stringhelper.php

<?php
 
namespace Bitrix\Translate\Text;
 
use Bitrix\Main;
use Bitrix\Translate;
 
class StringHelper
{
    // utf8 https://www.w3.org/International/questions/qa-forms-utf-8.en
    public const UTF8_REGEXP = '/(?:
              [\x09\x0A\x0D\x20-\x7E]            # ASCII
            | [\xC2-\xDF][\x80-\xBF]             # non-overlong 2-byte
            | \xE0[\xA0-\xBF][\x80-\xBF]         # excluding overlongs
            | [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}  # straight 3-byte
            | \xED[\x80-\x9F][\x80-\xBF]         # excluding surrogates
            | \xF0[\x90-\xBF][\x80-\xBF]{2}      # planes 1-3
            | [\xF1-\xF3][\x80-\xBF]{3}          # planes 4-15
            | \xF4[\x80-\x8F][\x80-\xBF]{2}      # plane 16
        )+/xs';
 
    public static function getLength($str, $encoding = null)
    {
        if (Translate\Config::isUtfMode())
        {
            if (empty($encoding))
            {
                $encoding = Main\Localization\Translation::getCurrentEncoding();
            }
            return \mb_strlen($str, $encoding);
        }
 
        return \strlen($str);
    }
 
    public static function getSubstring($str, $start, $length, $encoding = null)
    {
        if (Translate\Config::isUtfMode())
        {
            if (empty($encoding))
            {
                $encoding = Main\Localization\Translation::getCurrentEncoding();
            }
            return \mb_substr($str, $start, $length, $encoding);
        }
 
        return \substr($str, $start, $length);
    }
 
    public static function getPosition($haystack, $needle, $offset = 0, $encoding = null)
    {
        if (\function_exists('mb_strpos'))
        {
            if (empty($encoding))
            {
                $encoding = Main\Localization\Translation::getCurrentEncoding();
            }
            return \mb_strpos($haystack, $needle, $offset, $encoding);
        }
 
        return \strpos($haystack, $needle, $offset);
    }
 
    public static function changeCaseToLower($str, $encoding = null)
    {
        if (\function_exists('mb_strtolower'))
        {
            if (empty($encoding))
            {
                $encoding = Main\Localization\Translation::getCurrentEncoding();
            }
            return \mb_strtolower($str, $encoding);
        }
 
        return \mb_strtolower($str);
    }
 
    public static function changeCaseToUpper($str, $encoding = null)
    {
        if (\function_exists('mb_strtoupper'))
        {
            if (empty($encoding))
            {
                $encoding = Main\Localization\Translation::getCurrentEncoding();
            }
            return \mb_strtoupper($str, $encoding);
        }
 
        return \mb_strtoupper($str);
    }
 
    public static function htmlSpecialChars($string, $flags = ENT_COMPAT, $encoding = null)
    {
        if (empty($encoding))
        {
            $encoding = Main\Localization\Translation::getCurrentEncoding();
        }
        return \htmlspecialchars($string, $flags, $encoding, true);
    }
 
    public static function validateUtf8OctetSequences($string)
    {
        return Main\Text\Encoding::detectUtf8($string, false);
    }
 
    public static function escapePhp($str, $enclosure = '"', $additional = ''): string
    {
        $w = '[a-zA-Z_\x7f-\xff][a-zA-Z0-9_\x7f-\xff]*';
        //Lookaround negative lookbehind (?<!ASD)
        if ($enclosure === "'")
        {
            $str = \preg_replace("/((?<![\\\\])['{$additional}]{1})/", "\\\\$1", $str);
            // \${end of str} -> \\
            $str = \preg_replace("/((?<![\\\\])\\\\‍)$/", "\\\\$1", $str);
        }
        elseif ($enclosure === '"')
        {
            // " -> \"
            $str = \preg_replace("/((?<![\\\\])[\"{$additional}]{1})/", "\\\\$1", $str);
            // $x -> \$x
            $str = \preg_replace("/((?<![\\\\])[\$]{1}$w)/", "\\\\$1", $str);
            // ${ -> \${
            $str = \preg_replace("/((?<![\\\\])[\$]{1}\s*\{)/", "\\\\$1", $str);
            // \${end of str} -> \\
            $str = \preg_replace("/((?<![\\\\])\\\\‍)$/", "\\\\$1", $str);
        }
        elseif ($enclosure === '<<<')
        {
            // $x -> \$x
            $str = \preg_replace("/((?<![\\\\])[\$]{1}$w)/", "\\\\$1", $str);
            // ${ -> \${
            $str = \preg_replace("/((?<![\\\\])[\$]{1}\s*\{)/", "\\\\$1", $str);
        }
 
        return $str;
    }
 
    public static function unescapePhp($str, $enclosure = '"'): string
    {
        //Lookaround positive lookbehind (?<=ASD)
        // (?<=[\\]+)['\"\\\$]{1}
 
        if ($enclosure == "'")
        {
            $from = ["\\'"];
            $to = ["'"];
        }
        else
        {
            $from = ["\\\$", "\\\""];
            $to = ["\$", "\""];
        }
 
        return \str_replace($from, $to, $str);
    }
 
    public static function hasPhpTokens($str, $enclosure = '"'): bool
    {
        $result = false;
        if (!empty($str) && is_string($str))
        {
            if ($enclosure == '<<<')
            {
                $validTokens = [\T_CONSTANT_ENCAPSED_STRING, \T_START_HEREDOC, \T_ENCAPSED_AND_WHITESPACE, \T_END_HEREDOC];
                $validChars = [];
                $tokens = \token_get_all('<'. "?php \$MESS = <<<'HTML'\n".  $str. "\nHTML;");
            }
            else
            {
                $validTokens = [\T_CONSTANT_ENCAPSED_STRING];
                $validChars = [$enclosure];
                $tokens = \token_get_all('<'. '?php $MESS = '. $enclosure. $str. $enclosure . ';');
            }
            $cnt = count($tokens);
            if ($cnt <= 5 || $cnt > 10)
            {
                return true;
            }
 
            for ($inx = 5, $cnt--; $inx < $cnt ; $inx++)
            {
                $token = $tokens[$inx];
                if (is_array($token))
                {
                    $token[] = \token_name($token[0]);
                    if (!in_array($token[0], $validTokens))
                    {
                        $result = true;
                        break;
                    }
                }
                elseif (is_string($token))
                {
                    if (!in_array($token, $validChars))
                    {
                        $result = true;
                        break;
                    }
                }
 
            }
        }
 
        return $result;
    }
}