[deliverable/linux.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 <linux/crc-itu-t.h>
#include <linux/slab.h>

#include "udf_sb.h"

static int udf_translate_to_linux(uint8_t *, int, 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));
		pr_err("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 pr_fmt("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, len;


	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));
		pr_err("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++];

		len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
				    UDF_NAME_LEN - utf_o->u_len);
		/* Valid character? */
		if (len >= 0)
			utf_o->u_len += len;
		else
			utf_o->u_name[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)
{
	int len;
	unsigned 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)
			continue;
		/* Invalid character, deal with it */
		if (len < 0) {
			len = 1;
			uni_char = '?';
		}

		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, int slen,
		     uint8_t *dname, int dlen)
{
	struct ustr *filename, *unifilename;
	int len = 0;

	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	if (!filename)
		return 0;

	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	if (!unifilename)
		goto out1;

	if (udf_build_ustr_exact(unifilename, sname, slen))
		goto out2;

	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
		if (!udf_CS0toUTF8(filename, unifilename)) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			goto out2;
		}
	} 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);
			goto out2;
		}
	} else
		goto out2;

	len = udf_translate_to_linux(dname, dlen,
				     filename->u_name, filename->u_len,
				     unifilename->u_name, unifilename->u_len);
out2:
	kfree(unifilename);
out1:
	kfree(filename);
	return len;
}

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
/* Number of chars we need to store generated CRC to make filename unique */
#define CRC_LEN			5

static int udf_translate_to_linux(uint8_t *newName, int newLen,
				  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;

	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 < newLen)
				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 = newLen - CRC_LEN - localExtIndex;
			if (newIndex > maxFilenameLen)
				newIndex = maxFilenameLen;
			else
				newIndex = newExtIndex;
		} else if (newIndex > newLen - CRC_LEN)
			newIndex = newLen - CRC_LEN;
		newName[newIndex++] = CRC_MARK;
		valueCRC = crc_itu_t(0, fidName, fidNameLen);
		newName[newIndex++] = hex_asc_upper_hi(valueCRC >> 8);
		newName[newIndex++] = hex_asc_upper_lo(valueCRC >> 8);
		newName[newIndex++] = hex_asc_upper_hi(valueCRC);
		newName[newIndex++] = hex_asc_upper_lo(valueCRC);

		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>
f845fced	26	#include <linux/crc-itu-t.h>
5a0e3ad6	27	#include <linux/slab.h>
1da177e4 LT	28
	29	#include "udf_sb.h"
	30
0e5cc9a4 JK	31	static int udf_translate_to_linux(uint8_t , int, uint8_t , int, uint8_t *,
0e5cc9a4 JK	32	int);
1da177e4	33
28de7948	34	static int udf_char_to_ustr(struct ustr dest, const uint8_t src, int strlen)
1da177e4	35	{
cb00ea35	36	if ((!dest) \|\| (!src) \|\| (!strlen) \|\| (strlen > UDF_NAME_LEN - 2))
1da177e4	37	return 0;
28de7948	38
1da177e4 LT	39	memset(dest, 0, sizeof(struct ustr));
	40	memcpy(dest->u_name, src, strlen);
	41	dest->u_cmpID = 0x08;
	42	dest->u_len = strlen;
28de7948	43
1da177e4 LT	44	return strlen;
	45	}
	46
	47	/*
	48	* udf_build_ustr
	49	*/
28de7948	50	int udf_build_ustr(struct ustr dest, dstring ptr, int size)
1da177e4 LT	51	{
	52	int usesize;
	53
6305a0a9	54	if (!dest \|\| !ptr \|\| !size)
1da177e4	55	return -1;
6305a0a9	56	BUG_ON(size < 2);
1da177e4	57
6305a0a9 MS	58	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
6305a0a9 MS	59	usesize = min(usesize, size - 2);
cb00ea35	60	dest->u_cmpID = ptr[0];
6305a0a9 MS	61	dest->u_len = usesize;
	62	memcpy(dest->u_name, ptr + 1, usesize);
	63	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);
28de7948	64
1da177e4 LT	65	return 0;
	66	}
	67
	68	/*
	69	* udf_build_ustr_exact
	70	*/
28de7948	71	static int udf_build_ustr_exact(struct ustr dest, dstring ptr, int exactsize)
1da177e4	72	{
cb00ea35	73	if ((!dest) \|\| (!ptr) \|\| (!exactsize))
1da177e4 LT	74	return -1;
	75
	76	memset(dest, 0, sizeof(struct ustr));
cb00ea35 CG	77	dest->u_cmpID = ptr[0];
	78	dest->u_len = exactsize - 1;
	79	memcpy(dest->u_name, ptr + 1, exactsize - 1);
28de7948	80
1da177e4 LT	81	return 0;
	82	}
	83
	84	/*
	85	* udf_ocu_to_utf8
	86	*
	87	* PURPOSE
	88	* Convert OSTA Compressed Unicode to the UTF-8 equivalent.
	89	*
1da177e4 LT	90	* PRE-CONDITIONS
	91	* utf Pointer to UTF-8 output buffer.
	92	* ocu Pointer to OSTA Compressed Unicode input buffer
	93	* of size UDF_NAME_LEN bytes.
	94	* both of type "struct ustr *"
	95	*
	96	* POST-CONDITIONS
	97	* <return> Zero on success.
	98	*
	99	* HISTORY
	100	* November 12, 1997 - Andrew E. Mileski
	101	* Written, tested, and released.
	102	*/
79cfe0ff	103	int udf_CS0toUTF8(struct ustr utf_o, const struct ustr ocu_i)
1da177e4	104	{
79cfe0ff	105	const uint8_t *ocu;
1da177e4 LT	106	uint8_t cmp_id, ocu_len;
	107	int i;
	108
1da177e4	109	ocu_len = ocu_i->u_len;
cb00ea35	110	if (ocu_len == 0) {
1da177e4	111	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	112	return 0;
	113	}
	114
79cfe0ff	115	cmp_id = ocu_i->u_cmpID;
	116	if (cmp_id != 8 && cmp_id != 16) {
	117	memset(utf_o, 0, sizeof(struct ustr));
78ace70c	118	pr_err("unknown compression code (%d) stri=%s\n",
cb00ea35	119	cmp_id, ocu_i->u_name);
1da177e4 LT	120	return 0;
	121	}
	122
79cfe0ff	123	ocu = ocu_i->u_name;
79cfe0ff	124	utf_o->u_len = 0;
cb00ea35	125	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	126
1da177e4 LT	127	/* Expand OSTA compressed Unicode to Unicode */
79cfe0ff	128	uint32_t c = ocu[i++];
1da177e4 LT	129	if (cmp_id == 16)
	130	c = (c << 8) \| ocu[i++];
	131
	132	/* Compress Unicode to UTF-8 */
79cfe0ff	133	if (c < 0x80U)
28de7948	134	utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
79cfe0ff	135	else if (c < 0x800U) {
4b11111a MS	136	utf_o->u_name[utf_o->u_len++] =
	137	(uint8_t)(0xc0 \| (c >> 6));
	138	utf_o->u_name[utf_o->u_len++] =
	139	(uint8_t)(0x80 \| (c & 0x3f));
cb00ea35	140	} else {
4b11111a MS	141	utf_o->u_name[utf_o->u_len++] =
	142	(uint8_t)(0xe0 \| (c >> 12));
	143	utf_o->u_name[utf_o->u_len++] =
	144	(uint8_t)(0x80 \|
	145	((c >> 6) & 0x3f));
	146	utf_o->u_name[utf_o->u_len++] =
	147	(uint8_t)(0x80 \| (c & 0x3f));
1da177e4 LT	148	}
1da177e4 LT	149	}
cb00ea35	150	utf_o->u_cmpID = 8;
1da177e4 LT	151
	152	return utf_o->u_len;
	153	}
	154
	155	/*
	156	*
	157	* udf_utf8_to_ocu
	158	*
	159	* PURPOSE
	160	* Convert UTF-8 to the OSTA Compressed Unicode equivalent.
	161	*
	162	* DESCRIPTION
	163	* This routine is only called by udf_lookup().
	164	*
	165	* PRE-CONDITIONS
	166	* ocu Pointer to OSTA Compressed Unicode output
	167	* buffer of size UDF_NAME_LEN bytes.
	168	* utf Pointer to UTF-8 input buffer.
	169	* utf_len Length of UTF-8 input buffer in bytes.
	170	*
	171	* POST-CONDITIONS
	172	* <return> Zero on success.
	173	*
	174	* HISTORY
	175	* November 12, 1997 - Andrew E. Mileski
	176	* Written, tested, and released.
	177	*/
28de7948	178	static int udf_UTF8toCS0(dstring ocu, struct ustr utf, int length)
1da177e4 LT	179	{
	180	unsigned c, i, max_val, utf_char;
	181	int utf_cnt, u_len;
	182
	183	memset(ocu, 0, sizeof(dstring) * length);
	184	ocu[0] = 8;
	185	max_val = 0xffU;
	186
28de7948	187	try_again:
1da177e4 LT	188	u_len = 0U;
	189	utf_char = 0U;
	190	utf_cnt = 0U;
cb00ea35	191	for (i = 0U; i < utf->u_len; i++) {
28de7948	192	c = (uint8_t)utf->u_name[i];
1da177e4 LT	193
1da177e4 LT	194	/* Complete a multi-byte UTF-8 character */
cb00ea35	195	if (utf_cnt) {
1da177e4 LT	196	utf_char = (utf_char << 6) \| (c & 0x3fU);
	197	if (--utf_cnt)
	198	continue;
cb00ea35	199	} else {
1da177e4	200	/* Check for a multi-byte UTF-8 character */
cb00ea35	201	if (c & 0x80U) {
1da177e4	202	/* Start a multi-byte UTF-8 character */
cb00ea35	203	if ((c & 0xe0U) == 0xc0U) {
1da177e4 LT	204	utf_char = c & 0x1fU;
1da177e4 LT	205	utf_cnt = 1;
cb00ea35	206	} else if ((c & 0xf0U) == 0xe0U) {
1da177e4 LT	207	utf_char = c & 0x0fU;
1da177e4 LT	208	utf_cnt = 2;
cb00ea35	209	} else if ((c & 0xf8U) == 0xf0U) {
1da177e4 LT	210	utf_char = c & 0x07U;
1da177e4 LT	211	utf_cnt = 3;
cb00ea35	212	} else if ((c & 0xfcU) == 0xf8U) {
1da177e4 LT	213	utf_char = c & 0x03U;
1da177e4 LT	214	utf_cnt = 4;
cb00ea35	215	} else if ((c & 0xfeU) == 0xfcU) {
1da177e4 LT	216	utf_char = c & 0x01U;
1da177e4 LT	217	utf_cnt = 5;
28de7948	218	} else {
1da177e4	219	goto error_out;
28de7948	220	}
1da177e4	221	continue;
28de7948	222	} else {
1da177e4 LT	223	/* Single byte UTF-8 character (most common) */
1da177e4 LT	224	utf_char = c;
28de7948	225	}
1da177e4 LT	226	}
	227
	228	/* Choose no compression if necessary */
cb00ea35	229	if (utf_char > max_val) {
28de7948	230	if (max_val == 0xffU) {
1da177e4	231	max_val = 0xffffU;
28de7948	232	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	233	goto try_again;
	234	}
	235	goto error_out;
	236	}
	237
4b11111a	238	if (max_val == 0xffffU)
28de7948	239	ocu[++u_len] = (uint8_t)(utf_char >> 8);
28de7948	240	ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
1da177e4 LT	241	}
1da177e4 LT	242
cb00ea35	243	if (utf_cnt) {
28de7948	244	error_out:
1da177e4	245	ocu[++u_len] = '?';
78ace70c	246	printk(KERN_DEBUG pr_fmt("bad UTF-8 character\n"));
1da177e4 LT	247	}
1da177e4 LT	248
28de7948 CG	249	ocu[length - 1] = (uint8_t)u_len + 1;
28de7948 CG	250
1da177e4 LT	251	return u_len + 1;
	252	}
	253
cb00ea35	254	static int udf_CS0toNLS(struct nls_table nls, struct ustr utf_o,
34f953dd	255	const struct ustr *ocu_i)
1da177e4	256	{
34f953dd	257	const uint8_t *ocu;
1da177e4	258	uint8_t cmp_id, ocu_len;
59285c28	259	int i, len;
1da177e4	260
1da177e4 LT	261
1da177e4 LT	262	ocu_len = ocu_i->u_len;
cb00ea35	263	if (ocu_len == 0) {
1da177e4	264	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	265	return 0;
	266	}
	267
34f953dd	268	cmp_id = ocu_i->u_cmpID;
	269	if (cmp_id != 8 && cmp_id != 16) {
	270	memset(utf_o, 0, sizeof(struct ustr));
78ace70c	271	pr_err("unknown compression code (%d) stri=%s\n",
cb00ea35	272	cmp_id, ocu_i->u_name);
1da177e4 LT	273	return 0;
	274	}
	275
34f953dd	276	ocu = ocu_i->u_name;
34f953dd	277	utf_o->u_len = 0;
cb00ea35	278	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4	279	/* Expand OSTA compressed Unicode to Unicode */
34f953dd	280	uint32_t c = ocu[i++];
1da177e4 LT	281	if (cmp_id == 16)
	282	c = (c << 8) \| ocu[i++];
	283
59285c28 JK	284	len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
	285	UDF_NAME_LEN - utf_o->u_len);
	286	/* Valid character? */
	287	if (len >= 0)
	288	utf_o->u_len += len;
	289	else
	290	utf_o->u_name[utf_o->u_len++] = '?';
1da177e4	291	}
cb00ea35	292	utf_o->u_cmpID = 8;
1da177e4 LT	293
	294	return utf_o->u_len;
	295	}
	296
28de7948	297	static int udf_NLStoCS0(struct nls_table nls, dstring ocu, struct ustr *uni,
cb00ea35	298	int length)
1da177e4	299	{
59285c28 JK	300	int len;
59285c28 JK	301	unsigned i, max_val;
1da177e4 LT	302	uint16_t uni_char;
	303	int u_len;
	304
	305	memset(ocu, 0, sizeof(dstring) * length);
	306	ocu[0] = 8;
	307	max_val = 0xffU;
	308
28de7948	309	try_again:
1da177e4	310	u_len = 0U;
cb00ea35 CG	311	for (i = 0U; i < uni->u_len; i++) {
cb00ea35 CG	312	len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
59285c28	313	if (!len)
1da177e4	314	continue;
59285c28 JK	315	/* Invalid character, deal with it */
	316	if (len < 0) {
	317	len = 1;
	318	uni_char = '?';
	319	}
1da177e4	320
cb00ea35	321	if (uni_char > max_val) {
1da177e4	322	max_val = 0xffffU;
28de7948	323	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	324	goto try_again;
1da177e4 LT	325	}
cb00ea35	326
1da177e4	327	if (max_val == 0xffffU)
28de7948 CG	328	ocu[++u_len] = (uint8_t)(uni_char >> 8);
28de7948 CG	329	ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
1da177e4 LT	330	i += len - 1;
	331	}
	332
28de7948	333	ocu[length - 1] = (uint8_t)u_len + 1;
1da177e4 LT	334	return u_len + 1;
	335	}
	336
0e5cc9a4 JK	337	int udf_get_filename(struct super_block sb, uint8_t sname, int slen,
0e5cc9a4 JK	338	uint8_t *dname, int dlen)
1da177e4	339	{
530f1a5e MS	340	struct ustr filename, unifilename;
530f1a5e MS	341	int len = 0;
1da177e4	342
530f1a5e MS	343	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
530f1a5e MS	344	if (!filename)
1da177e4	345	return 0;
1da177e4	346
530f1a5e MS	347	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	348	if (!unifilename)
	349	goto out1;
	350
0e5cc9a4	351	if (udf_build_ustr_exact(unifilename, sname, slen))
530f1a5e MS	352	goto out2;
530f1a5e MS	353
cb00ea35	354	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
530f1a5e	355	if (!udf_CS0toUTF8(filename, unifilename)) {
4b11111a MS	356	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	357	sname);
530f1a5e	358	goto out2;
1da177e4	359	}
cb00ea35	360	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
530f1a5e MS	361	if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
530f1a5e MS	362	unifilename)) {
4b11111a MS	363	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	364	sname);
530f1a5e	365	goto out2;
1da177e4	366	}
4b11111a	367	} else
530f1a5e MS	368	goto out2;
530f1a5e MS	369
0e5cc9a4 JK	370	len = udf_translate_to_linux(dname, dlen,
0e5cc9a4 JK	371	filename->u_name, filename->u_len,
530f1a5e MS	372	unifilename->u_name, unifilename->u_len);
	373	out2:
	374	kfree(unifilename);
	375	out1:
	376	kfree(filename);
	377	return len;
1da177e4 LT	378	}
1da177e4 LT	379
28de7948 CG	380	int udf_put_filename(struct super_block sb, const uint8_t sname,
28de7948 CG	381	uint8_t *dname, int flen)
1da177e4 LT	382	{
	383	struct ustr unifilename;
	384	int namelen;
	385
4b11111a	386	if (!udf_char_to_ustr(&unifilename, sname, flen))
1da177e4	387	return 0;
1da177e4	388
cb00ea35	389	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
28de7948	390	namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
4b11111a	391	if (!namelen)
1da177e4	392	return 0;
cb00ea35	393	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
4b11111a MS	394	namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
	395	&unifilename, UDF_NAME_LEN);
	396	if (!namelen)
1da177e4	397	return 0;
4b11111a	398	} else
1da177e4 LT	399	return 0;
	400
	401	return namelen;
	402	}
	403
	404	#define ILLEGAL_CHAR_MARK '_'
28de7948 CG	405	#define EXT_MARK '.'
	406	#define CRC_MARK '#'
	407	#define EXT_SIZE 5
0e5cc9a4 JK	408	/* Number of chars we need to store generated CRC to make filename unique */
0e5cc9a4 JK	409	#define CRC_LEN 5
1da177e4	410
0e5cc9a4 JK	411	static int udf_translate_to_linux(uint8_t *newName, int newLen,
	412	uint8_t *udfName, int udfLen,
	413	uint8_t *fidName, int fidNameLen)
1da177e4	414	{
cb00ea35	415	int index, newIndex = 0, needsCRC = 0;
1da177e4 LT	416	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	417	unsigned short valueCRC;
	418	uint8_t curr;
1da177e4	419
28de7948 CG	420	if (udfName[0] == '.' &&
28de7948 CG	421	(udfLen == 1 \|\| (udfLen == 2 && udfName[1] == '.'))) {
1da177e4 LT	422	needsCRC = 1;
	423	newIndex = udfLen;
	424	memcpy(newName, udfName, udfLen);
cb00ea35 CG	425	} else {
cb00ea35 CG	426	for (index = 0; index < udfLen; index++) {
1da177e4	427	curr = udfName[index];
cb00ea35	428	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	429	needsCRC = 1;
1da177e4 LT	430	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	431	while (index + 1 < udfLen &&
	432	(udfName[index + 1] == '/' \|\|
	433	udfName[index + 1] == 0))
1da177e4	434	index++;
4b11111a MS	435	}
	436	if (curr == EXT_MARK &&
	437	(udfLen - index - 1) <= EXT_SIZE) {
	438	if (udfLen == index + 1)
1da177e4	439	hasExt = 0;
4b11111a	440	else {
1da177e4 LT	441	hasExt = 1;
	442	extIndex = index;
	443	newExtIndex = newIndex;
	444	}
	445	}
0e5cc9a4	446	if (newIndex < newLen)
1da177e4 LT	447	newName[newIndex++] = curr;
	448	else
	449	needsCRC = 1;
	450	}
	451	}
cb00ea35	452	if (needsCRC) {
1da177e4 LT	453	uint8_t ext[EXT_SIZE];
	454	int localExtIndex = 0;
	455
cb00ea35	456	if (hasExt) {
1da177e4	457	int maxFilenameLen;
4b11111a MS	458	for (index = 0;
	459	index < EXT_SIZE && extIndex + index + 1 < udfLen;
	460	index++) {
1da177e4 LT	461	curr = udfName[extIndex + index + 1];
1da177e4 LT	462
cb00ea35	463	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	464	needsCRC = 1;
1da177e4 LT	465	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	466	while (extIndex + index + 2 < udfLen &&
	467	(index + 1 < EXT_SIZE &&
	468	(udfName[extIndex + index + 2] == '/' \|\|
	469	udfName[extIndex + index + 2] == 0)))
1da177e4 LT	470	index++;
	471	}
	472	ext[localExtIndex++] = curr;
	473	}
0e5cc9a4	474	maxFilenameLen = newLen - CRC_LEN - localExtIndex;
1da177e4 LT	475	if (newIndex > maxFilenameLen)
	476	newIndex = maxFilenameLen;
	477	else
	478	newIndex = newExtIndex;
0e5cc9a4 JK	479	} else if (newIndex > newLen - CRC_LEN)
0e5cc9a4 JK	480	newIndex = newLen - CRC_LEN;
1da177e4	481	newName[newIndex++] = CRC_MARK;
f845fced	482	valueCRC = crc_itu_t(0, fidName, fidNameLen);
c7ff4821 AS	483	newName[newIndex++] = hex_asc_upper_hi(valueCRC >> 8);
	484	newName[newIndex++] = hex_asc_upper_lo(valueCRC >> 8);
	485	newName[newIndex++] = hex_asc_upper_hi(valueCRC);
	486	newName[newIndex++] = hex_asc_upper_lo(valueCRC);
1da177e4	487
cb00ea35	488	if (hasExt) {
1da177e4	489	newName[newIndex++] = EXT_MARK;
cb00ea35	490	for (index = 0; index < localExtIndex; index++)
1da177e4 LT	491	newName[newIndex++] = ext[index];
	492	}
	493	}
28de7948	494
1da177e4 LT	495	return newIndex;
1da177e4 LT	496	}