tokenize with substring

tokenize.php

<?php 

function tokenize($str, $minLength = 3) {
  $strLength = mb_strlen($str);
  if ($strLength <= $minLength) {
    return [
      "[{$str}",
      "{$str}]",
      "[{$str}]"
    ];
  }
  
  $subLength = $minLength;
  $tokens = [];
  while ($subLength <= $strLength) {
    $start = 0;
    while (($start + $subLength) <= $strLength) {
      $token = mb_substr($str, $start, $subLength);
      if ($str !== $token) {
        if ($start === 0) {
          $tokens[] = '[' . $token;
        } elseif (($start + $subLength) === $strLength) {
          $tokens[] = $token . ']';
        } else {
          $tokens[] = $token;
        }
      } else {
        $tokens[] = "[{$token}";
        $tokens[] = "{$token}]";
        $tokens[] = "[{$token}]";
      }
      $start++;
    }
    $subLength++;
  }
  
  return $tokens;
}