*	The caller must ensure that the source buffer is long enough that this
 *	routine does not run off the end and dereference non-existent memory
 *	looking for trail bytes. If the source buffer is known to be '\0'
 *	terminated, this cannot happen. Otherwise, the caller should call
 *	Tcl_UtfCharComplete() before calling this routine to ensure that
 *	enough bytes remain in the string.
 *
 *	Special handling of Surrogate pairs is handled as follows:
 *	For any UTF-8 string containing a character outside of the BMP, the
 *	first call to this function will fill *chPtr with the high surrogate
 *	and generate a return value of 1. Calling Tcl_UtfToUniChar again
 *	will produce the low surrogate and a return value of 3. Because *chPtr
 *	is used to remember whether the high surrogate is already produced, it
 *	is recommended to initialize the variable it points to as 0 before
 *	the first call to Tcl_UtfToUniChar is done.

    0x90, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014,
   0x2DC, 0x2122, 0x0161, 0x203A, 0x0153,   0x9D, 0x017E, 0x0178
};

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

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

 *	None.
 *
 *---------------------------------------------------------------------------
 */

int
Tcl_NumUtfChars(
    register const char *src,	/* The UTF-8 string to measure. */
    int length)			/* The length of the string in bytes, or -1
				 * for strlen(string). */
{
    Tcl_UniChar ch = 0;
    register int i = 0;

    /*
     * The separate implementations are faster.
     *
     * Since this is a time-sensitive function, we also do the check for the
     * single-byte char case specially.
     */

    if (length < 0) {
	while (*src != '\0') {
	    src += TclUtfToUniChar(src, &ch);
	    i++;
	}
	if (i < 0) i = INT_MAX; /* Bug [2738427] */
    } else {
	register const char *endPtr = src + length - 4;

	while (src < endPtr) {
	    src += TclUtfToUniChar(src, &ch);
	    i++;
	}
	endPtr += 4;
	while ((src < endPtr) && Tcl_UtfCharComplete(src, endPtr - src)) {

    int len, fullchar;
    Tcl_UniChar find = 0;

    while (1) {
	len = TclUtfToUniChar(src, &find);
	fullchar = find;
#if TCL_UTF_MAX <= 3
	if ((ch >= 0xD800) && (len < 3)) {
	    len += TclUtfToUniChar(src + len, &find);
	    fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000;
	}
#endif
	if (fullchar == ch) {
	    return src;
	}

    const char *last;

    last = NULL;
    while (1) {
	len = TclUtfToUniChar(src, &find);
	fullchar = find;
#if TCL_UTF_MAX <= 3
	if ((ch >= 0xD800) && (len < 3)) {
	    len += TclUtfToUniChar(src + len, &find);
	    fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000;
	}
#endif
	if (fullchar == ch) {
	    last = src;
	}

 *	None.
 *
 *---------------------------------------------------------------------------
 */

int
Tcl_UniCharAtIndex(
    register const char *src,	/* The UTF-8 string to dereference. */
    register int index)		/* The position of the desired character. */
{
    Tcl_UniChar ch = 0;
    int fullchar = 0;
#if TCL_UTF_MAX <= 3
	int len = 0;
#endif

 *	None.
 *
 *---------------------------------------------------------------------------
 */

const char *
Tcl_UtfAtIndex(
    register const char *src,	/* The UTF-8 string. */
    register int index)		/* The position of the desired character. */
{
    Tcl_UniChar ch = 0;
    int len = 0;

    while (index-- > 0) {
	len = TclUtfToUniChar(src, &ch);
	src += len;

{
    /*
     * We can't simply call 'memcmp(cs, ct, numBytes);' because we need to
     * check for Tcl's \xC0\x80 non-utf-8 null encoding. Otherwise utf-8 lexes
     * fine in the strcmp manner.
     */

    register int result = 0;

    for ( ; numBytes != 0; numBytes--, cs++, ct++) {
	if (*cs != *ct) {
	    result = UCHAR(*cs) - UCHAR(*ct);
	    break;
	}
    }