*
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */

#undef Tcl_UniCharToUtfDString
char *
Tcl_UniCharToUtfDString(
    const Tcl_UniChar *uniStr,	/* Unicode string to convert to UTF-8. */
    size_t uniLength,		/* Length of Unicode string in Tcl_UniChars
    const int *uniStr,	/* Unicode string to convert to UTF-8. */
    size_t uniLength,		/* Length of Unicode string. */
				 * (must be >= 0). */
    Tcl_DString *dsPtr)		/* UTF-8 representation of string is appended
				 * to this previously initialized DString. */
{
    const Tcl_UniChar *w, *wEnd;
    const int *w, *wEnd;
    char *p, *string;
    size_t oldLength;
    int len = 1;

    /*
     * UTF-8 string length in bytes will be <= Unicode string length * 4.
     */

    if (uniStr == NULL) {
	return NULL;
    }
    if (uniLength < 0) {
	uniLength = 0;
	w = uniStr;
	while (*w != '\0') {
	    uniLength++;
	    w++;
	}
    }
    oldLength = Tcl_DStringLength(dsPtr);
    Tcl_DStringSetLength(dsPtr, oldLength + (uniLength + 1) * 4);
    string = Tcl_DStringValue(dsPtr) + oldLength;

    p = string;
    wEnd = uniStr + uniLength;
    for (w = uniStr; w < wEnd; ) {
	if (!len && ((*w & 0xFC00) != 0xDC00)) {
	    /* Special case for handling high surrogates. */
	    p += Tcl_UniCharToUtf(-1, p);
	p += Tcl_UniCharToUtf(*w, p);
	}
	len = Tcl_UniCharToUtf(*w, p);
	p += len;
	if ((*w >= 0xD800) && (len < 3)) {
	    len = 0; /* Indication that high surrogate was found */
	}
	w++;
    }
    if (!len) {
	/* Special case for handling high surrogates. */
	p += Tcl_UniCharToUtf(-1, p);
    }
    Tcl_DStringSetLength(dsPtr, oldLength + (p - string));

    return string;
}

#if (TCL_UTF_MAX > 4) && (defined(__CYGWIN__) || defined(_WIN32))
char *
TclWCharToUtfDString(
    const WCHAR *uniStr,	/* WCHAR string to convert to UTF-8. */
    size_t uniLength,		/* Length of WCHAR string in Tcl_UniChars
Tcl_Char16ToUtfDString(
    const unsigned short *uniStr,/* Utf-16 string to convert to UTF-8. */
    size_t uniLength,		/* Length of Utf-16 string. */
				 * (must be >= 0). */
    Tcl_DString *dsPtr)		/* UTF-8 representation of string is appended
				 * to this previously initialized DString. */
{
    const WCHAR *w, *wEnd;
    const unsigned short *w, *wEnd;
    char *p, *string;
    size_t oldLength;
    int len = 1;

    /*
     * UTF-8 string length in bytes will be <= Unicode string length * 4.
     * UTF-8 string length in bytes will be <= Utf16 string length * 3.
     */

    if (uniStr == NULL) {
	return NULL;
    }
    if (uniLength < 0) {

	uniLength = 0;
	w = uniStr;
	while (*w != '\0') {
	    uniLength++;
	    w++;
	}
    }
    oldLength = Tcl_DStringLength(dsPtr);
    Tcl_DStringSetLength(dsPtr, oldLength + (uniLength + 1) * 4);
    Tcl_DStringSetLength(dsPtr, oldLength + (uniLength + 1) * 3);
    string = Tcl_DStringValue(dsPtr) + oldLength;

    p = string;
    wEnd = uniStr + uniLength;
    for (w = uniStr; w < wEnd; ) {
	if (!len && ((*w & 0xFC00) != 0xDC00)) {
	    /* Special case for handling high surrogates. */

	/* Special case for handling high surrogates. */
	p += Tcl_UniCharToUtf(-1, p);
    }
    Tcl_DStringSetLength(dsPtr, oldLength + (p - string));

    return string;
}
#endif
/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfToUniChar --
 *
 *	Extract the Tcl_UniChar represented by the UTF-8 string. Bad UTF-8
 *	sequences are converted to valid Tcl_UniChars and processing

static const unsigned short cp1252[32] = {
  0x20ac,   0x81, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021,
  0x02C6, 0x2030, 0x0160, 0x2039, 0x0152,   0x8D, 0x017D,   0x8F,
    0x90, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014,
   0x2DC, 0x2122, 0x0161, 0x203A, 0x0153,   0x9D, 0x017E, 0x0178
};

#undef Tcl_UtfToUniChar
int
Tcl_UtfToUniChar(
    const char *src,	/* The UTF-8 string. */
    Tcl_UniChar *chPtr)/* Filled with the Tcl_UniChar represented by
    int *chPtr)/* Filled with the unsigned int represented by
				 * the UTF-8 string. */
{
    Tcl_UniChar byte;
    int byte;

    /*
     * Unroll 1 to 4 byte UTF-8 sequences.
     */

    byte = *((unsigned char *) src);
    if (byte < 0xC0) {
	/*
	 * Handles properly formed UTF-8 characters between 0x01 and 0x7F.
	 * Treats naked trail bytes 0x80 to 0x9F as valid characters from
	 * the cp1252 table. See: <https://en.wikipedia.org/wiki/UTF-8>
	 * Also treats \0 and other naked trail bytes 0xA0 to 0xBF as valid
	 * characters representing themselves.
	 */

#if TCL_UTF_MAX <= 4
	/* If *chPtr contains a high surrogate (produced by a previous
	 * Tcl_UtfToUniChar() call) and the next 3 bytes are UTF-8 continuation
	 * bytes, then we must produce a follow-up low surrogate. We only
	 * do that if the high surrogate matches the bits we encounter.
	 */
	if ((byte >= 0x80)
		&& (((((byte - 0x10) << 2) & 0xFC) | 0xD800) == (*chPtr & 0xFCFC))
		&& ((src[1] & 0xF0) == (((*chPtr << 4) & 0x30) | 0x80))
		&& ((src[2] & 0xC0) == 0x80)) {
	    *chPtr = ((src[1] & 0x0F) << 6) + (src[2] & 0x3F) + 0xDC00;
	    return 3;
	}
#endif
	if ((unsigned)(byte-0x80) < (unsigned)0x20) {
	    *chPtr = cp1252[byte-0x80];
	} else {
	    *chPtr = byte;
	}
	return 1;
    } else if (byte < 0xE0) {

	 */
    }
    else if (byte < 0xF8) {
	if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80) && ((src[3] & 0xC0) == 0x80)) {
	    /*
	     * Four-byte-character lead byte followed by three trail bytes.
	     */
#if TCL_UTF_MAX <= 4
	    Tcl_UniChar high = (((byte & 0x07) << 8) | ((src[1] & 0x3F) << 2)
		    | ((src[2] & 0x3F) >> 4)) - 0x40;
	    if (high >= 0x400) {
		/* out of range, < 0x10000 or > 0x10ffff */
	    } else {
		/* produce high surrogate, advance source pointer */
		*chPtr = 0xD800 + high;
		return 1;
	    }
#else
	    *chPtr = (((byte & 0x07) << 18) | ((src[1] & 0x3F) << 12)
		    | ((src[2] & 0x3F) << 6) | (src[3] & 0x3F));
	    if ((unsigned)(*chPtr - 0x10000) <= 0xFFFFF) {
		return 4;
	    }
#endif
	}

	/*
	 * A four-byte-character lead-byte not followed by three trail-bytes
	 * represents itself.
	 */
    }

    *chPtr = byte;
    return 1;
}

#if (TCL_UTF_MAX > 4) && (defined(__CYGWIN__) || defined(_WIN32))
int
TclUtfToWChar(
Tcl_UtfToChar16(
    const char *src,	/* The UTF-8 string. */
    WCHAR *chPtr)/* Filled with the WCHAR represented by
    unsigned short *chPtr)/* Filled with the unsigned short represented by
				 * the UTF-8 string. */
{
    WCHAR byte;
    unsigned short byte;

    /*
     * Unroll 1 to 4 byte UTF-8 sequences.
     */

    byte = *((unsigned char *) src);
    if (byte < 0xC0) {

	 */
    }
    else if (byte < 0xF8) {
	if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80) && ((src[3] & 0xC0) == 0x80)) {
	    /*
	     * Four-byte-character lead byte followed by three trail bytes.
	     */
	    WCHAR high = (((byte & 0x07) << 8) | ((src[1] & 0x3F) << 2)
	    unsigned short high = (((byte & 0x07) << 8) | ((src[1] & 0x3F) << 2)
		    | ((src[2] & 0x3F) >> 4)) - 0x40;
	    if (high >= 0x400) {
		/* out of range, < 0x10000 or > 0x10ffff */
	    } else {
		/* produce high surrogate, advance source pointer */
		*chPtr = 0xD800 + high;
		return 1;
	    }
	}

	/*
	 * A four-byte-character lead-byte not followed by three trail-bytes
	 * represents itself.
	 */
    }

    *chPtr = byte;
    return 1;
}
#endif

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfToUniCharDString --
 *
 *	Convert the UTF-8 string to Unicode.
 *
 * Results:
 *	The return value is a pointer to the Unicode representation of the
 *	UTF-8 string. Storage for the return value is appended to the end of
 *	dsPtr. The Unicode string is terminated with a Unicode NULL character.
 *
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */

Tcl_UniChar *
#undef Tcl_UtfToUniCharDString
int *
Tcl_UtfToUniCharDString(
    const char *src,		/* UTF-8 string to convert to Unicode. */
    size_t length,			/* Length of UTF-8 string in bytes, or -1 for
				 * strlen(). */
    Tcl_DString *dsPtr)		/* Unicode representation of string is
				 * appended to this previously initialized
				 * DString. */
{
    Tcl_UniChar ch = 0, *w, *wString;
    int ch = 0, *w, *wString;
    const char *p, *end;
    size_t oldLength;

    if (src == NULL) {
	return NULL;
    }
    if (length == TCL_AUTO_LENGTH) {
	length = strlen(src);
    }

    /*
     * Unicode string length in Tcl_UniChars will be <= UTF-8 string length in
     * bytes.
     */

    oldLength = Tcl_DStringLength(dsPtr);

    Tcl_DStringSetLength(dsPtr,
	    oldLength + (int) ((length + 1) * sizeof(Tcl_UniChar)));
    wString = (Tcl_UniChar *) (Tcl_DStringValue(dsPtr) + oldLength);
	    oldLength + ((length + 1) * sizeof(int)));
    wString = (int *) (Tcl_DStringValue(dsPtr) + oldLength);

    w = wString;
    p = src;
    end = src + length - 4;
    while (p < end) {
	p += TclUtfToUniChar(p, &ch);
	p += Tcl_UtfToUniChar(p, &ch);
	*w++ = ch;
    }
    end += 4;
    while (p < end) {
	if (Tcl_UtfCharComplete(p, end-p)) {
	    p += TclUtfToUniChar(p, &ch);
	    p += Tcl_UtfToUniChar(p, &ch);
	} else {
	    ch = UCHAR(*p++);
	}
	*w++ = ch;
    }
    *w = '\0';
    Tcl_DStringSetLength(dsPtr,
	    oldLength + ((char *) w - (char *) wString));

    return wString;
}

#if (TCL_UTF_MAX > 4) && (defined(__CYGWIN__) || defined(_WIN32))
unsigned short *
WCHAR *
TclUtfToWCharDString(
Tcl_UtfToChar16DString(
    const char *src,		/* UTF-8 string to convert to Unicode. */
    size_t length,			/* Length of UTF-8 string in bytes, or -1 for
				 * strlen(). */
    Tcl_DString *dsPtr)		/* Unicode representation of string is
				 * appended to this previously initialized
				 * DString. */
{
    unsigned short ch = 0;
    WCHAR ch = 0, *w, *wString;
    unsigned short *w, *wString;
    const char *p, *end;
    size_t oldLength;

    if (src == NULL) {
	return NULL;
    }
    if (length == TCL_AUTO_LENGTH) {
	length = strlen(src);
    }

    /*
     * Unicode string length in WCHARs will be <= UTF-8 string length in
     * bytes.
     */

    oldLength = Tcl_DStringLength(dsPtr);

    Tcl_DStringSetLength(dsPtr,
	    oldLength + (int) ((length + 1) * sizeof(WCHAR)));
    wString = (WCHAR *) (Tcl_DStringValue(dsPtr) + oldLength);
	    oldLength + ((length + 1) * sizeof(unsigned short)));
    wString = (unsigned short *) (Tcl_DStringValue(dsPtr) + oldLength);

    w = wString;
    p = src;
    end = src + length - 4;
    while (p < end) {
	p += TclUtfToWChar(p, &ch);
	p += Tcl_UtfToChar16(p, &ch);
	*w++ = ch;
    }
    end += 4;
    while (p < end) {
	if (Tcl_UtfCharComplete(p, end-p)) {
	    p += TclUtfToWChar(p, &ch);
	    p += Tcl_UtfToChar16(p, &ch);
	} else {
	    ch = UCHAR(*p++);
	}
	*w++ = ch;
    }
    *w = '\0';
    Tcl_DStringSetLength(dsPtr,
	    oldLength + ((char *) w - (char *) wString));

    return wString;
}
#endif
/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfCharComplete --
 *
 *	Determine if the UTF-8 string of the given length is long enough to be
 *	decoded by Tcl_UtfToUniChar(). This does not ensure that the UTF-8

		 * quickly if the next char in the pattern isn't a special
		 * character
		 */

		if ((p != '[') && (p != '?') && (p != '\\')) {
		    if (nocase) {
			while (*uniStr && (p != *uniStr)
				&& (p != (Tcl_UniChar)Tcl_UniCharToLower(*uniStr))) {
				&& (p != Tcl_UniCharToLower(*uniStr))) {
			    uniStr++;
			}
		    } else {
			while (*uniStr && (p != *uniStr)) {
			    uniStr++;
			}
		    }

	 * characters separated by "-").
	 */

	if (p == '[') {
	    Tcl_UniChar startChar, endChar;

	    uniPattern++;
	    ch1 = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*uniStr) : *uniStr);
	    ch1 = (nocase ? Tcl_UniCharToLower(*uniStr) : *uniStr);
	    uniStr++;
	    while (1) {
		if ((*uniPattern == ']') || (*uniPattern == 0)) {
		    return 0;
		}
		startChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*uniPattern)
		startChar = (nocase ? Tcl_UniCharToLower(*uniPattern)
			: *uniPattern);
		uniPattern++;
		if (*uniPattern == '-') {
		    uniPattern++;
		    if (*uniPattern == 0) {
			return 0;
		    }
		    endChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*uniPattern)
		    endChar = (nocase ? Tcl_UniCharToLower(*uniPattern)
			    : *uniPattern);
		    uniPattern++;
		    if (((startChar <= ch1) && (ch1 <= endChar))
			    || ((endChar <= ch1) && (ch1 <= startChar))) {
			/*
			 * Matches ranges of form [a-z] or [z-a].
			 */

		 * quickly if the next char in the pattern isn't a special
		 * character.
		 */

		if ((p != '[') && (p != '?') && (p != '\\')) {
		    if (nocase) {
			while ((string < stringEnd) && (p != *string)
				&& (p != (Tcl_UniChar)Tcl_UniCharToLower(*string))) {
				&& (p != Tcl_UniCharToLower(*string))) {
			    string++;
			}
		    } else {
			while ((string < stringEnd) && (p != *string)) {
			    string++;
			}
		    }

	 * characters separated by "-").
	 */

	if (p == '[') {
	    Tcl_UniChar ch1, startChar, endChar;

	    pattern++;
	    ch1 = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*string) : *string);
	    ch1 = (nocase ? Tcl_UniCharToLower(*string) : *string);
	    string++;
	    while (1) {
		if ((*pattern == ']') || (pattern == patternEnd)) {
		    return 0;
		}
		startChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*pattern) : *pattern);
		startChar = (nocase ? Tcl_UniCharToLower(*pattern) : *pattern);
		pattern++;
		if (*pattern == '-') {
		    pattern++;
		    if (pattern == patternEnd) {
			return 0;
		    }
		    endChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*pattern)
		    endChar = (nocase ? Tcl_UniCharToLower(*pattern)
			    : *pattern);
		    pattern++;
		    if (((startChar <= ch1) && (ch1 <= endChar))
			    || ((endChar <= ch1) && (ch1 <= startChar))) {
			/*
			 * Matches ranges of form [a-z] or [z-a].
			 */