And email validation is done with a regex… This low-effort post really doesn't deserve the front page…
The format for emails is surprisingly complex, and while you can use a regex to catch 99% of the right values, it's very difficult to get it 100% right. Often the best thing to do is just to try and send an email to the address a validation link.
Check that its a string containing '@' with at least one character before and after the @ symbol is usually the only sensible thing to do.
This would reject valid email addresses using quoted parts like john."@".doe@valid.example.
Edit: Looking at it closely it looks like the attempt is to reject anything with an '@' in the user part, which would be an error (e.g. john."@".doe@valid.example). Except since "\S+" also includes '@' this regex will still work, but by accident. It would reject the valid email john."\ ".doe@valid.example
/^[^@\s]+@\S+$/.test("john.\"@\".doe@valid.example") == true
- use a library that implements RFC 5322 properly.
- don't validate the email address (that is, don't try to reject most invalid addresses) and just check that the field contains at least one `@` to be sure the user didn't enter something else than his email address by accident.
Tying to perform email address validation with regex you're writing yourself is the best way to reject perfectly acceptable email address and get frustrated users.
https://lexi-lambda.github.io/blog/2019/11/05/parse-don-t-va...