Function returns only allowed charachters from input string. Allowed charachters passes as array with ranges. Ranges are strings with start and end UFT-8 codes (as preg_match expects them).

sanitizeText.php

<?php
/**
 * Function filters out not-allowed symbols from string and return 'cleaned' string.
 * Allowed symbols specified as array of UTF-8 address ranges.
 * When it needs to change allowed range, consider this resource for all UTF-8 characters code sets: https://unicode-table.com/en/blocks/
 * @param string $text input string to filter
 * @param bool $stripTags remove HTML tags or not
 * @param array $allowed_utf8_ranges array with ranges of allowed symbols
 * @return mixed|null|string
 * @throws Exception
 */
function sanitizeText($text, $stripTags = TRUE, $allowed_utf8_ranges = ['\x{0000}-\x{007E}', '\x{00A1}-\x{017F}',]) {
  if (!is_array($allowed_utf8_ranges)) {
    throw new Exception('Third argument in sanitizeText function need to be an array.');
  }

  if (is_array($text)) {
    throw new Exception('First argument in sanitizeText need to be string, NOT array.');
  }

  if ($stripTags) {$text = filter_var($text, FILTER_SANITIZE_STRING, FILTER_FLAG_NO_ENCODE_QUOTES);} //Remove ALL html tags from string.

  //Here we will dynamically construct pattern for preg_replace. Example of needed pattern: '/([^(\x{0020}-\x{007E})|(\x{00A1}-\x{017F})])+/u'
  $dynamicPattern = '';
  foreach ($allowed_utf8_ranges as $range) {
    $dynamicPattern .= '('.$range.')|';
  }
  $dynamicPattern = rtrim($dynamicPattern, '|');
  $dynamicPattern = '/([^'.$dynamicPattern.'])+/u';

  $text = preg_replace($dynamicPattern, '', $text);

  return $text;
}