[net-next-2.6.git] / fs / udf / unicode.c

/*
 * unicode.c
 *
 * PURPOSE
 *	Routines for converting between UTF-8 and OSTA Compressed Unicode.
 *      Also handles filename mangling
 *
 * DESCRIPTION
 *	OSTA Compressed Unicode is explained in the OSTA UDF specification.
 *		http://www.osta.org/
 *	UTF-8 is explained in the IETF RFC XXXX.
 *		ftp://ftp.internic.net/rfc/rfcxxxx.txt
 *
 * COPYRIGHT
 *	This file is distributed under the terms of the GNU General Public
 *	License (GPL). Copies of the GPL can be obtained from:
 *		ftp://prep.ai.mit.edu/pub/gnu/GPL
 *	Each contributing author retains all rights to their own work.
 */

#include "udfdecl.h"

#include <linux/kernel.h>
#include <linux/string.h>	/* for memset */
#include <linux/nls.h>

#include "udf_sb.h"

static int udf_translate_to_linux(uint8_t *, uint8_t *, int, uint8_t *, int);

static int udf_char_to_ustr(struct ustr *dest, const uint8_t *src, int strlen)
{
	if ((!dest) || (!src) || (!strlen) || (strlen > UDF_NAME_LEN - 2))
		return 0;

	memset(dest, 0, sizeof(struct ustr));
	memcpy(dest->u_name, src, strlen);
	dest->u_cmpID = 0x08;
	dest->u_len = strlen;

	return strlen;
}

/*
 * udf_build_ustr
 */
int udf_build_ustr(struct ustr *dest, dstring *ptr, int size)
{
	int usesize;

	if (!dest || !ptr || !size)
		return -1;
	BUG_ON(size < 2);

	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
	usesize = min(usesize, size - 2);
	dest->u_cmpID = ptr[0];
	dest->u_len = usesize;
	memcpy(dest->u_name, ptr + 1, usesize);
	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);

	return 0;
}

/*
 * udf_build_ustr_exact
 */
static int udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize)
{
	if ((!dest) || (!ptr) || (!exactsize))
		return -1;

	memset(dest, 0, sizeof(struct ustr));
	dest->u_cmpID = ptr[0];
	dest->u_len = exactsize - 1;
	memcpy(dest->u_name, ptr + 1, exactsize - 1);

	return 0;
}

/*
 * udf_ocu_to_utf8
 *
 * PURPOSE
 *	Convert OSTA Compressed Unicode to the UTF-8 equivalent.
 *
 * PRE-CONDITIONS
 *	utf			Pointer to UTF-8 output buffer.
 *	ocu			Pointer to OSTA Compressed Unicode input buffer
 *				of size UDF_NAME_LEN bytes.
 * 				both of type "struct ustr *"
 *
 * POST-CONDITIONS
 *	<return>		Zero on success.
 *
 * HISTORY
 *	November 12, 1997 - Andrew E. Mileski
 *	Written, tested, and released.
 */
int udf_CS0toUTF8(struct ustr *utf_o, const struct ustr *ocu_i)
{
	const uint8_t *ocu;
	uint8_t cmp_id, ocu_len;
	int i;

	ocu_len = ocu_i->u_len;
	if (ocu_len == 0) {
		memset(utf_o, 0, sizeof(struct ustr));
		return 0;
	}

	cmp_id = ocu_i->u_cmpID;
	if (cmp_id != 8 && cmp_id != 16) {
		memset(utf_o, 0, sizeof(struct ustr));
		printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
		       cmp_id, ocu_i->u_name);
		return 0;
	}

	ocu = ocu_i->u_name;
	utf_o->u_len = 0;
	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {

		/* Expand OSTA compressed Unicode to Unicode */
		uint32_t c = ocu[i++];
		if (cmp_id == 16)
			c = (c << 8) | ocu[i++];

		/* Compress Unicode to UTF-8 */
		if (c < 0x80U)
			utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
		else if (c < 0x800U) {
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0xc0 | (c >> 6));
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0x80 | (c & 0x3f));
		} else {
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0xe0 | (c >> 12));
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0x80 |
							  ((c >> 6) & 0x3f));
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0x80 | (c & 0x3f));
		}
	}
	utf_o->u_cmpID = 8;

	return utf_o->u_len;
}

/*
 *
 * udf_utf8_to_ocu
 *
 * PURPOSE
 *	Convert UTF-8 to the OSTA Compressed Unicode equivalent.
 *
 * DESCRIPTION
 *	This routine is only called by udf_lookup().
 *
 * PRE-CONDITIONS
 *	ocu			Pointer to OSTA Compressed Unicode output
 *				buffer of size UDF_NAME_LEN bytes.
 *	utf			Pointer to UTF-8 input buffer.
 *	utf_len			Length of UTF-8 input buffer in bytes.
 *
 * POST-CONDITIONS
 *	<return>		Zero on success.
 *
 * HISTORY
 *	November 12, 1997 - Andrew E. Mileski
 *	Written, tested, and released.
 */
static int udf_UTF8toCS0(dstring *ocu, struct ustr *utf, int length)
{
	unsigned c, i, max_val, utf_char;
	int utf_cnt, u_len;

	memset(ocu, 0, sizeof(dstring) * length);
	ocu[0] = 8;
	max_val = 0xffU;

try_again:
	u_len = 0U;
	utf_char = 0U;
	utf_cnt = 0U;
	for (i = 0U; i < utf->u_len; i++) {
		c = (uint8_t)utf->u_name[i];

		/* Complete a multi-byte UTF-8 character */
		if (utf_cnt) {
			utf_char = (utf_char << 6) | (c & 0x3fU);
			if (--utf_cnt)
				continue;
		} else {
			/* Check for a multi-byte UTF-8 character */
			if (c & 0x80U) {
				/* Start a multi-byte UTF-8 character */
				if ((c & 0xe0U) == 0xc0U) {
					utf_char = c & 0x1fU;
					utf_cnt = 1;
				} else if ((c & 0xf0U) == 0xe0U) {
					utf_char = c & 0x0fU;
					utf_cnt = 2;
				} else if ((c & 0xf8U) == 0xf0U) {
					utf_char = c & 0x07U;
					utf_cnt = 3;
				} else if ((c & 0xfcU) == 0xf8U) {
					utf_char = c & 0x03U;
					utf_cnt = 4;
				} else if ((c & 0xfeU) == 0xfcU) {
					utf_char = c & 0x01U;
					utf_cnt = 5;
				} else {
					goto error_out;
				}
				continue;
			} else {
				/* Single byte UTF-8 character (most common) */
				utf_char = c;
			}
		}

		/* Choose no compression if necessary */
		if (utf_char > max_val) {
			if (max_val == 0xffU) {
				max_val = 0xffffU;
				ocu[0] = (uint8_t)0x10U;
				goto try_again;
			}
			goto error_out;
		}

		if (max_val == 0xffffU)
			ocu[++u_len] = (uint8_t)(utf_char >> 8);
		ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
	}

	if (utf_cnt) {
error_out:
		ocu[++u_len] = '?';
		printk(KERN_DEBUG "udf: bad UTF-8 character\n");
	}

	ocu[length - 1] = (uint8_t)u_len + 1;

	return u_len + 1;
}

static int udf_CS0toNLS(struct nls_table *nls, struct ustr *utf_o,
			const struct ustr *ocu_i)
{
	const uint8_t *ocu;
	uint8_t cmp_id, ocu_len;
	int i;


	ocu_len = ocu_i->u_len;
	if (ocu_len == 0) {
		memset(utf_o, 0, sizeof(struct ustr));
		return 0;
	}

	cmp_id = ocu_i->u_cmpID;
	if (cmp_id != 8 && cmp_id != 16) {
		memset(utf_o, 0, sizeof(struct ustr));
		printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
		       cmp_id, ocu_i->u_name);
		return 0;
	}

	ocu = ocu_i->u_name;
	utf_o->u_len = 0;
	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
		/* Expand OSTA compressed Unicode to Unicode */
		uint32_t c = ocu[i++];
		if (cmp_id == 16)
			c = (c << 8) | ocu[i++];

		utf_o->u_len += nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
					      UDF_NAME_LEN - utf_o->u_len);
	}
	utf_o->u_cmpID = 8;

	return utf_o->u_len;
}

static int udf_NLStoCS0(struct nls_table *nls, dstring *ocu, struct ustr *uni,
			int length)
{
	unsigned len, i, max_val;
	uint16_t uni_char;
	int u_len;

	memset(ocu, 0, sizeof(dstring) * length);
	ocu[0] = 8;
	max_val = 0xffU;

try_again:
	u_len = 0U;
	for (i = 0U; i < uni->u_len; i++) {
		len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
		if (len <= 0)
			continue;

		if (uni_char > max_val) {
			max_val = 0xffffU;
			ocu[0] = (uint8_t)0x10U;
			goto try_again;
		}

		if (max_val == 0xffffU)
			ocu[++u_len] = (uint8_t)(uni_char >> 8);
		ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
		i += len - 1;
	}

	ocu[length - 1] = (uint8_t)u_len + 1;
	return u_len + 1;
}

int udf_get_filename(struct super_block *sb, uint8_t *sname, uint8_t *dname,
		     int flen)
{
	struct ustr filename, unifilename;
	int len;

	if (udf_build_ustr_exact(&unifilename, sname, flen))
		return 0;

	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
		if (!udf_CS0toUTF8(&filename, &unifilename)) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			return 0;
		}
	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
		if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, &filename,
				  &unifilename)) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			return 0;
		}
	} else
		return 0;

	len = udf_translate_to_linux(dname, filename.u_name, filename.u_len,
				     unifilename.u_name, unifilename.u_len);
	if (len)
		return len;

	return 0;
}

int udf_put_filename(struct super_block *sb, const uint8_t *sname,
		     uint8_t *dname, int flen)
{
	struct ustr unifilename;
	int namelen;

	if (!udf_char_to_ustr(&unifilename, sname, flen))
		return 0;

	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
		namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
		if (!namelen)
			return 0;
	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
		namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
					&unifilename, UDF_NAME_LEN);
		if (!namelen)
			return 0;
	} else
		return 0;

	return namelen;
}

#define ILLEGAL_CHAR_MARK	'_'
#define EXT_MARK		'.'
#define CRC_MARK		'#'
#define EXT_SIZE 		5

static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
				  int udfLen, uint8_t *fidName,
				  int fidNameLen)
{
	int index, newIndex = 0, needsCRC = 0;
	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	unsigned short valueCRC;
	uint8_t curr;
	const uint8_t hexChar[] = "0123456789ABCDEF";

	if (udfName[0] == '.' &&
	    (udfLen == 1 || (udfLen == 2 && udfName[1] == '.'))) {
		needsCRC = 1;
		newIndex = udfLen;
		memcpy(newName, udfName, udfLen);
	} else {
		for (index = 0; index < udfLen; index++) {
			curr = udfName[index];
			if (curr == '/' || curr == 0) {
				needsCRC = 1;
				curr = ILLEGAL_CHAR_MARK;
				while (index + 1 < udfLen &&
						(udfName[index + 1] == '/' ||
						 udfName[index + 1] == 0))
					index++;
			}
			if (curr == EXT_MARK &&
					(udfLen - index - 1) <= EXT_SIZE) {
				if (udfLen == index + 1)
					hasExt = 0;
				else {
					hasExt = 1;
					extIndex = index;
					newExtIndex = newIndex;
				}
			}
			if (newIndex < 256)
				newName[newIndex++] = curr;
			else
				needsCRC = 1;
		}
	}
	if (needsCRC) {
		uint8_t ext[EXT_SIZE];
		int localExtIndex = 0;

		if (hasExt) {
			int maxFilenameLen;
			for (index = 0;
			     index < EXT_SIZE && extIndex + index + 1 < udfLen;
			     index++) {
				curr = udfName[extIndex + index + 1];

				if (curr == '/' || curr == 0) {
					needsCRC = 1;
					curr = ILLEGAL_CHAR_MARK;
					while (extIndex + index + 2 < udfLen &&
					      (index + 1 < EXT_SIZE &&
						(udfName[extIndex + index + 2] == '/' ||
						 udfName[extIndex + index + 2] == 0)))
						index++;
				}
				ext[localExtIndex++] = curr;
			}
			maxFilenameLen = 250 - localExtIndex;
			if (newIndex > maxFilenameLen)
				newIndex = maxFilenameLen;
			else
				newIndex = newExtIndex;
		} else if (newIndex > 250)
			newIndex = 250;
		newName[newIndex++] = CRC_MARK;
		valueCRC = udf_crc(fidName, fidNameLen, 0);
		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];

		if (hasExt) {
			newName[newIndex++] = EXT_MARK;
			for (index = 0; index < localExtIndex; index++)
				newName[newIndex++] = ext[index];
		}
	}

	return newIndex;
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* unicode.c
	3	*
	4	* PURPOSE
	5	* Routines for converting between UTF-8 and OSTA Compressed Unicode.
	6	* Also handles filename mangling
	7	*
	8	* DESCRIPTION
	9	* OSTA Compressed Unicode is explained in the OSTA UDF specification.
	10	* http://www.osta.org/
	11	* UTF-8 is explained in the IETF RFC XXXX.
	12	* ftp://ftp.internic.net/rfc/rfcxxxx.txt
	13	*
1da177e4 LT	14	* COPYRIGHT
	15	* This file is distributed under the terms of the GNU General Public
	16	* License (GPL). Copies of the GPL can be obtained from:
	17	* ftp://prep.ai.mit.edu/pub/gnu/GPL
	18	* Each contributing author retains all rights to their own work.
	19	*/
	20
	21	#include "udfdecl.h"
	22
	23	#include <linux/kernel.h>
	24	#include <linux/string.h> /* for memset */
	25	#include <linux/nls.h>
1da177e4 LT	26
	27	#include "udf_sb.h"
	28
	29	static int udf_translate_to_linux(uint8_t , uint8_t , int, uint8_t *, int);
	30
28de7948	31	static int udf_char_to_ustr(struct ustr dest, const uint8_t src, int strlen)
1da177e4	32	{
cb00ea35	33	if ((!dest) \|\| (!src) \|\| (!strlen) \|\| (strlen > UDF_NAME_LEN - 2))
1da177e4	34	return 0;
28de7948	35
1da177e4 LT	36	memset(dest, 0, sizeof(struct ustr));
	37	memcpy(dest->u_name, src, strlen);
	38	dest->u_cmpID = 0x08;
	39	dest->u_len = strlen;
28de7948	40
1da177e4 LT	41	return strlen;
	42	}
	43
	44	/*
	45	* udf_build_ustr
	46	*/
28de7948	47	int udf_build_ustr(struct ustr dest, dstring ptr, int size)
1da177e4 LT	48	{
	49	int usesize;
	50
6305a0a9	51	if (!dest \|\| !ptr \|\| !size)
1da177e4	52	return -1;
6305a0a9	53	BUG_ON(size < 2);
1da177e4	54
6305a0a9 MS	55	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
6305a0a9 MS	56	usesize = min(usesize, size - 2);
cb00ea35	57	dest->u_cmpID = ptr[0];
6305a0a9 MS	58	dest->u_len = usesize;
	59	memcpy(dest->u_name, ptr + 1, usesize);
	60	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);
28de7948	61
1da177e4 LT	62	return 0;
	63	}
	64
	65	/*
	66	* udf_build_ustr_exact
	67	*/
28de7948	68	static int udf_build_ustr_exact(struct ustr dest, dstring ptr, int exactsize)
1da177e4	69	{
cb00ea35	70	if ((!dest) \|\| (!ptr) \|\| (!exactsize))
1da177e4 LT	71	return -1;
	72
	73	memset(dest, 0, sizeof(struct ustr));
cb00ea35 CG	74	dest->u_cmpID = ptr[0];
	75	dest->u_len = exactsize - 1;
	76	memcpy(dest->u_name, ptr + 1, exactsize - 1);
28de7948	77
1da177e4 LT	78	return 0;
	79	}
	80
	81	/*
	82	* udf_ocu_to_utf8
	83	*
	84	* PURPOSE
	85	* Convert OSTA Compressed Unicode to the UTF-8 equivalent.
	86	*
1da177e4 LT	87	* PRE-CONDITIONS
	88	* utf Pointer to UTF-8 output buffer.
	89	* ocu Pointer to OSTA Compressed Unicode input buffer
	90	* of size UDF_NAME_LEN bytes.
	91	* both of type "struct ustr *"
	92	*
	93	* POST-CONDITIONS
	94	* <return> Zero on success.
	95	*
	96	* HISTORY
	97	* November 12, 1997 - Andrew E. Mileski
	98	* Written, tested, and released.
	99	*/
79cfe0ff	100	int udf_CS0toUTF8(struct ustr utf_o, const struct ustr ocu_i)
1da177e4	101	{
79cfe0ff	102	const uint8_t *ocu;
1da177e4 LT	103	uint8_t cmp_id, ocu_len;
	104	int i;
	105
1da177e4	106	ocu_len = ocu_i->u_len;
cb00ea35	107	if (ocu_len == 0) {
1da177e4	108	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	109	return 0;
	110	}
	111
79cfe0ff	112	cmp_id = ocu_i->u_cmpID;
	113	if (cmp_id != 8 && cmp_id != 16) {
	114	memset(utf_o, 0, sizeof(struct ustr));
cb00ea35 CG	115	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
cb00ea35 CG	116	cmp_id, ocu_i->u_name);
1da177e4 LT	117	return 0;
	118	}
	119
79cfe0ff	120	ocu = ocu_i->u_name;
79cfe0ff	121	utf_o->u_len = 0;
cb00ea35	122	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	123
1da177e4 LT	124	/* Expand OSTA compressed Unicode to Unicode */
79cfe0ff	125	uint32_t c = ocu[i++];
1da177e4 LT	126	if (cmp_id == 16)
	127	c = (c << 8) \| ocu[i++];
	128
	129	/* Compress Unicode to UTF-8 */
79cfe0ff	130	if (c < 0x80U)
28de7948	131	utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
79cfe0ff	132	else if (c < 0x800U) {
4b11111a MS	133	utf_o->u_name[utf_o->u_len++] =
	134	(uint8_t)(0xc0 \| (c >> 6));
	135	utf_o->u_name[utf_o->u_len++] =
	136	(uint8_t)(0x80 \| (c & 0x3f));
cb00ea35	137	} else {
4b11111a MS	138	utf_o->u_name[utf_o->u_len++] =
	139	(uint8_t)(0xe0 \| (c >> 12));
	140	utf_o->u_name[utf_o->u_len++] =
	141	(uint8_t)(0x80 \|
	142	((c >> 6) & 0x3f));
	143	utf_o->u_name[utf_o->u_len++] =
	144	(uint8_t)(0x80 \| (c & 0x3f));
1da177e4 LT	145	}
1da177e4 LT	146	}
cb00ea35	147	utf_o->u_cmpID = 8;
1da177e4 LT	148
	149	return utf_o->u_len;
	150	}
	151
	152	/*
	153	*
	154	* udf_utf8_to_ocu
	155	*
	156	* PURPOSE
	157	* Convert UTF-8 to the OSTA Compressed Unicode equivalent.
	158	*
	159	* DESCRIPTION
	160	* This routine is only called by udf_lookup().
	161	*
	162	* PRE-CONDITIONS
	163	* ocu Pointer to OSTA Compressed Unicode output
	164	* buffer of size UDF_NAME_LEN bytes.
	165	* utf Pointer to UTF-8 input buffer.
	166	* utf_len Length of UTF-8 input buffer in bytes.
	167	*
	168	* POST-CONDITIONS
	169	* <return> Zero on success.
	170	*
	171	* HISTORY
	172	* November 12, 1997 - Andrew E. Mileski
	173	* Written, tested, and released.
	174	*/
28de7948	175	static int udf_UTF8toCS0(dstring ocu, struct ustr utf, int length)
1da177e4 LT	176	{
	177	unsigned c, i, max_val, utf_char;
	178	int utf_cnt, u_len;
	179
	180	memset(ocu, 0, sizeof(dstring) * length);
	181	ocu[0] = 8;
	182	max_val = 0xffU;
	183
28de7948	184	try_again:
1da177e4 LT	185	u_len = 0U;
	186	utf_char = 0U;
	187	utf_cnt = 0U;
cb00ea35	188	for (i = 0U; i < utf->u_len; i++) {
28de7948	189	c = (uint8_t)utf->u_name[i];
1da177e4 LT	190
1da177e4 LT	191	/* Complete a multi-byte UTF-8 character */
cb00ea35	192	if (utf_cnt) {
1da177e4 LT	193	utf_char = (utf_char << 6) \| (c & 0x3fU);
	194	if (--utf_cnt)
	195	continue;
cb00ea35	196	} else {
1da177e4	197	/* Check for a multi-byte UTF-8 character */
cb00ea35	198	if (c & 0x80U) {
1da177e4	199	/* Start a multi-byte UTF-8 character */
cb00ea35	200	if ((c & 0xe0U) == 0xc0U) {
1da177e4 LT	201	utf_char = c & 0x1fU;
1da177e4 LT	202	utf_cnt = 1;
cb00ea35	203	} else if ((c & 0xf0U) == 0xe0U) {
1da177e4 LT	204	utf_char = c & 0x0fU;
1da177e4 LT	205	utf_cnt = 2;
cb00ea35	206	} else if ((c & 0xf8U) == 0xf0U) {
1da177e4 LT	207	utf_char = c & 0x07U;
1da177e4 LT	208	utf_cnt = 3;
cb00ea35	209	} else if ((c & 0xfcU) == 0xf8U) {
1da177e4 LT	210	utf_char = c & 0x03U;
1da177e4 LT	211	utf_cnt = 4;
cb00ea35	212	} else if ((c & 0xfeU) == 0xfcU) {
1da177e4 LT	213	utf_char = c & 0x01U;
1da177e4 LT	214	utf_cnt = 5;
28de7948	215	} else {
1da177e4	216	goto error_out;
28de7948	217	}
1da177e4	218	continue;
28de7948	219	} else {
1da177e4 LT	220	/* Single byte UTF-8 character (most common) */
1da177e4 LT	221	utf_char = c;
28de7948	222	}
1da177e4 LT	223	}
	224
	225	/* Choose no compression if necessary */
cb00ea35	226	if (utf_char > max_val) {
28de7948	227	if (max_val == 0xffU) {
1da177e4	228	max_val = 0xffffU;
28de7948	229	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	230	goto try_again;
	231	}
	232	goto error_out;
	233	}
	234
4b11111a	235	if (max_val == 0xffffU)
28de7948	236	ocu[++u_len] = (uint8_t)(utf_char >> 8);
28de7948	237	ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
1da177e4 LT	238	}
1da177e4 LT	239
cb00ea35	240	if (utf_cnt) {
28de7948	241	error_out:
1da177e4 LT	242	ocu[++u_len] = '?';
	243	printk(KERN_DEBUG "udf: bad UTF-8 character\n");
	244	}
	245
28de7948 CG	246	ocu[length - 1] = (uint8_t)u_len + 1;
28de7948 CG	247
1da177e4 LT	248	return u_len + 1;
	249	}
	250
cb00ea35	251	static int udf_CS0toNLS(struct nls_table nls, struct ustr utf_o,
34f953dd	252	const struct ustr *ocu_i)
1da177e4	253	{
34f953dd	254	const uint8_t *ocu;
1da177e4 LT	255	uint8_t cmp_id, ocu_len;
	256	int i;
	257
1da177e4 LT	258
1da177e4 LT	259	ocu_len = ocu_i->u_len;
cb00ea35	260	if (ocu_len == 0) {
1da177e4	261	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	262	return 0;
	263	}
	264
34f953dd	265	cmp_id = ocu_i->u_cmpID;
	266	if (cmp_id != 8 && cmp_id != 16) {
	267	memset(utf_o, 0, sizeof(struct ustr));
cb00ea35 CG	268	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
cb00ea35 CG	269	cmp_id, ocu_i->u_name);
1da177e4 LT	270	return 0;
	271	}
	272
34f953dd	273	ocu = ocu_i->u_name;
34f953dd	274	utf_o->u_len = 0;
cb00ea35	275	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4	276	/* Expand OSTA compressed Unicode to Unicode */
34f953dd	277	uint32_t c = ocu[i++];
1da177e4 LT	278	if (cmp_id == 16)
	279	c = (c << 8) \| ocu[i++];
	280
cb00ea35 CG	281	utf_o->u_len += nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
cb00ea35 CG	282	UDF_NAME_LEN - utf_o->u_len);
1da177e4	283	}
cb00ea35	284	utf_o->u_cmpID = 8;
1da177e4 LT	285
	286	return utf_o->u_len;
	287	}
	288
28de7948	289	static int udf_NLStoCS0(struct nls_table nls, dstring ocu, struct ustr *uni,
cb00ea35	290	int length)
1da177e4 LT	291	{
	292	unsigned len, i, max_val;
	293	uint16_t uni_char;
	294	int u_len;
	295
	296	memset(ocu, 0, sizeof(dstring) * length);
	297	ocu[0] = 8;
	298	max_val = 0xffU;
	299
28de7948	300	try_again:
1da177e4	301	u_len = 0U;
cb00ea35 CG	302	for (i = 0U; i < uni->u_len; i++) {
cb00ea35 CG	303	len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
1da177e4 LT	304	if (len <= 0)
	305	continue;
	306
cb00ea35	307	if (uni_char > max_val) {
1da177e4	308	max_val = 0xffffU;
28de7948	309	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	310	goto try_again;
1da177e4 LT	311	}
cb00ea35	312
1da177e4	313	if (max_val == 0xffffU)
28de7948 CG	314	ocu[++u_len] = (uint8_t)(uni_char >> 8);
28de7948 CG	315	ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
1da177e4 LT	316	i += len - 1;
	317	}
	318
28de7948	319	ocu[length - 1] = (uint8_t)u_len + 1;
1da177e4 LT	320	return u_len + 1;
	321	}
	322
28de7948	323	int udf_get_filename(struct super_block sb, uint8_t sname, uint8_t *dname,
cb00ea35	324	int flen)
1da177e4 LT	325	{
	326	struct ustr filename, unifilename;
	327	int len;
	328
4b11111a	329	if (udf_build_ustr_exact(&unifilename, sname, flen))
1da177e4	330	return 0;
1da177e4	331
cb00ea35 CG	332	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
cb00ea35 CG	333	if (!udf_CS0toUTF8(&filename, &unifilename)) {
4b11111a MS	334	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	335	sname);
1da177e4 LT	336	return 0;
1da177e4 LT	337	}
cb00ea35	338	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
4b11111a MS	339	if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, &filename,
	340	&unifilename)) {
	341	udf_debug("Failed in udf_get_filename: sname = %s\n",
	342	sname);
1da177e4 LT	343	return 0;
1da177e4 LT	344	}
4b11111a	345	} else
1da177e4 LT	346	return 0;
1da177e4 LT	347
28de7948 CG	348	len = udf_translate_to_linux(dname, filename.u_name, filename.u_len,
28de7948 CG	349	unifilename.u_name, unifilename.u_len);
4b11111a	350	if (len)
1da177e4	351	return len;
28de7948	352
1da177e4 LT	353	return 0;
	354	}
	355
28de7948 CG	356	int udf_put_filename(struct super_block sb, const uint8_t sname,
28de7948 CG	357	uint8_t *dname, int flen)
1da177e4 LT	358	{
	359	struct ustr unifilename;
	360	int namelen;
	361
4b11111a	362	if (!udf_char_to_ustr(&unifilename, sname, flen))
1da177e4	363	return 0;
1da177e4	364
cb00ea35	365	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
28de7948	366	namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
4b11111a	367	if (!namelen)
1da177e4	368	return 0;
cb00ea35	369	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
4b11111a MS	370	namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
	371	&unifilename, UDF_NAME_LEN);
	372	if (!namelen)
1da177e4	373	return 0;
4b11111a	374	} else
1da177e4 LT	375	return 0;
	376
	377	return namelen;
	378	}
	379
	380	#define ILLEGAL_CHAR_MARK '_'
28de7948 CG	381	#define EXT_MARK '.'
	382	#define CRC_MARK '#'
	383	#define EXT_SIZE 5
1da177e4	384
4b11111a MS	385	static int udf_translate_to_linux(uint8_t newName, uint8_t udfName,
	386	int udfLen, uint8_t *fidName,
	387	int fidNameLen)
1da177e4	388	{
cb00ea35	389	int index, newIndex = 0, needsCRC = 0;
1da177e4 LT	390	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	391	unsigned short valueCRC;
	392	uint8_t curr;
	393	const uint8_t hexChar[] = "0123456789ABCDEF";
	394
28de7948 CG	395	if (udfName[0] == '.' &&
28de7948 CG	396	(udfLen == 1 \|\| (udfLen == 2 && udfName[1] == '.'))) {
1da177e4 LT	397	needsCRC = 1;
	398	newIndex = udfLen;
	399	memcpy(newName, udfName, udfLen);
cb00ea35 CG	400	} else {
cb00ea35 CG	401	for (index = 0; index < udfLen; index++) {
1da177e4	402	curr = udfName[index];
cb00ea35	403	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	404	needsCRC = 1;
1da177e4 LT	405	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	406	while (index + 1 < udfLen &&
	407	(udfName[index + 1] == '/' \|\|
	408	udfName[index + 1] == 0))
1da177e4	409	index++;
4b11111a MS	410	}
	411	if (curr == EXT_MARK &&
	412	(udfLen - index - 1) <= EXT_SIZE) {
	413	if (udfLen == index + 1)
1da177e4	414	hasExt = 0;
4b11111a	415	else {
1da177e4 LT	416	hasExt = 1;
	417	extIndex = index;
	418	newExtIndex = newIndex;
	419	}
	420	}
	421	if (newIndex < 256)
	422	newName[newIndex++] = curr;
	423	else
	424	needsCRC = 1;
	425	}
	426	}
cb00ea35	427	if (needsCRC) {
1da177e4 LT	428	uint8_t ext[EXT_SIZE];
	429	int localExtIndex = 0;
	430
cb00ea35	431	if (hasExt) {
1da177e4	432	int maxFilenameLen;
4b11111a MS	433	for (index = 0;
	434	index < EXT_SIZE && extIndex + index + 1 < udfLen;
	435	index++) {
1da177e4 LT	436	curr = udfName[extIndex + index + 1];
1da177e4 LT	437
cb00ea35	438	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	439	needsCRC = 1;
1da177e4 LT	440	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	441	while (extIndex + index + 2 < udfLen &&
	442	(index + 1 < EXT_SIZE &&
	443	(udfName[extIndex + index + 2] == '/' \|\|
	444	udfName[extIndex + index + 2] == 0)))
1da177e4 LT	445	index++;
	446	}
	447	ext[localExtIndex++] = curr;
	448	}
	449	maxFilenameLen = 250 - localExtIndex;
	450	if (newIndex > maxFilenameLen)
	451	newIndex = maxFilenameLen;
	452	else
	453	newIndex = newExtIndex;
4b11111a	454	} else if (newIndex > 250)
1da177e4 LT	455	newIndex = 250;
	456	newName[newIndex++] = CRC_MARK;
	457	valueCRC = udf_crc(fidName, fidNameLen, 0);
	458	newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
	459	newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
	460	newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
	461	newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
	462
cb00ea35	463	if (hasExt) {
1da177e4	464	newName[newIndex++] = EXT_MARK;
cb00ea35	465	for (index = 0; index < localExtIndex; index++)
1da177e4 LT	466	newName[newIndex++] = ext[index];
	467	}
	468	}
28de7948	469
1da177e4 LT	470	return newIndex;
1da177e4 LT	471	}