Tcl Source Code

				 * to this previously initialized DString. */
{
    const Tcl_UniChar *w, *wEnd;
    char *p, *string;
    int oldLength;

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

     */

    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; ) {
	p += Tcl_UniCharToUtf(*w, p);
	w++;


/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfAtIndex --
 *
 *	Returns a pointer to the specified character (not byte) position in
 *	the UTF-8 string. If TCL_UTF_MAX <= 4, characters > U+FFFF count as
 *      2 positions, but then the pointer should never be placed between
 *      the two positions.
 *
 * Results:
 *	As above.
 *
 * Side effects:
 *	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 = 1;

    while (index-- > 0) {
	len = TclUtfToUniChar(src, &ch);
	src += len;
    }
#if TCL_UTF_MAX <= 4
     if (!len) {
	/* Index points at character following High Surrogate */
	src += TclUtfToUniChar(src, &ch);
    }
#endif
    return src;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfBackslash --

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

    src = dst = str;
    while (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	upChar = ch;
#if TCL_UTF_MAX <= 4
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    upChar = (((upChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
#endif
	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.
	 */

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

    src = dst = str;
    while (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	lowChar = ch;
#if TCL_UTF_MAX <= 4
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    lowChar = (((lowChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
#endif
	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.
	 */

     */

    src = dst = str;

    if (*src) {
	bytes = TclUtfToUniChar(src, &ch);
	titleChar = ch;
#if TCL_UTF_MAX <= 4
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    titleChar = (((titleChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
#endif
	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 TCL_UTF_MAX <= 4
	if (!bytes) {
	    /* TclUtfToUniChar only returns 0 for chars > 0xffff ! */
	    bytes = TclUtfToUniChar(src, &ch);
	    /* Combine surrogates */
	    lowChar = (((lowChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
	}
#endif
	lowChar = Tcl_UniCharToLower(lowChar);

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

proc makeShared {s} {uplevel 1 [list lappend copy $s]; return $s}

# Some tests require the testobj command

testConstraint testobj [expr {[info commands testobj] ne {}}]
testConstraint testindexobj [expr {[info commands testindexobj] ne {}}]
testConstraint testevalex [expr {[info commands testevalex] ne {}}]
testConstraint tip389 [expr {[string length \U010000] == 2}]

# Used for constraining memory leak tests
testConstraint memory [llength [info commands memory]]
if {[testConstraint memory]} {
    proc getbytes {} {
        set lines [split [memory info] \n]
        return [lindex $lines 3 3]

} -match glob -result {1 {*}}
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.$noComp {string index, surrogates, bug [11ae2be95dac9417]} tip389 {
    run {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

    list $input_hex $rxBuffer_hex $rxCRC_hex
} {000341 000341 0341}
test string-12.22.$noComp {string range, shimmering binary/index} {
    set s 0000000001
    binary scan $s a* x
    run {string range $s $s end}
} 000000001
test string-12.23.$noComp {string range, surrogates, bug [11ae2be95dac9417]} tip389 {
    run {list [string range a\U100000b 1 1] [string range a\U100000b 2 2] [string range a\U100000b 3 3]}
} [list \U100000 {} b]

test string-13.1.$noComp {string repeat} {
    list [catch {run {string repeat}} msg] $msg
} {1 {wrong # args: should be "string repeat string count"}}
test string-13.2.$noComp {string repeat} {

} "Abcabc\xe7\xe7"
test string-17.7.$noComp {string totitle, unicode} {
    run {string totitle \u01f3BCabc\xc7\xe7}
} "\u01f2bcabc\xe7\xe7"
test string-17.8.$noComp {string totitle, compiled} {
    lindex [run {string totitle [list aa bb [list cc]]}] 0
} Aa
test string-17.9.$noComp {string totitle, surrogates, bug [11ae2be95dac9417]} tip389 {
    run {list [string totitle a\U118c0c 1 1] [string totitle a\U118c0c 2 2] \
	[string totitle a\U118c0c 3 3]}
} [list a\U118a0c a\U118c0C a\U118c0C]

test string-18.1.$noComp {string trim} {
    list [catch {run {string trim}} msg] $msg
} {1 {wrong # args: should be "string trim string ?chars?"}}
test string-18.2.$noComp {string trim} {
    list [catch {run {string trim a b c}} msg] $msg
} {1 {wrong # args: should be "string trim string ?chars?"}}