Tcl Source Code

.sp
Tcl_UniChar *
\fBTcl_GetUnicodeFromObj\fR(\fIobjPtr, lengthPtr\fR)
.sp
Tcl_UniChar *
\fBTcl_GetUnicode\fR(\fIobjPtr\fR)
.sp
int
\fBTcl_GetUniChar\fR(\fIobjPtr, index\fR)
.sp
int
\fBTcl_GetCharLength\fR(\fIobjPtr\fR)
.sp
Tcl_Obj *
\fBTcl_GetRange\fR(\fIobjPtr, first, last\fR)

\fIlengthPtr\fR if it is non-NULL.  The storage referenced by the returned
byte pointer is owned by the value manager and should not be modified by
the caller.  The procedure \fBTcl_GetUnicode\fR is used in the common case
where the caller does not need the length of the unicode string
representation.
.PP
\fBTcl_GetUniChar\fR returns the \fIindex\fR'th character in the
value's Unicode representation. If the index is out of range or
it references a low surrogate preceded by a high surrogate, it returns -1;
.PP
\fBTcl_GetRange\fR returns a newly created value comprised of the
characters between \fIfirst\fR and \fIlast\fR (inclusive) in the
value's Unicode representation.  If the value's Unicode
representation is invalid, the Unicode representation is regenerated
from the value's string representation.
.PP

.BS
.SH NAME
Tcl_UniCharToUpper, Tcl_UniCharToLower, Tcl_UniCharToTitle, Tcl_UtfToUpper, Tcl_UtfToLower, Tcl_UtfToTitle \- routines for manipulating the case of Unicode characters and UTF-8 strings
.SH SYNOPSIS
.nf
\fB#include <tcl.h>\fR
.sp
int
\fBTcl_UniCharToUpper\fR(\fIch\fR)
.sp
int
\fBTcl_UniCharToLower\fR(\fIch\fR)
.sp
int
\fBTcl_UniCharToTitle\fR(\fIch\fR)
.sp
int
\fBTcl_UtfToUpper\fR(\fIstr\fR)
.sp
int
\fBTcl_UtfToLower\fR(\fIstr\fR)

.sp
const char *
\fBTcl_UtfNext\fR(\fIsrc\fR)
.sp
const char *
\fBTcl_UtfPrev\fR(\fIsrc, start\fR)
.sp
int
\fBTcl_UniCharAtIndex\fR(\fIsrc, index\fR)
.sp
const char *
\fBTcl_UtfAtIndex\fR(\fIsrc, index\fR)
.sp
int
\fBTcl_UtfBackslash\fR(\fIsrc, readPtr, dst\fR)

.PP
\fBTcl_UtfToUniChar\fR reads one UTF-8 character starting at \fIsrc\fR
and stores it as a Tcl_UniChar in \fI*chPtr\fR.  The return value is the
number of bytes read from \fIsrc\fR.  The caller must ensure that the
source buffer is long enough such that this routine does not run off the
end and dereference non-existent or random memory; if the source buffer
is known to be null-terminated, this will not happen.  If the input is
a byte in the range 0x80 - 0x9F, \fBTcl_UtfToUniChar\fR assumes the
cp1252 encoding, stores the corresponding Tcl_UniChar in \fI*chPtr\fR
and returns 1. If the input is otherwise
not in proper UTF-8 format, \fBTcl_UtfToUniChar\fR will store the first
byte of \fIsrc\fR in \fI*chPtr\fR as a Tcl_UniChar between 0x0000 and
0x00ff and return 1.
.PP
\fBTcl_UniCharToUtfDString\fR converts the given Unicode string
to UTF-8, storing the result in a previously initialized \fBTcl_DString\fR.
You must specify \fIuniLength\fR, the length of the given Unicode string.

    void Tcl_ThreadAlert(Tcl_ThreadId threadId)
}
declare 319 {
    void Tcl_ThreadQueueEvent(Tcl_ThreadId threadId, Tcl_Event *evPtr,
	    Tcl_QueuePosition position)
}
declare 320 {
    int Tcl_UniCharAtIndex(const char *src, int index)
}
declare 321 {
    int Tcl_UniCharToLower(int ch)
}
declare 322 {
    int Tcl_UniCharToTitle(int ch)
}
declare 323 {
    int Tcl_UniCharToUpper(int ch)
}
declare 324 {
    int Tcl_UniCharToUtf(int ch, char *buf)
}
declare 325 {
    const char *Tcl_UtfAtIndex(const char *src, int index)
}

    void Tcl_SetUnicodeObj(Tcl_Obj *objPtr, const Tcl_UniChar *unicode,
	    int numChars)
}
declare 380 {
    int Tcl_GetCharLength(Tcl_Obj *objPtr)
}
declare 381 {
    int Tcl_GetUniChar(Tcl_Obj *objPtr, int index)
}
declare 382 {
    Tcl_UniChar *Tcl_GetUnicode(Tcl_Obj *objPtr)
}
declare 383 {
    Tcl_Obj *Tcl_GetRange(Tcl_Obj *objPtr, int first, int last)
}

 * 4, then Tcl_UniChar must be 2-bytes in size (UCS-2) (the default). If 6,
 * then Tcl_UniChar must be 4-bytes in size (UCS-4). At this time UCS-2 mode
 * is the default and recommended mode. UCS-4 is experimental and not
 * recommended. It works for the core, but most extensions expect UCS-2.
 */

#ifndef TCL_UTF_MAX
#define TCL_UTF_MAX		4
#endif

/*
 * This represents a Unicode character. Any changes to this should also be
 * reflected in regcustom.h.
 */

typedef int (Tcl_NRPostProc) (ClientData data[], Tcl_Interp *interp,
				int result);

/*
 *----------------------------------------------------------------------------
 * The following constant is used to test for older versions of Tcl in the
 * stubs tables. If TCL_UTF_MAX>4 use a different value.
 */

#define TCL_STUB_MAGIC		((int) 0xFCA3BACF + (TCL_UTF_MAX>4))

/*
 * The following function is required to be defined in all stubs aware
 * extensions. The function is actually implemented in the stub library, not
 * the main Tcl library, although there is a trivial implementation in the
 * main library in case an extension is statically linked into an application.
 */

    Tcl_Obj *const objv[])	/* Argument objects. */
{
    int idx, result, cflags, all, wlen, wsublen, numMatches, offset;
    int start, end, subStart, subEnd, match, command, numParts;
    Tcl_RegExp regExpr;
    Tcl_RegExpInfo info;
    Tcl_Obj *resultPtr, *subPtr, *objPtr, *startIndex = NULL;
    Tcl_UniChar ch, *wsrc, *wfirstChar, *wstring, *wsubspec = 0, *wend;

    static const char *const options[] = {
	"-all",		"-command",	"-expanded",	"-line",
	"-linestop",	"-lineanchor",	"-nocase",	"-start",
	"--",		NULL
    };
    enum options {

	Tcl_InitHashTable(&charReuseTable, TCL_ONE_WORD_KEYS);

	for ( ; stringPtr < end; stringPtr += len) {
		int fullchar;
	    len = TclUtfToUniChar(stringPtr, &ch);
	    fullchar = ch;

#if TCL_UTF_MAX <= 4
	    if (!len) {
		len += TclUtfToUniChar(stringPtr, &ch);
		fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	    }
#endif

	    /*

    if (objc != 3) {
	Tcl_WrongNumArgs(interp, 1, objv, "string charIndex");
	return TCL_ERROR;
    }

    /*
     * Get the char length to calculate what 'end' means.
     */

    length = Tcl_GetCharLength(objv[1]);
    if (TclGetIntForIndexM(interp, objv[2], length-1, &index) != TCL_OK) {
	return TCL_ERROR;
    }

    if ((index >= 0) && (index < length)) {
	int ch = Tcl_GetUniChar(objv[1], index);

	if (ch == -1) {
	    return TCL_OK;
	}

	/*
	 * If we have a ByteArray object, we're careful to generate a new
	 * bytearray for a result.
	 */

	if (TclIsPureByteArray(objv[1])) {
	    unsigned char uch = (unsigned char) ch;

	    Tcl_SetObjResult(interp, Tcl_NewByteArrayObj(&uch, 1));
	} else {
	    char buf[4];

	    length = Tcl_UniCharToUtf(ch, buf);
	    Tcl_SetObjResult(interp, Tcl_NewStringObj(buf, length));
	}
    }
    return TCL_OK;
}

	    goto str_is_done;
	}
	end = string1 + length1;
	for (; string1 < end; string1 += length2, failat++) {
	    int fullchar;
	    length2 = TclUtfToUniChar(string1, &ch);
	    fullchar = ch;
#if TCL_UTF_MAX <= 4
	    if (!length2) {
	    	length2 = TclUtfToUniChar(string1, &ch);
	    	fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	    }
#endif
	    if (!chcomp(fullchar)) {
		result = 0;

				int flags);
/* 318 */
EXTERN void		Tcl_ThreadAlert(Tcl_ThreadId threadId);
/* 319 */
EXTERN void		Tcl_ThreadQueueEvent(Tcl_ThreadId threadId,
				Tcl_Event *evPtr, Tcl_QueuePosition position);
/* 320 */
EXTERN int		Tcl_UniCharAtIndex(const char *src, int index);
/* 321 */
EXTERN int		Tcl_UniCharToLower(int ch);
/* 322 */
EXTERN int		Tcl_UniCharToTitle(int ch);
/* 323 */
EXTERN int		Tcl_UniCharToUpper(int ch);
/* 324 */
EXTERN int		Tcl_UniCharToUtf(int ch, char *buf);
/* 325 */
EXTERN const char *	Tcl_UtfAtIndex(const char *src, int index);
/* 326 */
EXTERN int		Tcl_UtfCharComplete(const char *src, int length);
/* 327 */

				int numChars);
/* 379 */
EXTERN void		Tcl_SetUnicodeObj(Tcl_Obj *objPtr,
				const Tcl_UniChar *unicode, int numChars);
/* 380 */
EXTERN int		Tcl_GetCharLength(Tcl_Obj *objPtr);
/* 381 */
EXTERN int		Tcl_GetUniChar(Tcl_Obj *objPtr, int index);
/* 382 */
EXTERN Tcl_UniChar *	Tcl_GetUnicode(Tcl_Obj *objPtr);
/* 383 */
EXTERN Tcl_Obj *	Tcl_GetRange(Tcl_Obj *objPtr, int first, int last);
/* 384 */
EXTERN void		Tcl_AppendUnicodeToObj(Tcl_Obj *objPtr,
				const Tcl_UniChar *unicode, int length);

    int (*tcl_ReadChars) (Tcl_Channel channel, Tcl_Obj *objPtr, int charsToRead, int appendFlag); /* 313 */
    void (*tcl_RestoreResult) (Tcl_Interp *interp, Tcl_SavedResult *statePtr); /* 314 */
    void (*tcl_SaveResult) (Tcl_Interp *interp, Tcl_SavedResult *statePtr); /* 315 */
    int (*tcl_SetSystemEncoding) (Tcl_Interp *interp, const char *name); /* 316 */
    Tcl_Obj * (*tcl_SetVar2Ex) (Tcl_Interp *interp, const char *part1, const char *part2, Tcl_Obj *newValuePtr, int flags); /* 317 */
    void (*tcl_ThreadAlert) (Tcl_ThreadId threadId); /* 318 */
    void (*tcl_ThreadQueueEvent) (Tcl_ThreadId threadId, Tcl_Event *evPtr, Tcl_QueuePosition position); /* 319 */
    int (*tcl_UniCharAtIndex) (const char *src, int index); /* 320 */
    int (*tcl_UniCharToLower) (int ch); /* 321 */
    int (*tcl_UniCharToTitle) (int ch); /* 322 */
    int (*tcl_UniCharToUpper) (int ch); /* 323 */
    int (*tcl_UniCharToUtf) (int ch, char *buf); /* 324 */
    const char * (*tcl_UtfAtIndex) (const char *src, int index); /* 325 */
    int (*tcl_UtfCharComplete) (const char *src, int length); /* 326 */
    int (*tcl_UtfBackslash) (const char *src, int *readPtr, char *dst); /* 327 */
    const char * (*tcl_UtfFindFirst) (const char *src, int ch); /* 328 */
    const char * (*tcl_UtfFindLast) (const char *src, int ch); /* 329 */
    const char * (*tcl_UtfNext) (const char *src); /* 330 */

    int (*tcl_UniCharIsPrint) (int ch); /* 374 */
    int (*tcl_UniCharIsPunct) (int ch); /* 375 */
    int (*tcl_RegExpExecObj) (Tcl_Interp *interp, Tcl_RegExp regexp, Tcl_Obj *textObj, int offset, int nmatches, int flags); /* 376 */
    void (*tcl_RegExpGetInfo) (Tcl_RegExp regexp, Tcl_RegExpInfo *infoPtr); /* 377 */
    Tcl_Obj * (*tcl_NewUnicodeObj) (const Tcl_UniChar *unicode, int numChars); /* 378 */
    void (*tcl_SetUnicodeObj) (Tcl_Obj *objPtr, const Tcl_UniChar *unicode, int numChars); /* 379 */
    int (*tcl_GetCharLength) (Tcl_Obj *objPtr); /* 380 */
    int (*tcl_GetUniChar) (Tcl_Obj *objPtr, int index); /* 381 */
    Tcl_UniChar * (*tcl_GetUnicode) (Tcl_Obj *objPtr); /* 382 */
    Tcl_Obj * (*tcl_GetRange) (Tcl_Obj *objPtr, int first, int last); /* 383 */
    void (*tcl_AppendUnicodeToObj) (Tcl_Obj *objPtr, const Tcl_UniChar *unicode, int length); /* 384 */
    int (*tcl_RegExpMatchObj) (Tcl_Interp *interp, Tcl_Obj *textObj, Tcl_Obj *patternObj); /* 385 */
    void (*tcl_SetNotifier) (Tcl_NotifierProcs *notifierProcPtr); /* 386 */
    Tcl_Mutex * (*tcl_GetAllocMutex) (void); /* 387 */
    int (*tcl_GetChannelNames) (Tcl_Interp *interp); /* 388 */

	} else if (TclIsPureByteArray(valuePtr)) {
	    objResultPtr = Tcl_NewByteArrayObj(
		    Tcl_GetByteArrayFromObj(valuePtr, NULL)+index, 1);
	} else if (valuePtr->bytes && length == valuePtr->length) {
	    objResultPtr = Tcl_NewStringObj((const char *)
		    valuePtr->bytes+index, 1);
	} else {
	    char buf[4];
	    int ch = Tcl_GetUniChar(valuePtr, index);

	    /*
	     * This could be: Tcl_NewUnicodeObj((const Tcl_UniChar *)&ch, 1)
	     * but creating the object as a string seems to be faster in
	     * practical use.
	     */

	    length = (ch != -1) ? Tcl_UniCharToUtf(ch, buf) : 0;
	    objResultPtr = Tcl_NewStringObj(buf, length);
	}

	TRACE_APPEND(("\"%s\"\n", O2S(objResultPtr)));
	NEXT_INST_F(1, 2, 1);

    case INST_STR_RANGE:

}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_GetUniChar --
 *
 *	Get the index'th Unicode character from the String object. If index
 *	is out of range or it references a low surrogate preceded by a high
 *	surrogate, the result = -1;
 *
 * Results:
 *	Returns the index'th Unicode character in the Object.
 *
 * Side effects:
 *	Fills unichar with the index'th Unicode character.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_GetUniChar(
    Tcl_Obj *objPtr,		/* The object to get the Unicode charater
				 * from. */
    int index)			/* Get the index'th Unicode character. */
{
    String *stringPtr;
    int ch, length;

    if (index < 0) {
	return -1;
    }

    /*
     * Optimize the case where we're really dealing with a bytearray object
     * we don't need to convert to a string to perform the indexing operation.
     */

    if (TclIsPureByteArray(objPtr)) {
	unsigned char *bytes = Tcl_GetByteArrayFromObj(objPtr, &length);
	if (index >= length) {
		return -1;
	}

	return (int) bytes[index];
    }

    /*
     * OK, need to work with the object as a string.
     */

    SetStringFromAny(NULL, objPtr);

	}
	if (stringPtr->numChars == objPtr->length) {
	    return (Tcl_UniChar) objPtr->bytes[index];
	}
	FillUnicodeRep(objPtr);
	stringPtr = GET_STRING(objPtr);
    }

    if (index >= stringPtr->numChars) {
	return -1;
    }
	ch = stringPtr->unicode[index];
#if TCL_UTF_MAX <= 4
	/* See: bug [11ae2be95dac9417] */
	if ((ch&0xF800) == 0xD800) {
	    if (ch&0x400) {
		if ((index > 0) && ((stringPtr->unicode[index-1]&0xFC00) == 0xD800)) {
		    ch = -1; /* low surrogate preceded by high surrogate */
		}
	    } else if ((++index < stringPtr->numChars)
		    && ((stringPtr->unicode[index]&0xFC00) == 0xDC00)) {
		/* high surrogate followed by low surrogate */
		ch = (((ch & 0x3FF) << 10) | (stringPtr->unicode[index] & 0x3FF)) + 0x10000;
	    }
	}
#endif
    return ch;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_GetUnicode --
 *

Tcl_GetRange(
    Tcl_Obj *objPtr,		/* The Tcl object to find the range of. */
    int first,			/* First index of the range. */
    int last)			/* Last index of the range. */
{
    Tcl_Obj *newObjPtr;		/* The Tcl object to find the range of. */
    String *stringPtr;
    int length;

    if (first < 0) {
	first = 0;
	}
    /*
     * Optimize the case where we're really dealing with a bytearray object
     * we don't need to convert to a string to perform the substring operation.
     */

    if (TclIsPureByteArray(objPtr)) {
	unsigned char *bytes = Tcl_GetByteArrayFromObj(objPtr, &length);
	if (last >= length) {
	    last = length - 1;
	}
	if (last < first) {
	    return Tcl_NewObj();
	}
	return Tcl_NewByteArrayObj(bytes+first, last-first+1);
    }

    /*
     * OK, need to work with the object as a string.
     */

    SetStringFromAny(NULL, objPtr);
    stringPtr = GET_STRING(objPtr);

    if (stringPtr->hasUnicode == 0) {
	/*
	 * If numChars is unknown, compute it.
	 */

	if (stringPtr->numChars == -1) {
	    TclNumUtfChars(stringPtr->numChars, objPtr->bytes, objPtr->length);
	}
	if (stringPtr->numChars == objPtr->length) {
	    if (last >= stringPtr->numChars) {
		last = stringPtr->numChars - 1;
	    }
	    if (last < first) {
		return Tcl_NewObj();
	    }
	    newObjPtr = Tcl_NewStringObj(objPtr->bytes + first, last-first+1);

	    /*
	     * Since we know the char length of the result, store it.
	     */

	    SetStringFromAny(NULL, newObjPtr);
	    stringPtr = GET_STRING(newObjPtr);
	    stringPtr->numChars = newObjPtr->length;
	    return newObjPtr;
	}
	FillUnicodeRep(objPtr);
	stringPtr = GET_STRING(objPtr);
    }
	if (last > stringPtr->numChars) {
	    last = stringPtr->numChars;
	}
	if (last < first) {
	    return Tcl_NewObj();
	}
#if TCL_UTF_MAX <= 4
	/* See: bug [11ae2be95dac9417] */
	if ((first>0) && ((stringPtr->unicode[first]&0xFC00) == 0xDC00)
		&& ((stringPtr->unicode[first-1]&0xFC00) == 0xD800)) {
	    ++first;
	}
	if ((last+1<stringPtr->numChars) && ((stringPtr->unicode[last+1]&0xFC00) == 0xDC00)
		&& ((stringPtr->unicode[last]&0xFC00) == 0xD800)) {

		    numChars = precision;
		    Tcl_IncrRefCount(segment);
		    allocSegment = 1;
		}
	    }
	    break;
	case 'c': {
	    char buf[4];
	    int code, length;

	    if (TclGetIntFromObj(interp, segment, &code) != TCL_OK) {
		goto error;
	    }
	    length = Tcl_UniCharToUtf(code, buf);
	    segment = Tcl_NewStringObj(buf, length);

{
    if ((unsigned)(ch - 1) < (UNICODE_SELF - 1)) {
	return 1;
    }
    if (ch <= 0x7FF) {
	return 2;
    }

    if (((unsigned)(ch - 0x10000) <= 0xFFFFF)) {
	return 4;
    }

    return 3;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UniCharToUtf --

int
Tcl_UniCharToUtf(
    int ch,			/* The Tcl_UniChar to be stored in the
				 * buffer. */
    char *buf)			/* Buffer in which the UTF-8 representation of
				 * the Tcl_UniChar is stored. Buffer must be
				 * large enough to hold the UTF-8 character
				 * (at most 4 bytes). */
{
    if ((unsigned)(ch - 1) < (UNICODE_SELF - 1)) {
	buf[0] = (char) ch;
	return 1;
    }
    if (ch >= 0) {
	if (ch <= 0x7FF) {
	    buf[1] = (char) ((ch | 0x80) & 0xBF);
	    buf[0] = (char) ((ch >> 6) | 0xC0);
	    return 2;
	}
	if (ch <= 0xFFFF) {

	    if ((ch & 0xF800) == 0xD800) {
		if (ch & 0x0400) {
		    /* Low surrogate */
		    if (((buf[0] & 0xF8) == 0xF0) && ((buf[1] & 0xC0) == 0x80)
			    && ((buf[2] & 0xCF) == 0)) {
			/* Previous Tcl_UniChar was a High surrogate, so combine */
			buf[3] = (char) ((ch & 0x3F) | 0x80);
			buf[2] |= (char) (((ch >> 6) & 0x0F) | 0x80);
			return 4;
		    }
		    /* Previous Tcl_UniChar was not a High surrogate, so just output */
		} else {
		    /* High surrogate */
		    ch += 0x40;
		    /* Fill buffer with specific 3-byte (invalid) byte combination,
		       so following Low surrogate can recognize it and combine */
		    buf[2] = (char) ((ch << 4) & 0x30);
		    buf[1] = (char) (((ch >> 2) & 0x3F) | 0x80);
		    buf[0] = (char) (((ch >> 8) & 0x07) | 0xF0);
		    return 0;
		}
	    }

	    goto three;
	}


	if (ch <= 0x10FFFF) {
	    buf[3] = (char) ((ch | 0x80) & 0xBF);
	    buf[2] = (char) (((ch >> 6) | 0x80) & 0xBF);
	    buf[1] = (char) (((ch >> 12) | 0x80) & 0xBF);
	    buf[0] = (char) ((ch >> 18) | 0xF0);
	    return 4;
	}

    }

    ch = 0xFFFD;
three:
    buf[2] = (char) ((ch | 0x80) & 0xBF);
    buf[1] = (char) (((ch >> 6) | 0x80) & 0xBF);
    buf[0] = (char) ((ch >> 12) | 0xE0);

 *	number of bytes from the UTF-8 string that were consumed.
 *
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */

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
};

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

    /*
     * Unroll 1 to 3 (or 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 ((unsigned)(byte-0x80) < (unsigned) 0x20) {
	    *chPtr = (Tcl_UniChar) cp1252[byte-0x80];
	} else {
	    *chPtr = (Tcl_UniChar) byte;
	}
	return 1;
    } else if (byte < 0xE0) {
	if ((src[1] & 0xC0) == 0x80) {
	    /*
	     * Two-byte-character lead-byte followed by a trail-byte.
	     */

	 */
    }
    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 surrogate;

	    byte = (((byte & 0x07) << 18) | ((src[1] & 0x3F) << 12)
		    | ((src[2] & 0x3F) << 6) | (src[3] & 0x3F)) - 0x10000;
	    surrogate = (Tcl_UniChar) (0xD800 + (byte >> 10));
	    if (byte & 0x100000) {
		/* out of range, < 0x10000 or > 0x10ffff */

{
    int len, fullchar;
    Tcl_UniChar find = 0;

    while (1) {
	len = TclUtfToUniChar(src, &find);
	fullchar = find;
#if TCL_UTF_MAX <= 4
	if (!len) {
	    len += TclUtfToUniChar(src, &find);
	    fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000;
	}
#endif
	if (fullchar == ch) {
	    return src;

    Tcl_UniChar find = 0;
    const char *last;

    last = NULL;
    while (1) {
	len = TclUtfToUniChar(src, &find);
	fullchar = find;
#if TCL_UTF_MAX <= 4
	if (!len) {
	    len += TclUtfToUniChar(src, &find);
	    fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000;
	}
#endif
	if (fullchar == ch) {
	    last = src;

const char *
Tcl_UtfNext(
    const char *src)		/* The current location in the string. */
{
    Tcl_UniChar ch = 0;
    int len = TclUtfToUniChar(src, &ch);

#if TCL_UTF_MAX <= 4
    if (len == 0) {
      len = TclUtfToUniChar(src, &ch);
    }
#endif
    return src + len;
}


 *
 * Side effects:
 *	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 <= 4
	int len = 1;
#endif

    while (index-- >= 0) {
#if TCL_UTF_MAX <= 4
	src += (len = TclUtfToUniChar(src, &ch));
#else
	src += TclUtfToUniChar(src, &ch);
#endif
    }
    fullchar = ch;
#if TCL_UTF_MAX <= 4
     if (!len) {
	/* If last Tcl_UniChar was an upper surrogate, combine with lower surrogate */
	(void)TclUtfToUniChar(src, &ch);
	fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
    }
#endif
    return fullchar;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfAtIndex --
 *

 *----------------------------------------------------------------------
 */

int
Tcl_UtfToUpper(
    char *str)			/* String to convert in place. */
{
    Tcl_UniChar ch = 0;
    int upChar;
    char *src, *dst;
    int bytes;

    /*
     * Iterate over the string until we hit the terminating null.
     */

    src = dst = str;
    while (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	upChar = ch;
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    upChar = (((upChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
	upChar = Tcl_UniCharToUpper(upChar);

	/*
	 * To keep badly formed Utf strings from getting inflated by the
	 * conversion (thereby causing a segfault), only copy the upper case
	 * char to dst if its size is <= the original char.
	 */

 *----------------------------------------------------------------------
 */

int
Tcl_UtfToLower(
    char *str)			/* String to convert in place. */
{
    Tcl_UniChar ch = 0;
    int lowChar;
    char *src, *dst;
    int bytes;

    /*
     * Iterate over the string until we hit the terminating null.
     */

    src = dst = str;
    while (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	lowChar = ch;
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    lowChar = (((lowChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
	lowChar = Tcl_UniCharToLower(lowChar);

	/*
	 * To keep badly formed Utf strings from getting inflated by the
	 * conversion (thereby causing a segfault), only copy the lower case
	 * char to dst if its size is <= the original char.
	 */

 *----------------------------------------------------------------------
 */

int
Tcl_UtfToTitle(
    char *str)			/* String to convert in place. */
{
    Tcl_UniChar ch = 0;
    int titleChar, lowChar;
    char *src, *dst;
    int bytes;

    /*
     * Capitalize the first character and then lowercase the rest of the
     * characters until we get to a null.
     */

    src = dst = str;

    if (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	titleChar = ch;
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    titleChar = (((titleChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
	titleChar = Tcl_UniCharToTitle(titleChar);

	if (bytes < TclUtfCount(titleChar)) {
	    memcpy(dst, src, (size_t) bytes);
	    dst += bytes;
	} else {
	    dst += Tcl_UniCharToUtf(titleChar, dst);
	}
	src += bytes;
    }
    while (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	lowChar = ch;
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    lowChar = (((lowChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
	lowChar = Tcl_UniCharToLower(lowChar);

	if (bytes < TclUtfCount(lowChar)) {
	    memcpy(dst, src, (size_t) bytes);
	    dst += bytes;
	} else {
	    dst += Tcl_UniCharToUtf(lowChar, dst);
	}

	 * only when both strings are of at least n chars long (no need for \0
	 * check)
	 */

	cs += TclUtfToUniChar(cs, &ch1);
	ct += TclUtfToUniChar(ct, &ch2);
	if (ch1 != ch2) {
#if TCL_UTF_MAX <= 4
	    /* Surrogates always report higher than non-surrogates */
	    if (((ch1 & 0xFC00) == 0xD800)) {
	    if ((ch2 & 0xFC00) != 0xD800) {
		return ch1;
	    }
	    } else if ((ch2 & 0xFC00) == 0xD800) {
		return -ch2;

	 * n must be interpreted as chars, not bytes.
	 * This should be called only when both strings are of
	 * at least n chars long (no need for \0 check)
	 */
	cs += TclUtfToUniChar(cs, &ch1);
	ct += TclUtfToUniChar(ct, &ch2);
	if (ch1 != ch2) {
#if TCL_UTF_MAX <= 4
	    /* Surrogates always report higher than non-surrogates */
	    if (((ch1 & 0xFC00) == 0xD800)) {
	    if ((ch2 & 0xFC00) != 0xD800) {
		return ch1;
	    }
	    } else if ((ch2 & 0xFC00) == 0xD800) {
		return -ch2;

{
    Tcl_UniChar ch1 = 0, ch2 = 0;

    while (*cs && *ct) {
	cs += TclUtfToUniChar(cs, &ch1);
	ct += TclUtfToUniChar(ct, &ch2);
	if (ch1 != ch2) {
#if TCL_UTF_MAX <= 4
	    /* Surrogates always report higher than non-surrogates */
	    if (((ch1 & 0xFC00) == 0xD800)) {
	    if ((ch2 & 0xFC00) != 0xD800) {
		return ch1;
	    }
	    } else if ((ch2 & 0xFC00) == 0xD800) {
		return -ch2;

{
    Tcl_UniChar ch1 = 0, ch2 = 0;

    while (*cs && *ct) {
	cs += TclUtfToUniChar(cs, &ch1);
	ct += TclUtfToUniChar(ct, &ch2);
	if (ch1 != ch2) {
#if TCL_UTF_MAX <= 4
	    /* Surrogates always report higher than non-surrogates */
	    if (((ch1 & 0xFC00) == 0xD800)) {
	    if ((ch2 & 0xFC00) != 0xD800) {
		return ch1;
	    }
	    } else if ((ch2 & 0xFC00) == 0xD800) {
		return -ch2;

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

int
Tcl_UniCharToUpper(
    int ch)			/* Unicode character to convert. */
{
    if (!UNICODE_OUT_OF_RANGE(ch)) {
	int info = GetUniCharInfo(ch);

	if (GetCaseType(info) & 0x04) {
	    ch -= GetDelta(info);
	}
    }
    return ch & 0x1FFFFF;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharToLower --
 *
 *	Compute the lowercase equivalent of the given Unicode character.
 *
 * Results:
 *	Returns the lowercase Unicode character.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_UniCharToLower(
    int ch)			/* Unicode character to convert. */
{
    if (!UNICODE_OUT_OF_RANGE(ch)) {
	int info = GetUniCharInfo(ch);

	if (GetCaseType(info) & 0x02) {
	    ch += GetDelta(info);
	}
    }
    return ch & 0x1FFFFF;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharToTitle --
 *
 *	Compute the titlecase equivalent of the given Unicode character.
 *
 * Results:
 *	Returns the titlecase Unicode character.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_UniCharToTitle(
    int ch)			/* Unicode character to convert. */
{
    if (!UNICODE_OUT_OF_RANGE(ch)) {
	int info = GetUniCharInfo(ch);
	int mode = GetCaseType(info);

	if (mode & 0x1) {
	    /*
	     * Subtract or add one depending on the original case.
	     */

	    ch += ((mode & 0x4) ? -1 : 1);
	} else if (mode == 0x4) {
	    ch -= GetDelta(info);
	}
    }
    return ch & 0x1FFFFF;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharLen --
 *

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsAlnum(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return (((ALPHA_BITS | DIGIT_BITS) >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsAlpha --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsAlpha(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return ((ALPHA_BITS >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsControl --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsControl(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	ch &= 0x1FFFFF;
	if ((ch == 0xE0001) || ((ch >= 0xE0020) && (ch <= 0xE007f))) {
	    return 1;
	}
	if ((ch >= 0xF0000) && ((ch & 0xFFFF) <= 0xFFFD)) {
	    return 1;
	}
	return 0;
    }

    return ((CONTROL_BITS >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsDigit --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsDigit(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return (GetCategory(ch) == DECIMAL_DIGIT_NUMBER);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsGraph --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsGraph(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	ch &= 0x1FFFFF;
	return (ch >= 0xE0100) && (ch <= 0xE01EF);
    }

    return ((GRAPH_BITS >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsLower --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsLower(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return (GetCategory(ch) == LOWERCASE_LETTER);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsPrint --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsPrint(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	ch &= 0x1FFFFF;
	return (ch >= 0xE0100) && (ch <= 0xE01EF);
    }

    return (((GRAPH_BITS|SPACE_BITS) >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsPunct --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsPunct(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return ((PUNCT_BITS >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsSpace --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsSpace(
    int ch)			/* Unicode character to test. */
{

    /* Ignore upper 11 bits. */
    ch &= 0x1FFFFF;





    /*
     * If the character is within the first 127 characters, just use the
     * standard C function, otherwise consult the Unicode table.
     */

    if (ch < 0x80) {
	return TclIsSpaceProc((char) ch);

    } else if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;

    } else if (ch == 0x0085 || ch == 0x180E || ch == 0x200B
	    || ch == 0x202F || ch == 0x2060 || ch == 0xFEFF) {
	return 1;
    } else {
	return ((SPACE_BITS >> GetCategory(ch)) & 1);
    }
}

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsUpper(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return (GetCategory(ch) == UPPERCASE_LETTER);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharIsWordChar --

 *----------------------------------------------------------------------
 */

int
Tcl_UniCharIsWordChar(
    int ch)			/* Unicode character to test. */
{

    if (UNICODE_OUT_OF_RANGE(ch)) {
	return 0;
    }

    return ((WORD_BITS >> GetCategory(ch)) & 1);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_UniCharCaseMatch --

} -match glob -result {1 {*invalid octal number*}}
test string-5.19.$noComp {string index, bytearray object out of bounds} {
    run {string index [binary format I* {0x50515253 0x52}] -1}
} {}
test string-5.20.$noComp {string index, bytearray object out of bounds} {
    run {string index [binary format I* {0x50515253 0x52}] 20}
} {}
test string-5.21 {string index, surrogates, bug [11ae2be95dac9417]} fullutf {
    list [string index a\U100000b 1] [string index a\U100000b 2] [string index a\U100000b 3]
} [list \U100000 {} b]


proc largest_int {} {
    # This will give us what the largest valid int on this machine is,
    # so we can test for overflow properly below on >32 bit systems
    set int 1
    set exp 7; # assume we get at least 8 bits