Regex Email validation

Question

I use this

@"^([\w\.\-]+)@([\w\-]+)((\.(\w){2,3})+)$"

regexp to validate the email

([\w\.\-]+) - this is for the first-level domain (many letters and numbers, also point and hyphen)

([\w\-]+) - this is for second-level domain

((\.(\w){2,3})+) - and this is for other level domains(from 3 to infinity) which includes a point and 2 or 3 literals

what's wrong with this regex?

EDIT:it doesn't match the "[e-mail]" email

Answer 1

Try this, it's working for me:

public bool IsValidEmailAddress(string s)
{
    if (string.IsNullOrEmpty(s))
        return false;
    else
    {
        var regex = new Regex(@"\w+([-+.']\w+)*@\w+([-.]\w+)*\.\w+([-.]\w+)*");
        return regex.IsMatch(s) && !s.EndsWith(".");
    }
}

Answer 2

Why not use EF6 attribute based e-mail validation?

As you can see above, Regex validation for e-mail always has some hole in it. If you are using EF6 data annotations, you can easily achieve reliable and stronger e-mail validation with EmailAddress data annotation attribute available for that. I had to remove the regex validation I used before for e-mail when I got mobile device specific regex failure on e-mail input field. When the data annotation attribute used for e-mail validation, the issue on mobile was resolved.

public class LoginViewModel
{
    [EmailAddress(ErrorMessage = "The email format is not valid")]
    public string Email{ get; set; }

Answer 3

TLD's like .museum aren't matched this way, and there are a few other long TLD's. Also, you can validate email addresses using the MailAddress class as Microsoft explains here in a note:

Instead of using a regular expression to validate an email address, you can use the System.Net.Mail.MailAddress class. To determine whether an email address is valid, pass the email address to the MailAddress.MailAddress(String) class constructor.

public bool IsValid(string emailaddress)
{
    try
    {
        MailAddress m = new MailAddress(emailaddress);

        return true;
    }
    catch (FormatException)
    {
        return false;
    }
}

This saves you a lot af headaches because you don't have to write (or try to understand someone else's) regex.

Answer 4

This one prevents invalid emails mentioned by others in the comments:

[e-mail]
[e-mail]
name@hotmail
[e-mail]
[e-mail]

It also prevents emails with double dots:

hello..world@example..com

Try testing it with as many invalid email addresses as you can find.

using System.Text.RegularExpressions;

public static bool IsValidEmail(string email)
{
    return Regex.IsMatch(email, @"\A[a-z0-9]+([-._][a-z0-9]+)*@([a-z0-9]+(-[a-z0-9]+)*\.)+[a-z]{2,4}\z")
        && Regex.IsMatch(email, @"^(?=.{1,64}@.{4,64}$)(?=.{6,100}$).*");
}

See validate email address using regular expression in C#.

Answer 5

Just let me know IF it doesn't work :)

public static bool isValidEmail(this string email)
{

    string[] mail = email.Split(new string[] { "@" }, StringSplitOptions.None);

    if (mail.Length != 2)
        return false;

    //check part before ...@

    if (mail[0].Length < 1)
        return false;

    System.Text.RegularExpressions.Regex regex = new System.Text.RegularExpressions.Regex(@"^[a-zA-Z0-9_\-\.]+$");
    if (!regex.IsMatch(mail[0]))
        return false;

    //check part after @...

    string[] domain = mail[1].Split(new string[] { "." }, StringSplitOptions.None);

    if (domain.Length < 2)
        return false;

    regex = new System.Text.RegularExpressions.Regex(@"^[a-zA-Z0-9_\-]+$");

    foreach (string d in domain)
    {
        if (!regex.IsMatch(d))
            return false;
    }

    //get TLD
    if (domain[domain.Length - 1].Length < 2)
        return false;

    return true;

}

Answer 6

Try this on for size:

public static bool IsValidEmailAddress(this string s)
{
    var regex = new Regex(@"[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?");
    return regex.IsMatch(s);
}