problems with regex url validator

Question

I'm trying to create a regex to test if a url is valid or not. I had a good example to work off of, but I had to tweak it a bit to make it fit my purpose:

^(https?:\/\/)(www\.)?(\w*\.)+([\w\-_~:/?#[\]@!$&'()*+,;=.])*$

It works fine for the most part, but it matches the following, which drives me nuts:

http://www..example..com

I tried forever and I just can't get the magical combination of characters to get it to ignore the above use case. What am I doing wrong?

Here's a list of things I want the regex to match (all of them are matched):

http://www.example.com
https://www.example.com
https://www.example.com/
https://example.com/
https://blog.example.com/
https://my.blog.example.com/
https://my.blog.example.co.uk/
https://www.example.com/#test
https://www.example.com#test
https://www.example.com/test.php
https://www.example.com/test.php?test=yes&testmore=yesevenmore
https://www.example.com/test.php#test
https://www.example.com/test.php?test=yes&testmore2=yesevenmore&whatnumber=42#test
https://www.example.com/test
https://www.example.com/test/
https://www.example.com/test/?test=yes&testmore2=yesevenmore&whatnumber=42
https://www.example.com/test/#test
https://www.example.com/test/?test=yes&testmore=yesevenmore&whatnumber=42#test
https://www.example.com/test/?test=yes&testmore=yesevenmore&whatnumber=42#test
https://www.blog.example.com/test/?test=yes&testmore=yesevenmore&whatnumber=42#test
https://www.my.blog.example.com/test/?test=yes&testmore=yesevenmore&whatnumber=42#test
https://my.blog.example.co.uk/?test=yes&testmore=yesevenmore&whatnumber=42#test
http://255.255.255.255
http://www.example.com:8008
http://www.example.com:8008/test/?test=yes&testmore=yesevenmore&whatnumber=42#test

Here's a list of things I DON'T want it to match:

www.example.com
example.com
*http://www.blog..example..com
*http://www..example.com
*http://www...example.com
*http://www..example..com
http://www.example.com | not valid
http://www.example.com|
255.255.255.255

* still matched

How can I prevent regex from matching the multidots?

Do you mean like this? ^https?:\/\/(?:[-\w~:/?#[\]@!$&'()*+,;=]+\.)*[-\w~:/?#[\]@!$&'()*+,;=]+$ regex101.com/r/dF8zpI/1 — The fourth bird
– The fourth bird, Commented Sep 13, 2019 at 10:12
Just try to connect to the site and check the return status. — Toto
– Toto, Commented Sep 13, 2019 at 10:14
@Thefourthbird yes, thank you!!! That works for my use cases... — Synn Ko
– Synn Ko, Commented Sep 13, 2019 at 10:18
I think ^https?:\/\/(?:[^\s.|]+\.)*[^\s.|]+$ is what you need: only allow non-consecutive dots, no | and spaces in the URLs. See regex101.com/r/74BCXB/1 — Wiktor Stribiżew
– Wiktor Stribiżew, Commented Sep 13, 2019 at 10:31
Try ^https?:\/\/(?:www\.)?(?:[\w-]+\.)+[\w-]+(?:[:/#][-\w~:/?#[\]@!$&'()*+,;=.]*)?$, see regex101.com/r/HCB0Qt/1 — Wiktor Stribiżew
– Wiktor Stribiżew, Commented Sep 13, 2019 at 12:18

The fourth bird · Accepted Answer · 2019-09-13 12:34:06Z

1

Your pattern matches the dot literally \. as well as in the character class which is repeated 1+ times as a group and (\w*\.)+ also matches consecutive dots.

You could shorten the character class as some parts do not have to be escaped and \w also matches _

Using the characters from your character class that you accept to be valid you could repeat in a group matching what you want to allow excluding the dot and match a single dot at the end:

^https?:\/\/(?:[-\w~:/?#[\]@!$&'()*+,;=]+\.)*[-\w~:/?#[\]@!$&'()*+,;=]+$

That will match

^ Start of string
https?:\/\/ Match http:// or https://
(?: Non capturing group
- [-\w~:/?#[\]@!$&'()*+,;=]+\. Match 1+ times any of listed, then match a .
)* Close group and repeat 0+ times
[-\w~:/?#[\]@!$&'()*+,;=]+ Match any of the listed 1+ times (note that there is no .)
$ End of string

Regex demo

A more specific variant:

^https?:\/\/\w+(?:\.\w+)*(?:[/#:][-\w~:/?#[\]@!$&'()*+,;=.]*)?$

Regex demo

edited Sep 13, 2019 at 12:34

answered Sep 13, 2019 at 10:36

The fourth bird

165k16 gold badges61 silver badges75 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

Synn Ko Over a year ago

I think it's a good solution, but there is one flaw: you don't want to use the same character set for the domain part as for the rest of the URL, otherwise garbage like http://ww?w.example.com will be matched. I tried ^https?:\/\/(\w+.)*[-\w~:/?#[\]@!$&'()*+,;=]+$ and ^https?:\/\/([a-zA-Z0-9]+.)*[-\w~:/?#[\]@!$&'()*+,;=]+$, but they still match the ? for some reason...

Wiktor Stribiżew Over a year ago

@SynnKo You did not escape the dot.

The fourth bird Over a year ago

@SynnKo A more specific pattern could be ^https?:\/\/\w+(?:\.\w+)*(?:[/#:][-\w~:/?#[\]@!$&'()*+,;=.]*)?$ regex101.com/r/jDpE9P/1

Synn Ko Over a year ago

@Thefourthbird Right, forgot the dot... wouldn't have worked anyway, some of the previously matched test cases were no longer being matched. Your solution works like a charm, though! Many thanks.

Synn Ko Over a year ago

@Thefourthbird Small correction: the first * needs to be changed to a +, otherwise it will match stuff like www.

Collectives™ on Stack Overflow

problems with regex url validator

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related