xref: /kernel/linux/linux-5.10/fs/cifs/cifs_unicode.h (revision 8c2ecf20) 
1/* SPDX-License-Identifier: GPL-2.0-or-later */
2/*
3 * cifs_unicode:  Unicode kernel case support
4 *
5 * Function:
6 *     Convert a unicode character to upper or lower case using
7 *     compressed tables.
8 *
9 *   Copyright (c) International Business Machines  Corp., 2000,2009
10 *
11 * Notes:
12 *     These APIs are based on the C library functions.  The semantics
13 *     should match the C functions but with expanded size operands.
14 *
15 *     The upper/lower functions are based on a table created by mkupr.
16 *     This is a compressed table of upper and lower case conversion.
17 */
18#ifndef _CIFS_UNICODE_H
19#define _CIFS_UNICODE_H
20
21#include <asm/byteorder.h>
22#include <linux/types.h>
23#include <linux/nls.h>
24
25#define  UNIUPR_NOLOWER		/* Example to not expand lower case tables */
26
27/*
28 * Windows maps these to the user defined 16 bit Unicode range since they are
29 * reserved symbols (along with \ and /), otherwise illegal to store
30 * in filenames in NTFS
31 */
32#define UNI_ASTERISK    (__u16) ('*' + 0xF000)
33#define UNI_QUESTION    (__u16) ('?' + 0xF000)
34#define UNI_COLON       (__u16) (':' + 0xF000)
35#define UNI_GRTRTHAN    (__u16) ('>' + 0xF000)
36#define UNI_LESSTHAN    (__u16) ('<' + 0xF000)
37#define UNI_PIPE        (__u16) ('|' + 0xF000)
38#define UNI_SLASH       (__u16) ('\\' + 0xF000)
39
40/*
41 * Macs use an older "SFM" mapping of the symbols above. Fortunately it does
42 * not conflict (although almost does) with the mapping above.
43 */
44
45#define SFM_DOUBLEQUOTE ((__u16) 0xF020)
46#define SFM_ASTERISK    ((__u16) 0xF021)
47#define SFM_QUESTION    ((__u16) 0xF025)
48#define SFM_COLON       ((__u16) 0xF022)
49#define SFM_GRTRTHAN    ((__u16) 0xF024)
50#define SFM_LESSTHAN    ((__u16) 0xF023)
51#define SFM_PIPE        ((__u16) 0xF027)
52#define SFM_SLASH       ((__u16) 0xF026)
53#define SFM_SPACE	((__u16) 0xF028)
54#define SFM_PERIOD	((__u16) 0xF029)
55
56/*
57 * Mapping mechanism to use when one of the seven reserved characters is
58 * encountered.  We can only map using one of the mechanisms at a time
59 * since otherwise readdir could return directory entries which we would
60 * not be able to open
61 *
62 * NO_MAP_UNI_RSVD  = do not perform any remapping of the character
63 * SFM_MAP_UNI_RSVD = map reserved characters using SFM scheme (MAC compatible)
64 * SFU_MAP_UNI_RSVD = map reserved characters ala SFU ("mapchars" option)
65 *
66 */
67#define NO_MAP_UNI_RSVD		0
68#define SFM_MAP_UNI_RSVD	1
69#define SFU_MAP_UNI_RSVD	2
70
71/* Just define what we want from uniupr.h.  We don't want to define the tables
72 * in each source file.
73 */
74#ifndef	UNICASERANGE_DEFINED
75struct UniCaseRange {
76	wchar_t start;
77	wchar_t end;
78	signed char *table;
79};
80#endif				/* UNICASERANGE_DEFINED */
81
82#ifndef UNIUPR_NOUPPER
83extern signed char CifsUniUpperTable[512];
84extern const struct UniCaseRange CifsUniUpperRange[];
85#endif				/* UNIUPR_NOUPPER */
86
87#ifndef UNIUPR_NOLOWER
88extern signed char CifsUniLowerTable[512];
89extern const struct UniCaseRange CifsUniLowerRange[];
90#endif				/* UNIUPR_NOLOWER */
91
92#ifdef __KERNEL__
93int cifs_from_utf16(char *to, const __le16 *from, int tolen, int fromlen,
94		    const struct nls_table *cp, int map_type);
95int cifs_utf16_bytes(const __le16 *from, int maxbytes,
96		     const struct nls_table *codepage);
97int cifs_strtoUTF16(__le16 *, const char *, int, const struct nls_table *);
98char *cifs_strndup_from_utf16(const char *src, const int maxlen,
99			      const bool is_unicode,
100			      const struct nls_table *codepage);
101extern int cifsConvertToUTF16(__le16 *target, const char *source, int maxlen,
102			      const struct nls_table *cp, int mapChars);
103extern int cifs_remap(struct cifs_sb_info *cifs_sb);
104extern __le16 *cifs_strndup_to_utf16(const char *src, const int maxlen,
105				     int *utf16_len, const struct nls_table *cp,
106				     int remap);
107#endif
108
109wchar_t cifs_toupper(wchar_t in);
110
111/*
112 * UniStrcat:  Concatenate the second string to the first
113 *
114 * Returns:
115 *     Address of the first string
116 */
117static inline __le16 *
118UniStrcat(__le16 *ucs1, const __le16 *ucs2)
119{
120	__le16 *anchor = ucs1;	/* save a pointer to start of ucs1 */
121
122	while (*ucs1++) ;	/* To end of first string */
123	ucs1--;			/* Return to the null */
124	while ((*ucs1++ = *ucs2++)) ;	/* copy string 2 over */
125	return anchor;
126}
127
128/*
129 * UniStrchr:  Find a character in a string
130 *
131 * Returns:
132 *     Address of first occurrence of character in string
133 *     or NULL if the character is not in the string
134 */
135static inline wchar_t *
136UniStrchr(const wchar_t *ucs, wchar_t uc)
137{
138	while ((*ucs != uc) && *ucs)
139		ucs++;
140
141	if (*ucs == uc)
142		return (wchar_t *) ucs;
143	return NULL;
144}
145
146/*
147 * UniStrcmp:  Compare two strings
148 *
149 * Returns:
150 *     < 0:  First string is less than second
151 *     = 0:  Strings are equal
152 *     > 0:  First string is greater than second
153 */
154static inline int
155UniStrcmp(const wchar_t *ucs1, const wchar_t *ucs2)
156{
157	while ((*ucs1 == *ucs2) && *ucs1) {
158		ucs1++;
159		ucs2++;
160	}
161	return (int) *ucs1 - (int) *ucs2;
162}
163
164/*
165 * UniStrcpy:  Copy a string
166 */
167static inline wchar_t *
168UniStrcpy(wchar_t *ucs1, const wchar_t *ucs2)
169{
170	wchar_t *anchor = ucs1;	/* save the start of result string */
171
172	while ((*ucs1++ = *ucs2++)) ;
173	return anchor;
174}
175
176/*
177 * UniStrlen:  Return the length of a string (in 16 bit Unicode chars not bytes)
178 */
179static inline size_t
180UniStrlen(const wchar_t *ucs1)
181{
182	int i = 0;
183
184	while (*ucs1++)
185		i++;
186	return i;
187}
188
189/*
190 * UniStrnlen:  Return the length (in 16 bit Unicode chars not bytes) of a
191 *		string (length limited)
192 */
193static inline size_t
194UniStrnlen(const wchar_t *ucs1, int maxlen)
195{
196	int i = 0;
197
198	while (*ucs1++) {
199		i++;
200		if (i >= maxlen)
201			break;
202	}
203	return i;
204}
205
206/*
207 * UniStrncat:  Concatenate length limited string
208 */
209static inline wchar_t *
210UniStrncat(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
211{
212	wchar_t *anchor = ucs1;	/* save pointer to string 1 */
213
214	while (*ucs1++) ;
215	ucs1--;			/* point to null terminator of s1 */
216	while (n-- && (*ucs1 = *ucs2)) {	/* copy s2 after s1 */
217		ucs1++;
218		ucs2++;
219	}
220	*ucs1 = 0;		/* Null terminate the result */
221	return (anchor);
222}
223
224/*
225 * UniStrncmp:  Compare length limited string
226 */
227static inline int
228UniStrncmp(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
229{
230	if (!n)
231		return 0;	/* Null strings are equal */
232	while ((*ucs1 == *ucs2) && *ucs1 && --n) {
233		ucs1++;
234		ucs2++;
235	}
236	return (int) *ucs1 - (int) *ucs2;
237}
238
239/*
240 * UniStrncmp_le:  Compare length limited string - native to little-endian
241 */
242static inline int
243UniStrncmp_le(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
244{
245	if (!n)
246		return 0;	/* Null strings are equal */
247	while ((*ucs1 == __le16_to_cpu(*ucs2)) && *ucs1 && --n) {
248		ucs1++;
249		ucs2++;
250	}
251	return (int) *ucs1 - (int) __le16_to_cpu(*ucs2);
252}
253
254/*
255 * UniStrncpy:  Copy length limited string with pad
256 */
257static inline wchar_t *
258UniStrncpy(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
259{
260	wchar_t *anchor = ucs1;
261
262	while (n-- && *ucs2)	/* Copy the strings */
263		*ucs1++ = *ucs2++;
264
265	n++;
266	while (n--)		/* Pad with nulls */
267		*ucs1++ = 0;
268	return anchor;
269}
270
271/*
272 * UniStrncpy_le:  Copy length limited string with pad to little-endian
273 */
274static inline wchar_t *
275UniStrncpy_le(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
276{
277	wchar_t *anchor = ucs1;
278
279	while (n-- && *ucs2)	/* Copy the strings */
280		*ucs1++ = __le16_to_cpu(*ucs2++);
281
282	n++;
283	while (n--)		/* Pad with nulls */
284		*ucs1++ = 0;
285	return anchor;
286}
287
288/*
289 * UniStrstr:  Find a string in a string
290 *
291 * Returns:
292 *     Address of first match found
293 *     NULL if no matching string is found
294 */
295static inline wchar_t *
296UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2)
297{
298	const wchar_t *anchor1 = ucs1;
299	const wchar_t *anchor2 = ucs2;
300
301	while (*ucs1) {
302		if (*ucs1 == *ucs2) {
303			/* Partial match found */
304			ucs1++;
305			ucs2++;
306		} else {
307			if (!*ucs2)	/* Match found */
308				return (wchar_t *) anchor1;
309			ucs1 = ++anchor1;	/* No match */
310			ucs2 = anchor2;
311		}
312	}
313
314	if (!*ucs2)		/* Both end together */
315		return (wchar_t *) anchor1;	/* Match found */
316	return NULL;		/* No match */
317}
318
319#ifndef UNIUPR_NOUPPER
320/*
321 * UniToupper:  Convert a unicode character to upper case
322 */
323static inline wchar_t
324UniToupper(register wchar_t uc)
325{
326	register const struct UniCaseRange *rp;
327
328	if (uc < sizeof(CifsUniUpperTable)) {
329		/* Latin characters */
330		return uc + CifsUniUpperTable[uc];	/* Use base tables */
331	} else {
332		rp = CifsUniUpperRange;	/* Use range tables */
333		while (rp->start) {
334			if (uc < rp->start)	/* Before start of range */
335				return uc;	/* Uppercase = input */
336			if (uc <= rp->end)	/* In range */
337				return uc + rp->table[uc - rp->start];
338			rp++;	/* Try next range */
339		}
340	}
341	return uc;		/* Past last range */
342}
343
344/*
345 * UniStrupr:  Upper case a unicode string
346 */
347static inline __le16 *
348UniStrupr(register __le16 *upin)
349{
350	register __le16 *up;
351
352	up = upin;
353	while (*up) {		/* For all characters */
354		*up = cpu_to_le16(UniToupper(le16_to_cpu(*up)));
355		up++;
356	}
357	return upin;		/* Return input pointer */
358}
359#endif				/* UNIUPR_NOUPPER */
360
361#ifndef UNIUPR_NOLOWER
362/*
363 * UniTolower:  Convert a unicode character to lower case
364 */
365static inline wchar_t
366UniTolower(register wchar_t uc)
367{
368	register const struct UniCaseRange *rp;
369
370	if (uc < sizeof(CifsUniLowerTable)) {
371		/* Latin characters */
372		return uc + CifsUniLowerTable[uc];	/* Use base tables */
373	} else {
374		rp = CifsUniLowerRange;	/* Use range tables */
375		while (rp->start) {
376			if (uc < rp->start)	/* Before start of range */
377				return uc;	/* Uppercase = input */
378			if (uc <= rp->end)	/* In range */
379				return uc + rp->table[uc - rp->start];
380			rp++;	/* Try next range */
381		}
382	}
383	return uc;		/* Past last range */
384}
385
386/*
387 * UniStrlwr:  Lower case a unicode string
388 */
389static inline wchar_t *
390UniStrlwr(register wchar_t *upin)
391{
392	register wchar_t *up;
393
394	up = upin;
395	while (*up) {		/* For all characters */
396		*up = UniTolower(*up);
397		up++;
398	}
399	return upin;		/* Return input pointer */
400}
401
402#endif
403
404#endif /* _CIFS_UNICODE_H */
405