[linux-block.git] / fs / ksmbd / unicode.c

// SPDX-License-Identifier: GPL-2.0-or-later
/*
 *   Some of the source code in this file came from fs/cifs/cifs_unicode.c
 *
 *   Copyright (c) International Business Machines  Corp., 2000,2009
 *   Modified by Steve French (sfrench@us.ibm.com)
 *   Modified by Namjae Jeon (linkinjeon@kernel.org)
 */
#include <linux/fs.h>
#include <linux/slab.h>
#include <asm/unaligned.h>
#include "glob.h"
#include "unicode.h"
#include "uniupr.h"
#include "smb_common.h"

/*
 * smb_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:	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.
 *
 * Return:	string length after conversion
 */
static int smb_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).
 *
 * Return:	string length after conversion
 */
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;
}

/*
 * smb_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.
 *
 * Return:	string length after conversion
 */
static int smb_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;
}

/*
 * smb_strtoUTF16() - Convert character string to unicode string
 * @to:		destination buffer
 * @from:	source buffer
 * @len:	destination buffer size (in bytes)
 * @codepage:	codepage to which characters should be converted
 *
 * Return:	string length after conversion
 */
int smb_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 > 0 && *from; i++, from += charlen, len -= charlen) {
		charlen = codepage->char2uni(from, len, &wchar_to);
		if (charlen < 1) {
			/* A question mark */
			wchar_to = 0x003f;
			charlen = 1;
		}
		put_unaligned_le16(wchar_to, &to[i]);
	}

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

/*
 * smb_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.
 *
 * Return:	destination string buffer or error ptr
 */
char *smb_strndup_from_utf16(const char *src, const int maxlen,
			     const bool is_unicode,
			     const struct nls_table *codepage)
{
	int len, ret;
	char *dst;

	if (is_unicode) {
		len = smb_utf16_bytes((__le16 *)src, maxlen, codepage);
		len += nls_nullsize(codepage);
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return ERR_PTR(-ENOMEM);
		ret = smb_from_utf16(dst, (__le16 *)src, len, maxlen, codepage,
				     false);
		if (ret < 0) {
			kfree(dst);
			return ERR_PTR(-EINVAL);
		}
	} else {
		len = strnlen(src, maxlen);
		len++;
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return ERR_PTR(-ENOMEM);
		strscpy(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
 */
/*
 * smbConvertToUTF16() - convert string from local charset to utf16
 * @target:	destination buffer
 * @source:	source buffer
 * @srclen:	source buffer size (in bytes)
 * @cp:		codepage to which characters should be converted
 * @mapchar:	should characters be remapped according to the mapchars option?
 *
 * 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
 *
 * Return:	char length after conversion
 */
int smbConvertToUTF16(__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 smb_strtoUTF16(target, source, srclen, 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]);
			return j;
		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]);
	}

	return j;
}
Commit	Line	Data
e2f34481 NJ	1	// SPDX-License-Identifier: GPL-2.0-or-later
	2	/*
	3	* Some of the source code in this file came from fs/cifs/cifs_unicode.c
	4	*
	5	* Copyright (c) International Business Machines Corp., 2000,2009
	6	* Modified by Steve French (sfrench@us.ibm.com)
	7	* Modified by Namjae Jeon (linkinjeon@kernel.org)
	8	*/
	9	#include <linux/fs.h>
	10	#include <linux/slab.h>
	11	#include <asm/unaligned.h>
	12	#include "glob.h"
	13	#include "unicode.h"
	14	#include "uniupr.h"
	15	#include "smb_common.h"
	16
	17	/*
	18	* smb_utf16_bytes() - how long will a string be after conversion?
	19	* @from: pointer to input string
	20	* @maxbytes: don't go past this many bytes of input string
	21	* @codepage: destination codepage
	22	*
	23	* Walk a utf16le string and return the number of bytes that the string will
	24	* be after being converted to the given charset, not including any null
	25	* termination required. Don't walk past maxbytes in the source buffer.
	26	*
	27	* Return: string length after conversion
	28	*/
64b39f4a	29	static int smb_utf16_bytes(const __le16 *from, int maxbytes,
070fb21e	30	const struct nls_table *codepage)
e2f34481 NJ	31	{
	32	int i;
	33	int charlen, outlen = 0;
	34	int maxwords = maxbytes / 2;
	35	char tmp[NLS_MAX_CHARSET_SIZE];
	36	__u16 ftmp;
	37
	38	for (i = 0; i < maxwords; i++) {
	39	ftmp = get_unaligned_le16(&from[i]);
	40	if (ftmp == 0)
	41	break;
	42
	43	charlen = codepage->uni2char(ftmp, tmp, NLS_MAX_CHARSET_SIZE);
	44	if (charlen > 0)
	45	outlen += charlen;
	46	else
	47	outlen++;
	48	}
	49
	50	return outlen;
	51	}
	52
	53	/*
	54	* cifs_mapchar() - convert a host-endian char to proper char in codepage
	55	* @target: where converted character should be copied
	56	* @src_char: 2 byte host-endian source character
	57	* @cp: codepage to which character should be converted
	58	* @mapchar: should character be mapped according to mapchars mount option?
	59	*
	60	* This function handles the conversion of a single character. It is the
	61	* responsibility of the caller to ensure that the target buffer is large
	62	* enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
	63	*
	64	* Return: string length after conversion
	65	*/
	66	static int
	67	cifs_mapchar(char target, const __u16 src_char, const struct nls_table cp,
070fb21e	68	bool mapchar)
e2f34481 NJ	69	{
	70	int len = 1;
	71
	72	if (!mapchar)
	73	goto cp_convert;
	74
	75	/*
	76	* BB: Cannot handle remapping UNI_SLASH until all the calls to
	77	* build_path_from_dentry are modified, as they use slash as
	78	* separator.
	79	*/
	80	switch (src_char) {
	81	case UNI_COLON:
	82	*target = ':';
	83	break;
	84	case UNI_ASTERISK:
	85	target = '';
	86	break;
	87	case UNI_QUESTION:
	88	*target = '?';
	89	break;
	90	case UNI_PIPE:
	91	*target = '\|';
	92	break;
	93	case UNI_GRTRTHAN:
	94	*target = '>';
	95	break;
	96	case UNI_LESSTHAN:
	97	*target = '<';
	98	break;
	99	default:
	100	goto cp_convert;
	101	}
	102
	103	out:
	104	return len;
	105
	106	cp_convert:
	107	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
	108	if (len <= 0) {
	109	*target = '?';
	110	len = 1;
	111	}
	112
	113	goto out;
	114	}
	115
e2f34481 NJ	116	/*
	117	* smb_from_utf16() - convert utf16le string to local charset
	118	* @to: destination buffer
	119	* @from: source buffer
	120	* @tolen: destination buffer size (in bytes)
	121	* @fromlen: source buffer size (in bytes)
	122	* @codepage: codepage to which characters should be converted
	123	* @mapchar: should characters be remapped according to the mapchars option?
	124	*
	125	* Convert a little-endian utf16le string (as sent by the server) to a string
	126	* in the provided codepage. The tolen and fromlen parameters are to ensure
	127	* that the code doesn't walk off of the end of the buffer (which is always
	128	* a danger if the alignment of the source buffer is off). The destination
	129	* string is always properly null terminated and fits in the destination
	130	* buffer. Returns the length of the destination string in bytes (including
	131	* null terminator).
	132	*
	133	* Note that some windows versions actually send multiword UTF-16 characters
	134	* instead of straight UTF16-2. The linux nls routines however aren't able to
	135	* deal with those characters properly. In the event that we get some of
	136	* those characters, they won't be translated properly.
	137	*
	138	* Return: string length after conversion
	139	*/
64b39f4a	140	static int smb_from_utf16(char to, const __le16 from, int tolen, int fromlen,
070fb21e	141	const struct nls_table *codepage, bool mapchar)
e2f34481 NJ	142	{
	143	int i, charlen, safelen;
	144	int outlen = 0;
	145	int nullsize = nls_nullsize(codepage);
	146	int fromwords = fromlen / 2;
	147	char tmp[NLS_MAX_CHARSET_SIZE];
	148	__u16 ftmp;
	149
	150	/*
	151	* because the chars can be of varying widths, we need to take care
	152	* not to overflow the destination buffer when we get close to the
	153	* end of it. Until we get to this offset, we don't need to check
	154	* for overflow however.
	155	*/
	156	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);
	157
	158	for (i = 0; i < fromwords; i++) {
	159	ftmp = get_unaligned_le16(&from[i]);
	160	if (ftmp == 0)
	161	break;
	162
	163	/*
	164	* check to see if converting this character might make the
	165	* conversion bleed into the null terminator
	166	*/
	167	if (outlen >= safelen) {
	168	charlen = cifs_mapchar(tmp, ftmp, codepage, mapchar);
	169	if ((outlen + charlen) > (tolen - nullsize))
	170	break;
	171	}
	172
	173	/* put converted char into 'to' buffer */
	174	charlen = cifs_mapchar(&to[outlen], ftmp, codepage, mapchar);
	175	outlen += charlen;
	176	}
	177
	178	/* properly null-terminate string */
	179	for (i = 0; i < nullsize; i++)
	180	to[outlen++] = 0;
	181
	182	return outlen;
	183	}
	184
	185	/*
	186	* smb_strtoUTF16() - Convert character string to unicode string
	187	* @to: destination buffer
	188	* @from: source buffer
	189	* @len: destination buffer size (in bytes)
	190	* @codepage: codepage to which characters should be converted
	191	*
	192	* Return: string length after conversion
	193	*/
64b39f4a	194	int smb_strtoUTF16(__le16 to, const char from, int len,
070fb21e	195	const struct nls_table *codepage)
e2f34481 NJ	196	{
	197	int charlen;
	198	int i;
	199	wchar_t wchar_to; /* needed to quiet sparse */
	200
	201	/* special case for utf8 to handle no plane0 chars */
	202	if (!strcmp(codepage->charset, "utf8")) {
	203	/*
	204	* convert utf8 -> utf16, we assume we have enough space
	205	* as caller should have assumed conversion does not overflow
	206	* in destination len is length in wchar_t units (16bits)
	207	*/
	208	i = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN,
070fb21e	209	(wchar_t *)to, len);
e2f34481 NJ	210
	211	/* if success terminate and exit */
	212	if (i >= 0)
	213	goto success;
	214	/*
	215	* if fails fall back to UCS encoding as this
	216	* function should not return negative values
	217	* currently can fail only if source contains
	218	* invalid encoded characters
	219	*/
	220	}
	221
	222	for (i = 0; len > 0 && *from; i++, from += charlen, len -= charlen) {
	223	charlen = codepage->char2uni(from, len, &wchar_to);
	224	if (charlen < 1) {
	225	/* A question mark */
	226	wchar_to = 0x003f;
	227	charlen = 1;
	228	}
	229	put_unaligned_le16(wchar_to, &to[i]);
	230	}
	231
	232	success:
	233	put_unaligned_le16(0, &to[i]);
	234	return i;
	235	}
	236
	237	/*
	238	* smb_strndup_from_utf16() - copy a string from wire format to the local
	239	* codepage
	240	* @src: source string
	241	* @maxlen: don't walk past this many bytes in the source string
	242	* @is_unicode: is this a unicode string?
	243	* @codepage: destination codepage
	244	*
	245	* Take a string given by the server, convert it to the local codepage and
	246	* put it in a new buffer. Returns a pointer to the new string or NULL on
	247	* error.
	248	*
	249	* Return: destination string buffer or error ptr
	250	*/
64b39f4a	251	char smb_strndup_from_utf16(const char src, const int maxlen,
070fb21e NJ	252	const bool is_unicode,
070fb21e NJ	253	const struct nls_table *codepage)
e2f34481 NJ	254	{
	255	int len, ret;
	256	char *dst;
	257
	258	if (is_unicode) {
64b39f4a	259	len = smb_utf16_bytes((__le16 *)src, maxlen, codepage);
e2f34481 NJ	260	len += nls_nullsize(codepage);
	261	dst = kmalloc(len, GFP_KERNEL);
	262	if (!dst)
	263	return ERR_PTR(-ENOMEM);
64b39f4a	264	ret = smb_from_utf16(dst, (__le16 *)src, len, maxlen, codepage,
070fb21e	265	false);
e2f34481 NJ	266	if (ret < 0) {
	267	kfree(dst);
	268	return ERR_PTR(-EINVAL);
	269	}
	270	} else {
	271	len = strnlen(src, maxlen);
	272	len++;
	273	dst = kmalloc(len, GFP_KERNEL);
	274	if (!dst)
	275	return ERR_PTR(-ENOMEM);
	276	strscpy(dst, src, len);
	277	}
	278
	279	return dst;
	280	}
	281
	282	/*
	283	* Convert 16 bit Unicode pathname to wire format from string in current code
	284	* page. Conversion may involve remapping up the six characters that are
	285	* only legal in POSIX-like OS (if they are present in the string). Path
	286	* names are little endian 16 bit Unicode on the wire
	287	*/
	288	/*
	289	* smbConvertToUTF16() - convert string from local charset to utf16
	290	* @target: destination buffer
	291	* @source: source buffer
	292	* @srclen: source buffer size (in bytes)
	293	* @cp: codepage to which characters should be converted
	294	* @mapchar: should characters be remapped according to the mapchars option?
	295	*
	296	* Convert 16 bit Unicode pathname to wire format from string in current code
	297	* page. Conversion may involve remapping up the six characters that are
	298	* only legal in POSIX-like OS (if they are present in the string). Path
	299	* names are little endian 16 bit Unicode on the wire
	300	*
	301	* Return: char length after conversion
	302	*/
64b39f4a	303	int smbConvertToUTF16(__le16 target, const char source, int srclen,
070fb21e	304	const struct nls_table *cp, int mapchars)
e2f34481 NJ	305	{
	306	int i, j, charlen;
	307	char src_char;
	308	__le16 dst_char;
	309	wchar_t tmp;
	310
	311	if (!mapchars)
	312	return smb_strtoUTF16(target, source, srclen, cp);
	313
	314	for (i = 0, j = 0; i < srclen; j++) {
	315	src_char = source[i];
	316	charlen = 1;
	317	switch (src_char) {
	318	case 0:
	319	put_unaligned(0, &target[j]);
	320	return j;
	321	case ':':
	322	dst_char = cpu_to_le16(UNI_COLON);
	323	break;
	324	case '*':
	325	dst_char = cpu_to_le16(UNI_ASTERISK);
	326	break;
	327	case '?':
	328	dst_char = cpu_to_le16(UNI_QUESTION);
	329	break;
	330	case '<':
	331	dst_char = cpu_to_le16(UNI_LESSTHAN);
	332	break;
	333	case '>':
	334	dst_char = cpu_to_le16(UNI_GRTRTHAN);
	335	break;
	336	case '\|':
	337	dst_char = cpu_to_le16(UNI_PIPE);
	338	break;
	339	/*
	340	* FIXME: We can not handle remapping backslash (UNI_SLASH)
	341	* until all the calls to build_path_from_dentry are modified,
	342	* as they use backslash as separator.
	343	*/
	344	default:
	345	charlen = cp->char2uni(source + i, srclen - i, &tmp);
	346	dst_char = cpu_to_le16(tmp);
	347
	348	/*
	349	* if no match, use question mark, which at least in
	350	* some cases serves as wild card
	351	*/
	352	if (charlen < 1) {
	353	dst_char = cpu_to_le16(0x003f);
	354	charlen = 1;
	355	}
	356	}
	357	/*
	358	* character may take more than one byte in the source string,
	359	* but will take exactly two bytes in the target string
	360	*/
	361	i += charlen;
	362	put_unaligned(dst_char, &target[j]);
	363	}
	364
	365	return j;
	366	}