voku\helper\shim\Iconv

Complexity

Total Complexity

155

Size/Duplication

Total Lines	839
Duplicated Lines	4.41 %

Coupling/Cohesion

Components	1
Dependencies	1

Test Coverage

Coverage

84.88%

Importance

Changes	5
Bugs	2	Features	1

20 Methods

Rating	Name	Duplication	Size	Complexity
A	getData()	9	9	2
A	ob_iconv_handler()	0	4	1
A	iconv_strlen()	0	8	2
A	strlen1()	3	10	4
A	qp_byte_callback()	0	4	1
C	iconv_mime_decode_headers()	0	31	8
C	iconv_mime_decode()	0	53	14
F	iconv()	0	87	22
B	loadMap()	0	18	5
B	map_to_utf8()	0	17	5
C	utf8_to_utf8()	0	40	13
C	map_from_utf8()	0	55	17
A	iconv_get_encoding()	0	17	4
A	iconv_set_encoding()	0	19	4
D	iconv_mime_encode()	0	75	15
C	iconv_strpos()	9	23	8
D	iconv_substr()	5	53	14
B	strlen2()	3	22	6
A	preg_offset()	0	12	2
B	iconv_strrpos()	8	17	8

<?php

/*
 * Copyright (C) 2013 Nicolas Grekas - p@tchwork.com
 *
 * This library is free software; you can redistribute it and/or modify it
 * under the terms of the (at your option):
 * Apache License v2.0 (http://apache.org/licenses/LICENSE-2.0.txt), or
 * GNU General Public License v2.0 (http://gnu.org/licenses/gpl-2.0.txt).
 */

namespace voku\helper\shim;

/**
 * iconv implementation in pure PHP, UTF-8 centric.
 *
 * Implemented:
 * - iconv              - Convert string to requested character encoding
 * - iconv_mime_decode  - Decodes a MIME header field
 * - iconv_mime_decode_headers - Decodes multiple MIME header fields at once
 * - iconv_get_encoding - Retrieve internal configuration variables of iconv extension
 * - iconv_set_encoding - Set current setting for character encoding conversion
 * - iconv_mime_encode  - Composes a MIME header field
 * - ob_iconv_handler   - Convert character encoding as output buffer handler
 * - iconv_strlen       - Returns the character count of string
 * - iconv_strpos       - Finds position of first occurrence of a needle within a haystack
 * - iconv_strrpos      - Finds the last occurrence of a needle within a haystack
 * - iconv_substr       - Cut out part of a string
 *
 * Charsets available for convertion are defined by files
 * in the charset/ directory and by Iconv::$alias below.
 * You're welcome to send back any addition you make.
 *
 * @package voku\helper\shim
 */
class Iconv
{
  const ERROR_ILLEGAL_CHARACTER = 'iconv(): Detected an illegal character in input string';
  const ERROR_WRONG_CHARSET     = 'iconv(): Wrong charset, conversion from `%s\' to `%s\' is not allowed';

  public static $input_encoding    = 'utf-8';
  public static $output_encoding   = 'utf-8';
  public static $internal_encoding = 'utf-8';
  public static $translit_map      = array();
  public static $convert_map       = array();
  public static $error_handler, $last_error, $is_valid_utf8;

It is generally advisable to only define one property per statement.

  public static $ulen_mask         = array(
      "\xC0" => 2,
      "\xD0" => 2,
      "\xE0" => 3,
      "\xF0" => 4,
  );
  protected static $alias = array(
      'utf8'        => 'utf-8',
      'ascii'       => 'us-ascii',
      'tis-620'     => 'iso-8859-11',
      'cp1250'      => 'windows-1250',
      'cp1251'      => 'windows-1251',
      'cp1252'      => 'windows-1252',
      'cp1253'      => 'windows-1253',
      'cp1254'      => 'windows-1254',
      'cp1255'      => 'windows-1255',
      'cp1256'      => 'windows-1256',
      'cp1257'      => 'windows-1257',
      'cp1258'      => 'windows-1258',
      'shift-jis'   => 'cp932',
      'shift_jis'   => 'cp932',
      'latin1'      => 'iso-8859-1',
      'latin2'      => 'iso-8859-2',
      'latin3'      => 'iso-8859-3',
      'latin4'      => 'iso-8859-4',
      'latin5'      => 'iso-8859-9',
      'latin6'      => 'iso-8859-10',
      'latin7'      => 'iso-8859-13',
      'latin8'      => 'iso-8859-14',
      'latin9'      => 'iso-8859-15',
      'latin10'     => 'iso-8859-16',
      'iso8859-1'   => 'iso-8859-1',
      'iso8859-2'   => 'iso-8859-2',
      'iso8859-3'   => 'iso-8859-3',
      'iso8859-4'   => 'iso-8859-4',
      'iso8859-5'   => 'iso-8859-5',
      'iso8859-6'   => 'iso-8859-6',
      'iso8859-7'   => 'iso-8859-7',
      'iso8859-8'   => 'iso-8859-8',
      'iso8859-9'   => 'iso-8859-9',
      'iso8859-10'  => 'iso-8859-10',
      'iso8859-11'  => 'iso-8859-11',
      'iso8859-12'  => 'iso-8859-12',
      'iso8859-13'  => 'iso-8859-13',
      'iso8859-14'  => 'iso-8859-14',
      'iso8859-15'  => 'iso-8859-15',
      'iso8859-16'  => 'iso-8859-16',
      'iso_8859-1'  => 'iso-8859-1',
      'iso_8859-2'  => 'iso-8859-2',
      'iso_8859-3'  => 'iso-8859-3',
      'iso_8859-4'  => 'iso-8859-4',
      'iso_8859-5'  => 'iso-8859-5',
      'iso_8859-6'  => 'iso-8859-6',
      'iso_8859-7'  => 'iso-8859-7',
      'iso_8859-8'  => 'iso-8859-8',
      'iso_8859-9'  => 'iso-8859-9',
      'iso_8859-10' => 'iso-8859-10',
      'iso_8859-11' => 'iso-8859-11',
      'iso_8859-12' => 'iso-8859-12',
      'iso_8859-13' => 'iso-8859-13',
      'iso_8859-14' => 'iso-8859-14',
      'iso_8859-15' => 'iso-8859-15',
      'iso_8859-16' => 'iso-8859-16',
      'iso88591'    => 'iso-8859-1',
      'iso88592'    => 'iso-8859-2',
      'iso88593'    => 'iso-8859-3',
      'iso88594'    => 'iso-8859-4',
      'iso88595'    => 'iso-8859-5',
      'iso88596'    => 'iso-8859-6',
      'iso88597'    => 'iso-8859-7',
      'iso88598'    => 'iso-8859-8',
      'iso88599'    => 'iso-8859-9',
      'iso885910'   => 'iso-8859-10',
      'iso885911'   => 'iso-8859-11',
      'iso885912'   => 'iso-8859-12',
      'iso885913'   => 'iso-8859-13',
      'iso885914'   => 'iso-8859-14',
      'iso885915'   => 'iso-8859-15',
      'iso885916'   => 'iso-8859-16',
  );

  /**
   * @param string $str
   * @param int    $mode
   * @param string $charset
   *
   * @return array|bool
   */
  public static function iconv_mime_decode_headers($str, $mode = 0, $charset = INF)
  {
    INF === $charset && $charset = self::$internal_encoding;

    false !== strpos($str, "\r") && $str = strtr(str_replace("\r\n", "\n", $str), "\r", "\n");
    $str = explode("\n\n", $str, 2);

    $headers = array();

    $str = preg_split('/\n(?![ \t])/', $str[0]);
    // $str as $str ??

56% of this comment could be valid code. Did you maybe forget this after debugging?

    foreach ($str as $str) {
      $str = self::iconv_mime_decode($str, $mode, $charset);

      if (false === $str) {
        return false;
      }
      $str = explode(':', $str, 2);

      if (2 === count($str)) {
        if (isset($headers[$str[0]])) {
          is_array($headers[$str[0]]) || $headers[$str[0]] = array($headers[$str[0]]);
          $headers[$str[0]][] = ltrim($str[1]);
        } else {
          $headers[$str[0]] = ltrim($str[1]);
        }
      }
    }

    return $headers;
  }

  /**
   * @param string $str
   * @param int    $mode
   * @param string $charset
   *
   * @return false|string
   */
  public static function iconv_mime_decode($str, $mode = 0, $charset = INF)
  {
    INF === $charset && $charset = self::$internal_encoding;

    if (ICONV_MIME_DECODE_CONTINUE_ON_ERROR & $mode) {
      $charset .= '//IGNORE';
    }

    false !== strpos($str, "\r") && $str = strtr(str_replace("\r\n", "\n", $str), "\r", "\n");
    $str = preg_split('/\n(?![ \t])/', rtrim($str), 2);
    $str = preg_replace('/[ \t]*\n[ \t]+/', ' ', rtrim($str[0]));
    $str = preg_split('/=\?([^?]+)\?([bqBQ])\?(.*?)\?=/', $str, -1, PREG_SPLIT_DELIM_CAPTURE);

    $result = self::iconv('utf-8', $charset, $str[0]);
    if (false === $result) {
      return false;
    }

    $i = 1;
    $len = count($str);

    while ($i < $len) {
      $c = strtolower($str[$i]);
      if ((ICONV_MIME_DECODE_CONTINUE_ON_ERROR & $mode)
          && 'utf-8' !== $c
          && !isset(self::$alias[$c])
          && !static::loadMap('from.', $c, $d)

$d is of type string|false, but the function expects a array.

      ) {
        $d = false;
      } elseif ('B' === strtoupper($str[$i + 1])) {
        $d = base64_decode($str[$i + 2]);
      } else {
        $d = rawurldecode(strtr(str_replace('%', '%25', $str[$i + 2]), '=_', '% '));
      }

      if (false !== $d) {
        $result .= self::iconv($c, $charset, $d);
        $d = self::iconv('utf-8', $charset, $str[$i + 3]);
        if ('' !== trim($d)) {
          $result .= $d;
        }
      } elseif (ICONV_MIME_DECODE_CONTINUE_ON_ERROR & $mode) {
        $result .= "=?{$str[$i]}?{$str[$i + 1]}?{$str[$i + 2]}?={$str[$i + 3]}";
      } else {
        $result = false;
        break;
      }

      $i += 4;
    }

    return $result;
  }

  /**
   * @param string $in_charset
   * @param string $out_charset
   * @param string $str
   *
   * @return string|false
   */
  public static function iconv($in_charset, $out_charset, $str)

Please use __construct() instead of a PHP4-style constructor that is named after the class.

  {
    if ('' === $str .= '') {
      return '';
    }

    // Prepare for //IGNORE and //TRANSLIT

    $TRANSLIT = $IGNORE = '';

    $out_charset = strtolower($out_charset);
    $in_charset = strtolower($in_charset);

    '' === $out_charset && $out_charset = 'iso-8859-1';
    '' === $in_charset && $in_charset = 'iso-8859-1';

    if ('//translit' === substr($out_charset, -10)) {
      $TRANSLIT = '//TRANSLIT';
      $out_charset = substr($out_charset, 0, -10);
    }

    if ('//ignore' === substr($out_charset, -8)) {
      $IGNORE = '//IGNORE';
      $out_charset = substr($out_charset, 0, -8);
    }

    '//translit' === substr($in_charset, -10) && $in_charset = substr($in_charset, 0, -10);
    '//ignore' === substr($in_charset, -8) && $in_charset = substr($in_charset, 0, -8);

    isset(self::$alias[$in_charset]) && $in_charset = self::$alias[$in_charset];
    isset(self::$alias[$out_charset]) && $out_charset = self::$alias[$out_charset];

    // Load charset maps

    if (
        ('utf-8' !== $in_charset && !static::loadMap('from.', $in_charset, $in_map))
        ||
        ('utf-8' !== $out_charset && !static::loadMap('to.', $out_charset, $out_map))
    ) {
      user_error(sprintf(self::ERROR_WRONG_CHARSET, $in_charset, $out_charset));

      return false;
    }


    if ('utf-8' !== $in_charset) {
      // Convert input to UTF-8
      $result = '';
      /** @noinspection PhpUndefinedVariableInspection */
      if (self::map_to_utf8($result, $in_map, $str, $IGNORE)) {

The variable $in_map does not seem to be defined for all execution paths leading up to this point.

        $str = $result;
      } else {
        $str = false;
      }
      self::$is_valid_utf8 = true;
    } else {
      self::$is_valid_utf8 = preg_match('//u', $str);

      if (!self::$is_valid_utf8 && !$IGNORE) {
        user_error(self::ERROR_ILLEGAL_CHARACTER);

        return false;
      }

      if ('utf-8' === $out_charset) {
        // UTF-8 validation
        $str = self::utf8_to_utf8($str, $IGNORE);
      }
    }

    if (
        'utf-8' !== $out_charset
        &&
        false !== $str
    ) {
      // Convert output to UTF-8
      $result = '';
      /** @noinspection PhpUndefinedVariableInspection */
      if (self::map_from_utf8($result, $out_map, $str, $IGNORE, $TRANSLIT)) {

The variable $out_map does not seem to be defined for all execution paths leading up to this point.

        return $result;
      } else {
        return false;
      }
    } else {
      return $str;
    }
  }

  /**
   * @param string $type
   * @param string $charset
   * @param array  $map
   *
   * @return bool
   */
  protected static function loadMap($type, $charset, &$map)
  {
    if (!isset(self::$convert_map[$type . $charset])) {
      if (false === $map = static::getData($type . $charset)) {
        if ('to.' === $type && static::loadMap('from.', $charset, $map)) {

$map is of type boolean, but the function expects a array.

          $map = array_flip($map);
        } else {
          return false;
        }
      }

      self::$convert_map[$type . $charset] = $map;
    } else {
      $map = self::$convert_map[$type . $charset];
    }

    return true;
  }

  /**
   * get data
   *
   * @param string $file
   *
   * @return bool|mixed
   */
  protected static function getData($file)

This method seems to be duplicated in your project.

  {
    $file = __DIR__ . '/charset/' . $file . '.ser';
    if (file_exists($file)) {
      return unserialize(file_get_contents($file));
    } else {
      return false;
    }
  }

  /**
   * map to utf8
   *
   * @param string $result
   * @param array  $map
   * @param string $str
   * @param string $IGNORE
   *
   * @return bool
   */
  protected static function map_to_utf8(&$result, $map, $str, $IGNORE)
  {
    $len = strlen($str);
    for ($i = 0; $i < $len; ++$i) {
      if (isset($str[$i + 1], $map[$str[$i] . $str[$i + 1]])) {
        $result .= $map[$str[$i] . $str[++$i]];
      } elseif (isset($map[$str[$i]])) {
        $result .= $map[$str[$i]];
      } elseif (!$IGNORE) {
        user_error(self::ERROR_ILLEGAL_CHARACTER);

        return false;
      }
    }

    return true;
  }

  /**
   * utf8 to utf8
   *
   * @param string $str
   * @param string $IGNORE
   *
   * @return false|string
   */
  protected static function utf8_to_utf8($str, $IGNORE)
  {
    $ulen_mask = self::$ulen_mask;
    $valid = self::$is_valid_utf8;

    $u = $str;
    $i = $j = 0;
    $len = strlen($str);

    while ($i < $len) {
      if ($str[$i] < "\x80") {
        $u[$j++] = $str[$i++];
      } else {
        $ulen = $str[$i] & "\xF0";
        $ulen = isset($ulen_mask[$ulen]) ? $ulen_mask[$ulen] : 1;
        $uchr = substr($str, $i, $ulen);

        if (1 === $ulen || !($valid || preg_match('/^.$/us', $uchr))) {
          if ($IGNORE) {
            ++$i;
            continue;
          }

          user_error(self::ERROR_ILLEGAL_CHARACTER);

          return false;
        } else {
          $i += $ulen;
        }

        $u[$j++] = $uchr[0];

        isset($uchr[1]) && 0 !== ($u[$j++] = $uchr[1])

The strict comparison !== seems to always evaluate to true as the types of 0 (integer) and $u[$j++] = $uchr[1] (string) can never be identical. Maybe you want to use a loose comparison != instead?

        && isset($uchr[2]) && 0 !== ($u[$j++] = $uchr[2])

The strict comparison !== seems to always evaluate to true as the types of 0 (integer) and $u[$j++] = $uchr[2] (string) can never be identical. Maybe you want to use a loose comparison != instead?

        && isset($uchr[3]) && 0 !== ($u[$j++] = $uchr[3]);

The strict comparison !== seems to always evaluate to true as the types of 0 (integer) and $u[$j++] = $uchr[3] (string) can never be identical. Maybe you want to use a loose comparison != instead?

      }
    }

    return substr($u, 0, $j);
  }

  /**
   * map from utf8
   *
   * @param string $result
   * @param array  $map
   * @param string $str
   * @param string $IGNORE
   * @param string $TRANSLIT
   *
   * @return bool
   */
  protected static function map_from_utf8(&$result, $map, $str, $IGNORE, $TRANSLIT)
  {
    $ulen_mask = self::$ulen_mask;
    $valid = self::$is_valid_utf8;

    if ($TRANSLIT) {
      self::$translit_map or self::$translit_map = static::getData('translit');

It seems like static::getData('translit') of type * is incompatible with the declared type array of property $translit_map.

Using logical operators such as or instead of || is generally not recommended.

    }

    $i = 0;
    $len = strlen($str);

    while ($i < $len) {
      if ($str[$i] < "\x80") {
        $uchr = $str[$i++];
      } else {
        $ulen = $str[$i] & "\xF0";
        $ulen = isset($ulen_mask[$ulen]) ? $ulen_mask[$ulen] : 1;
        $uchr = substr($str, $i, $ulen);

        if ($IGNORE && (1 === $ulen || !($valid || preg_match('/^.$/us', $uchr)))) {
          ++$i;
          continue;
        } else {
          $i += $ulen;
        }
      }

      if (isset($map[$uchr])) {
        $result .= $map[$uchr];
      } elseif ($TRANSLIT) {
        if (isset(self::$translit_map[$uchr])) {
          $uchr = self::$translit_map[$uchr];
        } elseif ($uchr >= "\xC3\x80") {
          $uchr = \Normalizer::normalize($uchr, \Normalizer::NFD);

          if ($uchr[0] < "\x80") {
            $uchr = $uchr[0];
          } elseif ($IGNORE) {
            continue;
          } else {
            return false;
          }
        }

        $str = $uchr . substr($str, $i);
        $len = strlen($str);
        $i = 0;
      } elseif (!$IGNORE) {
        return false;
      }
    }

    return true;
  }

  /**
   * iconv: get encoding
   *
   * @param string $type
   *
   * @return array|string
   */
  public static function iconv_get_encoding($type = 'all')
  {
    switch ($type) {
      case 'input_encoding'   :

There must be no space before the colon in a CASE statement

        return self::$input_encoding;
      case 'output_encoding'  :

There must be no space before the colon in a CASE statement

        return self::$output_encoding;
      case 'internal_encoding':
        return self::$internal_encoding;
    }

    return array(
        'input_encoding'    => self::$input_encoding,
        'output_encoding'   => self::$output_encoding,
        'internal_encoding' => self::$internal_encoding,
    );
  }

  /**
   * iconv: set encoding
   *
   * @param string $type
   * @param string $charset
   *
   * @return bool
   */
  public static function iconv_set_encoding($type, $charset)
  {
    switch ($type) {
      case 'input_encoding'   :

There must be no space before the colon in a CASE statement

        self::$input_encoding = $charset;
        break;
      case 'output_encoding'  :

There must be no space before the colon in a CASE statement

        self::$output_encoding = $charset;
        break;
      case 'internal_encoding':
        self::$internal_encoding = $charset;
        break;

      default:
        return false;
    }

    return true;
  }

  /**
   * iconv: mime encode
   *
   * @param string $field_name
   * @param string $field_value
   * @param array  $pref
   *
   * @return false|string
   */
  public static function iconv_mime_encode($field_name, $field_value, $pref = INF)
  {
    if (is_array($pref) === false) {
      $pref = array();
    }

    $pref = array_merge(
        array(
            'scheme'           => 'B',
            'input-charset'    => self::$internal_encoding,
            'output-charset'   => self::$internal_encoding,
            'line-length'      => 76,
            'line-break-chars' => "\r\n",
        ),
        $pref
    );

    preg_match('/[\x80-\xFF]/', $field_name) && $field_name = '';

    $scheme = strtoupper(substr($pref['scheme'], 0, 1));
    $in = strtolower($pref['input-charset']);
    $out = strtolower($pref['output-charset']);

    if ('utf-8' !== $in && false === $field_value = self::iconv($in, 'utf-8', $field_value)) {
      return false;
    }

    preg_match_all('/./us', $field_value, $chars);

    $chars = isset($chars[0]) ? $chars[0] : array();

    $line_break = (int)$pref['line-length'];
    $line_start = "=?{$pref['output-charset']}?{$scheme}?";
    $line_length = strlen($field_name) + 2 + strlen($line_start) + 2;
    $line_offset = strlen($line_start) + 3;
    $line_data = '';

    $field_value = array();

    $Q = 'Q' === $scheme;

    foreach ($chars as $c) {
      if ('utf-8' !== $out && false === $c = self::iconv('utf-8', $out, $c)) {
        return false;
      }

      $o = $Q
          ? $c = preg_replace_callback(
              '/[=_\?\x00-\x1F\x80-\xFF]/',
              array(
                  __CLASS__,
                  'qp_byte_callback',
              ),
              $c
          )
          : base64_encode($line_data . $c);

      if (isset($o[$line_break - $line_length])) {
        $Q || $line_data = base64_encode($line_data);
        $field_value[] = $line_start . $line_data . '?=';
        $line_length = $line_offset;
        $line_data = '';
      }

      $line_data .= $c;
      $Q && $line_length += strlen($c);
    }

    if ('' !== $line_data) {
      $Q || $line_data = base64_encode($line_data);
      $field_value[] = $line_start . $line_data . '?=';
    }

    return $field_name . ': ' . implode($pref['line-break-chars'] . ' ', $field_value);
  }

  /**
   * @param string $buffer
   * @param mixed  $mode
   *
   * @return string|false
   */
  public static function ob_iconv_handler($buffer, /** @noinspection PhpUnusedParameterInspection */ $mode)

The parameter $mode is not used and could be removed.

  {
    return self::iconv(self::$internal_encoding, self::$output_encoding, $buffer);
  }

  /**
   * iconv: strpos
   *
   * @param string $haystack
   * @param string $needle
   * @param int    $offset
   * @param string $encoding
   *
   * @return bool|int
   */
  public static function iconv_strpos($haystack, $needle, $offset = 0, $encoding = INF)
  {
    INF === $encoding && $encoding = self::$internal_encoding;

    if (0 !== strncasecmp($encoding, 'utf-8', 5)) {

This code seems to be duplicated across your project.

      if (false === $haystack = self::iconv($encoding, 'utf-8', $haystack)) {
        return false;
      }

      if (false === $needle = self::iconv($encoding, 'utf-8', $needle)) {
        return false;
      }
    }

    $offset = (int)$offset;

    if ($offset) {
      $haystack = self::iconv_substr($haystack, $offset, 2147483647, 'utf-8');
    }
    $pos = strpos($haystack, $needle);

    return false === $pos ? false : ($offset + ($pos ? self::iconv_strlen(substr($haystack, 0, $pos), 'utf-8') : 0));
  }

  /**
   * iconv: substr
   *
   * @param string $s
   * @param int    $start
   * @param int    $length
   * @param string $encoding
   *
   * @return false|string
   */
  public static function iconv_substr($s, $start, $length = 2147483647, $encoding = INF)
  {
    INF === $encoding && $encoding = self::$internal_encoding;
    if (0 === strncasecmp($encoding, 'utf-8', 5)) {

This code seems to be duplicated across your project.

      $encoding = INF;
    } elseif (false === $s = self::iconv($encoding, 'utf-8', $s)) {
      return false;
    }

    $s .= '';
    $slen = self::iconv_strlen($s, 'utf-8');
    $start = (int)$start;

    if (0 > $start) {
      $start += $slen;
    }

    if (0 > $start) {
      return false;
    }

    if ($start >= $slen) {
      return false;
    }

    $rx = $slen - $start;

    if (0 > $length) {
      $length += $rx;
    }

    if (0 === $length) {
      return '';
    }

    if (0 > $length) {
      return false;
    }

    if ($length > $rx) {
      $length = $rx;
    }

    $rx = '/^' . ($start ? self::preg_offset($start) : '') . '(' . self::preg_offset($length) . ')/u';

    $s = preg_match($rx, $s, $s) ? $s[1] : '';

    if (INF === $encoding) {
      return $s;
    } else {
      return self::iconv('utf-8', $encoding, $s);
    }
  }

  /**
   * iconv: strlen
   *
   * @param string $s
   * @param string $encoding
   *
   * @return bool|int
   */
  public static function iconv_strlen($s, $encoding = INF)
  {
    if (extension_loaded('xml')) {
      return self::strlen1($s, $encoding);
    } else {
      return self::strlen2($s, $encoding);
    }
  }

  /**
   * @param string $s
   * @param string $encoding
   *
   * @return bool|int
   */
  public static function strlen1($s, $encoding = INF)
  {
    INF === $encoding && $encoding = self::$internal_encoding;

    if (0 !== strncasecmp($encoding, 'utf-8', 5) && false === $s = self::iconv($encoding, 'utf-8', $s)) {

This code seems to be duplicated across your project.

      return false;
    }

    return strlen(utf8_decode($s));

It seems like $s defined by self::iconv($encoding, 'utf-8', $s) on line 784 can also be of type false; however, utf8_decode() does only seem to accept string, did you maybe forget to handle an error condition?

  }

  /**
   * @param string $s
   * @param string $encoding
   *
   * @return bool|int
   */
  public static function strlen2($s, $encoding = INF)
  {
    INF === $encoding && $encoding = self::$internal_encoding;

    if (0 !== strncasecmp($encoding, 'utf-8', 5) && false === $s = self::iconv($encoding, 'utf-8', $s)) {

This code seems to be duplicated across your project.

      return false;
    }

    $ulen_mask = self::$ulen_mask;

    $i = 0;
    $j = 0;
    $len = strlen($s);

    while ($i < $len) {
      $u = $s[$i] & "\xF0";
      $i += isset($ulen_mask[$u]) ? $ulen_mask[$u] : 1;
      ++$j;
    }

    return $j;
  }

  /**
   * @param int $offset
   *
   * @return string
   */
  protected static function preg_offset($offset)
  {
    $rx = array();
    $offset = (int)$offset;

    while ($offset > 65535) {
      $rx[] = '.{65535}';
      $offset -= 65535;
    }

    return implode('', $rx) . '.{' . $offset . '}';
  }

  /**
   * iconv: strrpos
   *
   * @param string $haystack
   * @param string $needle
   * @param string $encoding
   *
   * @return bool|int
   */
  public static function iconv_strrpos($haystack, $needle, $encoding = INF)
  {
    INF === $encoding && $encoding = self::$internal_encoding;

    if (0 !== strncasecmp($encoding, 'utf-8', 5)) {

This code seems to be duplicated across your project.

      if (false === $haystack = self::iconv($encoding, 'utf-8', $haystack)) {
        return false;
      }
      if (false === $needle = self::iconv($encoding, 'utf-8', $needle)) {
        return false;
      }
    }

    $pos = isset($needle[0]) ? strrpos($haystack, $needle) : false;

    return false === $pos ? false : self::iconv_strlen($pos ? substr($haystack, 0, $pos) : $haystack, 'utf-8');
  }

  /**
   * @param array $m
   *
   * @return string
   */
  protected static function qp_byte_callback($m)
  {
    return '=' . strtoupper(dechex(ord($m[0])));
  }
}