<?php

/**
 * Truncate a UTF-8-encoded string safely to a number of bytes.
 *
 * If the end position is in the middle of a UTF-8 sequence, it scans backwards
 * until the beginning of the byte sequence.
 *
 * Use this function whenever you want to chop off a string at an unsure
 * location. On the other hand, if you're sure that you're splitting on a
 * character boundary (e.g. after using strpos() or similar), you can safely use
 * substr() instead.
 *
 * @param $string
 *   The string to truncate.
 * @param $len
 *   An upper limit on the returned string length.
 * @return
 *   The truncated string.
 */
function drupal_truncate_bytes($string, $len) {
  if (strlen($string) <= $len) {
     return $string;
  }
  if ((ord($string[$len]) < 0x80) || (ord($string[$len]) >= 0xC0)) {
    return substr($string, 0, $len);
  }
  while (--$len >= 0 && ord($string[$len]) >= 0x80 && ord($string[$len]) < 0xC0) {};
  return substr($string, 0, $len);
}


/**
 * Truncate a UTF-8-encoded string safely to a number of characters.
 *
 * @param $string
 *   The string to truncate.
 * @param $len
 *   An upper limit on the returned string length.
 * @param $wordsafe
 *   Flag to truncate at nearest space. Defaults to FALSE.
 * @param $dots
 *   Flag to add trailing dots. Defaults to FALSE.
 * @return
 *   The truncated string.
 */
function drupal_truncate_chars($string, $len, $wordsafe = FALSE, $dots = FALSE) {
  if (drupal_strlen($string) <= $len) {
    return $string;
  }
  if ($dots) {
    $len -= 4;
  }  
  if ($wordsafe) {
    $string = drupal_substr($string, 0, $len + 1); // leave one more character
    if ($last_space = strrpos($string, ' ')) { // space exists AND is not on position 0
      $string = substr($string, 0, $last_space);
    } else {
      $string = drupal_substr($string, 0, $len);
    }
  } else {
    $string = drupal_substr($string, 0, $len);
  }
  if ($dots) {
    $string .= ' ...';
  }  
  return $string;
}

// THIS IS ONLY FOR TESTING!!!
function truncate_utf8($string, $len, $wordsafe = FALSE, $dots = FALSE) {   
  return drupal_truncate_chars($string, $len, $wordsafe, $dots);
}