ASP: need regexp for various email formats

try this:

if( "^[a-z0-9_\+-]+(\.[a-z0-9_\+-]+)*@[a-z0-9-]+(\.[a-z0-9-]+)*\.([a-z]{2,4})$" )

taken from here:

http://www.markussipila.info/pub/emailvalidator.php?action=validate

Why not go by what is mentioned here:  http://www.regular-expressions.info/email.html

\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}\b

Select allOpen in new window

It's a bit generic, but I tend to agree with the author's logic regarding complex email validation. Besides, a regex can't tell you if the email address actually exists--only the receiving server can do that.

RegEx always takes me some time.   For validating emails I have just been checking for the @ and length greater than 3 with something on both sides of the @.   There are a lot of new gtld's coming on line this year to watch for.

http://www.regular-expressions.info/email.html

New gtld's http://www.newtldlist.com/

@Big Monty: this example wouldn't work on name.tac@fr.mac.moneys  but it did work for name.tac@fr.mac.mone

@kaufmed:      your solution unfortunately, did not work. not even for simple emails like test@aol.com  (weird)

Turn on case-insensitivity:

/\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}\b/i

Select allOpen in new window

Note the trailing "i" on the above expression.

e.g.

http://jsfiddle.net/d89gZ/

@kaufmed- that still didn't work. not even simple email addresses are accepted.

Did you check my sample code on JSFiddle? Every example you've given thus far validates in the sample. Updated to add additional prompts:

http://jsfiddle.net/d89gZ/1/

yes this worked for the types of emails i'm dealing with. thanks.

can you please explain what this regexp does and during what conditions it would fail?
i'm new to this.

^               - Matches start of string
[A-Z0-9._%+-]+  - Matches one or more ( + ) of any ( [...] ) letter ( A-Z ), number ( 0-9 ), period/full stop, underscore, percent, plus, or hypen ( ._%+- )
@               - Matches literal @
[A-Z0-9.-]+     - Matches one or more ( + ) of any ( [...] ) letter ( A-Z ), number ( 0-9 ), period/full stop ( . ), or hyphen ( - )
\.              - Matches literal period/full stop
[A-Z]{2,}       - Matches at least two ( {2,} ) of any ( [...] ) letter ( A-Z )
$               - Matches end of string
i               - Turns on case-insensitivity

Select allOpen in new window

The net effect is that you find one or more "words" each separated by a period/full stop (if present), then an @ symbol, then one or more "words" each separated by a period/full stop (if present), then the last period/full stop, then the TLD. The TLD must be at least two characters (e.g. .co, .us, .jp, etc.).

As far as when it "fails," I take it you mean what inputs will it reject? Anything with two @ symbols will be rejected. If the part trailing the @ does not have at least one dot, then that will be rejected. Anything with special characters other than period/full stop, underscore, percent, plus, or hyphen (e.g. $, ^, &, etc. ) will be rejected. (To be fair, & is actually a valid character in email addresses, as are many others. It is in less common usage, though.) If the TLD contains numbers, then it will be rejected.

Thank u for explanation

could i have also used
 if (/^\w+([\.-]?\w+)*@\w+([\.-]?\w+)*(\.\w{2,})+$/.test(args.Value)) {

seems to be this works just as well?

\w is shorthand for [a-zA-Z0-9_]. The equivalent to the pattern I suggested would be:

if (/^[\w.%+-]+@[\w.-]+\.[A-Z]{2,}$/i.test(args.Value)) {

Select allOpen in new window

thanks

@ kaufmed:
the only problem i find with
if (/^[\w.%+-]+@[\w.-]+\.[A-Z]{2,}$/i.test(args.Value)) {

is that this allows for multiple ".", so test...atol@aol.com is ok

whereas
 if (/^\w+([\.-]?\w+)*@\w+([\.-]?\w+)*(\.\w{2,})+$/.test(args.Value)) {
will catch that bc of  [\.-]?

For that reason, yes I agree. I was focused on the usage of \w, not the semantics of the overall pattern.

Thanks again.