Regular Expressions to find URL's in text

[SROSeaner]

I am working on an ASP page that parses text using the VBScript.RegExp
regular expression object. My reg expression right now is as follows:

[a-z]+\.[a-z]+\.[a-z]+/

And if find URL's no problem like: windowsupdate.microsoft.com,
www.cnn.com, etc.

But I need to also find any URL, like these:

www.amazon.com/books/atoz/index.html
OR
msdn.microsoft.com/newsgroups/default.aspx

Some URL with a deeper path than something.something.com if that makes
sense. Any ideas?

 

[David Morgan]

Nah...

What happens if someone writes a sentence and forgets to put a space between
the last word of the sentence, the period and the first word of the next
sentence?

URLs can take many forms and definitely don't need three parts. Some have
two some have four. What happens if someone puts in an IP address?

To get round the path/page name problem you should able to say where your
pattern matches anywhere in the string, not matches exactly.

Sorry to be the bearer of bad news.

 

[Larry Bud]

I am working on an ASP page that parses text using the VBScript.RegExp
regular expression object. My reg expression right now is as follows:

[a-z]+\.[a-z]+\.[a-z]+/

And if find URL's no problem like: windowsupdate.microsoft.com,
www.cnn.com, etc.

But I need to also find any URL, like these:

www.amazon.com/books/atoz/index.html
OR
msdn.microsoft.com/newsgroups/default.aspx

Some URL with a deeper path than something.something.com if that makes
sense. Any ideas?

Why don't you just parse it to the first / character, and see if that conforms?

 

[SROSeaner]

Thanks for your help. I got my parser to get all URL's in many forms
including IP addresses all from a disorganized html file. It is possible,
just a bugger to get going.

I am working on an ASP page that parses text using the VBScript.RegExp
regular expression object. My reg expression right now is as follows:

[a-z]+\.[a-z]+\.[a-z]+/

And if find URL's no problem like: windowsupdate.microsoft.com,
www.cnn.com, etc.

But I need to also find any URL, like these:

www.amazon.com/books/atoz/index.html
OR
msdn.microsoft.com/newsgroups/default.aspx

Some URL with a deeper path than something.something.com if that makes
sense. Any ideas?

Why don't you just parse it to the first / character, and see if that conforms?