Ruby Regular expression to match a url

Question

Possible Duplicates:
Regex to match URL
regex to remove the webpage part of a url in ruby

I am in search of a regular expression for parsing all the urls in a file.
i tried many of the regular expression i got after googling but it fails in one or the other case . my idea is to write one which checks the presense of http or https at the begening and it will match everything untill it sees a blank space .
any ideas ?
NOTE : i dont need to parse the url but erase all the urls from a file or atleast make it unreadable .

P.S.: where do you see Rails here? I delete this tag. Do you know difference between Ruby and Rails? — Nakilon, Jan 17 '11 at 18:37
Yeah, the possible duplicates questions are hardly duplicates although one could find the good answer there `URI.parse` or `URI::DEFAULT_PARSER.make_regexp`. And I don't even see a reopen vote here. — akostadinov, Jan 28 '23 at 18:08
This one is for rails but still not rails only answers: https://stackoverflow.com/q/161738/520567 — akostadinov, Jan 28 '23 at 18:08

score 69 · Answer 1 · edited Dec 20 '14 at 18:27

69

The standard URI library provides URI.regexp which is the regular expression for url string.

 require 'uri'
 string.scan(URI.regexp)

http://ruby-doc.org/stdlib/libdoc/uri/rdoc/index.html

edited Dec 20 '14 at 18:27

rogerdpack

62,887
36
269
388

answered Jan 17 '11 at 18:40

John Dyer

1,221
1
9
13

10

NB there is also a URI.extract method which basically parses all url's out of a string. Very useful. – rogerdpack Nov 23 '13 at 14:26
undefined local variable or method `string' for main:Object (NameError) Did you mean? String – Chris Vilches Sep 06 '17 at 21:14
7

URI.regexp seems to match strings that aren't really valid URLs such as `http://x`, or even just `http:`. – Jon Schneider Sep 25 '17 at 19:43
`string` is the variable you create, @FeloVilches – Jason Deppen Oct 31 '17 at 15:04
3

This matches things like `C:` too – localhostdotdev Mar 29 '19 at 16:11

score 26 · Accepted Answer · answered Jan 17 '11 at 18:31

26

You can try this:

/https?:\/\/[\S]+/

The \S means any non-whitespace character.

(Rubular)

answered Jan 17 '11 at 18:31

Mark Byers

811,555
193
1,581
1,452

2

Is that really a uri regex though? That's just looking for any string that starts with http(s)... which is not a uri. – Mohamad Dec 31 '13 at 00:46
It won't work for the url `[https://google.com//]`. `[` is not a valid url character. – Dhurba Baral Dec 10 '18 at 04:31
It won't work for urls with quotes as it will include the quotes – Tom Lobato Jul 04 '22 at 03:32

Ruby Regular expression to match a url

2 Answers2

Linked