[RESOLVED] ereg for multiple emails

ixalmida

Does anyone have a robust regular expression that will check a string with multiple email addresses? I'm way out of my league in trying to come up with one on my own. Or should I just continue to break the "TO" string into pieces and check each address individually?

The expression I've been using is:

ereg('^[!#$%&\'*+\\./0-9=?A-Z^_a-z{|}~]+'.'@'.'[-!#%&\'*+\\./0-9=?A-Z^_a-z{|}~]+\.'.'[-!#%&\'*+\\./0-9=?A-Z^_a-z{|}~]+$', $email)

It seems like maybe the current expression might even be overkill. I just want to rule out invalid characters and make sure there's an @ somewhere in the middle and a DOT-something at the end.

sneakyimp

If your current expression works for you, then stick with it! It's pretty easy to break a string into bits using [man]explode[/man] and then check each piece. Making a super-fancy regex pattern just to perform the same task just sounds like a pain in the ass.

nrg_alpha

I would first ditch ereg, as POSIX will not be included within the core of PHP as of version 6. Learn preg (PCRE) instead.

You can also have a look at this thread. Specifically, I provide links pointing to email validators..

What I always find bizarre is how people with regex problems always show the pattern, but not the example in which they are trying to extract from.. Provide a sample string of what you are dealing with (you can change the actual emails themselves for privacy sake), otherwise (and I seen it happen countless times), a back-and-forth problem solving dialogue exchanges due to poor communication from the get-go.

You can always do a quick and dirty method of extraction and use the links in the thread link I just provided to plug the extracted entries from and go from there.

example:

$emailString = 'Some text to:blah@blah.co.uk more text to:another-email@whatever.asia and finally, another email address: to:foo.bar@reallyFooBar.nz';
preg_match_all('#[^\s:]+@[^\s]+#', $emailString, $matches);
echo "<pre>".print_r($matches[0], true);

Granted, I'm making assumptions here (delimiting factors such as to: ) Without actual examples, it becomes a guessing game more than anything else.

ixalmida

Actually, it's just a list like you'd have in an actual To: line (no aliases):

someone@something.com; someone_else@something-else.com; someone.else@something_else.com; etc.

I will make a point to learn preg. The regex I'm using is just one I picked up off a Google search and it appears to choke on domain names with dashes.

ixalmida

Okay, I came up with this...

function validate_email($email)
  {
  // Get original number of addresses...
  $addr = explode(";", $email);
  $addr_count = count($addr);

  // Get number of valid addresses...
  preg_match_all('^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.(?:[A-Z]{2}|com|org|net|gov|mil|biz|info)$',
    $email, $matches);
  $good_count = count($matches);

  // Both counts should match...
  if($addr_count != $good_count)
    return False;
  else
    return True;
  }

Let me know what you think!

(Again, I had help coming up with the regex but this one actually made sense to me.)

Weedpacket

code$[/code]Top-level domains can be anything these days. Besides, you missed "int" anyway (as well as others from the current list).

No aliases? What about comments?

But here's something I came up with ... ah, seven years ago according to these notes ...
(?:(?:\r\n)?[ \t])(?:(?:(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t])))@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t])))|(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))<(?:(?:\r\n)?[ \t])(?:@(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}}](?:(?:\r\n)?[ \t])))(?:,@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t])))):(?:(?:\r\n)?[ \t]))?(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t])))@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t])))>(?:(?:\r\n)?[ \t]))|(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))}"(?:(?:\r\n)?[ \t])):(?:(?:\r\n)?[ \t])(?:(?:(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))}"(?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t])))@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t])))|(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))<(?:(?:\r\n)?[ \t])(?:@(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t])))(?:,@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}}](?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t])))):(?:(?:\r\n)?[ \t]))?(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t])))@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}}](?:(?:\r\n)?[ \t])))>(?:(?:\r\n)?[ \t]))(?:,\s(?:(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t])))@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t])))|(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))<(?:(?:\r\n)?[ \t])(?:@(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t])))(?:,@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t])))):(?:(?:\r\n)?[ \t]))?(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|"(?:[^{\"\r\]|\.|(?:(?:\r\n)?[} \t]))"(?:(?:\r\n)?[ \t])))@(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[]} \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)}](?:(?:\r\n)?[ \t]))(?:.(?:(?:\r\n)?[ \t])(?:[^{()<>@,;:\".[] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[["()<>@,;:\".[]]))|[([^{[]\r\]|\.)](?:(?:\r\n)?[} \t])))>(?:(?:\r\n)?[ \t]))))?;\s)

I figured it was better to write a proper parser instead.}}}}}}}

NogDog

Another option if the [man]filter[/man] extension is available:

$emails = preg_split('/\s*;\s*/', $email);
$valid = true;
$error = '';
foreach($emails as $addr)
{
   if(filter_var(filter_var($addr, FILTER_VALIDATE_EMAIL) == false)
   {
      $error .= "'$addr' is not a valid email address. ";
   }
}
if(!empty($error))
{
   echo $error;
}
else
{
   echo "The emails are OK.";
}

This would have the benefit of allowing you to specify which addresses are invalid.

nrg_alpha

preg_match_all('^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.(?:[A-Z]{2}|com|org|net|gov|mil|biz|info)$', 
    $email, $matches);

Problems with that pattern include (but are not necessarily limited to):

a) You are missing appropriate delimiters. After preg_match_all(' you start of with ^ (which the regex engine actually treats as the opening delimiter) while you intend to mean (from the start - and since you are doing a match all on a string, I would avoind ^ altogether). If you executed that script, you should get a 'No ending delimiter '^' found' warning. You can read more about delimiters here, and I would recommend testing your scripts.

b) What about email addresses ending in .co.uk, or .asia, or .mobi etc...

c) You pattern's character classes only accepts letters in uppercase? If you had proper delimiters set up, you could add the 'i' modifier to make letters case insensitive. You can read about modifiers here.

There's probably more issues. In either case, if you need something strict and reliable, I already posted the link to email parsers in one of my earlier posts.

There's simply too much room for error in my opinion.

ixalmida

Thanks for the detailed responses. Weedpacket, that regex is...well, wow!

I purposely cut out some of the odd and foreign domain extensions because it is strictly for North American enterprise businesses, which are more likely than not to be all .com's.

Regex has been one of the more difficult things for me to understand, particularly because there seems to be multiple versions with different syntax and multiple ways to accomplish the same thing even within a particular version. But I appreciate nrg_alpha's criticisms and links. I'll keep working on it.