Javascript Regular Expression Help Needed

[dwest100]

Hi,
I have the following e-mail validation code:

fValidate.prototype.email = function( level )
{
	if ( this.typeMismatch( 'text' ) ) return;
	if ( typeof level == 'undefined' ) level = 0;
	var emailPatterns = [
        /.+@.+\..+$/i,
		/^\w.+@\w.+\.[a-z]+$/i,
		/^\w[-_a-z~.]+@\w[-_a-z~.]+\.[a-z]{2}[a-z]*$/i,
		/^\w[\w\d]+(\.[\w\d]+)*@\w[\w\d]+(\.[\w\d]+)*\.[a-z]{2,7}$/i,
		];
	if ( ! emailPatterns[level].test( this.elem.value ) )
	{
		this.throwError();
	}	
}

It works fine but it does not validate ".con" as incorrect for ".com"

So, I found this regular expression which does validate for .com, .gov etc.:

/\b(^(\S+@).+((\.com)|(\.net)|(\.edu)|(\.mil)|(\.gov)|(\.org)|(\..{2,2}))$)\b/gi

However, if I simply replace the regular expressions in the first code sample with the second code sample, the first code sample breaks.

My question is how can I incorporate the extension checking capability of the second sample in to the first code sample?

Thanks much!

 

[BillyRayPreachersSon]

Would that validation routine work for addresses that don't end in 3 digit or 2,2 digit extensions? If not - consider domains things like ".tv", etc, too.

Hope this helps,
Dan

 

[dwest100]

Thanks. What I'm asking is how to incorporate all of those into what exists above. I don't know anything about reg ex.

Everything I've found on the web stops at the 2 digit, 3 digit mentality and considers it sufficient validation.

But that's not validated is it? After all one can't consider abc@def.con a valid e-mail address since it would not find a destination due to the mis-typing of the "m" at the end.

Any ideas??

 

[dwest100]

Well, as it turns out I just added the second code sample to the bottom of the first like so:

fValidate.prototype.email = function( level )
{
	if ( this.typeMismatch( 'text' ) ) return;
	if ( typeof level == 'undefined' ) level = 0;
	var emailPatterns = [
        /.+@.+\..+$/i,
		/^\w.+@\w.+\.[a-z]+$/i,
		/^\w[-_a-z~.]+@\w[-_a-z~.]+\.[a-z]{2}[a-z]*$/i,
		/^\w[\w\d]+(\.[\w\d]+)*@\w[\w\d]+(\.[\w\d]+)*\.[a-z]{2,7}$/i,
        /\b(^(\S+@).+((\.com)|(\.net)|(\.edu)|(\.mil)|(\.gov)|(\.org)|(\..{2,2}))$)\b/gi,
		];
	if ( ! emailPatterns[level].test( this.elem.value ) )
	{
		this.throwError();
	}	
}

And it works like a charm.

The first sample is set up so each filter is a "level" of strictness. So I set the strictness to level 4 after adding the last filter.

Don't know why but it works. ;-)

I just didn't want anyone wasting any time on this thread since I figured it out.

Thanks!!

 

[jemminger]

don't forget there are more top-level domains available, like ".family", ".shop" etc:

http://www.new.net/

=========================================================
-jeff

http://www.jeffemminger.com

try { succeed(); } catch(E) { tryAgain(); } finally { rtfm(); }

 

[dwest100]

Thanks!
I'll be sure to include them as well.
Regards