[J-u-boot.git] / lib / charset.c

// SPDX-License-Identifier: GPL-2.0+
/*
 *  charset conversion utils
 *
 *  Copyright (c) 2017 Rob Clark
 */

#include <common.h>
#include <charset.h>
#include <capitalization.h>
#include <cp437.h>
#include <efi_loader.h>
#include <malloc.h>

/**
 * codepage_437 - Unicode to codepage 437 translation table
 */
const u16 codepage_437[128] = CP437;

static struct capitalization_table capitalization_table[] =
#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
	UNICODE_CAPITALIZATION_TABLE;
#elif CONFIG_FAT_DEFAULT_CODEPAGE == 1250
	CP1250_CAPITALIZATION_TABLE;
#else
	CP437_CAPITALIZATION_TABLE;
#endif

/**
 * get_code() - read Unicode code point from UTF-8 stream
 *
 * @read_u8:	- stream reader
 * @src:	- string buffer passed to stream reader, optional
 * Return:	- Unicode code point
 */
static int get_code(u8 (*read_u8)(void *data), void *data)
{
	s32 ch = 0;

	ch = read_u8(data);
	if (!ch)
		return 0;
	if (ch >= 0xc2 && ch <= 0xf4) {
		int code = 0;

		if (ch >= 0xe0) {
			if (ch >= 0xf0) {
				/* 0xf0 - 0xf4 */
				ch &= 0x07;
				code = ch << 18;
				ch = read_u8(data);
				if (ch < 0x80 || ch > 0xbf)
					goto error;
				ch &= 0x3f;
			} else {
				/* 0xe0 - 0xef */
				ch &= 0x0f;
			}
			code += ch << 12;
			if ((code >= 0xD800 && code <= 0xDFFF) ||
			    code >= 0x110000)
				goto error;
			ch = read_u8(data);
			if (ch < 0x80 || ch > 0xbf)
				goto error;
		}
		/* 0xc0 - 0xdf or continuation byte (0x80 - 0xbf) */
		ch &= 0x3f;
		code += ch << 6;
		ch = read_u8(data);
		if (ch < 0x80 || ch > 0xbf)
			goto error;
		ch &= 0x3f;
		ch += code;
	} else if (ch >= 0x80) {
		goto error;
	}
	return ch;
error:
	return '?';
}

/**
 * read_string() - read byte from character string
 *
 * @data:	- pointer to string
 * Return:	- byte read
 *
 * The string pointer is incremented if it does not point to '\0'.
 */
static u8 read_string(void *data)

{
	const char **src = (const char **)data;
	u8 c;

	if (!src || !*src || !**src)
		return 0;
	c = **src;
	++*src;
	return c;
}

/**
 * read_console() - read byte from console
 *
 * @data	- not used, needed to match interface
 * Return:	- byte read or 0 on error
 */
static u8 read_console(void *data)
{
	int ch;

	ch = getchar();
	if (ch < 0)
		ch = 0;
	return ch;
}

int console_read_unicode(s32 *code)
{
	if (!tstc()) {
		/* No input available */
		return 1;
	}

	/* Read Unicode code */
	*code = get_code(read_console, NULL);
	return 0;
}

s32 utf8_get(const char **src)
{
	return get_code(read_string, src);
}

int utf8_put(s32 code, char **dst)
{
	if (!dst || !*dst)
		return -1;
	if ((code >= 0xD800 && code <= 0xDFFF) || code >= 0x110000)
		return -1;
	if (code <= 0x007F) {
		**dst = code;
	} else {
		if (code <= 0x07FF) {
			**dst = code >> 6 | 0xC0;
		} else {
			if (code < 0x10000) {
				**dst = code >> 12 | 0xE0;
			} else {
				**dst = code >> 18 | 0xF0;
				++*dst;
				**dst = (code >> 12 & 0x3F) | 0x80;
			}
			++*dst;
			**dst = (code >> 6 & 0x3F) | 0x80;
		}
		++*dst;
		**dst = (code & 0x3F) | 0x80;
	}
	++*dst;
	return 0;
}

size_t utf8_utf16_strnlen(const char *src, size_t count)
{
	size_t len = 0;

	for (; *src && count; --count)  {
		s32 code = utf8_get(&src);

		if (!code)
			break;
		if (code < 0) {
			/* Reserve space for a replacement character */
			len += 1;
		} else if (code < 0x10000) {
			len += 1;
		} else {
			len += 2;
		}
	}
	return len;
}

int utf8_utf16_strncpy(u16 **dst, const char *src, size_t count)
{
	if (!src || !dst || !*dst)
		return -1;

	for (; count && *src; --count) {
		s32 code = utf8_get(&src);

		if (code < 0)
			code = '?';
		utf16_put(code, dst);
	}
	**dst = 0;
	return 0;
}

s32 utf16_get(const u16 **src)
{
	s32 code, code2;

	if (!src || !*src)
		return -1;
	if (!**src)
		return 0;
	code = **src;
	++*src;
	if (code >= 0xDC00 && code <= 0xDFFF)
		return -1;
	if (code >= 0xD800 && code <= 0xDBFF) {
		if (!**src)
			return -1;
		code &= 0x3ff;
		code <<= 10;
		code += 0x10000;
		code2 = **src;
		++*src;
		if (code2 <= 0xDC00 || code2 >= 0xDFFF)
			return -1;
		code2 &= 0x3ff;
		code += code2;
	}
	return code;
}

int utf16_put(s32 code, u16 **dst)
{
	if (!dst || !*dst)
		return -1;
	if ((code >= 0xD800 && code <= 0xDFFF) || code >= 0x110000)
		return -1;
	if (code < 0x10000) {
		**dst = code;
	} else {
		code -= 0x10000;
		**dst = code >> 10 | 0xD800;
		++*dst;
		**dst = (code & 0x3ff) | 0xDC00;
	}
	++*dst;
	return 0;
}

size_t utf16_strnlen(const u16 *src, size_t count)
{
	size_t len = 0;

	for (; *src && count; --count)  {
		s32 code = utf16_get(&src);

		if (!code)
			break;
		/*
		 * In case of an illegal sequence still reserve space for a
		 * replacement character.
		 */
		++len;
	}
	return len;
}

size_t utf16_utf8_strnlen(const u16 *src, size_t count)
{
	size_t len = 0;

	for (; *src && count; --count)  {
		s32 code = utf16_get(&src);

		if (!code)
			break;
		if (code < 0)
			/* Reserve space for a replacement character */
			len += 1;
		else if (code < 0x80)
			len += 1;
		else if (code < 0x800)
			len += 2;
		else if (code < 0x10000)
			len += 3;
		else
			len += 4;
	}
	return len;
}

int utf16_utf8_strncpy(char **dst, const u16 *src, size_t count)
{
	if (!src || !dst || !*dst)
		return -1;

	for (; count && *src; --count) {
		s32 code = utf16_get(&src);

		if (code < 0)
			code = '?';
		utf8_put(code, dst);
	}
	**dst = 0;
	return 0;
}

s32 utf_to_lower(const s32 code)
{
	struct capitalization_table *pos = capitalization_table;
	s32 ret = code;

	if (code <= 0x7f) {
		if (code >= 'A' && code <= 'Z')
			ret += 0x20;
		return ret;
	}
	for (; pos->upper; ++pos) {
		if (pos->upper == code) {
			ret = pos->lower;
			break;
		}
	}
	return ret;
}

s32 utf_to_upper(const s32 code)
{
	struct capitalization_table *pos = capitalization_table;
	s32 ret = code;

	if (code <= 0x7f) {
		if (code >= 'a' && code <= 'z')
			ret -= 0x20;
		return ret;
	}
	for (; pos->lower; ++pos) {
		if (pos->lower == code) {
			ret = pos->upper;
			break;
		}
	}
	return ret;
}

/*
 * u16_strncmp() - compare two u16 string
 *
 * @s1:		first string to compare
 * @s2:		second string to compare
 * @n:		maximum number of u16 to compare
 * Return:	0  if the first n u16 are the same in s1 and s2
 *		< 0 if the first different u16 in s1 is less than the
 *		corresponding u16 in s2
 *		> 0 if the first different u16 in s1 is greater than the
 *		corresponding u16 in s2
 */
int u16_strncmp(const u16 *s1, const u16 *s2, size_t n)
{
	int ret = 0;

	for (; n; --n, ++s1, ++s2) {
		ret = *s1 - *s2;
		if (ret || !*s1)
			break;
	}

	return ret;
}

size_t u16_strlen(const void *in)
{
	const char *pos = in;
	size_t ret;

	for (; pos[0] || pos[1]; pos += 2)
		;
	ret = pos - (char *)in;
	ret >>= 1;
	return ret;
}

size_t __efi_runtime u16_strnlen(const u16 *in, size_t count)
{
	size_t i;
	for (i = 0; count-- && in[i]; i++);
	return i;
}

size_t u16_strsize(const void *in)
{
	return (u16_strlen(in) + 1) * sizeof(u16);
}

u16 *u16_strcpy(u16 *dest, const u16 *src)
{
	u16 *tmp = dest;

	for (;; dest++, src++) {
		*dest = *src;
		if (!*src)
			break;
	}

	return tmp;
}

u16 *u16_strdup(const void *src)
{
	u16 *new;
	size_t len;

	if (!src)
		return NULL;
	len = (u16_strlen(src) + 1) * sizeof(u16);
	new = malloc(len);
	if (!new)
		return NULL;
	memcpy(new, src, len);

	return new;
}

/* Convert UTF-16 to UTF-8.  */
uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size)
{
	uint32_t code_high = 0;

	while (size--) {
		uint32_t code = *src++;

		if (code_high) {
			if (code >= 0xDC00 && code <= 0xDFFF) {
				/* Surrogate pair.  */
				code = ((code_high - 0xD800) << 10) + (code - 0xDC00) + 0x10000;

				*dest++ = (code >> 18) | 0xF0;
				*dest++ = ((code >> 12) & 0x3F) | 0x80;
				*dest++ = ((code >> 6) & 0x3F) | 0x80;
				*dest++ = (code & 0x3F) | 0x80;
			} else {
				/* Error...  */
				*dest++ = '?';
				/* *src may be valid. Don't eat it.  */
				src--;
			}

			code_high = 0;
		} else {
			if (code <= 0x007F) {
				*dest++ = code;
			} else if (code <= 0x07FF) {
				*dest++ = (code >> 6) | 0xC0;
				*dest++ = (code & 0x3F) | 0x80;
			} else if (code >= 0xD800 && code <= 0xDBFF) {
				code_high = code;
				continue;
			} else if (code >= 0xDC00 && code <= 0xDFFF) {
				/* Error... */
				*dest++ = '?';
			} else if (code < 0x10000) {
				*dest++ = (code >> 12) | 0xE0;
				*dest++ = ((code >> 6) & 0x3F) | 0x80;
				*dest++ = (code & 0x3F) | 0x80;
			} else {
				*dest++ = (code >> 18) | 0xF0;
				*dest++ = ((code >> 12) & 0x3F) | 0x80;
				*dest++ = ((code >> 6) & 0x3F) | 0x80;
				*dest++ = (code & 0x3F) | 0x80;
			}
		}
	}

	return dest;
}
Commit	Line	Data
f739fcd8	1	// SPDX-License-Identifier: GPL-2.0+
78178bb0 RC	2	/*
	3	* charset conversion utils
	4	*
	5	* Copyright (c) 2017 Rob Clark
78178bb0 RC	6	*/
78178bb0 RC	7
35cbb796	8	#include <common.h>
78178bb0	9	#include <charset.h>
b5130a81	10	#include <capitalization.h>
70616a1e	11	#include <cp437.h>
6974a4a3	12	#include <efi_loader.h>
78178bb0 RC	13	#include <malloc.h>
78178bb0 RC	14
70616a1e HS	15	/**
	16	* codepage_437 - Unicode to codepage 437 translation table
	17	*/
	18	const u16 codepage_437[128] = CP437;
	19
b5130a81 HS	20	static struct capitalization_table capitalization_table[] =
	21	#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
	22	UNICODE_CAPITALIZATION_TABLE;
	23	#elif CONFIG_FAT_DEFAULT_CODEPAGE == 1250
	24	CP1250_CAPITALIZATION_TABLE;
	25	#else
	26	CP437_CAPITALIZATION_TABLE;
	27	#endif
	28
35cbb796 HS	29	/**
	30	* get_code() - read Unicode code point from UTF-8 stream
	31	*
	32	* @read_u8: - stream reader
	33	* @src: - string buffer passed to stream reader, optional
	34	* Return: - Unicode code point
	35	*/
	36	static int get_code(u8 (read_u8)(void data), void *data)
d8c28232	37	{
35cbb796	38	s32 ch = 0;
d8c28232	39
35cbb796 HS	40	ch = read_u8(data);
35cbb796 HS	41	if (!ch)
d8c28232	42	return 0;
35cbb796 HS	43	if (ch >= 0xc2 && ch <= 0xf4) {
	44	int code = 0;
	45
	46	if (ch >= 0xe0) {
	47	if (ch >= 0xf0) {
d8c28232	48	/* 0xf0 - 0xf4 */
35cbb796 HS	49	ch &= 0x07;
	50	code = ch << 18;
	51	ch = read_u8(data);
	52	if (ch < 0x80 \|\| ch > 0xbf)
	53	goto error;
	54	ch &= 0x3f;
d8c28232 HS	55	} else {
d8c28232 HS	56	/* 0xe0 - 0xef */
35cbb796	57	ch &= 0x0f;
d8c28232	58	}
35cbb796	59	code += ch << 12;
d8c28232 HS	60	if ((code >= 0xD800 && code <= 0xDFFF) \|\|
d8c28232 HS	61	code >= 0x110000)
35cbb796 HS	62	goto error;
	63	ch = read_u8(data);
	64	if (ch < 0x80 \|\| ch > 0xbf)
	65	goto error;
d8c28232 HS	66	}
d8c28232 HS	67	/* 0xc0 - 0xdf or continuation byte (0x80 - 0xbf) */
35cbb796 HS	68	ch &= 0x3f;
	69	code += ch << 6;
	70	ch = read_u8(data);
	71	if (ch < 0x80 \|\| ch > 0xbf)
	72	goto error;
	73	ch &= 0x3f;
	74	ch += code;
	75	} else if (ch >= 0x80) {
	76	goto error;
d8c28232	77	}
35cbb796 HS	78	return ch;
	79	error:
	80	return '?';
	81	}
	82
	83	/**
	84	* read_string() - read byte from character string
	85	*
	86	* @data: - pointer to string
	87	* Return: - byte read
	88	*
	89	* The string pointer is incremented if it does not point to '\0'.
	90	*/
	91	static u8 read_string(void *data)
	92
	93	{
	94	const char src = (const char )data;
	95	u8 c;
	96
	97	if (!src \|\| !src \|\| !*src)
	98	return 0;
	99	c = **src;
d8c28232	100	++*src;
35cbb796 HS	101	return c;
	102	}
	103
	104	/**
	105	* read_console() - read byte from console
	106	*
60d79876 HS	107	* @data - not used, needed to match interface
60d79876 HS	108	* Return: - byte read or 0 on error
35cbb796 HS	109	*/
	110	static u8 read_console(void *data)
	111	{
60d79876 HS	112	int ch;
60d79876 HS	113
c670aeee	114	ch = getchar();
60d79876 HS	115	if (ch < 0)
	116	ch = 0;
	117	return ch;
35cbb796 HS	118	}
	119
	120	int console_read_unicode(s32 *code)
	121	{
	122	if (!tstc()) {
	123	/* No input available */
	124	return 1;
	125	}
	126
	127	/* Read Unicode code */
	128	*code = get_code(read_console, NULL);
	129	return 0;
	130	}
	131
	132	s32 utf8_get(const char **src)
	133	{
	134	return get_code(read_string, src);
d8c28232 HS	135	}
	136
	137	int utf8_put(s32 code, char **dst)
	138	{
	139	if (!dst \|\| !*dst)
	140	return -1;
	141	if ((code >= 0xD800 && code <= 0xDFFF) \|\| code >= 0x110000)
	142	return -1;
	143	if (code <= 0x007F) {
	144	**dst = code;
	145	} else {
	146	if (code <= 0x07FF) {
	147	**dst = code >> 6 \| 0xC0;
	148	} else {
	149	if (code < 0x10000) {
	150	**dst = code >> 12 \| 0xE0;
	151	} else {
	152	**dst = code >> 18 \| 0xF0;
	153	++*dst;
	154	**dst = (code >> 12 & 0x3F) \| 0x80;
	155	}
	156	++*dst;
	157	**dst = (code >> 6 & 0x3F) \| 0x80;
	158	}
	159	++*dst;
	160	**dst = (code & 0x3F) \| 0x80;
	161	}
	162	++*dst;
	163	return 0;
	164	}
	165
	166	size_t utf8_utf16_strnlen(const char *src, size_t count)
	167	{
	168	size_t len = 0;
	169
	170	for (; *src && count; --count) {
	171	s32 code = utf8_get(&src);
	172
	173	if (!code)
	174	break;
	175	if (code < 0) {
	176	/* Reserve space for a replacement character */
	177	len += 1;
	178	} else if (code < 0x10000) {
	179	len += 1;
	180	} else {
	181	len += 2;
	182	}
	183	}
	184	return len;
	185	}
	186
	187	int utf8_utf16_strncpy(u16 *dst, const char src, size_t count)
	188	{
	189	if (!src \|\| !dst \|\| !*dst)
	190	return -1;
	191
	192	for (; count && *src; --count) {
	193	s32 code = utf8_get(&src);
	194
	195	if (code < 0)
	196	code = '?';
	197	utf16_put(code, dst);
	198	}
199	**dst = 0;
200	return 0;
201	}
202
203	s32 utf16_get(const u16 **src)
204	{
205	s32 code, code2;
206
207	if (!src \|\| !*src)
208	return -1;
209	if (!**src)
210	return 0;
211	code = **src;
212	++*src;
213	if (code >= 0xDC00 && code <= 0xDFFF)
214	return -1;
215	if (code >= 0xD800 && code <= 0xDBFF) {
216	if (!**src)
217	return -1;
218	code &= 0x3ff;
219	code <<= 10;
220	code += 0x10000;
221	code2 = **src;
222	++*src;
223	if (code2 <= 0xDC00 \|\| code2 >= 0xDFFF)
224	return -1;
225	code2 &= 0x3ff;
226	code += code2;
227	}
228	return code;
229	}
230
231	int utf16_put(s32 code, u16 **dst)
232	{
233	if (!dst \|\| !*dst)
234	return -1;
235	if ((code >= 0xD800 && code <= 0xDFFF) \|\| code >= 0x110000)
236	return -1;
237	if (code < 0x10000) {
238	**dst = code;
239	} else {
240	code -= 0x10000;
241	**dst = code >> 10 \| 0xD800;
242	++*dst;
243	**dst = (code & 0x3ff) \| 0xDC00;
244	}
245	++*dst;
246	return 0;
247	}
248
249	size_t utf16_strnlen(const u16 *src, size_t count)
250	{
251	size_t len = 0;
252
253	for (; *src && count; --count) {
254	s32 code = utf16_get(&src);
255
256	if (!code)
257	break;
258	/*
259	* In case of an illegal sequence still reserve space for a
260	* replacement character.
261	*/
262	++len;
263	}
264	return len;
265	}
266
267	size_t utf16_utf8_strnlen(const u16 *src, size_t count)
268	{
269	size_t len = 0;
270
271	for (; *src && count; --count) {
272	s32 code = utf16_get(&src);
273
274	if (!code)
275	break;
276	if (code < 0)
277	/* Reserve space for a replacement character */
278	len += 1;
279	else if (code < 0x80)
280	len += 1;
281	else if (code < 0x800)
282	len += 2;
283	else if (code < 0x10000)
284	len += 3;
285	else
286	len += 4;
287	}
288	return len;
289	}
290
291	int utf16_utf8_strncpy(char *dst, const u16 src, size_t count)
292	{
293	if (!src \|\| !dst \|\| !*dst)
294	return -1;
295
296	for (; count && *src; --count) {
297	s32 code = utf16_get(&src);
298
299	if (code < 0)
300	code = '?';
301	utf8_put(code, dst);
302	}
303	**dst = 0;
304	return 0;
305	}
306
b5130a81 HS	307	s32 utf_to_lower(const s32 code)
	308	{
	309	struct capitalization_table *pos = capitalization_table;
	310	s32 ret = code;
	311
	312	if (code <= 0x7f) {
	313	if (code >= 'A' && code <= 'Z')
	314	ret += 0x20;
	315	return ret;
	316	}
	317	for (; pos->upper; ++pos) {
	318	if (pos->upper == code) {
	319	ret = pos->lower;
	320	break;
	321	}
	322	}
	323	return ret;
	324	}
	325
	326	s32 utf_to_upper(const s32 code)
	327	{
	328	struct capitalization_table *pos = capitalization_table;
	329	s32 ret = code;
	330
	331	if (code <= 0x7f) {
	332	if (code >= 'a' && code <= 'z')
	333	ret -= 0x20;
	334	return ret;
	335	}
	336	for (; pos->lower; ++pos) {
	337	if (pos->lower == code) {
	338	ret = pos->upper;
	339	break;
	340	}
	341	}
	342	return ret;
	343	}
78178bb0	344
f8062c96 AT	345	/*
	346	* u16_strncmp() - compare two u16 string
	347	*
	348	* @s1: first string to compare
	349	* @s2: second string to compare
	350	* @n: maximum number of u16 to compare
	351	* Return: 0 if the first n u16 are the same in s1 and s2
	352	* < 0 if the first different u16 in s1 is less than the
	353	* corresponding u16 in s2
	354	* > 0 if the first different u16 in s1 is greater than the
	355	* corresponding u16 in s2
	356	*/
	357	int u16_strncmp(const u16 s1, const u16 s2, size_t n)
	358	{
	359	int ret = 0;
	360
	361	for (; n; --n, ++s1, ++s2) {
	362	ret = s1 - s2;
	363	if (ret \|\| !*s1)
	364	break;
	365	}
	366
	367	return ret;
	368	}
	369
317068b8	370	size_t u16_strlen(const void *in)
78178bb0	371	{
317068b8 HS	372	const char *pos = in;
	373	size_t ret;
	374
	375	for (; pos[0] \|\| pos[1]; pos += 2)
	376	;
	377	ret = pos - (char *)in;
	378	ret >>= 1;
	379	return ret;
78178bb0 RC	380	}
78178bb0 RC	381
6974a4a3	382	size_t __efi_runtime u16_strnlen(const u16 *in, size_t count)
78178bb0 RC	383	{
	384	size_t i;
	385	for (i = 0; count-- && in[i]; i++);
	386	return i;
	387	}
	388
4835d35a SG	389	size_t u16_strsize(const void *in)
	390	{
	391	return (u16_strlen(in) + 1) * sizeof(u16);
	392	}
	393
2a3537ae AT	394	u16 u16_strcpy(u16 dest, const u16 *src)
	395	{
	396	u16 *tmp = dest;
	397
	398	for (;; dest++, src++) {
	399	dest = src;
	400	if (!*src)
	401	break;
	402	}
	403
	404	return tmp;
	405	}
	406
317068b8	407	u16 u16_strdup(const void src)
2a3537ae AT	408	{
2a3537ae AT	409	u16 *new;
317068b8	410	size_t len;
2a3537ae AT	411
	412	if (!src)
	413	return NULL;
317068b8 HS	414	len = (u16_strlen(src) + 1) * sizeof(u16);
317068b8 HS	415	new = malloc(len);
2a3537ae AT	416	if (!new)
2a3537ae AT	417	return NULL;
317068b8	418	memcpy(new, src, len);
2a3537ae AT	419
	420	return new;
	421	}
	422
78178bb0 RC	423	/* Convert UTF-16 to UTF-8. */
	424	uint8_t utf16_to_utf8(uint8_t dest, const uint16_t *src, size_t size)
	425	{
	426	uint32_t code_high = 0;
	427
	428	while (size--) {
	429	uint32_t code = *src++;
	430
	431	if (code_high) {
	432	if (code >= 0xDC00 && code <= 0xDFFF) {
	433	/* Surrogate pair. */
	434	code = ((code_high - 0xD800) << 10) + (code - 0xDC00) + 0x10000;
	435
	436	*dest++ = (code >> 18) \| 0xF0;
	437	*dest++ = ((code >> 12) & 0x3F) \| 0x80;
	438	*dest++ = ((code >> 6) & 0x3F) \| 0x80;
	439	*dest++ = (code & 0x3F) \| 0x80;
	440	} else {
	441	/* Error... */
	442	*dest++ = '?';
	443	/* src may be valid. Don't eat it. /
	444	src--;
	445	}
	446
	447	code_high = 0;
	448	} else {
	449	if (code <= 0x007F) {
	450	*dest++ = code;
	451	} else if (code <= 0x07FF) {
	452	*dest++ = (code >> 6) \| 0xC0;
	453	*dest++ = (code & 0x3F) \| 0x80;
	454	} else if (code >= 0xD800 && code <= 0xDBFF) {
	455	code_high = code;
	456	continue;
	457	} else if (code >= 0xDC00 && code <= 0xDFFF) {
	458	/* Error... */
	459	*dest++ = '?';
	460	} else if (code < 0x10000) {
	461	*dest++ = (code >> 12) \| 0xE0;
	462	*dest++ = ((code >> 6) & 0x3F) \| 0x80;
	463	*dest++ = (code & 0x3F) \| 0x80;
	464	} else {
	465	*dest++ = (code >> 18) \| 0xF0;
	466	*dest++ = ((code >> 12) & 0x3F) \| 0x80;
	467	*dest++ = ((code >> 6) & 0x3F) \| 0x80;
	468	*dest++ = (code & 0x3F) \| 0x80;
	469	}
	470	}
	471	}
	472
	473	return dest;
	474	}