[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 <linux/crc-itu-t.h>
#include <linux/slab.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, 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));
		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++];

		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, uint8_t *dname,
		     int flen)
{
	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, flen))
		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, 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

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 = crc_itu_t(0, fidName, fidNameLen);
		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>
f845fced	26	#include <linux/crc-itu-t.h>
5a0e3ad6	27	#include <linux/slab.h>
1da177e4 LT	28
	29	#include "udf_sb.h"
	30
	31	static int udf_translate_to_linux(uint8_t , uint8_t , int, uint8_t *, int);
	32
28de7948	33	static int udf_char_to_ustr(struct ustr dest, const uint8_t src, int strlen)
1da177e4	34	{
cb00ea35	35	if ((!dest) \|\| (!src) \|\| (!strlen) \|\| (strlen > UDF_NAME_LEN - 2))
1da177e4	36	return 0;
28de7948	37
1da177e4 LT	38	memset(dest, 0, sizeof(struct ustr));
	39	memcpy(dest->u_name, src, strlen);
	40	dest->u_cmpID = 0x08;
	41	dest->u_len = strlen;
28de7948	42
1da177e4 LT	43	return strlen;
	44	}
	45
	46	/*
	47	* udf_build_ustr
	48	*/
28de7948	49	int udf_build_ustr(struct ustr dest, dstring ptr, int size)
1da177e4 LT	50	{
	51	int usesize;
	52
6305a0a9	53	if (!dest \|\| !ptr \|\| !size)
1da177e4	54	return -1;
6305a0a9	55	BUG_ON(size < 2);
1da177e4	56
6305a0a9 MS	57	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
6305a0a9 MS	58	usesize = min(usesize, size - 2);
cb00ea35	59	dest->u_cmpID = ptr[0];
6305a0a9 MS	60	dest->u_len = usesize;
	61	memcpy(dest->u_name, ptr + 1, usesize);
	62	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);
28de7948	63
1da177e4 LT	64	return 0;
	65	}
	66
	67	/*
	68	* udf_build_ustr_exact
	69	*/
28de7948	70	static int udf_build_ustr_exact(struct ustr dest, dstring ptr, int exactsize)
1da177e4	71	{
cb00ea35	72	if ((!dest) \|\| (!ptr) \|\| (!exactsize))
1da177e4 LT	73	return -1;
	74
	75	memset(dest, 0, sizeof(struct ustr));
cb00ea35 CG	76	dest->u_cmpID = ptr[0];
	77	dest->u_len = exactsize - 1;
	78	memcpy(dest->u_name, ptr + 1, exactsize - 1);
28de7948	79
1da177e4 LT	80	return 0;
	81	}
	82
	83	/*
	84	* udf_ocu_to_utf8
	85	*
	86	* PURPOSE
	87	* Convert OSTA Compressed Unicode to the UTF-8 equivalent.
	88	*
1da177e4 LT	89	* PRE-CONDITIONS
	90	* utf Pointer to UTF-8 output buffer.
	91	* ocu Pointer to OSTA Compressed Unicode input buffer
	92	* of size UDF_NAME_LEN bytes.
	93	* both of type "struct ustr *"
	94	*
	95	* POST-CONDITIONS
	96	* <return> Zero on success.
	97	*
	98	* HISTORY
	99	* November 12, 1997 - Andrew E. Mileski
	100	* Written, tested, and released.
	101	*/
79cfe0ff	102	int udf_CS0toUTF8(struct ustr utf_o, const struct ustr ocu_i)
1da177e4	103	{
79cfe0ff	104	const uint8_t *ocu;
1da177e4 LT	105	uint8_t cmp_id, ocu_len;
	106	int i;
	107
1da177e4	108	ocu_len = ocu_i->u_len;
cb00ea35	109	if (ocu_len == 0) {
1da177e4	110	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	111	return 0;
	112	}
	113
79cfe0ff	114	cmp_id = ocu_i->u_cmpID;
	115	if (cmp_id != 8 && cmp_id != 16) {
	116	memset(utf_o, 0, sizeof(struct ustr));
cb00ea35 CG	117	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
cb00ea35 CG	118	cmp_id, ocu_i->u_name);
1da177e4 LT	119	return 0;
	120	}
	121
79cfe0ff	122	ocu = ocu_i->u_name;
79cfe0ff	123	utf_o->u_len = 0;
cb00ea35	124	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	125
1da177e4 LT	126	/* Expand OSTA compressed Unicode to Unicode */
79cfe0ff	127	uint32_t c = ocu[i++];
1da177e4 LT	128	if (cmp_id == 16)
	129	c = (c << 8) \| ocu[i++];
	130
	131	/* Compress Unicode to UTF-8 */
79cfe0ff	132	if (c < 0x80U)
28de7948	133	utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
79cfe0ff	134	else if (c < 0x800U) {
4b11111a MS	135	utf_o->u_name[utf_o->u_len++] =
	136	(uint8_t)(0xc0 \| (c >> 6));
	137	utf_o->u_name[utf_o->u_len++] =
	138	(uint8_t)(0x80 \| (c & 0x3f));
cb00ea35	139	} else {
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	*
	156	* udf_utf8_to_ocu
	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	{
	179	unsigned c, i, max_val, utf_char;
	180	int utf_cnt, u_len;
	181
	182	memset(ocu, 0, sizeof(dstring) * length);
	183	ocu[0] = 8;
	184	max_val = 0xffU;
	185
28de7948	186	try_again:
1da177e4 LT	187	u_len = 0U;
	188	utf_char = 0U;
	189	utf_cnt = 0U;
cb00ea35	190	for (i = 0U; i < utf->u_len; i++) {
28de7948	191	c = (uint8_t)utf->u_name[i];
1da177e4 LT	192
1da177e4 LT	193	/* Complete a multi-byte UTF-8 character */
cb00ea35	194	if (utf_cnt) {
1da177e4 LT	195	utf_char = (utf_char << 6) \| (c & 0x3fU);
	196	if (--utf_cnt)
	197	continue;
cb00ea35	198	} else {
1da177e4	199	/* Check for a multi-byte UTF-8 character */
cb00ea35	200	if (c & 0x80U) {
1da177e4	201	/* Start a multi-byte UTF-8 character */
cb00ea35	202	if ((c & 0xe0U) == 0xc0U) {
1da177e4 LT	203	utf_char = c & 0x1fU;
1da177e4 LT	204	utf_cnt = 1;
cb00ea35	205	} else if ((c & 0xf0U) == 0xe0U) {
1da177e4 LT	206	utf_char = c & 0x0fU;
1da177e4 LT	207	utf_cnt = 2;
cb00ea35	208	} else if ((c & 0xf8U) == 0xf0U) {
1da177e4 LT	209	utf_char = c & 0x07U;
1da177e4 LT	210	utf_cnt = 3;
cb00ea35	211	} else if ((c & 0xfcU) == 0xf8U) {
1da177e4 LT	212	utf_char = c & 0x03U;
1da177e4 LT	213	utf_cnt = 4;
cb00ea35	214	} else if ((c & 0xfeU) == 0xfcU) {
1da177e4 LT	215	utf_char = c & 0x01U;
1da177e4 LT	216	utf_cnt = 5;
28de7948	217	} else {
1da177e4	218	goto error_out;
28de7948	219	}
1da177e4	220	continue;
28de7948	221	} else {
1da177e4 LT	222	/* Single byte UTF-8 character (most common) */
1da177e4 LT	223	utf_char = c;
28de7948	224	}
1da177e4 LT	225	}
	226
	227	/* Choose no compression if necessary */
cb00ea35	228	if (utf_char > max_val) {
28de7948	229	if (max_val == 0xffU) {
1da177e4	230	max_val = 0xffffU;
28de7948	231	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	232	goto try_again;
	233	}
	234	goto error_out;
	235	}
	236
4b11111a	237	if (max_val == 0xffffU)
28de7948	238	ocu[++u_len] = (uint8_t)(utf_char >> 8);
28de7948	239	ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
1da177e4 LT	240	}
1da177e4 LT	241
cb00ea35	242	if (utf_cnt) {
28de7948	243	error_out:
1da177e4 LT	244	ocu[++u_len] = '?';
	245	printk(KERN_DEBUG "udf: bad UTF-8 character\n");
	246	}
	247
28de7948 CG	248	ocu[length - 1] = (uint8_t)u_len + 1;
28de7948 CG	249
1da177e4 LT	250	return u_len + 1;
	251	}
	252
cb00ea35	253	static int udf_CS0toNLS(struct nls_table nls, struct ustr utf_o,
34f953dd	254	const struct ustr *ocu_i)
1da177e4	255	{
34f953dd	256	const uint8_t *ocu;
1da177e4	257	uint8_t cmp_id, ocu_len;
59285c28	258	int i, len;
1da177e4	259
1da177e4 LT	260
1da177e4 LT	261	ocu_len = ocu_i->u_len;
cb00ea35	262	if (ocu_len == 0) {
1da177e4	263	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	264	return 0;
	265	}
	266
34f953dd	267	cmp_id = ocu_i->u_cmpID;
	268	if (cmp_id != 8 && cmp_id != 16) {
	269	memset(utf_o, 0, sizeof(struct ustr));
cb00ea35 CG	270	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
cb00ea35 CG	271	cmp_id, ocu_i->u_name);
1da177e4 LT	272	return 0;
	273	}
	274
34f953dd	275	ocu = ocu_i->u_name;
34f953dd	276	utf_o->u_len = 0;
cb00ea35	277	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4	278	/* Expand OSTA compressed Unicode to Unicode */
34f953dd	279	uint32_t c = ocu[i++];
1da177e4 LT	280	if (cmp_id == 16)
	281	c = (c << 8) \| ocu[i++];
	282
59285c28 JK	283	len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
	284	UDF_NAME_LEN - utf_o->u_len);
	285	/* Valid character? */
	286	if (len >= 0)
	287	utf_o->u_len += len;
	288	else
	289	utf_o->u_name[utf_o->u_len++] = '?';
1da177e4	290	}
cb00ea35	291	utf_o->u_cmpID = 8;
1da177e4 LT	292
	293	return utf_o->u_len;
	294	}
	295
28de7948	296	static int udf_NLStoCS0(struct nls_table nls, dstring ocu, struct ustr *uni,
cb00ea35	297	int length)
1da177e4	298	{
59285c28 JK	299	int len;
59285c28 JK	300	unsigned i, max_val;
1da177e4 LT	301	uint16_t uni_char;
	302	int u_len;
	303
	304	memset(ocu, 0, sizeof(dstring) * length);
	305	ocu[0] = 8;
	306	max_val = 0xffU;
	307
28de7948	308	try_again:
1da177e4	309	u_len = 0U;
cb00ea35 CG	310	for (i = 0U; i < uni->u_len; i++) {
cb00ea35 CG	311	len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
59285c28	312	if (!len)
1da177e4	313	continue;
59285c28 JK	314	/* Invalid character, deal with it */
	315	if (len < 0) {
	316	len = 1;
	317	uni_char = '?';
	318	}
1da177e4	319
cb00ea35	320	if (uni_char > max_val) {
1da177e4	321	max_val = 0xffffU;
28de7948	322	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	323	goto try_again;
1da177e4 LT	324	}
cb00ea35	325
1da177e4	326	if (max_val == 0xffffU)
28de7948 CG	327	ocu[++u_len] = (uint8_t)(uni_char >> 8);
28de7948 CG	328	ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
1da177e4 LT	329	i += len - 1;
	330	}
	331
28de7948	332	ocu[length - 1] = (uint8_t)u_len + 1;
1da177e4 LT	333	return u_len + 1;
	334	}
	335
28de7948	336	int udf_get_filename(struct super_block sb, uint8_t sname, uint8_t *dname,
cb00ea35	337	int flen)
1da177e4	338	{
530f1a5e MS	339	struct ustr filename, unifilename;
530f1a5e MS	340	int len = 0;
1da177e4	341
530f1a5e MS	342	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
530f1a5e MS	343	if (!filename)
1da177e4	344	return 0;
1da177e4	345
530f1a5e MS	346	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	347	if (!unifilename)
	348	goto out1;
	349
	350	if (udf_build_ustr_exact(unifilename, sname, flen))
	351	goto out2;
	352
cb00ea35	353	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
530f1a5e	354	if (!udf_CS0toUTF8(filename, unifilename)) {
4b11111a MS	355	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	356	sname);
530f1a5e	357	goto out2;
1da177e4	358	}
cb00ea35	359	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
530f1a5e MS	360	if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
530f1a5e MS	361	unifilename)) {
4b11111a MS	362	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	363	sname);
530f1a5e	364	goto out2;
1da177e4	365	}
4b11111a	366	} else
530f1a5e MS	367	goto out2;
	368
	369	len = udf_translate_to_linux(dname, filename->u_name, filename->u_len,
	370	unifilename->u_name, unifilename->u_len);
	371	out2:
	372	kfree(unifilename);
	373	out1:
	374	kfree(filename);
	375	return len;
1da177e4 LT	376	}
1da177e4 LT	377
28de7948 CG	378	int udf_put_filename(struct super_block sb, const uint8_t sname,
28de7948 CG	379	uint8_t *dname, int flen)
1da177e4 LT	380	{
	381	struct ustr unifilename;
	382	int namelen;
	383
4b11111a	384	if (!udf_char_to_ustr(&unifilename, sname, flen))
1da177e4	385	return 0;
1da177e4	386
cb00ea35	387	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
28de7948	388	namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
4b11111a	389	if (!namelen)
1da177e4	390	return 0;
cb00ea35	391	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
4b11111a MS	392	namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
	393	&unifilename, UDF_NAME_LEN);
	394	if (!namelen)
1da177e4	395	return 0;
4b11111a	396	} else
1da177e4 LT	397	return 0;
	398
	399	return namelen;
	400	}
	401
	402	#define ILLEGAL_CHAR_MARK '_'
28de7948 CG	403	#define EXT_MARK '.'
	404	#define CRC_MARK '#'
	405	#define EXT_SIZE 5
1da177e4	406
4b11111a MS	407	static int udf_translate_to_linux(uint8_t newName, uint8_t udfName,
	408	int udfLen, uint8_t *fidName,
	409	int fidNameLen)
1da177e4	410	{
cb00ea35	411	int index, newIndex = 0, needsCRC = 0;
1da177e4 LT	412	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	413	unsigned short valueCRC;
	414	uint8_t curr;
	415	const uint8_t hexChar[] = "0123456789ABCDEF";
	416
28de7948 CG	417	if (udfName[0] == '.' &&
28de7948 CG	418	(udfLen == 1 \|\| (udfLen == 2 && udfName[1] == '.'))) {
1da177e4 LT	419	needsCRC = 1;
	420	newIndex = udfLen;
	421	memcpy(newName, udfName, udfLen);
cb00ea35 CG	422	} else {
cb00ea35 CG	423	for (index = 0; index < udfLen; index++) {
1da177e4	424	curr = udfName[index];
cb00ea35	425	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	426	needsCRC = 1;
1da177e4 LT	427	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	428	while (index + 1 < udfLen &&
	429	(udfName[index + 1] == '/' \|\|
	430	udfName[index + 1] == 0))
1da177e4	431	index++;
4b11111a MS	432	}
	433	if (curr == EXT_MARK &&
	434	(udfLen - index - 1) <= EXT_SIZE) {
	435	if (udfLen == index + 1)
1da177e4	436	hasExt = 0;
4b11111a	437	else {
1da177e4 LT	438	hasExt = 1;
	439	extIndex = index;
	440	newExtIndex = newIndex;
	441	}
	442	}
	443	if (newIndex < 256)
	444	newName[newIndex++] = curr;
	445	else
	446	needsCRC = 1;
	447	}
	448	}
cb00ea35	449	if (needsCRC) {
1da177e4 LT	450	uint8_t ext[EXT_SIZE];
	451	int localExtIndex = 0;
	452
cb00ea35	453	if (hasExt) {
1da177e4	454	int maxFilenameLen;
4b11111a MS	455	for (index = 0;
	456	index < EXT_SIZE && extIndex + index + 1 < udfLen;
	457	index++) {
1da177e4 LT	458	curr = udfName[extIndex + index + 1];
1da177e4 LT	459
cb00ea35	460	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	461	needsCRC = 1;
1da177e4 LT	462	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	463	while (extIndex + index + 2 < udfLen &&
	464	(index + 1 < EXT_SIZE &&
	465	(udfName[extIndex + index + 2] == '/' \|\|
	466	udfName[extIndex + index + 2] == 0)))
1da177e4 LT	467	index++;
	468	}
	469	ext[localExtIndex++] = curr;
	470	}
	471	maxFilenameLen = 250 - localExtIndex;
	472	if (newIndex > maxFilenameLen)
	473	newIndex = maxFilenameLen;
	474	else
	475	newIndex = newExtIndex;
4b11111a	476	} else if (newIndex > 250)
1da177e4 LT	477	newIndex = 250;
1da177e4 LT	478	newName[newIndex++] = CRC_MARK;
f845fced	479	valueCRC = crc_itu_t(0, fidName, fidNameLen);
1da177e4 LT	480	newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
	481	newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
	482	newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
	483	newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
	484
cb00ea35	485	if (hasExt) {
1da177e4	486	newName[newIndex++] = EXT_MARK;
cb00ea35	487	for (index = 0; index < localExtIndex; index++)
1da177e4 LT	488	newName[newIndex++] = ext[index];
	489	}
	490	}
28de7948	491
1da177e4 LT	492	return newIndex;
1da177e4 LT	493	}