public function testSplit()
{
// ASCII.
$res = CRegex::split("He,llo;th,ere!", "/[,;]/");
$this->assertTrue(CArray::length($res) == 4 && CString::equals($res[0], "He") && CString::equals($res[1], "llo") && CString::equals($res[2], "th") && CString::equals($res[3], "ere!"));
$res = CRegex::split("He,llo;th.ere!", CArray::fromElements("/[,;]/", "/\\./"));
$this->assertTrue(CArray::length($res) == 4 && CString::equals($res[0], "He") && CString::equals($res[1], "llo") && CString::equals($res[2], "th") && CString::equals($res[3], "ere!"));
// Special cases.
$res = CRegex::split("", "/[,;]/");
$this->assertTrue(CArray::length($res) == 1 && CString::equals($res[0], ""));
$res = CRegex::split("Hey", "//");
$this->assertTrue(CArray::length($res) == 3 && CString::equals($res[0], "H") && CString::equals($res[1], "e") && CString::equals($res[2], "y"));
$res = CRegex::split("", "//");
$this->assertTrue(CArray::length($res) == 1 && CString::equals($res[0], ""));
// Unicode.
$res = CRegex::split("¡He,llo·se,ñor!", "/[,·]/u");
$this->assertTrue(CArray::length($res) == 4 && CUString::equals($res[0], "¡He") && CUString::equals($res[1], "llo") && CUString::equals($res[2], "se") && CUString::equals($res[3], "ñor!"));
$res = CRegex::split("¡He,llo·se.ñor!", CArray::fromElements("/[,·]/u", "/\\./u"));
$this->assertTrue(CArray::length($res) == 4 && CUString::equals($res[0], "¡He") && CUString::equals($res[1], "llo") && CUString::equals($res[2], "se") && CUString::equals($res[3], "ñor!"));
// Special cases.
$res = CRegex::split("", "/[,·]/u");
$this->assertTrue(CArray::length($res) == 1 && CUString::equals($res[0], ""));
$res = CRegex::split("Héy", "//u");
$this->assertTrue(CArray::length($res) == 3 && CUString::equals($res[0], "H") && CUString::equals($res[1], "é") && CUString::equals($res[2], "y"));
$res = CRegex::split("", "//u");
$this->assertTrue(CArray::length($res) == 1 && CUString::equals($res[0], ""));
}
public function testSortOn()
{
$array = CArray::fromElements(new ClassForSorting("d"), new ClassForSorting("e"), new ClassForSorting("a"), new ClassForSorting("c"), new ClassForSorting("b"));
CArray::sortOn($array, "value", CComparator::ORDER_ASC);
$this->assertTrue($array[0]->value() === "a" && $array[1]->value() === "b" && $array[2]->value() === "c" && $array[3]->value() === "d" && $array[4]->value() === "e");
$array = CArray::fromElements(new ClassForSorting(5), new ClassForSorting(2), new ClassForSorting(1), new ClassForSorting(3), new ClassForSorting(4));
CArray::sortOn($array, "value", CComparator::ORDER_ASC);
$this->assertTrue($array[0]->value() == 1 && $array[1]->value() == 2 && $array[2]->value() == 3 && $array[3]->value() == 4 && $array[4]->value() == 5);
$array = CArray::fromElements(new ClassForSorting(u("d")), new ClassForSorting(u("e")), new ClassForSorting(u("a")), new ClassForSorting(u("c")), new ClassForSorting(u("b")));
CArray::sortOn($array, "value", CComparator::ORDER_ASC);
$this->assertTrue(CUString::equals($array[0]->value(), "a") && CUString::equals($array[1]->value(), "b") && CUString::equals($array[2]->value(), "c") && CUString::equals($array[3]->value(), "d") && CUString::equals($array[4]->value(), "e"));
}
/**
* Sorts the elements in an array of Unicode or ASCII strings, in the ascending order, case-insensitively, using
* natural order comparison.
*
* To illustrate natural order with an example, an array with strings "a100", "a20", "a3" would be sorted into the
* same array with `sortUStringsCi` method, but as "a3", "a20", "a100" with this method, which is the order a human
* being would choose.
*
* @param array $array The array to be sorted.
* @param bitfield $collationFlags **OPTIONAL. Default is** `CUString::COLLATION_DEFAULT`. The Unicode collation
* option(s) to be used for string comparison. See the [CUString](CUString.html) class for information on collation
* options.
* @param CULocale $inLocale **OPTIONAL. Default is** *the application's default locale*. The locale in which
* strings are to be compared with each other.
*
* @return void
*
* @link CUString.html CUString
*/
public static function sortUStringsNatCi($array, $collationFlags = CUString::COLLATION_DEFAULT, CULocale $inLocale = null)
{
assert('is_carray($array)', vs(isset($this), get_defined_vars()));
$array = splarray($array);
$locale = isset($inLocale) ? $inLocale->name() : CULocale::defaultLocaleName();
$coll = CUString::collatorObject(true, true, $locale, $collationFlags);
$pArray = self::toPArray($array);
$res = $coll->sort($pArray);
assert('$res', vs(isset($this), get_defined_vars()));
self::assignArrayToMap($array, $pArray);
}
protected function maybeWrapText($text)
{
if (!isset($this->m_bodyWordWrappingIsDisabled) || !$this->m_bodyWordWrappingIsDisabled) {
// Wrap the text.
$width;
if (!isset($this->m_bodyWordWrappingWidth)) {
$width = self::$ms_defaultWordWrappingWidth;
} else {
$width = $this->m_bodyWordWrappingWidth;
}
$text = CUString::wordWrap($text, $width);
}
return $text;
}
/**
* Splits a string into substrings using a specified pattern or patterns as the delimiter(s) and returns the
* resulting strings as an array.
*
* If no delimiter patterns were found, the resulting array contains just one element, which is the original
* string. If a delimiter is located at the very start or at the very end of the string or next to another
* delimiter, it will accordingly cause some string(s) in the resulting array to be empty.
*
* As a special case, the delimiter pattern can be empty, which will split the string into its constituting
* characters.
*
* @param string $string The string to be split.
* @param string|array|map $delimiterPatternOrPatterns The pattern or array of patterns to be recognized as the
* delimiter(s).
*
* @return CArray The resulting strings.
*/
public static function split($string, $delimiterPatternOrPatterns)
{
assert('is_cstring($string) && (is_cstring($delimiterPatternOrPatterns) || ' . 'is_collection($delimiterPatternOrPatterns))', vs(isset($this), get_defined_vars()));
if (is_cstring($delimiterPatternOrPatterns)) {
$numIdt = self::findGroups($delimiterPatternOrPatterns, "/^([^0-9A-Za-z\\s\\\\])(.*)\\1/", $foundGroups);
assert('$numIdt == 2', vs(isset($this), get_defined_vars()));
$idt = $foundGroups[1];
if (CString::isEmpty($idt)) {
// Special case.
if (CString::isEmpty($string)) {
$resStrings = CArray::fromElements("");
return $resStrings;
} else {
if (preg_match("/^([^0-9A-Za-z\\s\\\\])\\1[A-Za-z]*u[A-Za-z]*\\z/", $delimiterPatternOrPatterns) !== 1) {
$resStrings = CArray::make(strlen($string));
for ($i = 0; $i < strlen($string); $i++) {
$resStrings[$i] = $string[$i];
}
return $resStrings;
} else {
return CUString::splitIntoChars($string);
}
}
}
$paResStrings = preg_split($delimiterPatternOrPatterns, $string);
$qty = count($paResStrings);
$resStrings = CArray::make($qty);
for ($i = 0; $i < $qty; $i++) {
$resStrings[$i] = $paResStrings[$i];
}
return $resStrings;
} else {
$resStrings = CArray::fromElements($string);
foreach ($delimiterPatternOrPatterns as $delimiterPattern) {
assert('is_cstring($delimiterPattern)', vs(isset($this), get_defined_vars()));
$resStringsNew = CArray::make();
$len = CArray::length($resStrings);
for ($i = 0; $i < $len; $i++) {
CArray::pushArray($resStringsNew, self::split($resStrings[$i], $delimiterPattern));
}
$resStrings = $resStringsNew;
}
return $resStrings;
}
}
/**
* Determines if a string contains any characters from the Chinese, Japanese, or Korean scripts.
*
* @return bool `true` if the string contains at least one CJK character, `false` otherwise.
*/
public function hasCjkChar()
{
return CUString::hasCjkChar($this);
}
/**
* Filters a string or a collection of strings according to the expected output type(s) and returns the output
* value(s).
*
* @param mixed $inputStringOrDecodedCollection The string to be filtered or the array or map containing the
* strings to be filtered. If the parameter's value is a JSON-encoded string, the output value is going to be
* either an array or map.
* @param reference $success **OUTPUT.** After the method is called, the value of this parameter tells whether
* the filtering was successful.
*
* @return mixed The output value or a collection of values of the expected type(s) after having been put through
* the filter.
*/
public function filter($inputStringOrDecodedCollection, &$success)
{
assert('is_cstring($inputStringOrDecodedCollection) || is_collection($inputStringOrDecodedCollection)', vs(isset($this), get_defined_vars()));
$success = true;
if ($this->m_expectedType != self::CARRAY && $this->m_expectedType != self::CMAP) {
// The expected output type is not a collection; the input value must be of string type.
if (!is_cstring($inputStringOrDecodedCollection)) {
$success = false;
return oop_x($this->m_defaultValue);
}
$inputString = $inputStringOrDecodedCollection;
if ($this->m_expectedType == self::BOOL || $this->m_expectedType == self::INT || $this->m_expectedType == self::FLOAT || $this->m_expectedType == self::EMAIL || $this->m_expectedType == self::URL || $this->m_expectedType == self::IP) {
// Trim the input string on both sides from whitespace, including Unicode whitespace and control
// characters.
$trimmingSubjectRe = CUString::TRIMMING_AND_SPACING_NORM_SUBJECT_RE;
$inputString = CRegex::remove($inputString, "/^({$trimmingSubjectRe})+|({$trimmingSubjectRe})+\\z/u");
}
// Pre-process the string for integer and floating-point types.
$looksLikeHex;
if ($this->m_expectedType == self::INT || $this->m_expectedType == self::FLOAT) {
if (CString::startsWith($inputString, "+")) {
// Remove the plus sign.
$inputString = CString::substr($inputString, 1);
}
$looksLikeHex = CRegex::find($inputString, "/^-?0x/i");
if ($this->m_allowLeadingZeros && !($this->m_expectedType == self::INT && $this->m_allowHex && $looksLikeHex)) {
// Remove any leading zeros (except for special cases).
$inputString = CRegex::replace($inputString, "/^(\\D*)0*(?!\\b)/", "\$1");
}
if ($this->m_allowComma) {
$inputString = CRegex::remove($inputString, "/,(?=\\d{3}\\b)/");
}
}
// Validate and sanitize the value according to its expected type.
if ($this->m_expectedType == self::BOOL) {
if (!CRegex::find($inputString, "/^(1|true|yes|on|0|false|no|off)\\z/i")) {
$success = false;
return $this->m_defaultValue;
}
return CString::equals($inputString, "1") || CString::equalsCi($inputString, "true") || CString::equalsCi($inputString, "yes") || CString::equalsCi($inputString, "on");
}
if ($this->m_expectedType == self::INT) {
$value;
if (!($this->m_allowHex && $looksLikeHex)) {
// Regular.
if (!CRegex::find($inputString, "/^-?(?!0(?!\\b))\\d+\\z/")) {
$success = false;
return $this->m_defaultValue;
}
$value = CString::toInt($inputString);
} else {
// Hex.
if (!CRegex::find($inputString, "/^-?0x[0-9A-F]+\\z/i")) {
$success = false;
return $this->m_defaultValue;
}
$value = CString::toIntFromHex($inputString);
}
if (isset($this->m_intValidMin) && $value < $this->m_intValidMin || isset($this->m_intValidMax) && $value > $this->m_intValidMax) {
$success = false;
return $this->m_defaultValue;
}
if (isset($this->m_intClampingMin) && $value < $this->m_intClampingMin) {
$value = $this->m_intClampingMin;
}
if (isset($this->m_intClampingMax) && $value > $this->m_intClampingMax) {
$value = $this->m_intClampingMax;
}
return $value;
}
if ($this->m_expectedType == self::FLOAT) {
if (!CRegex::find($inputString, "/^-?(?!0(?!\\b))\\d*\\.?\\d+(e[\\-+]?\\d+)?\\z/i")) {
$success = false;
return $this->m_defaultValue;
}
$value = CString::toFloat($inputString);
if (isset($this->m_floatValidMin) && $value < $this->m_floatValidMin || isset($this->m_floatValidMax) && $value > $this->m_floatValidMax) {
$success = false;
return $this->m_defaultValue;
}
if (isset($this->m_floatClampingMin) && $value < $this->m_floatClampingMin) {
$value = $this->m_floatClampingMin;
}
if (isset($this->m_floatClampingMax) && $value > $this->m_floatClampingMax) {
$value = $this->m_floatClampingMax;
}
return $value;
}
if ($this->m_expectedType == self::CSTRING) {
$value = $inputString;
if (!CString::isValid($value)) {
$success = false;
return $this->m_defaultValue;
}
if (!$this->m_keepAbnormalNewlines) {
$value = CString::normNewlines($value);
}
if (!$this->m_keepNonPrintable) {
if (!$this->m_keepTabsAndNewlines) {
$value = CRegex::remove($value, "/[\\x00-\\x1F\\x7F-\\xFF]/");
} else {
$value = CRegex::remove($value, "/[\\x00-\\x1F\\x7F-\\xFF](?<![\\x09\\x0A\\x0D])/");
}
} else {
if (!$this->m_keepTabsAndNewlines) {
$value = CRegex::remove($value, "/[\\x09\\x0A\\x0D]/");
}
}
if (!$this->m_keepSideSpacing) {
$value = CString::trim($value);
}
if (!$this->m_keepExtraSpacing) {
$value = CString::normSpacing($value);
}
return $value;
}
if ($this->m_expectedType == self::CUSTRING) {
$value = $inputString;
if (!CUString::isValid($value)) {
$success = false;
return $this->m_defaultValue;
}
if (!$this->m_keepAbnormalNewlines) {
$value = CUString::normNewlines($value);
}
if (!$this->m_keepNonPrintable) {
if (!$this->m_keepTabsAndNewlines) {
$value = CRegex::remove($value, "/\\p{C}|\\p{Zl}|\\p{Zp}/u");
} else {
$value = CRegex::remove($value, "/\\p{C}(?<!\\x{0009}|\\x{000A}|\\x{000D})/u");
}
} else {
if (!$this->m_keepTabsAndNewlines) {
$value = CRegex::remove($value, "/\\x{0009}|\\x{000A}|\\x{000D}|\\p{Zl}|\\p{Zp}/u");
}
}
if (!$this->m_keepSideSpacing) {
$value = CUString::trim($value);
}
if (!$this->m_keepExtraSpacing) {
$value = CUString::normSpacing($value);
}
return $value;
}
if ($this->m_expectedType == self::EMAIL) {
$value = filter_var($inputString, FILTER_VALIDATE_EMAIL);
if (!is_cstring($value)) {
$success = false;
return $this->m_defaultValue;
}
return $value;
}
if ($this->m_expectedType == self::URL) {
$value = $inputString;
if (!CUrl::isValid($value, $this->m_ignoreProtocolAbsence)) {
$success = false;
return $this->m_defaultValue;
}
if ($this->m_ignoreProtocolAbsence) {
$value = CUrl::ensureProtocol($value);
}
return $value;
}
if ($this->m_expectedType == self::IP) {
$value = $inputString;
$options = CBitField::ALL_UNSET;
if (!$this->m_allowPrivateRange) {
$options |= CIp::DISALLOW_PRIVATE_RANGE;
}
if (!$this->m_allowReservedRange) {
$options |= CIp::DISALLOW_RESERVED_RANGE;
}
$isValid;
if (!$this->m_ipV6 && !$this->m_ipV4OrV6) {
$isValid = CIp::isValidV4($value, $options);
} else {
if (!$this->m_ipV4OrV6) {
$isValid = CIp::isValidV6($value, $options);
} else {
$isValid = CIp::isValidV4($value, $options) || CIp::isValidV6($value, $options);
}
}
if (!$isValid) {
$success = false;
return $this->m_defaultValue;
}
return $value;
}
} else {
if ($this->m_expectedType == self::CARRAY) {
if (!is_cstring($inputStringOrDecodedCollection) && !is_carray($inputStringOrDecodedCollection)) {
$success = false;
return oop_x($this->m_defaultValue);
}
$value;
if (is_cstring($inputStringOrDecodedCollection)) {
// Assume JSON format for the input string.
$json = new CJson($inputStringOrDecodedCollection, $this->m_jsonStrictness);
$value = $json->decode($success);
if (!$success) {
return oop_x($this->m_defaultValue);
}
if (!is_carray($value)) {
$success = false;
return oop_x($this->m_defaultValue);
}
} else {
$value = $inputStringOrDecodedCollection;
}
$value = self::recurseCollectionFiltering($value, $this->m_collectionInputFilters, $success, 0);
if (!$success) {
return oop_x($this->m_defaultValue);
}
return $value;
} else {
if (!is_cstring($inputStringOrDecodedCollection) && !is_cmap($inputStringOrDecodedCollection)) {
$success = false;
return oop_x($this->m_defaultValue);
}
$value;
if (is_cstring($inputStringOrDecodedCollection)) {
// Assume JSON format for the input string.
$json = new CJson($inputStringOrDecodedCollection, $this->m_jsonStrictness);
$value = $json->decode($success);
if (!$success) {
return oop_x($this->m_defaultValue);
}
if (!is_cmap($value)) {
$success = false;
return oop_x($this->m_defaultValue);
}
} else {
$value = $inputStringOrDecodedCollection;
}
$value = self::recurseCollectionFiltering($value, $this->m_collectionInputFilters, $success, 0);
if (!$success) {
return oop_x($this->m_defaultValue);
}
return $value;
}
}
}
/**
* Decodes the JSON-encoded string provided earlier to the decoder and returns the result.
*
* @param reference $success **OPTIONAL. OUTPUT.** After the method is called with this parameter provided, the
* parameter's value tells whether the decoding was successful.
*
* @return mixed The decoded value of type `CMapObject` or `CArrayObject`.
*/
public function decode(&$success = null)
{
assert('is_cstring($this->m_source)', vs(isset($this), get_defined_vars()));
$success = true;
$source = $this->m_source;
if ($this->m_decodingStrictness == self::LENIENT && !CUString::isValid($source)) {
// Change the character encoding or try fixing it.
if (CEString::looksLikeLatin1($source)) {
$source = CEString::convertLatin1ToUtf8($source);
} else {
$source = CEString::fixUtf8($source);
}
}
if ($this->m_decodingStrictness == self::STRICT_WITH_COMMENTS || $this->m_decodingStrictness == self::LENIENT) {
if (CRegex::find($source, "/\\/\\/|\\/\\*/u")) {
// Remove "//..." and "/*...*/" comments.
$source = CRegex::remove($source, "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"(*SKIP)(*FAIL)|" . "\\/\\/.*|\\/\\*\\C*?\\*\\//u");
}
}
if ($this->m_decodingStrictness == self::LENIENT) {
if (CRegex::find($source, "/[:\\[,]\\s*'([^\\\\']++|\\\\{2}|\\\\\\C)*'(?=\\s*[,}\\]])/u")) {
// Convert single-quoted string values into double-quoted, taking care of double quotes within such
// strings before and single quotes after. This needs to go in front of the rest of the leniency fixes.
while (true) {
$prevSource = $source;
$source = CRegex::replace($source, "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"(*SKIP)(*FAIL)|" . "([:\\[,]\\s*'(?:[^\\\\'\"]++|\\\\{2}|\\\\\\C)*)\"((?:[^\\\\']++|\\\\{2}|\\\\\\C)*')/u", "\$1\\\"\$2");
if (CString::equals($source, $prevSource) || is_null($source)) {
break;
}
}
if (is_null($source)) {
$source = "";
}
$source = CRegex::replace($source, "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"(*SKIP)(*FAIL)|" . "([:\\[,]\\s*)'((?:[^\\\\']++|\\\\{2}|\\\\\\C)*)'(?=\\s*[,}\\]])/u", "\$1\"\$2\"");
while (true) {
$prevSource = $source;
$source = CRegex::replace($source, "/([:\\[,]\\s*\"(?:[^\\\\\"]++|\\\\{2}|\\\\[^'])*)\\\\'((?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\")" . "(?=\\s*[,}\\]])/u", "\$1'\$2");
if (CString::equals($source, $prevSource) || is_null($source)) {
break;
}
}
if (is_null($source)) {
$source = "";
}
}
if (CRegex::find($source, "/[{,]\\s*[\\w\\-.]+\\s*:/u")) {
// Put property names in double quotes.
$source = CRegex::replace($source, "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"(*SKIP)(*FAIL)|" . "([{,]\\s*)([\\w\\-.]+)(\\s*:)/u", "\$1\"\$2\"\$3");
}
if (CRegex::find($source, "/[{,]\\s*'[\\w\\-.]+'\\s*:/u")) {
// Put property names that are in single quotes in double quotes.
$source = CRegex::replace($source, "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"(*SKIP)(*FAIL)|" . "([{,]\\s*)'([\\w\\-.]+)'(\\s*:)/u", "\$1\"\$2\"\$3");
}
if (CRegex::find($source, "/,\\s*[}\\]]/u")) {
// Remove trailing commas.
$source = CRegex::remove($source, "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"(*SKIP)(*FAIL)|" . ",(?=\\s*[}\\]])/u");
}
// Within string values, convert byte values for BS, FF, LF, CR, and HT, which are prohibited in JSON,
// to their escaped equivalents.
$stringValueSubjectRe = "/(?<!\\\\)\"(?:[^\\\\\"]++|\\\\{2}|\\\\\\C)*\"/u";
$source = CRegex::replaceWithCallback($source, $stringValueSubjectRe, function ($matches) {
return CRegex::replace($matches[0], "/\\x{0008}/u", "\\b");
});
$source = CRegex::replaceWithCallback($source, $stringValueSubjectRe, function ($matches) {
return CRegex::replace($matches[0], "/\\x{000C}/u", "\\f");
});
$source = CRegex::replaceWithCallback($source, $stringValueSubjectRe, function ($matches) {
return CRegex::replace($matches[0], "/\\x{000A}/u", "\\n");
});
$source = CRegex::replaceWithCallback($source, $stringValueSubjectRe, function ($matches) {
return CRegex::replace($matches[0], "/\\x{000D}/u", "\\r");
});
$source = CRegex::replaceWithCallback($source, $stringValueSubjectRe, function ($matches) {
return CRegex::replace($matches[0], "/\\x{0009}/u", "\\t");
});
}
$decodedValue = @json_decode($source, false, self::$ms_maxRecursionDepth);
if (is_null($decodedValue)) {
if ($this->m_decodingStrictness == self::STRICT || $this->m_decodingStrictness == self::STRICT_WITH_COMMENTS) {
$success = false;
} else {
if (CRegex::find($source, "/^\\s*[\\w.]+\\s*\\(/u")) {
// The source string appears to be a JSONP. Extract the function's argument and try decoding again.
$source = CRegex::replace($source, "/^\\s*[\\w.]+\\s*\\((\\C+)\\)/u", "\$1");
$decodedValue = @json_decode($source, false, self::$ms_maxRecursionDepth);
if (is_null($decodedValue)) {
$success = false;
}
}
}
}
if (!$success) {
return;
}
if ($this->m_decodingStrictness == self::STRICT || $this->m_decodingStrictness == self::STRICT_WITH_COMMENTS) {
if (!is_object($decodedValue) && !is_array($decodedValue)) {
$success = false;
return;
}
}
// Recursively convert any object into a CMapObject/CMap and any PHP array into a CArrayObject/CArray.
$decodedValue = self::recurseValueAfterDecoding($decodedValue, 0);
return $decodedValue;
}
