optimizing a php function that trims strings

i programmed this php function that takes any text/html string and trims it.

For example:

gen_string("Hello, how are you today?",10);

Returns: Hello, how...

The problem ar开发者_Python百科ises when the function string limit is the same as the position of a special character such as: á, ñ, etc...

In which case:

gen_string("Helló my friend",5);

Returns: Hell�...

Any ideas on how to solve this issue? This is the current function:

# string: advanced substr
function gen_string($string,$min,$clean=false) {
 $text = trim(strip_tags($string));
 if(strlen($text)>$min) {
  $blank = strpos($text,' ');
  if($blank) {
   # limit plus last word
   $extra = strpos(substr($text,$min),' ');
   $max = $min+$extra;
   $r = substr($text,0,$max);
   if(strlen($text)>=$max && !$clean) $r=trim($r,'.').'...';
  } else {
   # if there are no spaces
   $r = substr($text,0,$min).'...';
  }
 } else {
  # if original length is lower than limit
  $r = $text;
 }
 return trim($r);
}

Thanks!

---

You should use the multibyte string functions to correctly handle unicode characters.

For example you could try using mb_strimwidth to truncate a string to a specified length.

---

You could also take a different approach and make use of the PCRE regex extension's UTF-8 capabilities (assuming your strings are UTF-8!).

function gen_string($string, $length)
{
    $str    = trim(strip_tags($string));
    $strlen = strlen(utf8_decode($str));
    // String is less than limit
    if ($strlen <= $length) return $str;
    // Shorten string, preserving whole "words" (non-whitespace)
    preg_match('/^.{'.($length-1).'}\S*/su', $str, $match);
    // Append ellipsis if needed (bytes length is OK to check)
    if (strlen($match[0]) !== strlen($str)) $match[0] .= '...';
    return $match[0];
}

---

Aside from the multibyte issue, maybe you can write it shorter

function gen_string($str, $limit) {
    if ($str >= strlen($limit)) 
      return $str;
    $offset = -(strlen($str) - $limit);
    return substr($str, 0, strrpos($str, ' ', $offset)).'...';
}

It will limit the length of the string, so rather than cut it after the first word beyond the limit, it ensures that the length is never larger than the limit.

---

strlen() cannot be used for UTF-8 string, because it would count also the continuation characters, which should not be counted.

You can try with the following code:

define('PREG_CLASS_UNICODE_WORD_BOUNDARY', 
  '\x{0}-\x{2F}\x{3A}-\x{40}\x{5B}-\x{60}\x{7B}-\x{A9}\x{AB}-\x{B1}\x{B4}' .
  '\x{B6}-\x{B8}\x{BB}\x{BF}\x{D7}\x{F7}\x{2C2}-\x{2C5}\x{2D2}-\x{2DF}' .
  '\x{2E5}-\x{2EB}\x{2ED}\x{2EF}-\x{2FF}\x{375}\x{37E}-\x{385}\x{387}\x{3F6}' .
  '\x{482}\x{55A}-\x{55F}\x{589}-\x{58A}\x{5BE}\x{5C0}\x{5C3}\x{5C6}' .
  '\x{5F3}-\x{60F}\x{61B}-\x{61F}\x{66A}-\x{66D}\x{6D4}\x{6DD}\x{6E9}' .
  '\x{6FD}-\x{6FE}\x{700}-\x{70F}\x{7F6}-\x{7F9}\x{830}-\x{83E}' .
  '\x{964}-\x{965}\x{970}\x{9F2}-\x{9F3}\x{9FA}-\x{9FB}\x{AF1}\x{B70}' .
  '\x{BF3}-\x{BFA}\x{C7F}\x{CF1}-\x{CF2}\x{D79}\x{DF4}\x{E3F}\x{E4F}' .
  '\x{E5A}-\x{E5B}\x{F01}-\x{F17}\x{F1A}-\x{F1F}\x{F34}\x{F36}\x{F38}' .
  '\x{F3A}-\x{F3D}\x{F85}\x{FBE}-\x{FC5}\x{FC7}-\x{FD8}\x{104A}-\x{104F}' .
  '\x{109E}-\x{109F}\x{10FB}\x{1360}-\x{1368}\x{1390}-\x{1399}\x{1400}' .
  '\x{166D}-\x{166E}\x{1680}\x{169B}-\x{169C}\x{16EB}-\x{16ED}' .
  '\x{1735}-\x{1736}\x{17B4}-\x{17B5}\x{17D4}-\x{17D6}\x{17D8}-\x{17DB}' .
  '\x{1800}-\x{180A}\x{180E}\x{1940}-\x{1945}\x{19DE}-\x{19FF}' .
  '\x{1A1E}-\x{1A1F}\x{1AA0}-\x{1AA6}\x{1AA8}-\x{1AAD}\x{1B5A}-\x{1B6A}' .
  '\x{1B74}-\x{1B7C}\x{1C3B}-\x{1C3F}\x{1C7E}-\x{1C7F}\x{1CD3}\x{1FBD}' .
  '\x{1FBF}-\x{1FC1}\x{1FCD}-\x{1FCF}\x{1FDD}-\x{1FDF}\x{1FED}-\x{1FEF}' .
  '\x{1FFD}-\x{206F}\x{207A}-\x{207E}\x{208A}-\x{208E}\x{20A0}-\x{20B8}' .
  '\x{2100}-\x{2101}\x{2103}-\x{2106}\x{2108}-\x{2109}\x{2114}' .
  '\x{2116}-\x{2118}\x{211E}-\x{2123}\x{2125}\x{2127}\x{2129}\x{212E}' .
  '\x{213A}-\x{213B}\x{2140}-\x{2144}\x{214A}-\x{214D}\x{214F}' .
  '\x{2190}-\x{244A}\x{249C}-\x{24E9}\x{2500}-\x{2775}\x{2794}-\x{2B59}' .
  '\x{2CE5}-\x{2CEA}\x{2CF9}-\x{2CFC}\x{2CFE}-\x{2CFF}\x{2E00}-\x{2E2E}' .
  '\x{2E30}-\x{3004}\x{3008}-\x{3020}\x{3030}\x{3036}-\x{3037}' .
  '\x{303D}-\x{303F}\x{309B}-\x{309C}\x{30A0}\x{30FB}\x{3190}-\x{3191}' .
  '\x{3196}-\x{319F}\x{31C0}-\x{31E3}\x{3200}-\x{321E}\x{322A}-\x{3250}' .
  '\x{3260}-\x{327F}\x{328A}-\x{32B0}\x{32C0}-\x{33FF}\x{4DC0}-\x{4DFF}' .
  '\x{A490}-\x{A4C6}\x{A4FE}-\x{A4FF}\x{A60D}-\x{A60F}\x{A673}\x{A67E}' .
  '\x{A6F2}-\x{A716}\x{A720}-\x{A721}\x{A789}-\x{A78A}\x{A828}-\x{A82B}' .
  '\x{A836}-\x{A839}\x{A874}-\x{A877}\x{A8CE}-\x{A8CF}\x{A8F8}-\x{A8FA}' .
  '\x{A92E}-\x{A92F}\x{A95F}\x{A9C1}-\x{A9CD}\x{A9DE}-\x{A9DF}' .
  '\x{AA5C}-\x{AA5F}\x{AA77}-\x{AA79}\x{AADE}-\x{AADF}\x{ABEB}' .
  '\x{D800}-\x{F8FF}\x{FB29}\x{FD3E}-\x{FD3F}\x{FDFC}-\x{FDFD}' .
  '\x{FE10}-\x{FE19}\x{FE30}-\x{FE6B}\x{FEFF}-\x{FF0F}\x{FF1A}-\x{FF20}' .
  '\x{FF3B}-\x{FF40}\x{FF5B}-\x{FF65}\x{FFE0}-\x{FFFD}');

function utf8_strlen($text) {
  if (function_exists('mb_strlen')) {
    return mb_strlen($text);
  }

  // Do not count UTF-8 continuation bytes.
  return strlen(preg_replace("/[\x80-\xBF]/", '', $text));
}

function utf8_truncate($string, $max_length, $wordsafe = FALSE, $add_ellipsis = FALSE, $min_wordsafe_length = 1) {
  $ellipsis = '';
  $max_length = max($max_length, 0);
  $min_wordsafe_length = max($min_wordsafe_length, 0);

  if (utf8_strlen($string) <= $max_length) {
    // No truncation needed, so don't add ellipsis, just return.
    return $string;
  }

  if ($add_ellipsis) {
    // Truncate ellipsis in case $max_length is small.
    $ellipsis = utf8_substr('...', 0, $max_length);
    $max_length -= utf8_strlen($ellipsis);
    $max_length = max($max_length, 0);
  }

  if ($max_length <= $min_wordsafe_length) {
    // Do not attempt word-safe if lengths are bad.
    $wordsafe = FALSE;
  }

  if ($wordsafe) {
    $matches = array();
    // Find the last word boundary, if there is one within $min_wordsafe_length
    // to $max_length characters. preg_match() is always greedy, so it will
    // find the longest string possible.
    $found = preg_match('/^(.{' . $min_wordsafe_length . ',' . $max_length . '})[' . PREG_CLASS_UNICODE_WORD_BOUNDARY . ']/u', $string, $matches);
    if ($found) {
      $string = $matches[1];
    }
    else {
      $string = utf8_substr($string, 0, $max_length);
    }
  }
  else {
    $string = utf8_substr($string, 0, $max_length);
  }

  if ($add_ellipsis) {
    $string .= $ellipsis;
  }

  return $string;
}

function utf8_substr($text, $start, $length = NULL) {
  if (function_exists('mb_substr')) {
    return $length === NULL ? mb_substr($text, $start) : mb_substr($text, $start, $length);
  }
  else {
    $strlen = strlen($text);
    // Find the starting byte offset.
    $bytes = 0;
    if ($start > 0) {
      // Count all the continuation bytes from the start until we have found
      // $start characters or the end of the string.
      $bytes = -1;
      $chars = -1;
      while ($bytes < $strlen - 1 && $chars < $start) {
        $bytes++;
        $c = ord($text[$bytes]);
        if ($c < 0x80 || $c >= 0xC0) {
          $chars++;
        }
      }
    }
    elseif ($start < 0) {
      // Count all the continuation bytes from the end until we have found
      // abs($start) characters.
      $start = abs($start);
      $bytes = $strlen;
      $chars = 0;
      while ($bytes > 0 && $chars < $start) {
        $bytes--;
        $c = ord($text[$bytes]);
        if ($c < 0x80 || $c >= 0xC0) {
          $chars++;
        }
      }
    }
    $istart = $bytes;

    // Find the ending byte offset.
    if ($length === NULL) {
      $iend = $strlen;
    }
    elseif ($length > 0) {
      // Count all the continuation bytes from the starting index until we have
      // found $length characters or reached the end of the string, then
      // backtrace one byte.
      $iend = $istart - 1;
      $chars = -1;
      $last_real = FALSE;
      while ($iend < $strlen - 1 && $chars < $length) {
        $iend++;
        $c = ord($text[$iend]);
        $last_real = FALSE;
        if ($c < 0x80 || $c >= 0xC0) {
          $chars++;
          $last_real = TRUE;
        }
      }
      // Backtrace one byte if the last character we found was a real character
      // and we don't need it.
      if ($last_real && $chars >= $length) {
        $iend--;
      }
    }
    elseif ($length < 0) {
      // Count all the continuation bytes from the end until we have found
      // abs($start) characters, then backtrace one byte.
      $length = abs($length);
      $iend = $strlen;
      $chars = 0;
      while ($iend > 0 && $chars < $length) {
        $iend--;
        $c = ord($text[$iend]);
        if ($c < 0x80 || $c >= 0xC0) {
          $chars++;
        }
      }
      // Backtrace one byte if we are not at the beginning of the string.
      if ($iend > 0) {
        $iend--;
      }
    }
    else {
      // $length == 0, return an empty string.
      return '';
    }

    return substr($text, $istart, max(0, $iend - $istart + 1));
  }
}

---

For your return statement you could try:

return htmlspecialchars(trim($r));

EDIT: I tried your code as you provided it and it ran fine for me without having to use htmlspecialchars(). This is probably due to the face that in the <head> of the page the code was running on, the charset was set to UTF-8. So your options could be to set the encoding of the page like this:

<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">

or to use htmlspecialchars() as above.