How to match URL in c#?

Question

I have found many examples of how to match particular types of URL-s in PHP and other languages. I need to match any URL from my C# application. How to do this? When I talk about URL I talk about links to any sites or to files on sites and subdirectiories and so on.

I have a text like this: "Go to my awsome website http:\www.google.pl\something\blah\?lang=5" or else and I need to get this link from this message. Links can start only with www. too.

Solution 1

If you need to test your regex to find URLs you can try this resource

http://gskinner.com/RegExr/

It will test your regex while you're writing it.

In C# you can use regex for example as below:

Regex r = new Regex(@"(?<Protocol>\w+):\/\/(?<Domain>[\w@][\w.:@]+)\/?[\w\.?=%&=\-@/$,]*");
// Match the regular expression pattern against a text string.
Match m = r.Match(text);
while (m.Success) 
{
   //do things with your matching text 
   m = m.NextMatch();
}

Solution 2

Microsoft has a nice page of some regular expressions...this is what they say (works pretty good too)

^(ht|f)tp(s?)\:\/\/[0-9a-zA-Z]([-.\w]*[0-9a-zA-Z])*(:(0-9)*)*(\/?)([a-zA-Z0-9\-\.\?\,\'\/\\\+&%\$#_]*)?$

http://msdn.microsoft.com/en-us/library/ff650303.aspx#paght000001_commonregularexpressions

Solution 3

I am not sure exactly what you are asking, but a good start would be the Uri class, which will parse the url for you.

Solution 4

Here's one defined for URL's.

^http(s?)\:\/\/[0-9a-zA-Z]([-.\w]*[0-9a-zA-Z])*(:(0-9)*)*(\/?)([a-zA-Z0-9\-\.\?\,\'\/\\\+&%\$#_]*)?$

http://msdn.microsoft.com/en-us/library/ms998267.aspx

Solution 5

Regex regx = new Regex("http(s)?://([\\w+?\\.\\w+])+([a-zA-Z0-9\\~\\!\\@\\#\\$\\%\\^\\&\\*\\(\\)_\\-\\=\\+\\\\\\/\\?\\.\\:\\;\\'\\,]*)?", RegexOptions.IgnoreCase); 

View more solutions

Comments

[Patrick Klug]

+1 : although you forgot to add the @ symbol in front of the string.

[michele]

@Rumplin: can you explain the error you get? if i try with gskinner, the regex matches correctly test.com or do you intend th whole string with the '' ?

[Rumplin]

you have to look at the actual url, not the one shown. Stackoverflow corrects links... 'h t t p : / / h t t p : / /' without spaces

[Jared Updike]

This could be more specific: for example: var myUri = null; Uri.TryCreate(str, UriKind.Absolute, out myVar); and check if myUri changed from null to non-null.

[Wolf5]

That one does not include the query string behind the url. it stops at the first '=' letter. To fix, just add the single character '=' so it ends like this "....&%\$#_=]*)?$". Also, that regex pattern will not find an url in string. It will only tell you that the string is an url or not. To find it, ommit "^" at the start and "$" at the end. Regex.Matches(text, pattern) should then return all urls in the text.

[user]

Does it include mailto urls?

[Tom Rutchik]

this case fails: example.com:81/path/index.php?query=toto+le+heros#top It can be corrected with (ht|f)tp(s?)\:\/\/[0-9a-zA-Z]([-.\w]*[0-9a-zA-Z])*(:(0-9)*)*‌​(\/?)([a-zA-Z0-9\-\.‌​\?\,\'\/\\\+&%\$‌​#=_]*)? It also doesn't work for url containing a username@hostname