[mirror_ubuntu-bionic-kernel.git] / fs / cifs / cifs_unicode.c

/*
 *   fs/cifs/cifs_unicode.c
 *
 *   Copyright (c) International Business Machines  Corp., 2000,2009
 *   Modified by Steve French (sfrench@us.ibm.com)
 *
 *   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_unicode.h"
#include "cifs_uniupr.h"
#include "cifspdu.h"
#include "cifsglob.h"
#include "cifs_debug.h"

/*
 * 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;

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

		charlen = codepage->uni2char(ftmp, tmp, NLS_MAX_CHARSET_SIZE);
		if (charlen > 0)
			outlen += charlen;
		else
			outlen++;
	}

	return outlen;
}

/*
 * 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
 * @mapchar - should character be mapped according to mapchars mount option?
 *
 * 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 src_char, const struct nls_table *cp,
	     bool mapchar)
{
	int len = 1;

	if (!mapchar)
		goto cp_convert;

	/*
	 * 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:
		goto cp_convert;
	}

out:
	return len;

cp_convert:
	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
	if (len <= 0) {
		*target = '?';
		len = 1;
	}
	goto out;
}

/*
 * 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, bool mapchar)
{
	int i, charlen, safelen;
	int outlen = 0;
	int nullsize = nls_nullsize(codepage);
	int fromwords = fromlen / 2;
	char tmp[NLS_MAX_CHARSET_SIZE];
	__u16 ftmp;

	/*
	 * 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 = get_unaligned_le16(&from[i]);
		if (ftmp == 0)
			break;

		/*
		 * 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, mapchar);
			if ((outlen + charlen) > (tolen - nullsize))
				break;
		}

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

	/* 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 */

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

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

/*
 * 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,
			       false);
	} else {
		len = strnlen(src, maxlen);
		len++;
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return NULL;
		strlcpy(dst, src, len);
	}

	return dst;
}

/*
 * 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 mapChars)
{
	int i, j, charlen;
	char src_char;
	__le16 dst_char;
	wchar_t tmp;

	if (!mapChars)
		return cifs_strtoUTF16(target, source, PATH_MAX, cp);

	for (i = 0, j = 0; i < srclen; j++) {
		src_char = source[i];
		charlen = 1;
		switch (src_char) {
		case 0:
			put_unaligned(0, &target[j]);
			goto ctoUTF16_out;
		case ':':
			dst_char = cpu_to_le16(UNI_COLON);
			break;
		case '*':
			dst_char = cpu_to_le16(UNI_ASTERISK);
			break;
		case '?':
			dst_char = cpu_to_le16(UNI_QUESTION);
			break;
		case '<':
			dst_char = cpu_to_le16(UNI_LESSTHAN);
			break;
		case '>':
			dst_char = cpu_to_le16(UNI_GRTRTHAN);
			break;
		case '|':
			dst_char = cpu_to_le16(UNI_PIPE);
			break;
		/*
		 * 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.
		 */
		default:
			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 < 1) {
				dst_char = cpu_to_le16(0x003f);
				charlen = 1;
			}
		}
		/*
		 * 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:
	return i;
}
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 (sfrench@us.ibm.com)
	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>
1da177e4 LT	23	#include "cifs_unicode.h"
	24	#include "cifs_uniupr.h"
	25	#include "cifspdu.h"
3979877e	26	#include "cifsglob.h"
1da177e4 LT	27	#include "cifs_debug.h"
1da177e4 LT	28
69f801fc	29	/*
acbbb76a SF	30	* cifs_utf16_bytes - how long will a string be after conversion?
acbbb76a SF	31	* @utf16 - pointer to input string
69f801fc JL	32	* @maxbytes - don't go past this many bytes of input string
	33	* @codepage - destination codepage
	34	*
acbbb76a	35	* Walk a utf16le string and return the number of bytes that the string will
69f801fc JL	36	* be after being converted to the given charset, not including any null
	37	* termination required. Don't walk past maxbytes in the source buffer.
	38	*/
	39	int
acbbb76a	40	cifs_utf16_bytes(const __le16 *from, int maxbytes,
69f801fc JL	41	const struct nls_table *codepage)
	42	{
	43	int i;
	44	int charlen, outlen = 0;
	45	int maxwords = maxbytes / 2;
	46	char tmp[NLS_MAX_CHARSET_SIZE];
ba2dbf30	47	__u16 ftmp;
69f801fc	48
ba2dbf30 JL	49	for (i = 0; i < maxwords; i++) {
	50	ftmp = get_unaligned_le16(&from[i]);
	51	if (ftmp == 0)
	52	break;
	53
	54	charlen = codepage->uni2char(ftmp, tmp, NLS_MAX_CHARSET_SIZE);
69f801fc JL	55	if (charlen > 0)
	56	outlen += charlen;
	57	else
	58	outlen++;
	59	}
	60
	61	return outlen;
	62	}
	63
7fabf0c9	64	/*
ba2dbf30	65	* cifs_mapchar - convert a host-endian char to proper char in codepage
7fabf0c9	66	* @target - where converted character should be copied
ba2dbf30	67	* @src_char - 2 byte host-endian source character
7fabf0c9 JL	68	* @cp - codepage to which character should be converted
	69	* @mapchar - should character be mapped according to mapchars mount option?
	70	*
	71	* This function handles the conversion of a single character. It is the
	72	* responsibility of the caller to ensure that the target buffer is large
	73	* enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
	74	*/
	75	static int
ba2dbf30	76	cifs_mapchar(char target, const __u16 src_char, const struct nls_table cp,
7fabf0c9 JL	77	bool mapchar)
	78	{
	79	int len = 1;
	80
	81	if (!mapchar)
	82	goto cp_convert;
	83
	84	/*
	85	* BB: Cannot handle remapping UNI_SLASH until all the calls to
	86	* build_path_from_dentry are modified, as they use slash as
	87	* separator.
	88	*/
ba2dbf30	89	switch (src_char) {
7fabf0c9 JL	90	case UNI_COLON:
	91	*target = ':';
	92	break;
581ade4d	93	case UNI_ASTERISK:
7fabf0c9 JL	94	target = '';
	95	break;
	96	case UNI_QUESTION:
	97	*target = '?';
	98	break;
	99	case UNI_PIPE:
	100	*target = '\|';
	101	break;
	102	case UNI_GRTRTHAN:
	103	*target = '>';
	104	break;
	105	case UNI_LESSTHAN:
	106	*target = '<';
	107	break;
	108	default:
	109	goto cp_convert;
	110	}
	111
	112	out:
	113	return len;
	114
	115	cp_convert:
ba2dbf30	116	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
7fabf0c9 JL	117	if (len <= 0) {
	118	*target = '?';
	119	len = 1;
	120	}
	121	goto out;
	122	}
	123
	124	/*
acbbb76a	125	* cifs_from_utf16 - convert utf16le string to local charset
7fabf0c9 JL	126	* @to - destination buffer
	127	* @from - source buffer
	128	* @tolen - destination buffer size (in bytes)
	129	* @fromlen - source buffer size (in bytes)
	130	* @codepage - codepage to which characters should be converted
	131	* @mapchar - should characters be remapped according to the mapchars option?
	132	*
acbbb76a	133	* Convert a little-endian utf16le string (as sent by the server) to a string
7fabf0c9 JL	134	* in the provided codepage. The tolen and fromlen parameters are to ensure
	135	* that the code doesn't walk off of the end of the buffer (which is always
	136	* a danger if the alignment of the source buffer is off). The destination
	137	* string is always properly null terminated and fits in the destination
	138	* buffer. Returns the length of the destination string in bytes (including
	139	* null terminator).
	140	*
	141	* Note that some windows versions actually send multiword UTF-16 characters
acbbb76a	142	* instead of straight UTF16-2. The linux nls routines however aren't able to
7fabf0c9 JL	143	* deal with those characters properly. In the event that we get some of
	144	* those characters, they won't be translated properly.
	145	*/
	146	int
acbbb76a	147	cifs_from_utf16(char to, const __le16 from, int tolen, int fromlen,
7fabf0c9 JL	148	const struct nls_table *codepage, bool mapchar)
	149	{
	150	int i, charlen, safelen;
	151	int outlen = 0;
	152	int nullsize = nls_nullsize(codepage);
	153	int fromwords = fromlen / 2;
	154	char tmp[NLS_MAX_CHARSET_SIZE];
ba2dbf30	155	__u16 ftmp;
7fabf0c9 JL	156
	157	/*
	158	* because the chars can be of varying widths, we need to take care
	159	* not to overflow the destination buffer when we get close to the
	160	* end of it. Until we get to this offset, we don't need to check
	161	* for overflow however.
	162	*/
	163	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);
	164
ba2dbf30 JL	165	for (i = 0; i < fromwords; i++) {
	166	ftmp = get_unaligned_le16(&from[i]);
	167	if (ftmp == 0)
	168	break;
	169
7fabf0c9 JL	170	/*
	171	* check to see if converting this character might make the
	172	* conversion bleed into the null terminator
	173	*/
	174	if (outlen >= safelen) {
ba2dbf30	175	charlen = cifs_mapchar(tmp, ftmp, codepage, mapchar);
7fabf0c9 JL	176	if ((outlen + charlen) > (tolen - nullsize))
	177	break;
	178	}
	179
	180	/* put converted char into 'to' buffer */
ba2dbf30	181	charlen = cifs_mapchar(&to[outlen], ftmp, codepage, mapchar);
7fabf0c9 JL	182	outlen += charlen;
	183	}
	184
	185	/* properly null-terminate string */
	186	for (i = 0; i < nullsize; i++)
	187	to[outlen++] = 0;
	188
	189	return outlen;
	190	}
	191
1da177e4	192	/*
acbbb76a	193	* NAME: cifs_strtoUTF16()
1da177e4 LT	194	*
	195	* FUNCTION: Convert character string to unicode string
	196	*
	197	*/
	198	int
acbbb76a	199	cifs_strtoUTF16(__le16 to, const char from, int len,
1da177e4 LT	200	const struct nls_table *codepage)
	201	{
	202	int charlen;
	203	int i;
ba2dbf30	204	wchar_t wchar_to; /* needed to quiet sparse */
1da177e4 LT	205
1da177e4 LT	206	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
ba2dbf30	207	charlen = codepage->char2uni(from, len, &wchar_to);
1da177e4	208	if (charlen < 1) {
acbbb76a	209	cERROR(1, "strtoUTF16: char2uni of 0x%x returned %d",
ba2dbf30	210	*from, charlen);
69114089	211	/* A question mark */
ba2dbf30	212	wchar_to = 0x003f;
1da177e4	213	charlen = 1;
ba2dbf30 JL	214	}
ba2dbf30 JL	215	put_unaligned_le16(wchar_to, &to[i]);
1da177e4 LT	216	}
1da177e4 LT	217
ba2dbf30	218	put_unaligned_le16(0, &to[i]);
1da177e4 LT	219	return i;
	220	}
	221
066ce689	222	/*
acbbb76a SF	223	* cifs_strndup_from_utf16 - copy a string from wire format to the local
acbbb76a SF	224	* codepage
066ce689 JL	225	* @src - source string
	226	* @maxlen - don't walk past this many bytes in the source string
	227	* @is_unicode - is this a unicode string?
	228	* @codepage - destination codepage
	229	*
	230	* Take a string given by the server, convert it to the local codepage and
	231	* put it in a new buffer. Returns a pointer to the new string or NULL on
	232	* error.
	233	*/
	234	char *
acbbb76a SF	235	cifs_strndup_from_utf16(const char *src, const int maxlen,
acbbb76a SF	236	const bool is_unicode, const struct nls_table *codepage)
066ce689 JL	237	{
	238	int len;
	239	char *dst;
	240
	241	if (is_unicode) {
acbbb76a	242	len = cifs_utf16_bytes((__le16 *) src, maxlen, codepage);
066ce689 JL	243	len += nls_nullsize(codepage);
	244	dst = kmalloc(len, GFP_KERNEL);
	245	if (!dst)
	246	return NULL;
acbbb76a	247	cifs_from_utf16(dst, (__le16 *) src, len, maxlen, codepage,
066ce689 JL	248	false);
	249	} else {
	250	len = strnlen(src, maxlen);
	251	len++;
	252	dst = kmalloc(len, GFP_KERNEL);
	253	if (!dst)
	254	return NULL;
	255	strlcpy(dst, src, len);
	256	}
	257
	258	return dst;
	259	}
	260
84cdf74e JL	261	/*
	262	* Convert 16 bit Unicode pathname to wire format from string in current code
	263	* page. Conversion may involve remapping up the six characters that are
	264	* only legal in POSIX-like OS (if they are present in the string). Path
	265	* names are little endian 16 bit Unicode on the wire
	266	*/
	267	int
acbbb76a	268	cifsConvertToUTF16(__le16 target, const char source, int srclen,
84cdf74e JL	269	const struct nls_table *cp, int mapChars)
	270	{
	271	int i, j, charlen;
84cdf74e	272	char src_char;
581ade4d JL	273	__le16 dst_char;
581ade4d JL	274	wchar_t tmp;
84cdf74e JL	275
84cdf74e JL	276	if (!mapChars)
acbbb76a	277	return cifs_strtoUTF16(target, source, PATH_MAX, cp);
84cdf74e	278
581ade4d	279	for (i = 0, j = 0; i < srclen; j++) {
84cdf74e	280	src_char = source[i];
11379b5e	281	charlen = 1;
84cdf74e JL	282	switch (src_char) {
84cdf74e JL	283	case 0:
581ade4d	284	put_unaligned(0, &target[j]);
acbbb76a	285	goto ctoUTF16_out;
84cdf74e	286	case ':':
581ade4d	287	dst_char = cpu_to_le16(UNI_COLON);
84cdf74e JL	288	break;
84cdf74e JL	289	case '*':
581ade4d	290	dst_char = cpu_to_le16(UNI_ASTERISK);
84cdf74e JL	291	break;
84cdf74e JL	292	case '?':
581ade4d	293	dst_char = cpu_to_le16(UNI_QUESTION);
84cdf74e JL	294	break;
84cdf74e JL	295	case '<':
581ade4d	296	dst_char = cpu_to_le16(UNI_LESSTHAN);
84cdf74e JL	297	break;
84cdf74e JL	298	case '>':
581ade4d	299	dst_char = cpu_to_le16(UNI_GRTRTHAN);
84cdf74e JL	300	break;
84cdf74e JL	301	case '\|':
581ade4d	302	dst_char = cpu_to_le16(UNI_PIPE);
84cdf74e JL	303	break;
	304	/*
	305	* FIXME: We can not handle remapping backslash (UNI_SLASH)
	306	* until all the calls to build_path_from_dentry are modified,
	307	* as they use backslash as separator.
	308	*/
	309	default:
581ade4d JL	310	charlen = cp->char2uni(source + i, srclen - i, &tmp);
	311	dst_char = cpu_to_le16(tmp);
	312
84cdf74e JL	313	/*
	314	* if no match, use question mark, which at least in
	315	* some cases serves as wild card
	316	*/
	317	if (charlen < 1) {
581ade4d	318	dst_char = cpu_to_le16(0x003f);
84cdf74e JL	319	charlen = 1;
84cdf74e JL	320	}
84cdf74e	321	}
11379b5e JL	322	/*
	323	* character may take more than one byte in the source string,
	324	* but will take exactly two bytes in the target string
	325	*/
	326	i += charlen;
581ade4d	327	put_unaligned(dst_char, &target[j]);
84cdf74e JL	328	}
84cdf74e JL	329
acbbb76a	330	ctoUTF16_out:
84cdf74e JL	331	return i;
	332	}
	333