framework/helpers/BaseStringHelper.php
<?php
/**
* @link https://www.yiiframework.com/
* @copyright Copyright (c) 2008 Yii Software LLC
* @license https://www.yiiframework.com/license/
*/
namespace yii\helpers;
use Yii;
/**
* BaseStringHelper provides concrete implementation for [[StringHelper]].
*
* Do not use BaseStringHelper. Use [[StringHelper]] instead.
*
* @author Qiang Xue <qiang.xue@gmail.com>
* @author Alex Makarov <sam@rmcreative.ru>
* @since 2.0
*/
class BaseStringHelper
{
/**
* Returns the number of bytes in the given string.
* This method ensures the string is treated as a byte array by using `mb_strlen()`.
*
* @param string $string the string being measured for length
* @return int the number of bytes in the given string.
*/
public static function byteLength($string)
{
return mb_strlen((string)$string, '8bit');
}
/**
* Returns the portion of string specified by the start and length parameters.
* This method ensures the string is treated as a byte array by using `mb_substr()`.
*
* @param string $string the input string. Must be one character or longer.
* @param int $start the starting position
* @param int|null $length the desired portion length. If not specified or `null`, there will be
* no limit on length i.e. the output will be until the end of the string.
* @return string the extracted part of string, or FALSE on failure or an empty string.
* @see https://www.php.net/manual/en/function.substr.php
*/
public static function byteSubstr($string, $start, $length = null)
{
if ($length === null) {
$length = static::byteLength($string);
}
return mb_substr((string)$string, $start, $length, '8bit');
}
/**
* Returns the trailing name component of a path.
* This method is similar to the php function `basename()` except that it will
* treat both \ and / as directory separators, independent of the operating system.
* This method was mainly created to work on php namespaces. When working with real
* file paths, php's `basename()` should work fine for you.
* Note: this method is not aware of the actual filesystem, or path components such as "..".
*
* @param string $path A path string.
* @param string $suffix If the name component ends in suffix this will also be cut off.
* @return string the trailing name component of the given path.
* @see https://www.php.net/manual/en/function.basename.php
*/
public static function basename($path, $suffix = '')
{
$path = (string)$path;
$len = mb_strlen($suffix);
if ($len > 0 && mb_substr($path, -$len) === $suffix) {
$path = mb_substr($path, 0, -$len);
}
$path = rtrim(str_replace('\\', '/', $path), '/');
$pos = mb_strrpos($path, '/');
if ($pos !== false) {
return mb_substr($path, $pos + 1);
}
return $path;
}
/**
* Returns parent directory's path.
* This method is similar to `dirname()` except that it will treat
* both \ and / as directory separators, independent of the operating system.
*
* @param string $path A path string.
* @return string the parent directory's path.
* @see https://www.php.net/manual/en/function.basename.php
*/
public static function dirname($path)
{
$normalizedPath = rtrim(
str_replace('\\', '/', (string)$path),
'/'
);
$separatorPosition = mb_strrpos($normalizedPath, '/');
if ($separatorPosition !== false) {
return mb_substr($path, 0, $separatorPosition);
}
return '';
}
/**
* Truncates a string to the number of characters specified.
*
* In order to truncate for an exact length, the $suffix char length must be counted towards the $length. For example
* to have a string which is exactly 255 long with $suffix `...` of 3 chars, then `StringHelper::truncate($string, 252, '...')`
* must be used to ensure you have 255 long string afterwards.
*
* @param string $string The string to truncate.
* @param int $length How many characters from original string to include into truncated string.
* @param string $suffix String to append to the end of truncated string.
* @param string|null $encoding The charset to use, defaults to charset currently used by application.
* @param bool $asHtml Whether to treat the string being truncated as HTML and preserve proper HTML tags.
* This parameter is available since version 2.0.1.
* @return string the truncated string.
*/
public static function truncate($string, $length, $suffix = '...', $encoding = null, $asHtml = false)
{
$string = (string)$string;
if ($encoding === null) {
$encoding = Yii::$app ? Yii::$app->charset : 'UTF-8';
}
if ($asHtml) {
return static::truncateHtml($string, $length, $suffix, $encoding);
}
if (mb_strlen($string, $encoding) > $length) {
return rtrim(mb_substr($string, 0, $length, $encoding)) . $suffix;
}
return $string;
}
/**
* Truncates a string to the number of words specified.
*
* @param string $string The string to truncate.
* @param int $count How many words from original string to include into truncated string.
* @param string $suffix String to append to the end of truncated string.
* @param bool $asHtml Whether to treat the string being truncated as HTML and preserve proper HTML tags.
* This parameter is available since version 2.0.1.
* @return string the truncated string.
*/
public static function truncateWords($string, $count, $suffix = '...', $asHtml = false)
{
if ($asHtml) {
return static::truncateHtml($string, $count, $suffix);
}
$words = preg_split('/(\s+)/u', trim($string), 0, PREG_SPLIT_DELIM_CAPTURE);
if (count($words) / 2 > $count) {
return implode('', array_slice($words, 0, ($count * 2) - 1)) . $suffix;
}
return $string;
}
/**
* Truncate a string while preserving the HTML.
*
* @param string $string The string to truncate
* @param int $count The counter
* @param string $suffix String to append to the end of the truncated string.
* @param string|bool $encoding Encoding flag or charset.
* @return string
* @since 2.0.1
*/
protected static function truncateHtml($string, $count, $suffix, $encoding = false)
{
$config = \HTMLPurifier_Config::create(null);
if (Yii::$app !== null) {
$config->set('Cache.SerializerPath', Yii::$app->getRuntimePath());
}
$lexer = \HTMLPurifier_Lexer::create($config);
$tokens = $lexer->tokenizeHTML($string, $config, new \HTMLPurifier_Context());
$openTokens = [];
$totalCount = 0;
$depth = 0;
$truncated = [];
foreach ($tokens as $token) {
if ($token instanceof \HTMLPurifier_Token_Start) { //Tag begins
$openTokens[$depth] = $token->name;
$truncated[] = $token;
++$depth;
} elseif ($token instanceof \HTMLPurifier_Token_Text && $totalCount <= $count) { //Text
if (false === $encoding) {
preg_match('/^(\s*)/um', $token->data, $prefixSpace) ?: $prefixSpace = ['', ''];
$token->data = $prefixSpace[1] . self::truncateWords(ltrim($token->data), $count - $totalCount, '');
$currentCount = self::countWords($token->data);
} else {
$token->data = self::truncate($token->data, $count - $totalCount, '', $encoding);
$currentCount = mb_strlen($token->data, $encoding);
}
$totalCount += $currentCount;
$truncated[] = $token;
} elseif ($token instanceof \HTMLPurifier_Token_End) { //Tag ends
if ($token->name === $openTokens[$depth - 1]) {
--$depth;
unset($openTokens[$depth]);
$truncated[] = $token;
}
} elseif ($token instanceof \HTMLPurifier_Token_Empty) { //Self contained tags, i.e. <img/> etc.
$truncated[] = $token;
}
if ($totalCount >= $count) {
if (0 < count($openTokens)) {
krsort($openTokens);
foreach ($openTokens as $name) {
$truncated[] = new \HTMLPurifier_Token_End($name);
}
}
break;
}
}
$context = new \HTMLPurifier_Context();
$generator = new \HTMLPurifier_Generator($config, $context);
return $generator->generateFromTokens($truncated) . ($totalCount >= $count ? $suffix : '');
}
/**
* Check if given string starts with specified substring. Binary and multibyte safe.
*
* @param string $string Input string
* @param string $with Part to search inside the $string
* @param bool $caseSensitive Case sensitive search. Default is true. When case sensitive is enabled, `$with` must
* exactly match the starting of the string in order to get a true value.
* @return bool Returns true if first input starts with second input, false otherwise
*/
public static function startsWith($string, $with, $caseSensitive = true)
{
$string = (string)$string;
$with = (string)$with;
if (!$bytes = static::byteLength($with)) {
return true;
}
if ($caseSensitive) {
return strncmp($string, $with, $bytes) === 0;
}
$encoding = Yii::$app ? Yii::$app->charset : 'UTF-8';
$string = static::byteSubstr($string, 0, $bytes);
return mb_strtolower($string, $encoding) === mb_strtolower($with, $encoding);
}
/**
* Check if given string ends with specified substring. Binary and multibyte safe.
*
* @param string $string Input string to check
* @param string $with Part to search inside of the `$string`.
* @param bool $caseSensitive Case sensitive search. Default is true. When case sensitive is enabled, `$with` must
* exactly match the ending of the string in order to get a true value.
* @return bool Returns true if first input ends with second input, false otherwise
*/
public static function endsWith($string, $with, $caseSensitive = true)
{
$string = (string)$string;
$with = (string)$with;
if (!$bytes = static::byteLength($with)) {
return true;
}
if ($caseSensitive) {
// Warning check, see https://php.net/substr-compare#refsect1-function.substr-compare-returnvalues
if (static::byteLength($string) < $bytes) {
return false;
}
return substr_compare($string, $with, -$bytes, $bytes) === 0;
}
$encoding = Yii::$app ? Yii::$app->charset : 'UTF-8';
$string = static::byteSubstr($string, -$bytes);
return mb_strtolower($string, $encoding) === mb_strtolower($with, $encoding);
}
/**
* Explodes string into array, optionally trims values and skips empty ones.
*
* @param string $string String to be exploded.
* @param string $delimiter Delimiter. Default is ','.
* @param mixed $trim Whether to trim each element. Can be:
* - boolean - to trim normally;
* - string - custom characters to trim. Will be passed as a second argument to `trim()` function.
* - callable - will be called for each value instead of trim. Takes the only argument - value.
* @param bool $skipEmpty Whether to skip empty strings between delimiters. Default is false.
* @return array
* @since 2.0.4
*/
public static function explode($string, $delimiter = ',', $trim = true, $skipEmpty = false)
{
$result = explode($delimiter, $string);
if ($trim !== false) {
if ($trim === true) {
$trim = 'trim';
} elseif (!is_callable($trim)) {
$trim = function ($v) use ($trim) {
return trim($v, $trim);
};
}
$result = array_map($trim, $result);
}
if ($skipEmpty) {
// Wrapped with array_values to make array keys sequential after empty values removing
$result = array_values(array_filter($result, function ($value) {
return $value !== '';
}));
}
return $result;
}
/**
* Counts words in a string.
*
* @param string $string the text to calculate
* @return int
* @since 2.0.8
*/
public static function countWords($string)
{
return count(preg_split('/\s+/u', $string, 0, PREG_SPLIT_NO_EMPTY));
}
/**
* Returns string representation of number value with replaced commas to dots, if decimal point
* of current locale is comma.
*
* @param int|float|string $value the value to normalize.
* @return string
* @since 2.0.11
*/
public static function normalizeNumber($value)
{
$value = (string) $value;
$localeInfo = localeconv();
$decimalSeparator = isset($localeInfo['decimal_point']) ? $localeInfo['decimal_point'] : null;
if ($decimalSeparator !== null && $decimalSeparator !== '.') {
$value = str_replace($decimalSeparator, '.', $value);
}
return $value;
}
/**
* Encodes string into "Base 64 Encoding with URL and Filename Safe Alphabet" (RFC 4648).
*
* > Note: Base 64 padding `=` may be at the end of the returned string.
* > `=` is not transparent to URL encoding.
*
* @param string $input the string to encode.
* @return string encoded string.
* @see https://tools.ietf.org/html/rfc4648#page-7
* @since 2.0.12
*/
public static function base64UrlEncode($input)
{
return strtr(base64_encode($input), '+/', '-_');
}
/**
* Decodes "Base 64 Encoding with URL and Filename Safe Alphabet" (RFC 4648).
*
* @param string $input encoded string.
* @return string decoded string.
* @see https://tools.ietf.org/html/rfc4648#page-7
* @since 2.0.12
*/
public static function base64UrlDecode($input)
{
return base64_decode(strtr($input, '-_', '+/'));
}
/**
* Safely casts a float to string independent of the current locale.
* The decimal separator will always be `.`.
*
* @param float|int $number a floating point number or integer.
* @return string the string representation of the number.
* @since 2.0.13
*/
public static function floatToString($number)
{
// . and , are the only decimal separators known in ICU data,
// so its safe to call str_replace here
return str_replace(',', '.', (string) $number);
}
/**
* Checks if the passed string would match the given shell wildcard pattern.
* This function emulates [[fnmatch()]], which may be unavailable at certain environment, using PCRE.
*
* @param string $pattern the shell wildcard pattern.
* @param string $string the tested string.
* @param array $options options for matching. Valid options are:
*
* - caseSensitive: bool, whether pattern should be case sensitive. Defaults to `true`.
* - escape: bool, whether backslash escaping is enabled. Defaults to `true`.
* - filePath: bool, whether slashes in string only matches slashes in the given pattern. Defaults to `false`.
*
* @return bool whether the string matches pattern or not.
* @since 2.0.14
*/
public static function matchWildcard($pattern, $string, $options = [])
{
if ($pattern === '*' && empty($options['filePath'])) {
return true;
}
$replacements = [
'\\\\\\\\' => '\\\\',
'\\\\\\*' => '[*]',
'\\\\\\?' => '[?]',
'\*' => '.*',
'\?' => '.',
'\[\!' => '[^',
'\[' => '[',
'\]' => ']',
'\-' => '-',
];
if (isset($options['escape']) && !$options['escape']) {
unset($replacements['\\\\\\\\']);
unset($replacements['\\\\\\*']);
unset($replacements['\\\\\\?']);
}
if (!empty($options['filePath'])) {
$replacements['\*'] = '[^/\\\\]*';
$replacements['\?'] = '[^/\\\\]';
}
$pattern = strtr(preg_quote($pattern, '#'), $replacements);
$pattern = '#^' . $pattern . '$#us';
if (isset($options['caseSensitive']) && !$options['caseSensitive']) {
$pattern .= 'i';
}
return preg_match($pattern, (string)$string) === 1;
}
/**
* This method provides a unicode-safe implementation of built-in PHP function `ucfirst()`.
*
* @param string $string the string to be proceeded
* @param string $encoding Optional, defaults to "UTF-8"
* @return string
* @see https://www.php.net/manual/en/function.ucfirst.php
* @since 2.0.16
* @phpcs:disable PSR1.Methods.CamelCapsMethodName.NotCamelCaps
*/
public static function mb_ucfirst($string, $encoding = 'UTF-8')
{
$firstChar = mb_substr((string)$string, 0, 1, $encoding);
$rest = mb_substr((string)$string, 1, null, $encoding);
return mb_strtoupper($firstChar, $encoding) . $rest;
}
/**
* This method provides a unicode-safe implementation of built-in PHP function `ucwords()`.
*
* @param string $string the string to be proceeded
* @param string $encoding Optional, defaults to "UTF-8"
* @return string
* @see https://www.php.net/manual/en/function.ucwords
* @since 2.0.16
* @phpcs:disable PSR1.Methods.CamelCapsMethodName.NotCamelCaps
*/
public static function mb_ucwords($string, $encoding = 'UTF-8')
{
$string = (string) $string;
if (empty($string)) {
return $string;
}
$parts = preg_split('/(\s+\W+\s+|^\W+\s+|\s+)/u', $string, -1, PREG_SPLIT_NO_EMPTY | PREG_SPLIT_DELIM_CAPTURE);
$ucfirstEven = trim(mb_substr($parts[0], -1, 1, $encoding)) === '';
foreach ($parts as $key => $value) {
$isEven = (bool)($key % 2);
if ($ucfirstEven === $isEven) {
$parts[$key] = static::mb_ucfirst($value, $encoding);
}
}
return implode('', $parts);
}
/**
* Masks a portion of a string with a repeated character.
* This method is multibyte-safe.
*
* @param string $string The input string.
* @param int $start The starting position from where to begin masking.
* This can be a positive or negative integer.
* Positive values count from the beginning,
* negative values count from the end of the string.
* @param int $length The length of the section to be masked.
* The masking will start from the $start position
* and continue for $length characters.
* @param string $mask The character to use for masking. The default is '*'.
* @return string The masked string.
*/
public static function mask($string, $start, $length, $mask = '*')
{
$strLength = mb_strlen($string, 'UTF-8');
// Return original string if start position is out of bounds
if ($start >= $strLength || $start < -$strLength) {
return $string;
}
$masked = mb_substr($string, 0, $start, 'UTF-8');
$masked .= str_repeat($mask, abs($length));
$masked .= mb_substr($string, $start + abs($length), null, 'UTF-8');
return $masked;
}
/**
* Returns the portion of the string that lies between the first occurrence of the start string
* and the last occurrence of the end string after that.
*
* @param string $string The input string.
* @param string $start The string marking the start of the portion to extract.
* @param string $end The string marking the end of the portion to extract.
* @return string|null The portion of the string between the first occurrence of
* start and the last occurrence of end, or null if either start or end cannot be found.
*/
public static function findBetween($string, $start, $end)
{
$startPos = mb_strpos($string, $start);
if ($startPos === false) {
return null;
}
$startPos += mb_strlen($start);
$endPos = mb_strrpos($string, $end, $startPos);
if ($endPos === false) {
return null;
}
return mb_substr($string, $startPos, $endPos - $startPos);
}
}