PHP REGEX: Ignore + and #

Question

So here is what I'm doing exactly:

I have a form with a caption on the right of it, I want to write in the form only A-Z,0-9 and whitespaces, and for the caption I want to do the opposite, so if the user write's something wrong I can show what's the problem for example: "Invalid Charachter"

But I'm stuck with + and # I want to ignore them too from the form with regular expression so I can show the "Invalid character" message for these too, as I saw php thinks that + sign is = to space ( ) or what, but I need to ignore + and # signs too. This is my current code:

preg_match_all("/[^\w\s]/",$string,$matches);


foreach($matches[0] as $ic){
     if(strpos($str,$ic) || $str[0] == $ic){
          $fullname_error = "Invalid Character";
     }  
}

Valid strings:

• John Doe
• Mary Sue

Not valid strings:

• J#ohn Doe
• John&Doe
• John+Doe
• Mar@y+Sue
• !Mary Sue
• Mary Sue!

Answer 1

Try this:

<?
function checkString($str)
 {
 echo "Testing ".$str."<br />";
 // Check if there are invalid characters
 if (!preg_match("/^[a-zA-Z0-9\s]+$/", $str))
  {
  echo "Oh no! There are invalid characters! :(<br />";
  }
 else
  {
  echo "There is no invalid character!!! :)<br />";
  }

 // What are the invalid characters?
 if (preg_match("/[^(a-zA-Z0-9\s)]/", $str, $matches))
  {
  echo "Invalid character: ".$matches[0]."<br />";
  }
 }

checkString("This is a good string");
checkString("This is a not a good string$%#@#$"); 
?>

Answer 2

You could do something like this to handle the invalid characters:

$str = 'gum@#+boo';
if (preg_match_all('/[^\w\s]/u', $str, $matches)) {
    echo sprintf(
        '<p>Your input <b>%s</b> contains %d invalid character%s: <b>%s</b>.</p>',
        htmlspecialchars($str),
        count($matches[0]),
        count($matches[0]) > 1 ? '' : 's',
        implode('</b>, <b>', array_map('htmlspecialchars', array_unique($matches[0])))
    );
    echo '<p>Please choose a different input value.</p>';
}