mb_detect_encoding - PHP code example

PHP Function MB_Detect_Encoding

php function mb_detect_encoding is a multibyte string (mbstring) extension that provides conversions and judgments between different character encodings. The multi-byte string (mbstring) extension allows manipulation of strings that contain characters from Asian scripts, Emojis, and thousands of other symbols that cannot be assigned or fit into a single byte. It also supports converting to and from multiple character encodings, including Unicode-based encodings.

mb_detect_encoding() detects the most likely character encoding for a stringstring. It does this by looking at sequences of bytes in a string and then judging which ones are most likely to represent a valid character in the given encoding. The detection is not 100% reliable, however, since not every sequence of bytes can be recognized as a valid sequence in any encoding. It is always preferable to use an indication of character encoding stored or transmitted with the data, such as in a content-type HTTP header.

In PHP 8.2, mb_detect_encoding() is deprecated and replaced by the dedicated mb_convert_encoding() function, as well as the built-in functions for encoding/decoding Base64, Quoted-Printable, Uuencode, and HTML Entities. These new dedicated functions do not require the mbstring extension, so there are no deprecation warnings when using them.

mb_detect_encoding() can reveal sensitive parameters in stack traces when used without sanitization. PHP 8.2 introduces a new function to mask these values, making it possible to debug a function without showing its callers' credentials. This is a valuable feature to have in any application, especially when dealing with user-supplied data.

mb_detect_encoding example

$ofs += ('A' === $c[2]) ? -1 : 1;
                    $ofs = ($numMatches + $ofs) % $numMatches;
                }
            } elseif (\ord($c) < 32) {
                if ("\t" === $c || "\n" === $c) {
                    if ($numMatches > 0 && -1 !== $ofs) {
                        $ret = (string) $matches[$ofs];
                        // Echo out remaining chars for current match                         $remainingCharacters = substr($ret, \strlen(trim($this->mostRecentlyEnteredValue($fullChoice))));
                        $output->write($remainingCharacters);
                        $fullChoice .= $remainingCharacters;
                        $i = (false === $encoding = mb_detect_encoding($fullChoice, null, true)) ? \strlen($fullChoice) : mb_strlen($fullChoice, $encoding);

                        $matches = array_filter(
                            $autocomplete($ret),
                            fn ($match) => '' === $ret || str_starts_with($match, $ret)
                        );
                        $numMatches = \count($matches);
                        $ofs = -1;
                    }

                    if ("\n" === $c) {
                        $output->write($c);

    public function setContentDisposition(string $disposition, string $filename = '', string $filenameFallback = ''): static
    {
        if ('' === $filename) {
            $filename = $this->file->getFilename();
        }

        if ('' === $filenameFallback && (!preg_match('/^[\x20-\x7e]*$/', $filename) || str_contains($filename, '%'))) {
            $encoding = mb_detect_encoding($filename, null, true) ?: '8bit';

            for ($i = 0, $filenameLength = mb_strlen($filename, $encoding); $i < $filenameLength; ++$i) {
                $char = mb_substr($filename, $i, 1, $encoding);

                if ('%' === $char || \ord($char) < 32 || \ord($char) > 126) {
                    $filenameFallback .= '_';
                } else {
                    $filenameFallback .= $char;
                }
            }
        }

    public function isSingleCommand(): bool
    {
        return $this->singleCommand;
    }

    private function splitStringByWidth(string $string, int $width): array
    {
        // str_split is not suitable for multi-byte characters, we should use preg_split to get char array properly.         // additionally, array_slice() is not enough as some character has doubled width.         // we need a function to split string not by character count but by string width         if (false === $encoding = mb_detect_encoding($string, null, true)) {
            return str_split($string, $width);
        }

        $utf8String = mb_convert_encoding($string, 'utf8', $encoding);
        $lines = [];
        $line = '';

        $offset = 0;
        while (preg_match('/.{1,10000}/u', $utf8String, $m, 0, $offset)) {
            $offset += \strlen($m[0]);

if ($this->isDecorated()) {
            foreach ($lines as $i => $line) {
                $lines[$i] = $this->styleStack->getCurrent()->apply($line);
            }
        }

        return implode("\n", $lines);
    }

    private function addLineBreaks(string $text, int $width): string
    {
        $encoding = mb_detect_encoding($text, null, true) ?: 'UTF-8';

        return b($text)->toCodePointString($encoding)->wordwrap($width, "\n", true)->toByteString($encoding);
    }
}

'µ', 'ſ', "\xCD\x85", 'ς', "\xCF\x90", "\xCF\x91", "\xCF\x95", "\xCF\x96", "\xCF\xB0", "\xCF\xB1", "\xCF\xB5", "\xE1\xBA\x9B", "\xE1\xBE\xBE"],
        ['μ', 's', 'ι',        'σ', 'β',        'θ',        'φ',        'π',        'κ',        'ρ',        'ε',        "\xE1\xB9\xA1", 'ι'],
    ];

    private static $encodingList = ['ASCII', 'UTF-8'];
    private static $language = 'neutral';
    private static $internalEncoding = 'UTF-8';

    public static function mb_convert_encoding($s, $toEncoding, $fromEncoding = null)
    {
        if (\is_array($fromEncoding) || (null !== $fromEncoding && false !== strpos($fromEncoding, ','))) {
            $fromEncoding = self::mb_detect_encoding($s, $fromEncoding);
        } else {
            $fromEncoding = self::getEncoding($fromEncoding);
        }

        $toEncoding = self::getEncoding($toEncoding);

        if ('BASE64' === $fromEncoding) {
            $s = base64_decode($s);
            $fromEncoding = $toEncoding;
        }

if (\is_array($value)) {
                $array[$key] = $this->convert($value);
            }

            /** @var list<string> $encodings */
            $encodings = mb_detect_order();
            // NEXT-21735 - This is covered randomly             // @codeCoverageIgnoreStart             if (\is_string($value)) {
                if (!ctype_print($value) && mb_strlen($value) === 16) {
                    $array[$key] = sprintf('ATTENTION: Converted binary string by the "%s": %s', self::class, bin2hex($value));
                } elseif (!mb_detect_encoding($value, $encodings, true)) {
                    $array[$key] = mb_convert_encoding($value, 'UTF-8', 'ISO-8859-1');
                }
            }
            // @codeCoverageIgnoreEnd         }

        return $array;
    }
}

foreach ($messages as $message) {
            ++$count[$message['state']];
        }

        return $count;
    }

    private function sanitizeString(string $string, int $length = 80): string
    {
        $string = trim(preg_replace('/\s+/', ' ', $string));

        if (false !== $encoding = mb_detect_encoding($string, null, true)) {
            if (mb_strlen($string, $encoding) > $length) {
                return mb_substr($string, 0, $length - 3, $encoding).'...';
            }
        } elseif (\strlen($string) > $length) {
            return substr($string, 0, $length - 3).'...';
        }

        return $string;
    }
}

/** * Retrieve token type. Also processes the token value if necessary. * * @param string $value * @throws \InvalidArgumentException * @return integer */
    protected function getType(&$value): int
    {
        $encoded = $value;

        if (mb_detect_encoding($value, 'auto', true) !== 'UTF-8') {
            $encoded = mb_convert_encoding($value, 'UTF-8', 'Windows-1252');
        }

        if ($this->isValid($encoded)) {
            return $this->charValue[$encoded];
        }

        if ($this->isNullType($encoded)) {
            return self::C_NUL;
        }

function _wp_json_convert_string( $input_string ) {
    static $use_mb = null;
    if ( is_null( $use_mb ) ) {
        $use_mb = function_exists( 'mb_convert_encoding' );
    }

    if ( $use_mb ) {
        $encoding = mb_detect_encoding( $input_string, mb_detect_order(), true );
        if ( $encoding ) {
            return mb_convert_encoding( $input_string, 'UTF-8', $encoding );
        } else {
            return mb_convert_encoding( $input_string, 'UTF-8', 'UTF-8' );
        }
    } else {
        return wp_check_invalid_utf8( $input_string, true );
    }
}

/** * Prepares response data to be serialized to JSON. * * This supports the JsonSerializable interface for PHP 5.2-5.3 as well. * * @ignore * @since 4.4.0 * @deprecated 5.3.0 This function is no longer needed as support for PHP 5.2-5.3 * has been dropped. * @access private * * @param mixed $data Native representation. * @return bool|int|float|null|string|array Data ready for `json_encode()`. */

if (\in_array($fromEncoding, [null, 'utf8', 'utf-8', 'UTF8', 'UTF-8'], true) && preg_match('//u', $this->string)) {
            $u->string = $this->string;

            return $u;
        }

        set_error_handler(static fn ($t, $m) => throw new InvalidArgumentException($m));

        try {
            try {
                $validEncoding = false !== mb_detect_encoding($this->string, $fromEncoding ?? 'Windows-1252', true);
            } catch (InvalidArgumentException $e) {
                if (!\function_exists('iconv')) {
                    throw $e;
                }

                $u->string = iconv($fromEncoding ?? 'Windows-1252', 'UTF-8', $this->string);

                return $u;
            }
        } finally {
            restore_error_handler();
        }

if (null !== $title) {
            $titleLength = Helper::width(Helper::removeDecoration($formatter = $this->output->getFormatter(), $formattedTitle = sprintf($titleFormat, $title)));
            $markupLength = Helper::width($markup);
            if ($titleLength > $limit = $markupLength - 4) {
                $titleLength = $limit;
                $formatLength = Helper::width(Helper::removeDecoration($formatter, sprintf($titleFormat, '')));
                $formattedTitle = sprintf($titleFormat, Helper::substr($title, 0, $limit - $formatLength - 3).'...');
            }

            $titleStart = intdiv($markupLength - $titleLength, 2);
            if (false === mb_detect_encoding($markup, null, true)) {
                $markup = substr_replace($markup, $formattedTitle, $titleStart, $titleLength);
            } else {
                $markup = mb_substr($markup, 0, $titleStart).$formattedTitle.mb_substr($markup, $titleStart + $titleLength);
            }
        }

        $this->output->writeln(sprintf($this->style->getBorderFormat(), $markup));
    }

    /** * Renders vertical column separator. */

public function getLocale(): string
    {
        return $this->translator->getLocale();
    }

    private function getParts(string $originalTrans): array
    {
        if (!$this->parseHTML) {
            return [[true, true, $originalTrans]];
        }

        $html = mb_encode_numericentity($originalTrans, [0x80, 0xFFFF, 0, 0xFFFF], mb_detect_encoding($originalTrans, null, true) ?: 'UTF-8');

        $useInternalErrors = libxml_use_internal_errors(true);

        $dom = new \DOMDocument();
        $dom->loadHTML('<trans>'.$html.'</trans>');

        libxml_clear_errors();
        libxml_use_internal_errors($useInternalErrors);

        return $this->parseNode($dom->childNodes->item(1)->childNodes->item(0)->childNodes->item(0));
    }

if (intl_is_failure($formatter->getErrorCode())) {
            throw new TransformationFailedException($formatter->getErrorMessage());
        }

        if ($result >= \PHP_INT_MAX || $result <= -\PHP_INT_MAX) {
            throw new TransformationFailedException('I don\'t have a clear idea what infinity looks like.');
        }

        $result = $this->castParsedValue($result);

        if (false !== $encoding = mb_detect_encoding($value, null, true)) {
            $length = mb_strlen($value, $encoding);
            $remainder = mb_substr($value, $position, $length, $encoding);
        } else {
            $length = \strlen($value);
            $remainder = substr($value, $position, $length);
        }

        // After parsing, position holds the index of the character where the         // parsing stopped         if ($position < $length) {
            // Check if there are unrecognized characters at the end of the