* true will skip the use of mb_check_encoding, this is mostly intended for
* unit testing our internal implementation.
*
- * @since 1.21
* @note In MediaWiki 1.21, this function did not provide proper UTF-8 validation.
* In particular, the pure PHP code path did not in fact check for overlong forms.
* Beware of this when backporting code to that version of MediaWiki.
*
+ * @since 1.21
* @param string $value String to check
* @param bool $disableMbstring Whether to use the pure PHP
- * implementation instead of trying mb_check_encoding. Intended for unit
- * testing. Default: false
- *
+ * implementation instead of trying mb_check_encoding. Intended for unit
+ * testing. Default: false
* @return bool Whether the given $value is a valid UTF-8 encoded string
*/
static function isUtf8( $value, $disableMbstring = false ) {
if ( $regexes === null ) {
$cont = "[\x80-\xbf]";
$after = "(?!$cont)"; // "(?:[^\x80-\xbf]|$)" would work here
- $regexes = array(
+ $regexes = [
// Continuation byte at the start
"/^$cont/",
"/\xf0(?![\x90-\xbf]$cont{2}$after)/",
"/[\xf1-\xf3](?!$cont{3}$after)/S",
"/\xf4(?![\x80-\x8f]$cont{2}$after)/",
- );
+ ];
}
foreach ( $regexes as $regex ) {
}
/**
- * Perform an operation equivalent to
+ * Perform an operation equivalent to `preg_replace()`
*
- * preg_replace( "!$startDelim(.*?)$endDelim!", $replace, $subject );
+ * Matches this code:
*
- * except that it's worst-case O(N) instead of O(N^2)
+ * preg_replace( "!$startDelim(.*?)$endDelim!", $replace, $subject );
*
- * Compared to delimiterReplace(), this implementation is fast but memory-
- * hungry and inflexible. The memory requirements are such that I don't
- * recommend using it on anything but guaranteed small chunks of text.
+ * ..except that it's worst-case O(N) instead of O(N^2). Compared to delimiterReplace(), this
+ * implementation is fast but memory-hungry and inflexible. The memory requirements are such
+ * that I don't recommend using it on anything but guaranteed small chunks of text.
*
* @param string $startDelim
* @param string $endDelim
* @param string $replace
* @param string $subject
- *
* @return string
*/
static function hungryDelimiterReplace( $startDelim, $endDelim, $replace, $subject ) {
}
/**
- * Perform an operation equivalent to
+ * Perform an operation equivalent to `preg_replace_callback()`
*
- * preg_replace_callback( "!$startDelim(.*)$endDelim!s$flags", $callback, $subject )
+ * Matches this code:
*
- * This implementation is slower than hungryDelimiterReplace but uses far less
- * memory. The delimiters are literal strings, not regular expressions.
+ * preg_replace_callback( "!$startDelim(.*)$endDelim!s$flags", $callback, $subject );
*
* If the start delimiter ends with an initial substring of the end delimiter,
* e.g. in the case of C-style comments, the behavior differs from the model
* regex. In this implementation, the end must share no characters with the
- * start, so e.g. /*\/ is not considered to be both the start and end of a
- * comment. /*\/xy/*\/ is considered to be a single comment with contents /xy/.
+ * start, so e.g. `/*\/` is not considered to be both the start and end of a
+ * comment. `/*\/xy/*\/` is considered to be a single comment with contents `/xy/`.
+ *
+ * The implementation of delimiterReplaceCallback() is slower than hungryDelimiterReplace()
+ * but uses far less memory. The delimiters are literal strings, not regular expressions.
*
* @param string $startDelim Start delimiter
* @param string $endDelim End delimiter
$encEnd = preg_quote( $endDelim, '!' );
$strcmp = strpos( $flags, 'i' ) === false ? 'strcmp' : 'strcasecmp';
$endLength = strlen( $endDelim );
- $m = array();
+ $m = [];
while ( $inputPos < strlen( $subject ) &&
preg_match( "!($encStart)|($encEnd)!S$flags", $subject, $m, PREG_OFFSET_CAPTURE, $inputPos )
} elseif ( $tokenType == 'end' ) {
if ( $foundStart ) {
# Found match
- $output .= call_user_func( $callback, array(
+ $output .= call_user_func( $callback, [
substr( $subject, $outputPos, $tokenOffset + $tokenLength - $outputPos ),
substr( $subject, $contentPos, $tokenOffset - $contentPos )
- ) );
+ ] );
$foundStart = false;
} else {
# Non-matching end, write it out
}
/**
- * Perform an operation equivalent to
+ * Perform an operation equivalent to `preg_replace()` with flags.
+ *
+ * Matches this code:
*
- * preg_replace( "!$startDelim(.*)$endDelim!$flags", $replace, $subject )
+ * preg_replace( "!$startDelim(.*)$endDelim!$flags", $replace, $subject );
*
* @param string $startDelim Start delimiter regular expression
* @param string $endDelim End delimiter regular expression
* @param string $replace Replacement string. May contain $1, which will be
- * replaced by the text between the delimiters
+ * replaced by the text between the delimiters
* @param string $subject String to search
* @param string $flags Regular expression flags
* @return string The string with the matches replaced
/**
* More or less "markup-safe" explode()
- * Ignores any instances of the separator inside <...>
+ * Ignores any instances of the separator inside `<...>`
* @param string $separator
* @param string $text
* @return array
static function escapeRegexReplacement( $string ) {
$string = str_replace( '\\', '\\\\', $string );
$string = str_replace( '$', '\\$', $string );
-
return $string;
}
/**
* Workalike for explode() with limited memory usage.
- * Returns an Iterator
+ *
* @param string $separator
* @param string $subject
* @return ArrayIterator|ExplodeIterator