/*
 * tclUtf.c --
 *
 *	Routines for manipulating UTF-8 strings.
 *
 * Copyright (c) 1997-1998 Sun Microsystems, Inc.
 *
 * See the file "license.terms" for information on usage and redistribution
 * of this file, and for a DISCLAIMER OF ALL WARRANTIES.
 *
 * RCS: @(#) $Id: tclUtf.c,v 1.28 2002/08/05 03:24:41 dgp Exp $
 * RCS: @(#) $Id: tclUtf.c,v 1.29 2002/11/12 02:26:29 hobbs Exp $
 */

#include "tclInt.h"

/*
 * Include the static character classification tables and macros.
 */

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

    w = wString;
    end = string + length;
    for (p = string; p < end; ) {
	p += Tcl_UtfToUniChar(p, w);
	p += TclUtfToUniChar(p, w);
	w++;
    }
    *w = '\0';
    Tcl_DStringSetLength(dsPtr,
	    (oldLength + ((char *) w - (char *) wString)));

    return wString;

Tcl_NumUtfChars(str, len)
    register CONST char *str;	/* The UTF-8 string to measure. */
    int len;			/* The length of the string in bytes, or -1
				 * for strlen(string). */
{
    Tcl_UniChar ch;
    register Tcl_UniChar *chPtr = &ch;
    register int n;
    register int i;
    int i;

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

    i = 0;
    if (len < 0) {
	while (1) {
	    str += Tcl_UtfToUniChar(str, chPtr);
	    str += TclUtfToUniChar(str, chPtr);
	    if (ch == '\0') {
		break;
	    }
	    i++;
	}
    } else {
	register int n;

	while (len > 0) {
	    if (UCHAR(*str) < 0xC0) {
		len--;
		str++;
	    } else {
	    n = Tcl_UtfToUniChar(str, chPtr);
	    len -= n;
	    str += n;
		n = Tcl_UtfToUniChar(str, chPtr);
		len -= n;
		str += n;
	    }
	    i++;
	}
    }
    return i;
}

/*

    CONST char *string;		/* The UTF-8 string to be searched. */
    int ch;			/* The Tcl_UniChar to search for. */
{
    int len;
    Tcl_UniChar find;
    
    while (1) {
	len = Tcl_UtfToUniChar(string, &find);
	len = TclUtfToUniChar(string, &find);
	if (find == ch) {
	    return string;
	}
	if (*string == '\0') {
	    return NULL;
	}
	string += len;

{
    int len;
    Tcl_UniChar find;
    CONST char *last;
	
    last = NULL;
    while (1) {
	len = Tcl_UtfToUniChar(string, &find);
	len = TclUtfToUniChar(string, &find);
	if (find == ch) {
	    last = string;
	}
	if (*string == '\0') {
	    break;
	}
	string += len;

 
CONST char *
Tcl_UtfNext(str) 
    CONST char *str;		    /* The current location in the string. */
{
    Tcl_UniChar ch;

    return str + Tcl_UtfToUniChar(str, &ch);
    return str + TclUtfToUniChar(str, &ch);
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_UtfPrev --
 *

    register CONST char *src;	/* The UTF-8 string to dereference. */
    register int index;		/* The position of the desired character. */
{
    Tcl_UniChar ch;

    while (index >= 0) {
	index--;
	src += Tcl_UtfToUniChar(src, &ch);
	src += TclUtfToUniChar(src, &ch);
    }
    return ch;
}

/*
 *---------------------------------------------------------------------------
 *

    register CONST char *src;	/* The UTF-8 string. */
    register int index;		/* The position of the desired character. */
{
    Tcl_UniChar ch;
    
    while (index > 0) {
	index--;
	src += Tcl_UtfToUniChar(src, &ch);
	src += TclUtfToUniChar(src, &ch);
    }
    return src;
}

/*
 *---------------------------------------------------------------------------
 *

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

    src = dst = str;
    while (*src) {
        bytes = Tcl_UtfToUniChar(src, &ch);
        bytes = TclUtfToUniChar(src, &ch);
	upChar = Tcl_UniCharToUpper(ch);

	/*
	 * 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 = Tcl_UtfToUniChar(src, &ch);
	bytes = TclUtfToUniChar(src, &ch);
	lowChar = Tcl_UniCharToLower(ch);

	/*
	 * 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.
	 */

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

    src = dst = str;

    if (*src) {
	bytes = Tcl_UtfToUniChar(src, &ch);
	bytes = TclUtfToUniChar(src, &ch);
	titleChar = Tcl_UniCharToTitle(ch);

	if (bytes < UtfCount(titleChar)) {
	    memcpy(dst, src, (size_t) bytes);
	    dst += bytes;
	} else {
	    dst += Tcl_UniCharToUtf(titleChar, dst);
	}
	src += bytes;
    }
    while (*src) {
	bytes = Tcl_UtfToUniChar(src, &ch);
	bytes = TclUtfToUniChar(src, &ch);
	lowChar = Tcl_UniCharToLower(ch);

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

     */
    while (n-- > 0) {
	/*
	 * 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 += Tcl_UtfToUniChar(cs, &ch1);
	ct += Tcl_UtfToUniChar(ct, &ch2);
	cs += TclUtfToUniChar(cs, &ch1);
	ct += TclUtfToUniChar(ct, &ch2);
	if (ch1 != ch2) {
	    return (ch1 - ch2);
	}
    }
    return 0;
}


    Tcl_UniChar ch1, ch2;
    while (n-- > 0) {
	/*
	 * 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 += Tcl_UtfToUniChar(cs, &ch1);
	ct += Tcl_UtfToUniChar(ct, &ch2);
	cs += TclUtfToUniChar(cs, &ch1);
	ct += TclUtfToUniChar(ct, &ch2);
	if (ch1 != ch2) {
	    ch1 = Tcl_UniCharToLower(ch1);
	    ch2 = Tcl_UniCharToLower(ch2);
	    if (ch1 != ch2) {
		return (ch1 - ch2);
	    }
	}