[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 void udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize)
{
	memset(dest, 0, sizeof(struct ustr));
	dest->u_cmpID = ptr[0];
	dest->u_len = exactsize - 1;
	memcpy(dest->u_name, ptr + 1, exactsize - 1);
}

/*
 * udf_CS0toUTF8
 *
 * 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>		>= 0 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 -EINVAL;
	}

	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) {
			if (utf_o->u_len > (UDF_NAME_LEN - 4))
				break;
			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 {
			if (utf_o->u_len > (UDF_NAME_LEN - 5))
				break;
			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_UTF8toCS0
 *
 * 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, u_ch;

	memset(ocu, 0, sizeof(dstring) * length);
	ocu[0] = 8;
	max_val = 0xffU;
	u_ch = 1;

try_again:
	u_len = 0U;
	utf_char = 0U;
	utf_cnt = 0U;
	for (i = 0U; i < utf->u_len; i++) {
		/* Name didn't fit? */
		if (u_len + 1 + u_ch >= length)
			return 0;

		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;
				u_ch = 2;
				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 -EINVAL;
	}

	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 - 2 - 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, u_ch;

	memset(ocu, 0, sizeof(dstring) * length);
	ocu[0] = 8;
	max_val = 0xffU;
	u_ch = 1;

try_again:
	u_len = 0U;
	for (i = 0U; i < uni->u_len; i++) {
		/* Name didn't fit? */
		if (u_len + 1 + u_ch >= length)
			return 0;
		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;
			u_ch = 2;
			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 ret;

	if (!slen)
		return -EIO;

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

	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	if (!unifilename) {
		ret = -ENOMEM;
		goto out1;
	}

	udf_build_ustr_exact(unifilename, sname, slen);
	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
		ret = udf_CS0toUTF8(filename, unifilename);
		if (ret < 0) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			goto out2;
		}
	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
		ret = udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
				   unifilename);
		if (ret < 0) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			goto out2;
		}
	} else
		BUG();

	ret = udf_translate_to_linux(dname, dlen,
				     filename->u_name, filename->u_len,
				     unifilename->u_name, unifilename->u_len);
	/* Zero length filename isn't valid... */
	if (ret == 0)
		ret = -EINVAL;
out2:
	kfree(unifilename);
out1:
	kfree(filename);
	return ret;
}

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	*/
31f2566f	71	static void udf_build_ustr_exact(struct ustr dest, dstring ptr, int exactsize)
1da177e4	72	{
1da177e4	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);
1da177e4 LT	77	}
	78
	79	/*
d67e4a48	80	* udf_CS0toUTF8
1da177e4 LT	81	*
	82	* PURPOSE
	83	* Convert OSTA Compressed Unicode to the UTF-8 equivalent.
	84	*
1da177e4 LT	85	* PRE-CONDITIONS
	86	* utf Pointer to UTF-8 output buffer.
	87	* ocu Pointer to OSTA Compressed Unicode input buffer
	88	* of size UDF_NAME_LEN bytes.
	89	* both of type "struct ustr *"
	90	*
	91	* POST-CONDITIONS
e9d4cf41	92	* <return> >= 0 on success.
1da177e4 LT	93	*
	94	* HISTORY
	95	* November 12, 1997 - Andrew E. Mileski
	96	* Written, tested, and released.
	97	*/
79cfe0ff	98	int udf_CS0toUTF8(struct ustr utf_o, const struct ustr ocu_i)
1da177e4	99	{
79cfe0ff	100	const uint8_t *ocu;
1da177e4 LT	101	uint8_t cmp_id, ocu_len;
	102	int i;
	103
1da177e4	104	ocu_len = ocu_i->u_len;
cb00ea35	105	if (ocu_len == 0) {
1da177e4	106	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	107	return 0;
	108	}
	109
79cfe0ff	110	cmp_id = ocu_i->u_cmpID;
	111	if (cmp_id != 8 && cmp_id != 16) {
	112	memset(utf_o, 0, sizeof(struct ustr));
78ace70c	113	pr_err("unknown compression code (%d) stri=%s\n",
cb00ea35	114	cmp_id, ocu_i->u_name);
e9d4cf41	115	return -EINVAL;
1da177e4 LT	116	}
1da177e4 LT	117
79cfe0ff	118	ocu = ocu_i->u_name;
79cfe0ff	119	utf_o->u_len = 0;
cb00ea35	120	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	121
1da177e4 LT	122	/* Expand OSTA compressed Unicode to Unicode */
79cfe0ff	123	uint32_t c = ocu[i++];
1da177e4 LT	124	if (cmp_id == 16)
	125	c = (c << 8) \| ocu[i++];
	126
	127	/* Compress Unicode to UTF-8 */
79cfe0ff	128	if (c < 0x80U)
28de7948	129	utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
79cfe0ff	130	else if (c < 0x800U) {
ad402b26 AG	131	if (utf_o->u_len > (UDF_NAME_LEN - 4))
ad402b26 AG	132	break;
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 {
ad402b26 AG	138	if (utf_o->u_len > (UDF_NAME_LEN - 5))
ad402b26 AG	139	break;
4b11111a MS	140	utf_o->u_name[utf_o->u_len++] =
	141	(uint8_t)(0xe0 \| (c >> 12));
	142	utf_o->u_name[utf_o->u_len++] =
	143	(uint8_t)(0x80 \|
	144	((c >> 6) & 0x3f));
	145	utf_o->u_name[utf_o->u_len++] =
	146	(uint8_t)(0x80 \| (c & 0x3f));
1da177e4 LT	147	}
1da177e4 LT	148	}
cb00ea35	149	utf_o->u_cmpID = 8;
1da177e4 LT	150
	151	return utf_o->u_len;
	152	}
	153
	154	/*
	155	*
d67e4a48	156	* udf_UTF8toCS0
1da177e4 LT	157	*
	158	* PURPOSE
	159	* Convert UTF-8 to the OSTA Compressed Unicode equivalent.
	160	*
	161	* DESCRIPTION
	162	* This routine is only called by udf_lookup().
	163	*
	164	* PRE-CONDITIONS
	165	* ocu Pointer to OSTA Compressed Unicode output
	166	* buffer of size UDF_NAME_LEN bytes.
	167	* utf Pointer to UTF-8 input buffer.
	168	* utf_len Length of UTF-8 input buffer in bytes.
	169	*
	170	* POST-CONDITIONS
	171	* <return> Zero on success.
	172	*
	173	* HISTORY
	174	* November 12, 1997 - Andrew E. Mileski
	175	* Written, tested, and released.
	176	*/
28de7948	177	static int udf_UTF8toCS0(dstring ocu, struct ustr utf, int length)
1da177e4 LT	178	{
1da177e4 LT	179	unsigned c, i, max_val, utf_char;
bb00c898	180	int utf_cnt, u_len, u_ch;
1da177e4 LT	181
	182	memset(ocu, 0, sizeof(dstring) * length);
	183	ocu[0] = 8;
	184	max_val = 0xffU;
bb00c898	185	u_ch = 1;
1da177e4	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++) {
bb00c898 AG	192	/* Name didn't fit? */
	193	if (u_len + 1 + u_ch >= length)
	194	return 0;
	195
28de7948	196	c = (uint8_t)utf->u_name[i];
1da177e4 LT	197
1da177e4 LT	198	/* Complete a multi-byte UTF-8 character */
cb00ea35	199	if (utf_cnt) {
1da177e4 LT	200	utf_char = (utf_char << 6) \| (c & 0x3fU);
	201	if (--utf_cnt)
	202	continue;
cb00ea35	203	} else {
1da177e4	204	/* Check for a multi-byte UTF-8 character */
cb00ea35	205	if (c & 0x80U) {
1da177e4	206	/* Start a multi-byte UTF-8 character */
cb00ea35	207	if ((c & 0xe0U) == 0xc0U) {
1da177e4 LT	208	utf_char = c & 0x1fU;
1da177e4 LT	209	utf_cnt = 1;
cb00ea35	210	} else if ((c & 0xf0U) == 0xe0U) {
1da177e4 LT	211	utf_char = c & 0x0fU;
1da177e4 LT	212	utf_cnt = 2;
cb00ea35	213	} else if ((c & 0xf8U) == 0xf0U) {
1da177e4 LT	214	utf_char = c & 0x07U;
1da177e4 LT	215	utf_cnt = 3;
cb00ea35	216	} else if ((c & 0xfcU) == 0xf8U) {
1da177e4 LT	217	utf_char = c & 0x03U;
1da177e4 LT	218	utf_cnt = 4;
cb00ea35	219	} else if ((c & 0xfeU) == 0xfcU) {
1da177e4 LT	220	utf_char = c & 0x01U;
1da177e4 LT	221	utf_cnt = 5;
28de7948	222	} else {
1da177e4	223	goto error_out;
28de7948	224	}
1da177e4	225	continue;
28de7948	226	} else {
1da177e4 LT	227	/* Single byte UTF-8 character (most common) */
1da177e4 LT	228	utf_char = c;
28de7948	229	}
1da177e4 LT	230	}
	231
	232	/* Choose no compression if necessary */
cb00ea35	233	if (utf_char > max_val) {
28de7948	234	if (max_val == 0xffU) {
1da177e4	235	max_val = 0xffffU;
28de7948	236	ocu[0] = (uint8_t)0x10U;
bb00c898	237	u_ch = 2;
1da177e4 LT	238	goto try_again;
	239	}
	240	goto error_out;
	241	}
	242
4b11111a	243	if (max_val == 0xffffU)
28de7948	244	ocu[++u_len] = (uint8_t)(utf_char >> 8);
28de7948	245	ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
1da177e4 LT	246	}
1da177e4 LT	247
cb00ea35	248	if (utf_cnt) {
28de7948	249	error_out:
1da177e4	250	ocu[++u_len] = '?';
78ace70c	251	printk(KERN_DEBUG pr_fmt("bad UTF-8 character\n"));
1da177e4 LT	252	}
1da177e4 LT	253
28de7948 CG	254	ocu[length - 1] = (uint8_t)u_len + 1;
28de7948 CG	255
1da177e4 LT	256	return u_len + 1;
	257	}
	258
cb00ea35	259	static int udf_CS0toNLS(struct nls_table nls, struct ustr utf_o,
34f953dd	260	const struct ustr *ocu_i)
1da177e4	261	{
34f953dd	262	const uint8_t *ocu;
1da177e4	263	uint8_t cmp_id, ocu_len;
59285c28	264	int i, len;
1da177e4	265
1da177e4 LT	266
1da177e4 LT	267	ocu_len = ocu_i->u_len;
cb00ea35	268	if (ocu_len == 0) {
1da177e4	269	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	270	return 0;
	271	}
	272
34f953dd	273	cmp_id = ocu_i->u_cmpID;
	274	if (cmp_id != 8 && cmp_id != 16) {
	275	memset(utf_o, 0, sizeof(struct ustr));
78ace70c	276	pr_err("unknown compression code (%d) stri=%s\n",
cb00ea35	277	cmp_id, ocu_i->u_name);
78fc2e69	278	return -EINVAL;
1da177e4 LT	279	}
1da177e4 LT	280
34f953dd	281	ocu = ocu_i->u_name;
34f953dd	282	utf_o->u_len = 0;
cb00ea35	283	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4	284	/* Expand OSTA compressed Unicode to Unicode */
34f953dd	285	uint32_t c = ocu[i++];
1da177e4 LT	286	if (cmp_id == 16)
	287	c = (c << 8) \| ocu[i++];
	288
59285c28	289	len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
ad402b26	290	UDF_NAME_LEN - 2 - utf_o->u_len);
59285c28 JK	291	/* Valid character? */
	292	if (len >= 0)
	293	utf_o->u_len += len;
	294	else
	295	utf_o->u_name[utf_o->u_len++] = '?';
1da177e4	296	}
cb00ea35	297	utf_o->u_cmpID = 8;
1da177e4 LT	298
	299	return utf_o->u_len;
	300	}
	301
28de7948	302	static int udf_NLStoCS0(struct nls_table nls, dstring ocu, struct ustr *uni,
cb00ea35	303	int length)
1da177e4	304	{
59285c28 JK	305	int len;
59285c28 JK	306	unsigned i, max_val;
1da177e4	307	uint16_t uni_char;
bb00c898	308	int u_len, u_ch;
1da177e4 LT	309
	310	memset(ocu, 0, sizeof(dstring) * length);
	311	ocu[0] = 8;
	312	max_val = 0xffU;
bb00c898	313	u_ch = 1;
1da177e4	314
28de7948	315	try_again:
1da177e4	316	u_len = 0U;
cb00ea35	317	for (i = 0U; i < uni->u_len; i++) {
bb00c898 AG	318	/* Name didn't fit? */
	319	if (u_len + 1 + u_ch >= length)
	320	return 0;
cb00ea35	321	len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
59285c28	322	if (!len)
1da177e4	323	continue;
59285c28 JK	324	/* Invalid character, deal with it */
	325	if (len < 0) {
	326	len = 1;
	327	uni_char = '?';
	328	}
1da177e4	329
cb00ea35	330	if (uni_char > max_val) {
1da177e4	331	max_val = 0xffffU;
28de7948	332	ocu[0] = (uint8_t)0x10U;
bb00c898	333	u_ch = 2;
1da177e4 LT	334	goto try_again;
1da177e4 LT	335	}
cb00ea35	336
1da177e4	337	if (max_val == 0xffffU)
28de7948 CG	338	ocu[++u_len] = (uint8_t)(uni_char >> 8);
28de7948 CG	339	ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
1da177e4 LT	340	i += len - 1;
	341	}
	342
28de7948	343	ocu[length - 1] = (uint8_t)u_len + 1;
1da177e4 LT	344	return u_len + 1;
	345	}
	346
0e5cc9a4 JK	347	int udf_get_filename(struct super_block sb, uint8_t sname, int slen,
0e5cc9a4 JK	348	uint8_t *dname, int dlen)
1da177e4	349	{
530f1a5e	350	struct ustr filename, unifilename;
6ce63836	351	int ret;
1da177e4	352
31f2566f FF	353	if (!slen)
	354	return -EIO;
	355
530f1a5e MS	356	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
530f1a5e MS	357	if (!filename)
5ceb8b55	358	return -ENOMEM;
1da177e4	359
530f1a5e	360	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
5ceb8b55 FF	361	if (!unifilename) {
5ceb8b55 FF	362	ret = -ENOMEM;
530f1a5e	363	goto out1;
5ceb8b55	364	}
530f1a5e	365
31f2566f	366	udf_build_ustr_exact(unifilename, sname, slen);
cb00ea35	367	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
e9d4cf41 FF	368	ret = udf_CS0toUTF8(filename, unifilename);
e9d4cf41 FF	369	if (ret < 0) {
4b11111a MS	370	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	371	sname);
530f1a5e	372	goto out2;
1da177e4	373	}
cb00ea35	374	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
78fc2e69 FF	375	ret = udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
	376	unifilename);
	377	if (ret < 0) {
4b11111a MS	378	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	379	sname);
530f1a5e	380	goto out2;
1da177e4	381	}
4b11111a	382	} else
5dce54b7	383	BUG();
530f1a5e	384
5ceb8b55	385	ret = udf_translate_to_linux(dname, dlen,
0e5cc9a4	386	filename->u_name, filename->u_len,
530f1a5e	387	unifilename->u_name, unifilename->u_len);
6ce63836 FF	388	/* Zero length filename isn't valid... */
	389	if (ret == 0)
	390	ret = -EINVAL;
530f1a5e MS	391	out2:
	392	kfree(unifilename);
	393	out1:
	394	kfree(filename);
5ceb8b55	395	return ret;
1da177e4 LT	396	}
1da177e4 LT	397
28de7948 CG	398	int udf_put_filename(struct super_block sb, const uint8_t sname,
28de7948 CG	399	uint8_t *dname, int flen)
1da177e4 LT	400	{
	401	struct ustr unifilename;
	402	int namelen;
	403
4b11111a	404	if (!udf_char_to_ustr(&unifilename, sname, flen))
1da177e4	405	return 0;
1da177e4	406
cb00ea35	407	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
28de7948	408	namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
4b11111a	409	if (!namelen)
1da177e4	410	return 0;
cb00ea35	411	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
4b11111a MS	412	namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
	413	&unifilename, UDF_NAME_LEN);
	414	if (!namelen)
1da177e4	415	return 0;
4b11111a	416	} else
1da177e4 LT	417	return 0;
	418
	419	return namelen;
	420	}
	421
	422	#define ILLEGAL_CHAR_MARK '_'
28de7948 CG	423	#define EXT_MARK '.'
	424	#define CRC_MARK '#'
	425	#define EXT_SIZE 5
0e5cc9a4 JK	426	/* Number of chars we need to store generated CRC to make filename unique */
0e5cc9a4 JK	427	#define CRC_LEN 5
1da177e4	428
0e5cc9a4 JK	429	static int udf_translate_to_linux(uint8_t *newName, int newLen,
	430	uint8_t *udfName, int udfLen,
	431	uint8_t *fidName, int fidNameLen)
1da177e4	432	{
cb00ea35	433	int index, newIndex = 0, needsCRC = 0;
1da177e4 LT	434	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	435	unsigned short valueCRC;
	436	uint8_t curr;
1da177e4	437
28de7948 CG	438	if (udfName[0] == '.' &&
28de7948 CG	439	(udfLen == 1 \|\| (udfLen == 2 && udfName[1] == '.'))) {
1da177e4 LT	440	needsCRC = 1;
	441	newIndex = udfLen;
	442	memcpy(newName, udfName, udfLen);
cb00ea35 CG	443	} else {
cb00ea35 CG	444	for (index = 0; index < udfLen; index++) {
1da177e4	445	curr = udfName[index];
cb00ea35	446	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	447	needsCRC = 1;
1da177e4 LT	448	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	449	while (index + 1 < udfLen &&
	450	(udfName[index + 1] == '/' \|\|
	451	udfName[index + 1] == 0))
1da177e4	452	index++;
4b11111a MS	453	}
	454	if (curr == EXT_MARK &&
	455	(udfLen - index - 1) <= EXT_SIZE) {
	456	if (udfLen == index + 1)
1da177e4	457	hasExt = 0;
4b11111a	458	else {
1da177e4 LT	459	hasExt = 1;
	460	extIndex = index;
	461	newExtIndex = newIndex;
	462	}
	463	}
0e5cc9a4	464	if (newIndex < newLen)
1da177e4 LT	465	newName[newIndex++] = curr;
	466	else
	467	needsCRC = 1;
	468	}
	469	}
cb00ea35	470	if (needsCRC) {
1da177e4 LT	471	uint8_t ext[EXT_SIZE];
	472	int localExtIndex = 0;
	473
cb00ea35	474	if (hasExt) {
1da177e4	475	int maxFilenameLen;
4b11111a MS	476	for (index = 0;
	477	index < EXT_SIZE && extIndex + index + 1 < udfLen;
	478	index++) {
1da177e4 LT	479	curr = udfName[extIndex + index + 1];
1da177e4 LT	480
cb00ea35	481	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	482	needsCRC = 1;
1da177e4 LT	483	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	484	while (extIndex + index + 2 < udfLen &&
	485	(index + 1 < EXT_SIZE &&
	486	(udfName[extIndex + index + 2] == '/' \|\|
	487	udfName[extIndex + index + 2] == 0)))
1da177e4 LT	488	index++;
	489	}
	490	ext[localExtIndex++] = curr;
	491	}
0e5cc9a4	492	maxFilenameLen = newLen - CRC_LEN - localExtIndex;
1da177e4 LT	493	if (newIndex > maxFilenameLen)
	494	newIndex = maxFilenameLen;
	495	else
	496	newIndex = newExtIndex;
0e5cc9a4 JK	497	} else if (newIndex > newLen - CRC_LEN)
0e5cc9a4 JK	498	newIndex = newLen - CRC_LEN;
1da177e4	499	newName[newIndex++] = CRC_MARK;
f845fced	500	valueCRC = crc_itu_t(0, fidName, fidNameLen);
c7ff4821 AS	501	newName[newIndex++] = hex_asc_upper_hi(valueCRC >> 8);
	502	newName[newIndex++] = hex_asc_upper_lo(valueCRC >> 8);
	503	newName[newIndex++] = hex_asc_upper_hi(valueCRC);
	504	newName[newIndex++] = hex_asc_upper_lo(valueCRC);
1da177e4	505
cb00ea35	506	if (hasExt) {
1da177e4	507	newName[newIndex++] = EXT_MARK;
cb00ea35	508	for (index = 0; index < localExtIndex; index++)
1da177e4 LT	509	newName[newIndex++] = ext[index];
	510	}
	511	}
28de7948	512
1da177e4 LT	513	return newIndex;
1da177e4 LT	514	}