more digits.

(?![^<]*(?:<\/\w+>|\/?>)): This is a negative lookahead that ensures that the URL is not part of an HTML tag. It checks if there are no characters before the URL that match the pattern (?:<\/\w+>|\/?>), which matches any closing HTML tag or an empty tag.

Please add some explanation to your answer such that others can learn from it. What did you change, and why? — Nico Haase, Apr 17 '23 at 06:57

How to accept port number in this regex?

1 Answers1