1 /*
2 * cifs_unicode: Unicode kernel case support
3 *
4 * Function:
5 * Convert a unicode character to upper or lower case using
6 * compressed tables.
7 *
8 * Copyright (c) International Business Machines Corp., 2000,2009
9 *
10 * This program is free software; you can redistribute it and/or modify
11 * it under the terms of the GNU General Public License as published by
12 * the Free Software Foundation; either version 2 of the License, or
13 * (at your option) any later version.
14 *
15 * This program is distributed in the hope that it will be useful,
16 * but WITHOUT ANY WARRANTY; without even the implied warranty of
17 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
18 * the GNU General Public License for more details.
19 *
20 * You should have received a copy of the GNU General Public License
21 * along with this program; if not, write to the Free Software
22 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
23 *
24 *
25 * Notes:
26 * These APIs are based on the C library functions. The semantics
27 * should match the C functions but with expanded size operands.
28 *
29 * The upper/lower functions are based on a table created by mkupr.
30 * This is a compressed table of upper and lower case conversion.
31 *
32 */
33 #ifndef _CIFS_UNICODE_H
34 #define _CIFS_UNICODE_H
35
36 #include <asm/byteorder.h>
37 #include <linux/types.h>
38 #include <linux/nls.h>
39
40 #define UNIUPR_NOLOWER /* Example to not expand lower case tables */
41
42 /*
43 * Windows maps these to the user defined 16 bit Unicode range since they are
44 * reserved symbols (along with \ and /), otherwise illegal to store
45 * in filenames in NTFS
46 */
47 #define UNI_ASTERISK (__u16) ('*' + 0xF000)
48 #define UNI_QUESTION (__u16) ('?' + 0xF000)
49 #define UNI_COLON (__u16) (':' + 0xF000)
50 #define UNI_GRTRTHAN (__u16) ('>' + 0xF000)
51 #define UNI_LESSTHAN (__u16) ('<' + 0xF000)
52 #define UNI_PIPE (__u16) ('|' + 0xF000)
53 #define UNI_SLASH (__u16) ('\\' + 0xF000)
54
55 /*
56 * Macs use an older "SFM" mapping of the symbols above. Fortunately it does
57 * not conflict (although almost does) with the mapping above.
58 */
59
60 #define SFM_DOUBLEQUOTE ((__u16) 0xF020)
61 #define SFM_ASTERISK ((__u16) 0xF021)
62 #define SFM_QUESTION ((__u16) 0xF025)
63 #define SFM_COLON ((__u16) 0xF022)
64 #define SFM_GRTRTHAN ((__u16) 0xF024)
65 #define SFM_LESSTHAN ((__u16) 0xF023)
66 #define SFM_PIPE ((__u16) 0xF027)
67 #define SFM_SLASH ((__u16) 0xF026)
68 #define SFM_SPACE ((__u16) 0xF028)
69 #define SFM_PERIOD ((__u16) 0xF029)
70
71 /*
72 * Mapping mechanism to use when one of the seven reserved characters is
73 * encountered. We can only map using one of the mechanisms at a time
74 * since otherwise readdir could return directory entries which we would
75 * not be able to open
76 *
77 * NO_MAP_UNI_RSVD = do not perform any remapping of the character
78 * SFM_MAP_UNI_RSVD = map reserved characters using SFM scheme (MAC compatible)
79 * SFU_MAP_UNI_RSVD = map reserved characters ala SFU ("mapchars" option)
80 *
81 */
82 #define NO_MAP_UNI_RSVD 0
83 #define SFM_MAP_UNI_RSVD 1
84 #define SFU_MAP_UNI_RSVD 2
85
86 /* Just define what we want from uniupr.h. We don't want to define the tables
87 * in each source file.
88 */
89 #ifndef UNICASERANGE_DEFINED
90 struct UniCaseRange {
91 wchar_t start;
92 wchar_t end;
93 signed char *table;
94 };
95 #endif /* UNICASERANGE_DEFINED */
96
97 #ifndef UNIUPR_NOUPPER
98 extern signed char CifsUniUpperTable[512];
99 extern const struct UniCaseRange CifsUniUpperRange[];
100 #endif /* UNIUPR_NOUPPER */
101
102 #ifndef UNIUPR_NOLOWER
103 extern signed char CifsUniLowerTable[512];
104 extern const struct UniCaseRange CifsUniLowerRange[];
105 #endif /* UNIUPR_NOLOWER */
106
107 #ifdef __KERNEL__
108 int cifs_from_utf16(char *to, const __le16 *from, int tolen, int fromlen,
109 const struct nls_table *cp, int map_type);
110 int cifs_utf16_bytes(const __le16 *from, int maxbytes,
111 const struct nls_table *codepage);
112 int cifs_strtoUTF16(__le16 *, const char *, int, const struct nls_table *);
113 char *cifs_strndup_from_utf16(const char *src, const int maxlen,
114 const bool is_unicode,
115 const struct nls_table *codepage);
116 extern int cifsConvertToUTF16(__le16 *target, const char *source, int maxlen,
117 const struct nls_table *cp, int mapChars);
118 extern int cifs_remap(struct cifs_sb_info *cifs_sb);
119 #ifdef CONFIG_CIFS_SMB2
120 extern __le16 *cifs_strndup_to_utf16(const char *src, const int maxlen,
121 int *utf16_len, const struct nls_table *cp,
122 int remap);
123 #endif /* CONFIG_CIFS_SMB2 */
124 #endif
125
126 wchar_t cifs_toupper(wchar_t in);
127
128 /*
129 * UniStrcat: Concatenate the second string to the first
130 *
131 * Returns:
132 * Address of the first string
133 */
134 static inline wchar_t *
UniStrcat(wchar_t * ucs1,const wchar_t * ucs2)135 UniStrcat(wchar_t *ucs1, const wchar_t *ucs2)
136 {
137 wchar_t *anchor = ucs1; /* save a pointer to start of ucs1 */
138
139 while (*ucs1++) ; /* To end of first string */
140 ucs1--; /* Return to the null */
141 while ((*ucs1++ = *ucs2++)) ; /* copy string 2 over */
142 return anchor;
143 }
144
145 /*
146 * UniStrchr: Find a character in a string
147 *
148 * Returns:
149 * Address of first occurrence of character in string
150 * or NULL if the character is not in the string
151 */
152 static inline wchar_t *
UniStrchr(const wchar_t * ucs,wchar_t uc)153 UniStrchr(const wchar_t *ucs, wchar_t uc)
154 {
155 while ((*ucs != uc) && *ucs)
156 ucs++;
157
158 if (*ucs == uc)
159 return (wchar_t *) ucs;
160 return NULL;
161 }
162
163 /*
164 * UniStrcmp: Compare two strings
165 *
166 * Returns:
167 * < 0: First string is less than second
168 * = 0: Strings are equal
169 * > 0: First string is greater than second
170 */
171 static inline int
UniStrcmp(const wchar_t * ucs1,const wchar_t * ucs2)172 UniStrcmp(const wchar_t *ucs1, const wchar_t *ucs2)
173 {
174 while ((*ucs1 == *ucs2) && *ucs1) {
175 ucs1++;
176 ucs2++;
177 }
178 return (int) *ucs1 - (int) *ucs2;
179 }
180
181 /*
182 * UniStrcpy: Copy a string
183 */
184 static inline wchar_t *
UniStrcpy(wchar_t * ucs1,const wchar_t * ucs2)185 UniStrcpy(wchar_t *ucs1, const wchar_t *ucs2)
186 {
187 wchar_t *anchor = ucs1; /* save the start of result string */
188
189 while ((*ucs1++ = *ucs2++)) ;
190 return anchor;
191 }
192
193 /*
194 * UniStrlen: Return the length of a string (in 16 bit Unicode chars not bytes)
195 */
196 static inline size_t
UniStrlen(const wchar_t * ucs1)197 UniStrlen(const wchar_t *ucs1)
198 {
199 int i = 0;
200
201 while (*ucs1++)
202 i++;
203 return i;
204 }
205
206 /*
207 * UniStrnlen: Return the length (in 16 bit Unicode chars not bytes) of a
208 * string (length limited)
209 */
210 static inline size_t
UniStrnlen(const wchar_t * ucs1,int maxlen)211 UniStrnlen(const wchar_t *ucs1, int maxlen)
212 {
213 int i = 0;
214
215 while (*ucs1++) {
216 i++;
217 if (i >= maxlen)
218 break;
219 }
220 return i;
221 }
222
223 /*
224 * UniStrncat: Concatenate length limited string
225 */
226 static inline wchar_t *
UniStrncat(wchar_t * ucs1,const wchar_t * ucs2,size_t n)227 UniStrncat(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
228 {
229 wchar_t *anchor = ucs1; /* save pointer to string 1 */
230
231 while (*ucs1++) ;
232 ucs1--; /* point to null terminator of s1 */
233 while (n-- && (*ucs1 = *ucs2)) { /* copy s2 after s1 */
234 ucs1++;
235 ucs2++;
236 }
237 *ucs1 = 0; /* Null terminate the result */
238 return (anchor);
239 }
240
241 /*
242 * UniStrncmp: Compare length limited string
243 */
244 static inline int
UniStrncmp(const wchar_t * ucs1,const wchar_t * ucs2,size_t n)245 UniStrncmp(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
246 {
247 if (!n)
248 return 0; /* Null strings are equal */
249 while ((*ucs1 == *ucs2) && *ucs1 && --n) {
250 ucs1++;
251 ucs2++;
252 }
253 return (int) *ucs1 - (int) *ucs2;
254 }
255
256 /*
257 * UniStrncmp_le: Compare length limited string - native to little-endian
258 */
259 static inline int
UniStrncmp_le(const wchar_t * ucs1,const wchar_t * ucs2,size_t n)260 UniStrncmp_le(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
261 {
262 if (!n)
263 return 0; /* Null strings are equal */
264 while ((*ucs1 == __le16_to_cpu(*ucs2)) && *ucs1 && --n) {
265 ucs1++;
266 ucs2++;
267 }
268 return (int) *ucs1 - (int) __le16_to_cpu(*ucs2);
269 }
270
271 /*
272 * UniStrncpy: Copy length limited string with pad
273 */
274 static inline wchar_t *
UniStrncpy(wchar_t * ucs1,const wchar_t * ucs2,size_t n)275 UniStrncpy(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++ = *ucs2++;
281
282 n++;
283 while (n--) /* Pad with nulls */
284 *ucs1++ = 0;
285 return anchor;
286 }
287
288 /*
289 * UniStrncpy_le: Copy length limited string with pad to little-endian
290 */
291 static inline wchar_t *
UniStrncpy_le(wchar_t * ucs1,const wchar_t * ucs2,size_t n)292 UniStrncpy_le(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
293 {
294 wchar_t *anchor = ucs1;
295
296 while (n-- && *ucs2) /* Copy the strings */
297 *ucs1++ = __le16_to_cpu(*ucs2++);
298
299 n++;
300 while (n--) /* Pad with nulls */
301 *ucs1++ = 0;
302 return anchor;
303 }
304
305 /*
306 * UniStrstr: Find a string in a string
307 *
308 * Returns:
309 * Address of first match found
310 * NULL if no matching string is found
311 */
312 static inline wchar_t *
UniStrstr(const wchar_t * ucs1,const wchar_t * ucs2)313 UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2)
314 {
315 const wchar_t *anchor1 = ucs1;
316 const wchar_t *anchor2 = ucs2;
317
318 while (*ucs1) {
319 if (*ucs1 == *ucs2) {
320 /* Partial match found */
321 ucs1++;
322 ucs2++;
323 } else {
324 if (!*ucs2) /* Match found */
325 return (wchar_t *) anchor1;
326 ucs1 = ++anchor1; /* No match */
327 ucs2 = anchor2;
328 }
329 }
330
331 if (!*ucs2) /* Both end together */
332 return (wchar_t *) anchor1; /* Match found */
333 return NULL; /* No match */
334 }
335
336 #ifndef UNIUPR_NOUPPER
337 /*
338 * UniToupper: Convert a unicode character to upper case
339 */
340 static inline wchar_t
UniToupper(register wchar_t uc)341 UniToupper(register wchar_t uc)
342 {
343 register const struct UniCaseRange *rp;
344
345 if (uc < sizeof(CifsUniUpperTable)) {
346 /* Latin characters */
347 return uc + CifsUniUpperTable[uc]; /* Use base tables */
348 } else {
349 rp = CifsUniUpperRange; /* Use range tables */
350 while (rp->start) {
351 if (uc < rp->start) /* Before start of range */
352 return uc; /* Uppercase = input */
353 if (uc <= rp->end) /* In range */
354 return uc + rp->table[uc - rp->start];
355 rp++; /* Try next range */
356 }
357 }
358 return uc; /* Past last range */
359 }
360
361 /*
362 * UniStrupr: Upper case a unicode string
363 */
364 static inline __le16 *
UniStrupr(register __le16 * upin)365 UniStrupr(register __le16 *upin)
366 {
367 register __le16 *up;
368
369 up = upin;
370 while (*up) { /* For all characters */
371 *up = cpu_to_le16(UniToupper(le16_to_cpu(*up)));
372 up++;
373 }
374 return upin; /* Return input pointer */
375 }
376 #endif /* UNIUPR_NOUPPER */
377
378 #ifndef UNIUPR_NOLOWER
379 /*
380 * UniTolower: Convert a unicode character to lower case
381 */
382 static inline wchar_t
UniTolower(register wchar_t uc)383 UniTolower(register wchar_t uc)
384 {
385 register const struct UniCaseRange *rp;
386
387 if (uc < sizeof(CifsUniLowerTable)) {
388 /* Latin characters */
389 return uc + CifsUniLowerTable[uc]; /* Use base tables */
390 } else {
391 rp = CifsUniLowerRange; /* Use range tables */
392 while (rp->start) {
393 if (uc < rp->start) /* Before start of range */
394 return uc; /* Uppercase = input */
395 if (uc <= rp->end) /* In range */
396 return uc + rp->table[uc - rp->start];
397 rp++; /* Try next range */
398 }
399 }
400 return uc; /* Past last range */
401 }
402
403 /*
404 * UniStrlwr: Lower case a unicode string
405 */
406 static inline wchar_t *
UniStrlwr(register wchar_t * upin)407 UniStrlwr(register wchar_t *upin)
408 {
409 register wchar_t *up;
410
411 up = upin;
412 while (*up) { /* For all characters */
413 *up = UniTolower(*up);
414 up++;
415 }
416 return upin; /* Return input pointer */
417 }
418
419 #endif
420
421 #endif /* _CIFS_UNICODE_H */
422