从字符串中检索完整的电子邮件地址

Question

I'm currently building a Slack bot using Laravel, and one of the features is that it can receive an email address and send a message to it.

The issue is that email addresses (eg bob@example.com ) come through as <mailto:bob@example.com|bob@example.com> from Slack.

I currently have a function that retrieves the email from this:

public function getEmail($string)
{
    $pattern = '/[a-z0-9_\-\+]+@[a-z0-9\-]+\.([a-z]{2,3})(?:\.[a-z]{2})?/i';
    preg_match_all($pattern, $string, $matches);
    $matches = array_filter($matches);

    return $matches[0][0];
}

This seemed to be working fine with email addresses like bob@example.com , however it seems to fail when working with email addresses like bob.jones@example.com (which would come through as <mailto:bob.jones@example.com|bob.jones@example.com> . In these cases, the function is returning jones@example.com as the email address.

I'm not great with regex, but is there something else I could use/change in my pattern, or a better way to fetch the email address from the string provided by Slack?

Answer 1

Could always take regex out of the equation if you know that's always the format it'll be in:

$testString = '<mailto:bob@example.com|bob@example.com>';

$testString = str_replace(['<mailto:', '>'], '', $testString);

$addresses = explode('|', $testString);

echo $addresses[0];

Answer 2

This method will do the job and you avoid to have regular expressions. and make sure the email being returned is a real email address by validating it with php functions.

function getEmailAddress($string) 
{
    $string = trim($string, '<>');
    $args = explode('|', $string);
    foreach ($args as $_ => $val) {
        if(filter_var($val, FILTER_VALIDATE_EMAIL) !== false) {
            return $val;
        }
    }

    return null;    
}

echo getEmailAddress('<mailto:bob@example.com|bob@example.com>');

Output

bob@example.com

Answer 3

You know the strings containing the e-mail address will always be of the form <mailto:bob@example.com|bob@example.com> , so use that. Specifically, you know the string will start with <mailto: , will contain a | , and will end with > .

An added difficulty though, is that the local part of an e-mail address may contain a pipe character as well, but the domain may not; see the following question.
What characters are allowed in an email address?

public function getEmail($string)
{
    $pattern = '/^<mailto:([^@]+@[^|]+)|(.*)>$/i';
    preg_match_all($pattern, $string, $matches);
    $matches = array_filter($matches);
    return $matches[1][0];
}

This matches the full line from beginning to end, but we capture the e-mail address within the first set of parentheses. $matches[1] contains all matches from the first capturing parentheses. You could use preg_match instead, since you're not looking for all matches, just the first one.