ObjFW  Diff

extern const of_char16_t of_iso_8859_15[128];
extern const of_char16_t of_windows_1252[128];
extern const of_char16_t of_codepage_437[128];

static inline int
memcasecmp(const char *first, const char *second, size_t length)
{


	for (size_t i = 0; i < length; i++) {
		if (tolower((int)first[i]) > tolower((int)second[i]))
			return OF_ORDERED_DESCENDING;
		if (tolower((int)first[i]) < tolower((int)second[i]))
			return OF_ORDERED_ASCENDING;
	}

	return OF_ORDERED_SAME;
}

int
of_string_utf8_check(const char *UTF8String, size_t UTF8Length, size_t *length)
{
	size_t tmpLength = UTF8Length;
	int isUTF8 = 0;

	for (size_t i = 0; i < UTF8Length; i++) {
		/* No sign of UTF-8 here */
		if OF_LIKELY (!(UTF8String[i] & 0x80))
			continue;

		isUTF8 = 1;

		/* We're missing a start byte here */

	return isUTF8;
}

size_t
of_string_utf8_get_index(const char *string, size_t position)
{
	size_t index = position;

	for (size_t i = 0; i < position; i++)
		if OF_UNLIKELY ((string[i] & 0xC0) == 0x80)
			index--;

	return index;
}

size_t
of_string_utf8_get_position(const char *string, size_t index, size_t length)
{


	for (size_t i = 0; i <= index; i++)
		if OF_UNLIKELY ((string[i] & 0xC0) == 0x80)
			if (++index > length)
				return OF_NOT_FOUND;

	return index;
}

- initWithCString: (const char*)cString
	 encoding: (of_string_encoding_t)encoding
	   length: (size_t)cStringLength
{
	self = [super init];

	@try {

		const of_char16_t *table;
		size_t j;

		if (encoding == OF_STRING_ENCODING_UTF_8 &&
		    cStringLength >= 3 &&
		    memcmp(cString, "\xEF\xBB\xBF", 3) == 0) {
			cString += 3;
			cStringLength -= 3;
		}

			return self;
		}

		/* All other encodings we support are single byte encodings */
		_s->length = cStringLength;

		if (encoding == OF_STRING_ENCODING_ISO_8859_1) {
			j = 0;
			for (size_t i = 0; i < cStringLength; i++) {
				char buffer[4];
				size_t bytes;

				if (!(cString[i] & 0x80)) {
					_s->cString[j++] = cString[i];
					continue;
				}

		case OF_STRING_ENCODING_CODEPAGE_437:
			table = of_codepage_437;
			break;
		default:
			@throw [OFInvalidEncodingException exception];
		}

		j = 0;
		for (size_t i = 0; i < cStringLength; i++) {
			char buffer[4];
			of_unichar_t character;
			size_t characterBytes;

			if (!(cString[i] & 0x80)) {
				_s->cString[j++] = cString[i];
				continue;

- initWithCharacters: (const of_unichar_t*)characters
	      length: (size_t)length
{
	self = [super init];

	@try {
		size_t j;

		_s = &_storage;

		_s->cString = [self allocMemoryWithSize: (length * 4) + 1];
		_s->length = length;

		j = 0;
		for (size_t i = 0; i < length; i++) {
			size_t len = of_string_utf8_encode(characters[i],
			    _s->cString + j);

			if (len == 0)
				@throw [OFInvalidEncodingException exception];

			if (len > 1)

- initWithUTF16String: (const of_char16_t*)string
	       length: (size_t)length
	    byteOrder: (of_byte_order_t)byteOrder
{
	self = [super init];

	@try {
		size_t j;
		bool swap = false;

		if (length > 0 && *string == 0xFEFF) {
			string++;
			length--;
		} else if (length > 0 && *string == 0xFFFE) {
			swap = true;
			string++;
			length--;
		} else if (byteOrder != OF_BYTE_ORDER_NATIVE)
			swap = true;

		_s = &_storage;

		_s->cString = [self allocMemoryWithSize: (length * 4) + 1];
		_s->length = length;

		j = 0;
		for (size_t i = 0; i < length; i++) {
			of_unichar_t character =
			    (swap ? OF_BSWAP16(string[i]) : string[i]);
			size_t len;

			/* Missing high surrogate */
			if ((character & 0xFC00) == 0xDC00)
				@throw [OFInvalidEncodingException exception];

- initWithUTF32String: (const of_char32_t*)characters
	       length: (size_t)length
	    byteOrder: (of_byte_order_t)byteOrder
{
	self = [super init];

	@try {
		size_t j;
		bool swap = false;

		if (length > 0 && *characters == 0xFEFF) {
			characters++;
			length--;
		} else if (length > 0 && *characters == 0xFFFE0000) {
			swap = true;
			characters++;
			length--;
		} else if (byteOrder != OF_BYTE_ORDER_NATIVE)
			swap = true;

		_s = &_storage;

		_s->cString = [self allocMemoryWithSize: (length * 4) + 1];
		_s->length = length;

		j = 0;
		for (size_t i = 0; i < length; i++) {
			char buffer[4];
			size_t len = of_string_utf8_encode(
			    (swap ? OF_BSWAP32(characters[i]) : characters[i]),
			    buffer);

			switch (len) {
			case 1:

		return OF_ORDERED_ASCENDING;

	return OF_ORDERED_SAME;
}

- (uint32_t)hash
{

	uint32_t hash;

	if (_s->hashed)
		return _s->hash;

	OF_HASH_INIT(hash);

	for (size_t i = 0; i < _s->cStringLength; i++) {
		of_unichar_t c;
		size_t length;

		if ((length = of_string_utf8_decode(_s->cString + i,
		    _s->cStringLength - i, &c)) == 0)
			@throw [OFInvalidEncodingException exception];

}

- (of_range_t)rangeOfString: (OFString*)string
		    options: (int)options
		      range: (of_range_t)range
{
	const char *cString = [string UTF8String];
	size_t cStringLength = [string UTF8StringLength];
	size_t rangeLocation, rangeLength;

	if (range.length > SIZE_MAX - range.location ||
	    range.location + range.length > _s->length)
		@throw [OFOutOfRangeException exception];

	if (_s->isUTF8) {

	if (cStringLength == 0)
		return of_range(0, 0);

	if (cStringLength > rangeLength)
		return of_range(OF_NOT_FOUND, 0);

	if (options & OF_STRING_SEARCH_BACKWARDS) {
		for (size_t i = rangeLength - cStringLength;; i--) {
			if (memcmp(_s->cString + rangeLocation + i, cString,
			    cStringLength) == 0) {
				range.location += of_string_utf8_get_index(
				    _s->cString + rangeLocation, i);
				range.length = [string length];

				return range;
			}

			/* Did not match and we're at the last char */
			if (i == 0)
				return of_range(OF_NOT_FOUND, 0);
		}
	} else {
		for (size_t i = 0; i <= rangeLength - cStringLength; i++) {
			if (memcmp(_s->cString + rangeLocation + i, cString,
			    cStringLength) == 0) {
				range.location += of_string_utf8_get_index(
				    _s->cString + rangeLocation, i);
				range.length = [string length];

				return range;
			}
		}
	}

	return of_range(OF_NOT_FOUND, 0);
}

- (bool)containsString: (OFString*)string
{
	const char *cString = [string UTF8String];
	size_t cStringLength = [string UTF8StringLength];

	if (cStringLength == 0)
		return true;

	if (cStringLength > _s->cStringLength)
		return false;

	for (size_t i = 0; i <= _s->cStringLength - cStringLength; i++)
		if (memcmp(_s->cString + i, cString, cStringLength) == 0)
			return true;

	return false;
}

- (OFString*)substringWithRange: (of_range_t)range

				options: (int)options
{
	void *pool;
	OFMutableArray *array;
	const char *cString = [delimiter UTF8String];
	size_t cStringLength = [delimiter UTF8StringLength];
	bool skipEmpty = (options & OF_STRING_SKIP_EMPTY);
	size_t last;
	OFString *component;

	array = [OFMutableArray array];
	pool = objc_autoreleasePoolPush();

	if (cStringLength > _s->cStringLength) {
		[array addObject: [[self copy] autorelease]];
		objc_autoreleasePoolPop(pool);

		return array;
	}

	last = 0;
	for (size_t i = 0; i <= _s->cStringLength - cStringLength; i++) {
		if (memcmp(_s->cString + i, cString, cStringLength) != 0)
			continue;

		component = [OFString stringWithUTF8String: _s->cString + last
						    length: i - last];
		if (!skipEmpty || [component length] > 0)
			[array addObject: component];

	return [OFString stringWithUTF8String: _s->cString + i
				       length: pathCStringLength - i];
}

- (OFString*)stringByDeletingLastPathComponent
{
	size_t pathCStringLength = _s->cStringLength;

	if (pathCStringLength == 0)
		return @"";

	if (OF_IS_PATH_DELIMITER(_s->cString[pathCStringLength - 1]))
		pathCStringLength--;

	if (pathCStringLength == 0)
		return [OFString stringWithUTF8String: _s->cString
					       length: 1];

	for (size_t i = pathCStringLength - 1; i >= 1; i--)
		if (OF_IS_PATH_DELIMITER(_s->cString[i]))
			return [OFString stringWithUTF8String: _s->cString
						       length: i];

	if (OF_IS_PATH_DELIMITER(_s->cString[0]))
		return [OFString stringWithUTF8String: _s->cString
					       length: 1];