[linux.git] / fs / cifs / cifs_unicode.c

/*
 *   fs/cifs/cifs_unicode.c
 *
 *   Copyright (c) International Business Machines  Corp., 2000,2009
 *   Modified by Steve French ([email protected])
 *
 *   This program is free software;  you can redistribute it and/or modify
 *   it under the terms of the GNU General Public License as published by
 *   the Free Software Foundation; either version 2 of the License, or
 *   (at your option) any later version.
 *
 *   This program is distributed in the hope that it will be useful,
 *   but WITHOUT ANY WARRANTY;  without even the implied warranty of
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
 *   the GNU General Public License for more details.
 *
 *   You should have received a copy of the GNU General Public License
 *   along with this program;  if not, write to the Free Software
 *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
 */
#include <linux/fs.h>
#include <linux/slab.h>
#include "cifs_fs_sb.h"
#include "cifs_unicode.h"
#include "cifs_uniupr.h"
#include "cifspdu.h"
#include "cifsglob.h"
#include "cifs_debug.h"

int cifs_remap(struct cifs_sb_info *cifs_sb)
{
	int map_type;

	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR)
		map_type = SFM_MAP_UNI_RSVD;
	else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
		map_type = SFU_MAP_UNI_RSVD;
	else
		map_type = NO_MAP_UNI_RSVD;

	return map_type;
}

/* Convert character using the SFU - "Services for Unix" remapping range */
static bool
convert_sfu_char(const __u16 src_char, char *target)
{
	/*
	 * BB: Cannot handle remapping UNI_SLASH until all the calls to
	 *     build_path_from_dentry are modified, as they use slash as
	 *     separator.
	 */
	switch (src_char) {
	case UNI_COLON:
		*target = ':';
		break;
	case UNI_ASTERISK:
		*target = '*';
		break;
	case UNI_QUESTION:
		*target = '?';
		break;
	case UNI_PIPE:
		*target = '|';
		break;
	case UNI_GRTRTHAN:
		*target = '>';
		break;
	case UNI_LESSTHAN:
		*target = '<';
		break;
	default:
		return false;
	}
	return true;
}

/* Convert character using the SFM - "Services for Mac" remapping range */
static bool
convert_sfm_char(const __u16 src_char, char *target)
{
	if (src_char >= 0xF001 && src_char <= 0xF01F) {
		*target = src_char - 0xF000;
		return true;
	}
	switch (src_char) {
	case SFM_COLON:
		*target = ':';
		break;
	case SFM_DOUBLEQUOTE:
		*target = '"';
		break;
	case SFM_ASTERISK:
		*target = '*';
		break;
	case SFM_QUESTION:
		*target = '?';
		break;
	case SFM_PIPE:
		*target = '|';
		break;
	case SFM_GRTRTHAN:
		*target = '>';
		break;
	case SFM_LESSTHAN:
		*target = '<';
		break;
	case SFM_SLASH:
		*target = '\\';
		break;
	case SFM_SPACE:
		*target = ' ';
		break;
	case SFM_PERIOD:
		*target = '.';
		break;
	default:
		return false;
	}
	return true;
}


/*
 * cifs_mapchar - convert a host-endian char to proper char in codepage
 * @target - where converted character should be copied
 * @src_char - 2 byte host-endian source character
 * @cp - codepage to which character should be converted
 * @map_type - How should the 7 NTFS/SMB reserved characters be mapped to UCS2?
 *
 * This function handles the conversion of a single character. It is the
 * responsibility of the caller to ensure that the target buffer is large
 * enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
 */
static int
cifs_mapchar(char *target, const __u16 *from, const struct nls_table *cp,
	     int maptype)
{
	int len = 1;
	__u16 src_char;

	src_char = *from;

	if ((maptype == SFM_MAP_UNI_RSVD) && convert_sfm_char(src_char, target))
		return len;
	else if ((maptype == SFU_MAP_UNI_RSVD) &&
		  convert_sfu_char(src_char, target))
		return len;

	/* if character not one of seven in special remap set */
	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
	if (len <= 0)
		goto surrogate_pair;

	return len;

surrogate_pair:
	/* convert SURROGATE_PAIR and IVS */
	if (strcmp(cp->charset, "utf8"))
		goto unknown;
	len = utf16s_to_utf8s(from, 3, UTF16_LITTLE_ENDIAN, target, 6);
	if (len <= 0)
		goto unknown;
	return len;

unknown:
	*target = '?';
	len = 1;
	return len;
}

/*
 * cifs_from_utf16 - convert utf16le string to local charset
 * @to - destination buffer
 * @from - source buffer
 * @tolen - destination buffer size (in bytes)
 * @fromlen - source buffer size (in bytes)
 * @codepage - codepage to which characters should be converted
 * @mapchar - should characters be remapped according to the mapchars option?
 *
 * Convert a little-endian utf16le string (as sent by the server) to a string
 * in the provided codepage. The tolen and fromlen parameters are to ensure
 * that the code doesn't walk off of the end of the buffer (which is always
 * a danger if the alignment of the source buffer is off). The destination
 * string is always properly null terminated and fits in the destination
 * buffer. Returns the length of the destination string in bytes (including
 * null terminator).
 *
 * Note that some windows versions actually send multiword UTF-16 characters
 * instead of straight UTF16-2. The linux nls routines however aren't able to
 * deal with those characters properly. In the event that we get some of
 * those characters, they won't be translated properly.
 */
int
cifs_from_utf16(char *to, const __le16 *from, int tolen, int fromlen,
		const struct nls_table *codepage, int map_type)
{
	int i, charlen, safelen;
	int outlen = 0;
	int nullsize = nls_nullsize(codepage);
	int fromwords = fromlen / 2;
	char tmp[NLS_MAX_CHARSET_SIZE];
	__u16 ftmp[3];		/* ftmp[3] = 3array x 2bytes = 6bytes UTF-16 */

	/*
	 * because the chars can be of varying widths, we need to take care
	 * not to overflow the destination buffer when we get close to the
	 * end of it. Until we get to this offset, we don't need to check
	 * for overflow however.
	 */
	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);

	for (i = 0; i < fromwords; i++) {
		ftmp[0] = get_unaligned_le16(&from[i]);
		if (ftmp[0] == 0)
			break;
		if (i + 1 < fromwords)
			ftmp[1] = get_unaligned_le16(&from[i + 1]);
		else
			ftmp[1] = 0;
		if (i + 2 < fromwords)
			ftmp[2] = get_unaligned_le16(&from[i + 2]);
		else
			ftmp[2] = 0;

		/*
		 * check to see if converting this character might make the
		 * conversion bleed into the null terminator
		 */
		if (outlen >= safelen) {
			charlen = cifs_mapchar(tmp, ftmp, codepage, map_type);
			if ((outlen + charlen) > (tolen - nullsize))
				break;
		}

		/* put converted char into 'to' buffer */
		charlen = cifs_mapchar(&to[outlen], ftmp, codepage, map_type);
		outlen += charlen;

		/* charlen (=bytes of UTF-8 for 1 character)
		 * 4bytes UTF-8(surrogate pair) is charlen=4
		 *   (4bytes UTF-16 code)
		 * 7-8bytes UTF-8(IVS) is charlen=3+4 or 4+4
		 *   (2 UTF-8 pairs divided to 2 UTF-16 pairs) */
		if (charlen == 4)
			i++;
		else if (charlen >= 5)
			/* 5-6bytes UTF-8 */
			i += 2;
	}

	/* properly null-terminate string */
	for (i = 0; i < nullsize; i++)
		to[outlen++] = 0;

	return outlen;
}

/*
 * NAME:	cifs_strtoUTF16()
 *
 * FUNCTION:	Convert character string to unicode string
 *
 */
int
cifs_strtoUTF16(__le16 *to, const char *from, int len,
	      const struct nls_table *codepage)
{
	int charlen;
	int i;
	wchar_t wchar_to; /* needed to quiet sparse */

	/* special case for utf8 to handle no plane0 chars */
	if (!strcmp(codepage->charset, "utf8")) {
		/*
		 * convert utf8 -> utf16, we assume we have enough space
		 * as caller should have assumed conversion does not overflow
		 * in destination len is length in wchar_t units (16bits)
		 */
		i  = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN,
				       (wchar_t *) to, len);

		/* if success terminate and exit */
		if (i >= 0)
			goto success;
		/*
		 * if fails fall back to UCS encoding as this
		 * function should not return negative values
		 * currently can fail only if source contains
		 * invalid encoded characters
		 */
	}

	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
		charlen = codepage->char2uni(from, len, &wchar_to);
		if (charlen < 1) {
			cifs_dbg(VFS, "strtoUTF16: char2uni of 0x%x returned %d\n",
				 *from, charlen);
			/* A question mark */
			wchar_to = 0x003f;
			charlen = 1;
		}
		put_unaligned_le16(wchar_to, &to[i]);
	}

success:
	put_unaligned_le16(0, &to[i]);
	return i;
}

/*
 * cifs_utf16_bytes - how long will a string be after conversion?
 * @utf16 - pointer to input string
 * @maxbytes - don't go past this many bytes of input string
 * @codepage - destination codepage
 *
 * Walk a utf16le string and return the number of bytes that the string will
 * be after being converted to the given charset, not including any null
 * termination required. Don't walk past maxbytes in the source buffer.
 */
int
cifs_utf16_bytes(const __le16 *from, int maxbytes,
		const struct nls_table *codepage)
{
	int i;
	int charlen, outlen = 0;
	int maxwords = maxbytes / 2;
	char tmp[NLS_MAX_CHARSET_SIZE];
	__u16 ftmp[3];

	for (i = 0; i < maxwords; i++) {
		ftmp[0] = get_unaligned_le16(&from[i]);
		if (ftmp[0] == 0)
			break;
		if (i + 1 < maxwords)
			ftmp[1] = get_unaligned_le16(&from[i + 1]);
		else
			ftmp[1] = 0;
		if (i + 2 < maxwords)
			ftmp[2] = get_unaligned_le16(&from[i + 2]);
		else
			ftmp[2] = 0;

		charlen = cifs_mapchar(tmp, ftmp, codepage, NO_MAP_UNI_RSVD);
		outlen += charlen;
	}

	return outlen;
}

/*
 * cifs_strndup_from_utf16 - copy a string from wire format to the local
 * codepage
 * @src - source string
 * @maxlen - don't walk past this many bytes in the source string
 * @is_unicode - is this a unicode string?
 * @codepage - destination codepage
 *
 * Take a string given by the server, convert it to the local codepage and
 * put it in a new buffer. Returns a pointer to the new string or NULL on
 * error.
 */
char *
cifs_strndup_from_utf16(const char *src, const int maxlen,
			const bool is_unicode, const struct nls_table *codepage)
{
	int len;
	char *dst;

	if (is_unicode) {
		len = cifs_utf16_bytes((__le16 *) src, maxlen, codepage);
		len += nls_nullsize(codepage);
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return NULL;
		cifs_from_utf16(dst, (__le16 *) src, len, maxlen, codepage,
			       NO_MAP_UNI_RSVD);
	} else {
		len = strnlen(src, maxlen);
		len++;
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return NULL;
		strlcpy(dst, src, len);
	}

	return dst;
}

static __le16 convert_to_sfu_char(char src_char)
{
	__le16 dest_char;

	switch (src_char) {
	case ':':
		dest_char = cpu_to_le16(UNI_COLON);
		break;
	case '*':
		dest_char = cpu_to_le16(UNI_ASTERISK);
		break;
	case '?':
		dest_char = cpu_to_le16(UNI_QUESTION);
		break;
	case '<':
		dest_char = cpu_to_le16(UNI_LESSTHAN);
		break;
	case '>':
		dest_char = cpu_to_le16(UNI_GRTRTHAN);
		break;
	case '|':
		dest_char = cpu_to_le16(UNI_PIPE);
		break;
	default:
		dest_char = 0;
	}

	return dest_char;
}

static __le16 convert_to_sfm_char(char src_char, bool end_of_string)
{
	__le16 dest_char;

	if (src_char >= 0x01 && src_char <= 0x1F) {
		dest_char = cpu_to_le16(src_char + 0xF000);
		return dest_char;
	}
	switch (src_char) {
	case ':':
		dest_char = cpu_to_le16(SFM_COLON);
		break;
	case '"':
		dest_char = cpu_to_le16(SFM_DOUBLEQUOTE);
		break;
	case '*':
		dest_char = cpu_to_le16(SFM_ASTERISK);
		break;
	case '?':
		dest_char = cpu_to_le16(SFM_QUESTION);
		break;
	case '<':
		dest_char = cpu_to_le16(SFM_LESSTHAN);
		break;
	case '>':
		dest_char = cpu_to_le16(SFM_GRTRTHAN);
		break;
	case '|':
		dest_char = cpu_to_le16(SFM_PIPE);
		break;
	case '.':
		if (end_of_string)
			dest_char = cpu_to_le16(SFM_PERIOD);
		else
			dest_char = 0;
		break;
	case ' ':
		if (end_of_string)
			dest_char = cpu_to_le16(SFM_SPACE);
		else
			dest_char = 0;
		break;
	default:
		dest_char = 0;
	}

	return dest_char;
}

/*
 * Convert 16 bit Unicode pathname to wire format from string in current code
 * page. Conversion may involve remapping up the six characters that are
 * only legal in POSIX-like OS (if they are present in the string). Path
 * names are little endian 16 bit Unicode on the wire
 */
int
cifsConvertToUTF16(__le16 *target, const char *source, int srclen,
		 const struct nls_table *cp, int map_chars)
{
	int i, charlen;
	int j = 0;
	char src_char;
	__le16 dst_char;
	wchar_t tmp;
	wchar_t *wchar_to;	/* UTF-16 */
	int ret;
	unicode_t u;

	if (map_chars == NO_MAP_UNI_RSVD)
		return cifs_strtoUTF16(target, source, PATH_MAX, cp);

	wchar_to = kzalloc(6, GFP_KERNEL);

	for (i = 0; i < srclen; j++) {
		src_char = source[i];
		charlen = 1;

		/* check if end of string */
		if (src_char == 0)
			goto ctoUTF16_out;

		/* see if we must remap this char */
		if (map_chars == SFU_MAP_UNI_RSVD)
			dst_char = convert_to_sfu_char(src_char);
		else if (map_chars == SFM_MAP_UNI_RSVD) {
			bool end_of_string;

			if (i == srclen - 1)
				end_of_string = true;
			else
				end_of_string = false;

			dst_char = convert_to_sfm_char(src_char, end_of_string);
		} else
			dst_char = 0;
		/*
		 * FIXME: We can not handle remapping backslash (UNI_SLASH)
		 * until all the calls to build_path_from_dentry are modified,
		 * as they use backslash as separator.
		 */
		if (dst_char == 0) {
			charlen = cp->char2uni(source + i, srclen - i, &tmp);
			dst_char = cpu_to_le16(tmp);

			/*
			 * if no match, use question mark, which at least in
			 * some cases serves as wild card
			 */
			if (charlen > 0)
				goto ctoUTF16;

			/* convert SURROGATE_PAIR */
			if (strcmp(cp->charset, "utf8") || !wchar_to)
				goto unknown;
			if (*(source + i) & 0x80) {
				charlen = utf8_to_utf32(source + i, 6, &u);
				if (charlen < 0)
					goto unknown;
			} else
				goto unknown;
			ret  = utf8s_to_utf16s(source + i, charlen,
					       UTF16_LITTLE_ENDIAN,
					       wchar_to, 6);
			if (ret < 0)
				goto unknown;

			i += charlen;
			dst_char = cpu_to_le16(*wchar_to);
			if (charlen <= 3)
				/* 1-3bytes UTF-8 to 2bytes UTF-16 */
				put_unaligned(dst_char, &target[j]);
			else if (charlen == 4) {
				/* 4bytes UTF-8(surrogate pair) to 4bytes UTF-16
				 * 7-8bytes UTF-8(IVS) divided to 2 UTF-16
				 *   (charlen=3+4 or 4+4) */
				put_unaligned(dst_char, &target[j]);
				dst_char = cpu_to_le16(*(wchar_to + 1));
				j++;
				put_unaligned(dst_char, &target[j]);
			} else if (charlen >= 5) {
				/* 5-6bytes UTF-8 to 6bytes UTF-16 */
				put_unaligned(dst_char, &target[j]);
				dst_char = cpu_to_le16(*(wchar_to + 1));
				j++;
				put_unaligned(dst_char, &target[j]);
				dst_char = cpu_to_le16(*(wchar_to + 2));
				j++;
				put_unaligned(dst_char, &target[j]);
			}
			continue;

unknown:
			dst_char = cpu_to_le16(0x003f);
			charlen = 1;
		}

ctoUTF16:
		/*
		 * character may take more than one byte in the source string,
		 * but will take exactly two bytes in the target string
		 */
		i += charlen;
		put_unaligned(dst_char, &target[j]);
	}

ctoUTF16_out:
	put_unaligned(0, &target[j]); /* Null terminate target unicode string */
	kfree(wchar_to);
	return j;
}

/*
 * cifs_local_to_utf16_bytes - how long will a string be after conversion?
 * @from - pointer to input string
 * @maxbytes - don't go past this many bytes of input string
 * @codepage - source codepage
 *
 * Walk a string and return the number of bytes that the string will
 * be after being converted to the given charset, not including any null
 * termination required. Don't walk past maxbytes in the source buffer.
 */

static int
cifs_local_to_utf16_bytes(const char *from, int len,
			  const struct nls_table *codepage)
{
	int charlen;
	int i;
	wchar_t wchar_to;

	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
		charlen = codepage->char2uni(from, len, &wchar_to);
		/* Failed conversion defaults to a question mark */
		if (charlen < 1)
			charlen = 1;
	}
	return 2 * i; /* UTF16 characters are two bytes */
}

/*
 * cifs_strndup_to_utf16 - copy a string to wire format from the local codepage
 * @src - source string
 * @maxlen - don't walk past this many bytes in the source string
 * @utf16_len - the length of the allocated string in bytes (including null)
 * @cp - source codepage
 * @remap - map special chars
 *
 * Take a string convert it from the local codepage to UTF16 and
 * put it in a new buffer. Returns a pointer to the new string or NULL on
 * error.
 */
__le16 *
cifs_strndup_to_utf16(const char *src, const int maxlen, int *utf16_len,
		      const struct nls_table *cp, int remap)
{
	int len;
	__le16 *dst;

	len = cifs_local_to_utf16_bytes(src, maxlen, cp);
	len += 2; /* NULL */
	dst = kmalloc(len, GFP_KERNEL);
	if (!dst) {
		*utf16_len = 0;
		return NULL;
	}
	cifsConvertToUTF16(dst, src, strlen(src), cp, remap);
	*utf16_len = len;
	return dst;
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* fs/cifs/cifs_unicode.c
	3	*
d185cda7	4	* Copyright (c) International Business Machines Corp., 2000,2009
1da177e4 LT	5	* Modified by Steve French ([email protected])
	6	*
	7	* This program is free software; you can redistribute it and/or modify
	8	* it under the terms of the GNU General Public License as published by
221601c3	9	* the Free Software Foundation; either version 2 of the License, or
1da177e4	10	* (at your option) any later version.
221601c3	11	*
1da177e4 LT	12	* This program is distributed in the hope that it will be useful,
	13	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	14	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
	15	* the GNU General Public License for more details.
	16	*
	17	* You should have received a copy of the GNU General Public License
221601c3	18	* along with this program; if not, write to the Free Software
1da177e4 LT	19	* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
	20	*/
	21	#include <linux/fs.h>
5a0e3ad6	22	#include <linux/slab.h>
2baa2682	23	#include "cifs_fs_sb.h"
1da177e4 LT	24	#include "cifs_unicode.h"
	25	#include "cifs_uniupr.h"
	26	#include "cifspdu.h"
3979877e	27	#include "cifsglob.h"
1da177e4 LT	28	#include "cifs_debug.h"
1da177e4 LT	29
2baa2682 SF	30	int cifs_remap(struct cifs_sb_info *cifs_sb)
	31	{
	32	int map_type;
	33
	34	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR)
	35	map_type = SFM_MAP_UNI_RSVD;
	36	else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
	37	map_type = SFU_MAP_UNI_RSVD;
	38	else
	39	map_type = NO_MAP_UNI_RSVD;
	40
	41	return map_type;
	42	}
	43
b693855f SF	44	/* Convert character using the SFU - "Services for Unix" remapping range */
	45	static bool
	46	convert_sfu_char(const __u16 src_char, char *target)
7fabf0c9	47	{
7fabf0c9 JL	48	/*
	49	* BB: Cannot handle remapping UNI_SLASH until all the calls to
	50	* build_path_from_dentry are modified, as they use slash as
	51	* separator.
	52	*/
ba2dbf30	53	switch (src_char) {
7fabf0c9 JL	54	case UNI_COLON:
	55	*target = ':';
	56	break;
581ade4d	57	case UNI_ASTERISK:
7fabf0c9 JL	58	target = '';
	59	break;
	60	case UNI_QUESTION:
	61	*target = '?';
	62	break;
	63	case UNI_PIPE:
	64	*target = '\|';
	65	break;
	66	case UNI_GRTRTHAN:
	67	*target = '>';
	68	break;
	69	case UNI_LESSTHAN:
	70	*target = '<';
	71	break;
	72	default:
b693855f	73	return false;
7fabf0c9	74	}
b693855f SF	75	return true;
b693855f SF	76	}
7fabf0c9	77
b693855f SF	78	/* Convert character using the SFM - "Services for Mac" remapping range */
	79	static bool
	80	convert_sfm_char(const __u16 src_char, char *target)
	81	{
7e46f090 BJ	82	if (src_char >= 0xF001 && src_char <= 0xF01F) {
	83	*target = src_char - 0xF000;
	84	return true;
	85	}
b693855f SF	86	switch (src_char) {
	87	case SFM_COLON:
	88	*target = ':';
	89	break;
85435d7a BJ	90	case SFM_DOUBLEQUOTE:
	91	*target = '"';
	92	break;
b693855f SF	93	case SFM_ASTERISK:
	94	target = '';
	95	break;
	96	case SFM_QUESTION:
	97	*target = '?';
	98	break;
	99	case SFM_PIPE:
	100	*target = '\|';
	101	break;
	102	case SFM_GRTRTHAN:
	103	*target = '>';
	104	break;
	105	case SFM_LESSTHAN:
	106	*target = '<';
	107	break;
	108	case SFM_SLASH:
	109	*target = '\\';
	110	break;
45e8a258 SF	111	case SFM_SPACE:
	112	*target = ' ';
	113	break;
	114	case SFM_PERIOD:
	115	*target = '.';
	116	break;
b693855f SF	117	default:
	118	return false;
	119	}
	120	return true;
	121	}
	122
	123
	124	/*
	125	* cifs_mapchar - convert a host-endian char to proper char in codepage
	126	* @target - where converted character should be copied
	127	* @src_char - 2 byte host-endian source character
	128	* @cp - codepage to which character should be converted
	129	* @map_type - How should the 7 NTFS/SMB reserved characters be mapped to UCS2?
	130	*
	131	* This function handles the conversion of a single character. It is the
	132	* responsibility of the caller to ensure that the target buffer is large
	133	* enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
	134	*/
	135	static int
b2910307	136	cifs_mapchar(char target, const __u16 from, const struct nls_table *cp,
b693855f SF	137	int maptype)
	138	{
	139	int len = 1;
b2910307 NA	140	__u16 src_char;
	141
	142	src_char = *from;
b693855f SF	143
	144	if ((maptype == SFM_MAP_UNI_RSVD) && convert_sfm_char(src_char, target))
	145	return len;
	146	else if ((maptype == SFU_MAP_UNI_RSVD) &&
	147	convert_sfu_char(src_char, target))
	148	return len;
7fabf0c9	149
b693855f	150	/* if character not one of seven in special remap set */
ba2dbf30	151	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
b2910307 NA	152	if (len <= 0)
	153	goto surrogate_pair;
	154
	155	return len;
	156
	157	surrogate_pair:
	158	/* convert SURROGATE_PAIR and IVS */
	159	if (strcmp(cp->charset, "utf8"))
	160	goto unknown;
	161	len = utf16s_to_utf8s(from, 3, UTF16_LITTLE_ENDIAN, target, 6);
	162	if (len <= 0)
	163	goto unknown;
	164	return len;
	165
	166	unknown:
	167	*target = '?';
	168	len = 1;
b693855f	169	return len;
7fabf0c9 JL	170	}
	171
	172	/*
acbbb76a	173	* cifs_from_utf16 - convert utf16le string to local charset
7fabf0c9 JL	174	* @to - destination buffer
	175	* @from - source buffer
	176	* @tolen - destination buffer size (in bytes)
	177	* @fromlen - source buffer size (in bytes)
	178	* @codepage - codepage to which characters should be converted
	179	* @mapchar - should characters be remapped according to the mapchars option?
	180	*
acbbb76a	181	* Convert a little-endian utf16le string (as sent by the server) to a string
7fabf0c9 JL	182	* in the provided codepage. The tolen and fromlen parameters are to ensure
	183	* that the code doesn't walk off of the end of the buffer (which is always
	184	* a danger if the alignment of the source buffer is off). The destination
	185	* string is always properly null terminated and fits in the destination
	186	* buffer. Returns the length of the destination string in bytes (including
	187	* null terminator).
	188	*
	189	* Note that some windows versions actually send multiword UTF-16 characters
acbbb76a	190	* instead of straight UTF16-2. The linux nls routines however aren't able to
7fabf0c9 JL	191	* deal with those characters properly. In the event that we get some of
	192	* those characters, they won't be translated properly.
	193	*/
	194	int
acbbb76a	195	cifs_from_utf16(char to, const __le16 from, int tolen, int fromlen,
b693855f	196	const struct nls_table *codepage, int map_type)
7fabf0c9 JL	197	{
	198	int i, charlen, safelen;
	199	int outlen = 0;
	200	int nullsize = nls_nullsize(codepage);
	201	int fromwords = fromlen / 2;
	202	char tmp[NLS_MAX_CHARSET_SIZE];
b2910307	203	__u16 ftmp[3]; /* ftmp[3] = 3array x 2bytes = 6bytes UTF-16 */
7fabf0c9 JL	204
	205	/*
	206	* because the chars can be of varying widths, we need to take care
	207	* not to overflow the destination buffer when we get close to the
	208	* end of it. Until we get to this offset, we don't need to check
	209	* for overflow however.
	210	*/
	211	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);
	212
ba2dbf30	213	for (i = 0; i < fromwords; i++) {
b2910307 NA	214	ftmp[0] = get_unaligned_le16(&from[i]);
b2910307 NA	215	if (ftmp[0] == 0)
ba2dbf30	216	break;
b2910307 NA	217	if (i + 1 < fromwords)
	218	ftmp[1] = get_unaligned_le16(&from[i + 1]);
	219	else
	220	ftmp[1] = 0;
	221	if (i + 2 < fromwords)
	222	ftmp[2] = get_unaligned_le16(&from[i + 2]);
	223	else
	224	ftmp[2] = 0;
ba2dbf30	225
7fabf0c9 JL	226	/*
	227	* check to see if converting this character might make the
	228	* conversion bleed into the null terminator
	229	*/
	230	if (outlen >= safelen) {
b693855f	231	charlen = cifs_mapchar(tmp, ftmp, codepage, map_type);
7fabf0c9 JL	232	if ((outlen + charlen) > (tolen - nullsize))
	233	break;
	234	}
	235
	236	/* put converted char into 'to' buffer */
b693855f	237	charlen = cifs_mapchar(&to[outlen], ftmp, codepage, map_type);
7fabf0c9	238	outlen += charlen;
b2910307 NA	239
	240	/* charlen (=bytes of UTF-8 for 1 character)
	241	* 4bytes UTF-8(surrogate pair) is charlen=4
	242	* (4bytes UTF-16 code)
	243	* 7-8bytes UTF-8(IVS) is charlen=3+4 or 4+4
	244	* (2 UTF-8 pairs divided to 2 UTF-16 pairs) */
	245	if (charlen == 4)
	246	i++;
	247	else if (charlen >= 5)
	248	/* 5-6bytes UTF-8 */
	249	i += 2;
7fabf0c9 JL	250	}
	251
	252	/* properly null-terminate string */
	253	for (i = 0; i < nullsize; i++)
	254	to[outlen++] = 0;
	255
	256	return outlen;
	257	}
	258
1da177e4	259	/*
acbbb76a	260	* NAME: cifs_strtoUTF16()
1da177e4 LT	261	*
	262	* FUNCTION: Convert character string to unicode string
	263	*
	264	*/
	265	int
acbbb76a	266	cifs_strtoUTF16(__le16 to, const char from, int len,
1da177e4 LT	267	const struct nls_table *codepage)
	268	{
	269	int charlen;
	270	int i;
ba2dbf30	271	wchar_t wchar_to; /* needed to quiet sparse */
1da177e4	272
fd3ba42c FZ	273	/* special case for utf8 to handle no plane0 chars */
	274	if (!strcmp(codepage->charset, "utf8")) {
	275	/*
	276	* convert utf8 -> utf16, we assume we have enough space
	277	* as caller should have assumed conversion does not overflow
	278	* in destination len is length in wchar_t units (16bits)
	279	*/
	280	i = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN,
	281	(wchar_t *) to, len);
	282
	283	/* if success terminate and exit */
	284	if (i >= 0)
	285	goto success;
	286	/*
	287	* if fails fall back to UCS encoding as this
	288	* function should not return negative values
	289	* currently can fail only if source contains
	290	* invalid encoded characters
	291	*/
	292	}
	293
1da177e4	294	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
ba2dbf30	295	charlen = codepage->char2uni(from, len, &wchar_to);
1da177e4	296	if (charlen < 1) {
f96637be JP	297	cifs_dbg(VFS, "strtoUTF16: char2uni of 0x%x returned %d\n",
f96637be JP	298	*from, charlen);
69114089	299	/* A question mark */
ba2dbf30	300	wchar_to = 0x003f;
1da177e4	301	charlen = 1;
ba2dbf30 JL	302	}
ba2dbf30 JL	303	put_unaligned_le16(wchar_to, &to[i]);
1da177e4 LT	304	}
1da177e4 LT	305
fd3ba42c	306	success:
ba2dbf30	307	put_unaligned_le16(0, &to[i]);
1da177e4 LT	308	return i;
	309	}
	310
b2910307 NA	311	/*
	312	* cifs_utf16_bytes - how long will a string be after conversion?
	313	* @utf16 - pointer to input string
	314	* @maxbytes - don't go past this many bytes of input string
	315	* @codepage - destination codepage
	316	*
	317	* Walk a utf16le string and return the number of bytes that the string will
	318	* be after being converted to the given charset, not including any null
	319	* termination required. Don't walk past maxbytes in the source buffer.
	320	*/
	321	int
	322	cifs_utf16_bytes(const __le16 *from, int maxbytes,
	323	const struct nls_table *codepage)
	324	{
	325	int i;
	326	int charlen, outlen = 0;
	327	int maxwords = maxbytes / 2;
	328	char tmp[NLS_MAX_CHARSET_SIZE];
	329	__u16 ftmp[3];
	330
	331	for (i = 0; i < maxwords; i++) {
	332	ftmp[0] = get_unaligned_le16(&from[i]);
	333	if (ftmp[0] == 0)
	334	break;
	335	if (i + 1 < maxwords)
	336	ftmp[1] = get_unaligned_le16(&from[i + 1]);
	337	else
	338	ftmp[1] = 0;
	339	if (i + 2 < maxwords)
	340	ftmp[2] = get_unaligned_le16(&from[i + 2]);
	341	else
	342	ftmp[2] = 0;
	343
	344	charlen = cifs_mapchar(tmp, ftmp, codepage, NO_MAP_UNI_RSVD);
	345	outlen += charlen;
	346	}
	347
	348	return outlen;
	349	}
	350
066ce689	351	/*
acbbb76a SF	352	* cifs_strndup_from_utf16 - copy a string from wire format to the local
acbbb76a SF	353	* codepage
066ce689 JL	354	* @src - source string
	355	* @maxlen - don't walk past this many bytes in the source string
	356	* @is_unicode - is this a unicode string?
	357	* @codepage - destination codepage
	358	*
	359	* Take a string given by the server, convert it to the local codepage and
	360	* put it in a new buffer. Returns a pointer to the new string or NULL on
	361	* error.
	362	*/
	363	char *
acbbb76a SF	364	cifs_strndup_from_utf16(const char *src, const int maxlen,
acbbb76a SF	365	const bool is_unicode, const struct nls_table *codepage)
066ce689 JL	366	{
	367	int len;
	368	char *dst;
	369
	370	if (is_unicode) {
acbbb76a	371	len = cifs_utf16_bytes((__le16 *) src, maxlen, codepage);
066ce689 JL	372	len += nls_nullsize(codepage);
	373	dst = kmalloc(len, GFP_KERNEL);
	374	if (!dst)
	375	return NULL;
acbbb76a	376	cifs_from_utf16(dst, (__le16 *) src, len, maxlen, codepage,
b693855f	377	NO_MAP_UNI_RSVD);
066ce689 JL	378	} else {
	379	len = strnlen(src, maxlen);
	380	len++;
	381	dst = kmalloc(len, GFP_KERNEL);
	382	if (!dst)
	383	return NULL;
	384	strlcpy(dst, src, len);
	385	}
	386
	387	return dst;
	388	}
	389
a4153cb1 SF	390	static __le16 convert_to_sfu_char(char src_char)
	391	{
	392	__le16 dest_char;
	393
	394	switch (src_char) {
	395	case ':':
	396	dest_char = cpu_to_le16(UNI_COLON);
	397	break;
	398	case '*':
	399	dest_char = cpu_to_le16(UNI_ASTERISK);
	400	break;
	401	case '?':
	402	dest_char = cpu_to_le16(UNI_QUESTION);
	403	break;
	404	case '<':
	405	dest_char = cpu_to_le16(UNI_LESSTHAN);
	406	break;
	407	case '>':
	408	dest_char = cpu_to_le16(UNI_GRTRTHAN);
	409	break;
	410	case '\|':
	411	dest_char = cpu_to_le16(UNI_PIPE);
	412	break;
	413	default:
	414	dest_char = 0;
	415	}
	416
	417	return dest_char;
	418	}
	419
45e8a258	420	static __le16 convert_to_sfm_char(char src_char, bool end_of_string)
a4153cb1 SF	421	{
	422	__le16 dest_char;
	423
7e46f090 BJ	424	if (src_char >= 0x01 && src_char <= 0x1F) {
	425	dest_char = cpu_to_le16(src_char + 0xF000);
	426	return dest_char;
	427	}
a4153cb1 SF	428	switch (src_char) {
	429	case ':':
	430	dest_char = cpu_to_le16(SFM_COLON);
	431	break;
85435d7a BJ	432	case '"':
	433	dest_char = cpu_to_le16(SFM_DOUBLEQUOTE);
	434	break;
a4153cb1 SF	435	case '*':
	436	dest_char = cpu_to_le16(SFM_ASTERISK);
	437	break;
	438	case '?':
	439	dest_char = cpu_to_le16(SFM_QUESTION);
	440	break;
	441	case '<':
	442	dest_char = cpu_to_le16(SFM_LESSTHAN);
	443	break;
	444	case '>':
	445	dest_char = cpu_to_le16(SFM_GRTRTHAN);
	446	break;
	447	case '\|':
	448	dest_char = cpu_to_le16(SFM_PIPE);
	449	break;
45e8a258 SF	450	case '.':
	451	if (end_of_string)
	452	dest_char = cpu_to_le16(SFM_PERIOD);
	453	else
	454	dest_char = 0;
	455	break;
	456	case ' ':
	457	if (end_of_string)
	458	dest_char = cpu_to_le16(SFM_SPACE);
	459	else
	460	dest_char = 0;
	461	break;
a4153cb1 SF	462	default:
	463	dest_char = 0;
	464	}
	465
	466	return dest_char;
	467	}
	468
84cdf74e JL	469	/*
	470	* Convert 16 bit Unicode pathname to wire format from string in current code
	471	* page. Conversion may involve remapping up the six characters that are
	472	* only legal in POSIX-like OS (if they are present in the string). Path
	473	* names are little endian 16 bit Unicode on the wire
	474	*/
	475	int
acbbb76a	476	cifsConvertToUTF16(__le16 target, const char source, int srclen,
a4153cb1	477	const struct nls_table *cp, int map_chars)
84cdf74e	478	{
ce36d9ab SF	479	int i, charlen;
ce36d9ab SF	480	int j = 0;
84cdf74e	481	char src_char;
581ade4d JL	482	__le16 dst_char;
581ade4d JL	483	wchar_t tmp;
b2910307 NA	484	wchar_t wchar_to; / UTF-16 */
	485	int ret;
	486	unicode_t u;
84cdf74e	487
a4153cb1	488	if (map_chars == NO_MAP_UNI_RSVD)
acbbb76a	489	return cifs_strtoUTF16(target, source, PATH_MAX, cp);
84cdf74e	490
b2910307 NA	491	wchar_to = kzalloc(6, GFP_KERNEL);
b2910307 NA	492
ce36d9ab	493	for (i = 0; i < srclen; j++) {
84cdf74e	494	src_char = source[i];
11379b5e	495	charlen = 1;
a4153cb1 SF	496
	497	/* check if end of string */
	498	if (src_char == 0)
acbbb76a	499	goto ctoUTF16_out;
a4153cb1 SF	500
	501	/* see if we must remap this char */
	502	if (map_chars == SFU_MAP_UNI_RSVD)
	503	dst_char = convert_to_sfu_char(src_char);
45e8a258 SF	504	else if (map_chars == SFM_MAP_UNI_RSVD) {
	505	bool end_of_string;
	506
	507	if (i == srclen - 1)
	508	end_of_string = true;
	509	else
	510	end_of_string = false;
	511
	512	dst_char = convert_to_sfm_char(src_char, end_of_string);
	513	} else
a4153cb1	514	dst_char = 0;
84cdf74e JL	515	/*
	516	* FIXME: We can not handle remapping backslash (UNI_SLASH)
	517	* until all the calls to build_path_from_dentry are modified,
	518	* as they use backslash as separator.
	519	*/
a4153cb1	520	if (dst_char == 0) {
581ade4d JL	521	charlen = cp->char2uni(source + i, srclen - i, &tmp);
	522	dst_char = cpu_to_le16(tmp);
	523
84cdf74e JL	524	/*
	525	* if no match, use question mark, which at least in
	526	* some cases serves as wild card
	527	*/
b2910307 NA	528	if (charlen > 0)
	529	goto ctoUTF16;
	530
	531	/* convert SURROGATE_PAIR */
	532	if (strcmp(cp->charset, "utf8") \|\| !wchar_to)
	533	goto unknown;
	534	if (*(source + i) & 0x80) {
	535	charlen = utf8_to_utf32(source + i, 6, &u);
	536	if (charlen < 0)
	537	goto unknown;
	538	} else
	539	goto unknown;
	540	ret = utf8s_to_utf16s(source + i, charlen,
	541	UTF16_LITTLE_ENDIAN,
	542	wchar_to, 6);
	543	if (ret < 0)
	544	goto unknown;
	545
	546	i += charlen;
	547	dst_char = cpu_to_le16(*wchar_to);
	548	if (charlen <= 3)
	549	/* 1-3bytes UTF-8 to 2bytes UTF-16 */
	550	put_unaligned(dst_char, &target[j]);
	551	else if (charlen == 4) {
	552	/* 4bytes UTF-8(surrogate pair) to 4bytes UTF-16
	553	* 7-8bytes UTF-8(IVS) divided to 2 UTF-16
	554	* (charlen=3+4 or 4+4) */
	555	put_unaligned(dst_char, &target[j]);
	556	dst_char = cpu_to_le16(*(wchar_to + 1));
	557	j++;
	558	put_unaligned(dst_char, &target[j]);
	559	} else if (charlen >= 5) {
	560	/* 5-6bytes UTF-8 to 6bytes UTF-16 */
	561	put_unaligned(dst_char, &target[j]);
	562	dst_char = cpu_to_le16(*(wchar_to + 1));
	563	j++;
	564	put_unaligned(dst_char, &target[j]);
	565	dst_char = cpu_to_le16(*(wchar_to + 2));
	566	j++;
	567	put_unaligned(dst_char, &target[j]);
84cdf74e	568	}
b2910307 NA	569	continue;
	570
	571	unknown:
	572	dst_char = cpu_to_le16(0x003f);
	573	charlen = 1;
84cdf74e	574	}
b2910307 NA	575
b2910307 NA	576	ctoUTF16:
11379b5e JL	577	/*
	578	* character may take more than one byte in the source string,
	579	* but will take exactly two bytes in the target string
	580	*/
	581	i += charlen;
581ade4d	582	put_unaligned(dst_char, &target[j]);
84cdf74e JL	583	}
84cdf74e JL	584
acbbb76a	585	ctoUTF16_out:
ce36d9ab	586	put_unaligned(0, &target[j]); /* Null terminate target unicode string */
b2910307	587	kfree(wchar_to);
c73f6939	588	return j;
84cdf74e	589	}
2503a0db	590
2503a0db PS	591	/*
	592	* cifs_local_to_utf16_bytes - how long will a string be after conversion?
	593	* @from - pointer to input string
	594	* @maxbytes - don't go past this many bytes of input string
	595	* @codepage - source codepage
	596	*
	597	* Walk a string and return the number of bytes that the string will
	598	* be after being converted to the given charset, not including any null
	599	* termination required. Don't walk past maxbytes in the source buffer.
	600	*/
	601
	602	static int
	603	cifs_local_to_utf16_bytes(const char *from, int len,
	604	const struct nls_table *codepage)
	605	{
	606	int charlen;
	607	int i;
	608	wchar_t wchar_to;
	609
	610	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
	611	charlen = codepage->char2uni(from, len, &wchar_to);
	612	/* Failed conversion defaults to a question mark */
	613	if (charlen < 1)
	614	charlen = 1;
	615	}
	616	return 2 * i; /* UTF16 characters are two bytes */
	617	}
	618
	619	/*
	620	* cifs_strndup_to_utf16 - copy a string to wire format from the local codepage
	621	* @src - source string
	622	* @maxlen - don't walk past this many bytes in the source string
	623	* @utf16_len - the length of the allocated string in bytes (including null)
	624	* @cp - source codepage
	625	* @remap - map special chars
	626	*
	627	* Take a string convert it from the local codepage to UTF16 and
	628	* put it in a new buffer. Returns a pointer to the new string or NULL on
	629	* error.
	630	*/
	631	__le16 *
	632	cifs_strndup_to_utf16(const char src, const int maxlen, int utf16_len,
	633	const struct nls_table *cp, int remap)
	634	{
	635	int len;
	636	__le16 *dst;
	637
	638	len = cifs_local_to_utf16_bytes(src, maxlen, cp);
	639	len += 2; /* NULL */
	640	dst = kmalloc(len, GFP_KERNEL);
	641	if (!dst) {
	642	*utf16_len = 0;
	643	return NULL;
	644	}
	645	cifsConvertToUTF16(dst, src, strlen(src), cp, remap);
	646	*utf16_len = len;
	647	return dst;
	648	}