[mirror_ubuntu-bionic-kernel.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/udf_fs.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;

	memset(dest, 0, sizeof(struct ustr));
	usesize = (size > UDF_NAME_LEN) ? UDF_NAME_LEN : size;
	dest->u_cmpID = ptr[0];
	dest->u_len = ptr[size - 1];
	memcpy(dest->u_name, ptr + 1, usesize - 1);

	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.
 *
 * DESCRIPTION
 *	This routine is only called by udf_filldir().
 *
 * 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, struct ustr *ocu_i)
{
	uint8_t *ocu;
	uint32_t c;
	uint8_t cmp_id, ocu_len;
	int i;

	ocu = ocu_i->u_name;

	ocu_len = ocu_i->u_len;
	cmp_id = ocu_i->u_cmpID;
	utf_o->u_len = 0;

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

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

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

		/* Expand OSTA compressed Unicode to Unicode */
		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,
			struct ustr *ocu_i)
{
	uint8_t *ocu;
	uint32_t c;
	uint8_t cmp_id, ocu_len;
	int i;

	ocu = ocu_i->u_name;

	ocu_len = ocu_i->u_len;
	cmp_id = ocu_i->u_cmpID;
	utf_o->u_len = 0;

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

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

	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
		/* Expand OSTA compressed Unicode to Unicode */
		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>
	26	#include <linux/udf_fs.h>
	27
	28	#include "udf_sb.h"
	29
	30	static int udf_translate_to_linux(uint8_t , uint8_t , int, uint8_t *, int);
	31
28de7948	32	static int udf_char_to_ustr(struct ustr dest, const uint8_t src, int strlen)
1da177e4	33	{
cb00ea35	34	if ((!dest) \|\| (!src) \|\| (!strlen) \|\| (strlen > UDF_NAME_LEN - 2))
1da177e4	35	return 0;
28de7948	36
1da177e4 LT	37	memset(dest, 0, sizeof(struct ustr));
	38	memcpy(dest->u_name, src, strlen);
	39	dest->u_cmpID = 0x08;
	40	dest->u_len = strlen;
28de7948	41
1da177e4 LT	42	return strlen;
	43	}
	44
	45	/*
	46	* udf_build_ustr
	47	*/
28de7948	48	int udf_build_ustr(struct ustr dest, dstring ptr, int size)
1da177e4 LT	49	{
	50	int usesize;
	51
cb00ea35	52	if ((!dest) \|\| (!ptr) \|\| (!size))
1da177e4 LT	53	return -1;
	54
	55	memset(dest, 0, sizeof(struct ustr));
cb00ea35 CG	56	usesize = (size > UDF_NAME_LEN) ? UDF_NAME_LEN : size;
	57	dest->u_cmpID = ptr[0];
	58	dest->u_len = ptr[size - 1];
	59	memcpy(dest->u_name, ptr + 1, usesize - 1);
28de7948	60
1da177e4 LT	61	return 0;
	62	}
	63
	64	/*
	65	* udf_build_ustr_exact
	66	*/
28de7948	67	static int udf_build_ustr_exact(struct ustr dest, dstring ptr, int exactsize)
1da177e4	68	{
cb00ea35	69	if ((!dest) \|\| (!ptr) \|\| (!exactsize))
1da177e4 LT	70	return -1;
	71
	72	memset(dest, 0, sizeof(struct ustr));
cb00ea35 CG	73	dest->u_cmpID = ptr[0];
	74	dest->u_len = exactsize - 1;
	75	memcpy(dest->u_name, ptr + 1, exactsize - 1);
28de7948	76
1da177e4 LT	77	return 0;
	78	}
	79
	80	/*
	81	* udf_ocu_to_utf8
	82	*
	83	* PURPOSE
	84	* Convert OSTA Compressed Unicode to the UTF-8 equivalent.
	85	*
	86	* DESCRIPTION
	87	* This routine is only called by udf_filldir().
	88	*
	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	*/
	102	int udf_CS0toUTF8(struct ustr utf_o, struct ustr ocu_i)
	103	{
	104	uint8_t *ocu;
	105	uint32_t c;
	106	uint8_t cmp_id, ocu_len;
	107	int i;
	108
	109	ocu = ocu_i->u_name;
	110
	111	ocu_len = ocu_i->u_len;
	112	cmp_id = ocu_i->u_cmpID;
	113	utf_o->u_len = 0;
	114
cb00ea35	115	if (ocu_len == 0) {
1da177e4 LT	116	memset(utf_o, 0, sizeof(struct ustr));
	117	utf_o->u_cmpID = 0;
	118	utf_o->u_len = 0;
	119	return 0;
	120	}
	121
cb00ea35 CG	122	if ((cmp_id != 8) && (cmp_id != 16)) {
	123	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
	124	cmp_id, ocu_i->u_name);
1da177e4 LT	125	return 0;
	126	}
	127
cb00ea35	128	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	129
	130	/* Expand OSTA compressed Unicode to Unicode */
	131	c = ocu[i++];
	132	if (cmp_id == 16)
	133	c = (c << 8) \| ocu[i++];
	134
	135	/* Compress Unicode to UTF-8 */
28de7948 CG	136	if (c < 0x80U) {
	137	utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
	138	} else if (c < 0x800U) {
	139	utf_o->u_name[utf_o->u_len++] = (uint8_t)(0xc0 \| (c >> 6));
	140	utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 \| (c & 0x3f));
cb00ea35	141	} else {
28de7948 CG	142	utf_o->u_name[utf_o->u_len++] = (uint8_t)(0xe0 \| (c >> 12));
	143	utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 \| ((c >> 6) & 0x3f));
	144	utf_o->u_name[utf_o->u_len++] = (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
cb00ea35	235	if (max_val == 0xffffU) {
28de7948	236	ocu[++u_len] = (uint8_t)(utf_char >> 8);
1da177e4	237	}
28de7948	238	ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
1da177e4 LT	239	}
1da177e4 LT	240
cb00ea35	241	if (utf_cnt) {
28de7948	242	error_out:
1da177e4 LT	243	ocu[++u_len] = '?';
	244	printk(KERN_DEBUG "udf: bad UTF-8 character\n");
	245	}
	246
28de7948 CG	247	ocu[length - 1] = (uint8_t)u_len + 1;
28de7948 CG	248
1da177e4 LT	249	return u_len + 1;
	250	}
	251
cb00ea35 CG	252	static int udf_CS0toNLS(struct nls_table nls, struct ustr utf_o,
cb00ea35 CG	253	struct ustr *ocu_i)
1da177e4 LT	254	{
	255	uint8_t *ocu;
	256	uint32_t c;
	257	uint8_t cmp_id, ocu_len;
	258	int i;
	259
	260	ocu = ocu_i->u_name;
	261
	262	ocu_len = ocu_i->u_len;
	263	cmp_id = ocu_i->u_cmpID;
	264	utf_o->u_len = 0;
	265
cb00ea35	266	if (ocu_len == 0) {
1da177e4 LT	267	memset(utf_o, 0, sizeof(struct ustr));
	268	utf_o->u_cmpID = 0;
	269	utf_o->u_len = 0;
	270	return 0;
	271	}
	272
cb00ea35 CG	273	if ((cmp_id != 8) && (cmp_id != 16)) {
	274	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
	275	cmp_id, ocu_i->u_name);
1da177e4 LT	276	return 0;
	277	}
	278
cb00ea35	279	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	280	/* Expand OSTA compressed Unicode to Unicode */
	281	c = ocu[i++];
	282	if (cmp_id == 16)
	283	c = (c << 8) \| ocu[i++];
	284
cb00ea35 CG	285	utf_o->u_len += nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
cb00ea35 CG	286	UDF_NAME_LEN - utf_o->u_len);
1da177e4	287	}
cb00ea35	288	utf_o->u_cmpID = 8;
1da177e4 LT	289
	290	return utf_o->u_len;
	291	}
	292
28de7948	293	static int udf_NLStoCS0(struct nls_table nls, dstring ocu, struct ustr *uni,
cb00ea35	294	int length)
1da177e4 LT	295	{
	296	unsigned len, i, max_val;
	297	uint16_t uni_char;
	298	int u_len;
	299
	300	memset(ocu, 0, sizeof(dstring) * length);
	301	ocu[0] = 8;
	302	max_val = 0xffU;
	303
28de7948	304	try_again:
1da177e4	305	u_len = 0U;
cb00ea35 CG	306	for (i = 0U; i < uni->u_len; i++) {
cb00ea35 CG	307	len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
1da177e4 LT	308	if (len <= 0)
	309	continue;
	310
cb00ea35	311	if (uni_char > max_val) {
1da177e4	312	max_val = 0xffffU;
28de7948	313	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	314	goto try_again;
1da177e4 LT	315	}
cb00ea35	316
1da177e4	317	if (max_val == 0xffffU)
28de7948 CG	318	ocu[++u_len] = (uint8_t)(uni_char >> 8);
28de7948 CG	319	ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
1da177e4 LT	320	i += len - 1;
	321	}
	322
28de7948	323	ocu[length - 1] = (uint8_t)u_len + 1;
1da177e4 LT	324	return u_len + 1;
	325	}
	326
28de7948	327	int udf_get_filename(struct super_block sb, uint8_t sname, uint8_t *dname,
cb00ea35	328	int flen)
1da177e4 LT	329	{
	330	struct ustr filename, unifilename;
	331	int len;
	332
cb00ea35	333	if (udf_build_ustr_exact(&unifilename, sname, flen)) {
1da177e4 LT	334	return 0;
	335	}
	336
cb00ea35 CG	337	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
cb00ea35 CG	338	if (!udf_CS0toUTF8(&filename, &unifilename)) {
28de7948	339	udf_debug("Failed in udf_get_filename: sname = %s\n", sname);
1da177e4 LT	340	return 0;
1da177e4 LT	341	}
cb00ea35	342	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
28de7948 CG	343	if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, &filename, &unifilename)) {
28de7948 CG	344	udf_debug("Failed in udf_get_filename: sname = %s\n", sname);
1da177e4 LT	345	return 0;
1da177e4 LT	346	}
28de7948	347	} else {
1da177e4	348	return 0;
28de7948	349	}
1da177e4	350
28de7948 CG	351	len = udf_translate_to_linux(dname, filename.u_name, filename.u_len,
	352	unifilename.u_name, unifilename.u_len);
	353	if (len) {
1da177e4 LT	354	return len;
1da177e4 LT	355	}
28de7948	356
1da177e4 LT	357	return 0;
	358	}
	359
28de7948 CG	360	int udf_put_filename(struct super_block sb, const uint8_t sname,
28de7948 CG	361	uint8_t *dname, int flen)
1da177e4 LT	362	{
	363	struct ustr unifilename;
	364	int namelen;
	365
cb00ea35	366	if (!(udf_char_to_ustr(&unifilename, sname, flen))) {
1da177e4 LT	367	return 0;
	368	}
	369
cb00ea35	370	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
28de7948 CG	371	namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
28de7948 CG	372	if (!namelen) {
1da177e4 LT	373	return 0;
1da177e4 LT	374	}
cb00ea35	375	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
28de7948 CG	376	namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname, &unifilename, UDF_NAME_LEN);
28de7948 CG	377	if (!namelen) {
1da177e4 LT	378	return 0;
1da177e4 LT	379	}
28de7948	380	} else {
1da177e4	381	return 0;
28de7948	382	}
1da177e4 LT	383
	384	return namelen;
	385	}
	386
	387	#define ILLEGAL_CHAR_MARK '_'
28de7948 CG	388	#define EXT_MARK '.'
	389	#define CRC_MARK '#'
	390	#define EXT_SIZE 5
1da177e4	391
28de7948 CG	392	static int udf_translate_to_linux(uint8_t newName, uint8_t udfName, int udfLen,
28de7948 CG	393	uint8_t *fidName, int fidNameLen)
1da177e4	394	{
cb00ea35	395	int index, newIndex = 0, needsCRC = 0;
1da177e4 LT	396	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	397	unsigned short valueCRC;
	398	uint8_t curr;
	399	const uint8_t hexChar[] = "0123456789ABCDEF";
	400
28de7948 CG	401	if (udfName[0] == '.' &&
28de7948 CG	402	(udfLen == 1 \|\| (udfLen == 2 && udfName[1] == '.'))) {
1da177e4 LT	403	needsCRC = 1;
	404	newIndex = udfLen;
	405	memcpy(newName, udfName, udfLen);
cb00ea35 CG	406	} else {
cb00ea35 CG	407	for (index = 0; index < udfLen; index++) {
1da177e4	408	curr = udfName[index];
cb00ea35	409	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	410	needsCRC = 1;
1da177e4 LT	411	curr = ILLEGAL_CHAR_MARK;
28de7948 CG	412	while (index + 1 < udfLen && (udfName[index + 1] == '/' \|\|
28de7948 CG	413	udfName[index + 1] == 0))
1da177e4	414	index++;
28de7948 CG	415	} if (curr == EXT_MARK && (udfLen - index - 1) <= EXT_SIZE) {
28de7948 CG	416	if (udfLen == index + 1) {
1da177e4	417	hasExt = 0;
28de7948	418	} else {
1da177e4 LT	419	hasExt = 1;
	420	extIndex = index;
	421	newExtIndex = newIndex;
	422	}
	423	}
	424	if (newIndex < 256)
	425	newName[newIndex++] = curr;
	426	else
	427	needsCRC = 1;
	428	}
	429	}
cb00ea35	430	if (needsCRC) {
1da177e4 LT	431	uint8_t ext[EXT_SIZE];
	432	int localExtIndex = 0;
	433
cb00ea35	434	if (hasExt) {
1da177e4	435	int maxFilenameLen;
28de7948	436	for(index = 0; index < EXT_SIZE && extIndex + index + 1 < udfLen; index++) {
1da177e4 LT	437	curr = udfName[extIndex + index + 1];
1da177e4 LT	438
cb00ea35	439	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	440	needsCRC = 1;
1da177e4 LT	441	curr = ILLEGAL_CHAR_MARK;
28de7948 CG	442	while(extIndex + index + 2 < udfLen &&
	443	(index + 1 < EXT_SIZE
	444	&& (udfName[extIndex + index + 2] == '/' \|\|
	445	udfName[extIndex + index + 2] == 0)))
1da177e4 LT	446	index++;
	447	}
	448	ext[localExtIndex++] = curr;
	449	}
	450	maxFilenameLen = 250 - localExtIndex;
	451	if (newIndex > maxFilenameLen)
	452	newIndex = maxFilenameLen;
	453	else
	454	newIndex = newExtIndex;
28de7948	455	} else if (newIndex > 250) {
1da177e4	456	newIndex = 250;
28de7948	457	}
1da177e4 LT	458	newName[newIndex++] = CRC_MARK;
	459	valueCRC = udf_crc(fidName, fidNameLen, 0);
	460	newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
	461	newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
	462	newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
	463	newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
	464
cb00ea35	465	if (hasExt) {
1da177e4	466	newName[newIndex++] = EXT_MARK;
cb00ea35	467	for (index = 0; index < localExtIndex; index++)
1da177e4 LT	468	newName[newIndex++] = ext[index];
	469	}
	470	}
28de7948	471
1da177e4 LT	472	return newIndex;
1da177e4 LT	473	}