shithub: libmujs

ref: c87898e63b24d89de26f68e70df00a73cc64554d
dir: /utf.c/

View raw version
/*
 * The authors of this software are Rob Pike and Ken Thompson.
 *              Copyright (c) 2002 by Lucent Technologies.
 * Permission to use, copy, modify, and distribute this software for any
 * purpose without fee is hereby granted, provided that this entire notice
 * is included in all copies of any software which is or includes a copy
 * or modification of this software and in all copies of the supporting
 * documentation for such software.
 * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
 * WARRANTY.  IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE
 * ANY REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
 * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
 */
#include <stdlib.h>
#include <string.h>

#include "utf.h"

typedef unsigned char uchar;

enum
{
	Bit1	= 7,
	Bitx	= 6,
	Bit2	= 5,
	Bit3	= 4,
	Bit4	= 3,

	T1	= ((1<<(Bit1+1))-1) ^ 0xFF,	/* 0000 0000 */
	Tx	= ((1<<(Bitx+1))-1) ^ 0xFF,	/* 1000 0000 */
	T2	= ((1<<(Bit2+1))-1) ^ 0xFF,	/* 1100 0000 */
	T3	= ((1<<(Bit3+1))-1) ^ 0xFF,	/* 1110 0000 */
	T4	= ((1<<(Bit4+1))-1) ^ 0xFF,	/* 1111 0000 */

	Rune1	= (1<<(Bit1+0*Bitx))-1,		/* 0000 0000 0111 1111 */
	Rune2	= (1<<(Bit2+1*Bitx))-1,		/* 0000 0111 1111 1111 */
	Rune3	= (1<<(Bit3+2*Bitx))-1,		/* 1111 1111 1111 1111 */

	Maskx	= (1<<Bitx)-1,			/* 0011 1111 */
	Testx	= Maskx ^ 0xFF,			/* 1100 0000 */

	Bad	= Runeerror,
};

unsigned int
chartorune(Rune *rune, const char *str)
{
	int c, c1, c2;
	int l;

	/*
	 * one character sequence
	 *	00000-0007F => T1
	 */
	c = *(uchar*)str;
	if(c < Tx) {
		*rune = c;
		return 1;
	}

	/*
	 * two character sequence
	 *	0080-07FF => T2 Tx
	 */
	c1 = *(uchar*)(str+1) ^ Tx;
	if(c1 & Testx)
		goto bad;
	if(c < T3) {
		if(c < T2)
			goto bad;
		l = ((c << Bitx) | c1) & Rune2;
		if(l <= Rune1)
			goto bad;
		*rune = l;
		return 2;
	}

	/*
	 * three character sequence
	 *	0800-FFFF => T3 Tx Tx
	 */
	c2 = *(uchar*)(str+2) ^ Tx;
	if(c2 & Testx)
		goto bad;
	if(c < T4) {
		l = ((((c << Bitx) | c1) << Bitx) | c2) & Rune3;
		if(l <= Rune2)
			goto bad;
		*rune = l;
		return 3;
	}

	/*
	 * bad decoding
	 */
bad:
	*rune = Bad;
	return 1;
}

unsigned int
runetochar(char *str, const Rune *rune)
{
	unsigned int c;

	/*
	 * one character sequence
	 *	00000-0007F => 00-7F
	 */
	c = *rune;
	if(c <= Rune1) {
		str[0] = c;
		return 1;
	}

	/*
	 * two character sequence
	 *	0080-07FF => T2 Tx
	 */
	if(c <= Rune2) {
		str[0] = T2 | (c >> 1*Bitx);
		str[1] = Tx | (c & Maskx);
		return 2;
	}

	/*
	 * three character sequence
	 *	0800-FFFF => T3 Tx Tx
	 */
	str[0] = T3 |  (c >> 2*Bitx);
	str[1] = Tx | ((c >> 1*Bitx) & Maskx);
	str[2] = Tx |  (c & Maskx);
	return 3;
}

unsigned int
runelen(int c)
{
	Rune rune;
	char str[10];

	rune = c;
	return runetochar(str, &rune);
}

unsigned int
utflen(const char *s)
{
	unsigned int c;
	unsigned int n;
	Rune rune;

	n = 0;
	for(;;) {
		c = *(uchar*)s;
		if(c < Runeself) {
			if(c == 0)
				return n;
			s++;
		} else
			s += chartorune(&rune, s);
		n++;
	}
}