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 extern __le16 *cifs_strndup_to_utf16(const char *src, const int maxlen,
120 int *utf16_len, const struct nls_table *cp,
121 int remap);
122 #endif
123
124 wchar_t cifs_toupper(wchar_t in);
125
126 /*
127 * UniStrcat: Concatenate the second string to the first
128 *
129 * Returns:
130 * Address of the first string
131 */
132 static inline __le16 *
UniStrcat(__le16 * ucs1,const __le16 * ucs2)133 UniStrcat(__le16 *ucs1, const __le16 *ucs2)
134 {
135 __le16 *anchor = ucs1; /* save a pointer to start of ucs1 */
136
137 while (*ucs1++) ; /* To end of first string */
138 ucs1--; /* Return to the null */
139 while ((*ucs1++ = *ucs2++)) ; /* copy string 2 over */
140 return anchor;
141 }
142
143 /*
144 * UniStrchr: Find a character in a string
145 *
146 * Returns:
147 * Address of first occurrence of character in string
148 * or NULL if the character is not in the string
149 */
150 static inline wchar_t *
UniStrchr(const wchar_t * ucs,wchar_t uc)151 UniStrchr(const wchar_t *ucs, wchar_t uc)
152 {
153 while ((*ucs != uc) && *ucs)
154 ucs++;
155
156 if (*ucs == uc)
157 return (wchar_t *) ucs;
158 return NULL;
159 }
160
161 /*
162 * UniStrcmp: Compare two strings
163 *
164 * Returns:
165 * < 0: First string is less than second
166 * = 0: Strings are equal
167 * > 0: First string is greater than second
168 */
169 static inline int
UniStrcmp(const wchar_t * ucs1,const wchar_t * ucs2)170 UniStrcmp(const wchar_t *ucs1, const wchar_t *ucs2)
171 {
172 while ((*ucs1 == *ucs2) && *ucs1) {
173 ucs1++;
174 ucs2++;
175 }
176 return (int) *ucs1 - (int) *ucs2;
177 }
178
179 /*
180 * UniStrcpy: Copy a string
181 */
182 static inline wchar_t *
UniStrcpy(wchar_t * ucs1,const wchar_t * ucs2)183 UniStrcpy(wchar_t *ucs1, const wchar_t *ucs2)
184 {
185 wchar_t *anchor = ucs1; /* save the start of result string */
186
187 while ((*ucs1++ = *ucs2++)) ;
188 return anchor;
189 }
190
191 /*
192 * UniStrlen: Return the length of a string (in 16 bit Unicode chars not bytes)
193 */
194 static inline size_t
UniStrlen(const wchar_t * ucs1)195 UniStrlen(const wchar_t *ucs1)
196 {
197 int i = 0;
198
199 while (*ucs1++)
200 i++;
201 return i;
202 }
203
204 /*
205 * UniStrnlen: Return the length (in 16 bit Unicode chars not bytes) of a
206 * string (length limited)
207 */
208 static inline size_t
UniStrnlen(const wchar_t * ucs1,int maxlen)209 UniStrnlen(const wchar_t *ucs1, int maxlen)
210 {
211 int i = 0;
212
213 while (*ucs1++) {
214 i++;
215 if (i >= maxlen)
216 break;
217 }
218 return i;
219 }
220
221 /*
222 * UniStrncat: Concatenate length limited string
223 */
224 static inline wchar_t *
UniStrncat(wchar_t * ucs1,const wchar_t * ucs2,size_t n)225 UniStrncat(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
226 {
227 wchar_t *anchor = ucs1; /* save pointer to string 1 */
228
229 while (*ucs1++) ;
230 ucs1--; /* point to null terminator of s1 */
231 while (n-- && (*ucs1 = *ucs2)) { /* copy s2 after s1 */
232 ucs1++;
233 ucs2++;
234 }
235 *ucs1 = 0; /* Null terminate the result */
236 return (anchor);
237 }
238
239 /*
240 * UniStrncmp: Compare length limited string
241 */
242 static inline int
UniStrncmp(const wchar_t * ucs1,const wchar_t * ucs2,size_t n)243 UniStrncmp(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 == *ucs2) && *ucs1 && --n) {
248 ucs1++;
249 ucs2++;
250 }
251 return (int) *ucs1 - (int) *ucs2;
252 }
253
254 /*
255 * UniStrncmp_le: Compare length limited string - native to little-endian
256 */
257 static inline int
UniStrncmp_le(const wchar_t * ucs1,const wchar_t * ucs2,size_t n)258 UniStrncmp_le(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
259 {
260 if (!n)
261 return 0; /* Null strings are equal */
262 while ((*ucs1 == __le16_to_cpu(*ucs2)) && *ucs1 && --n) {
263 ucs1++;
264 ucs2++;
265 }
266 return (int) *ucs1 - (int) __le16_to_cpu(*ucs2);
267 }
268
269 /*
270 * UniStrncpy: Copy length limited string with pad
271 */
272 static inline wchar_t *
UniStrncpy(wchar_t * ucs1,const wchar_t * ucs2,size_t n)273 UniStrncpy(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
274 {
275 wchar_t *anchor = ucs1;
276
277 while (n-- && *ucs2) /* Copy the strings */
278 *ucs1++ = *ucs2++;
279
280 n++;
281 while (n--) /* Pad with nulls */
282 *ucs1++ = 0;
283 return anchor;
284 }
285
286 /*
287 * UniStrncpy_le: Copy length limited string with pad to little-endian
288 */
289 static inline wchar_t *
UniStrncpy_le(wchar_t * ucs1,const wchar_t * ucs2,size_t n)290 UniStrncpy_le(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
291 {
292 wchar_t *anchor = ucs1;
293
294 while (n-- && *ucs2) /* Copy the strings */
295 *ucs1++ = __le16_to_cpu(*ucs2++);
296
297 n++;
298 while (n--) /* Pad with nulls */
299 *ucs1++ = 0;
300 return anchor;
301 }
302
303 /*
304 * UniStrstr: Find a string in a string
305 *
306 * Returns:
307 * Address of first match found
308 * NULL if no matching string is found
309 */
310 static inline wchar_t *
UniStrstr(const wchar_t * ucs1,const wchar_t * ucs2)311 UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2)
312 {
313 const wchar_t *anchor1 = ucs1;
314 const wchar_t *anchor2 = ucs2;
315
316 while (*ucs1) {
317 if (*ucs1 == *ucs2) {
318 /* Partial match found */
319 ucs1++;
320 ucs2++;
321 } else {
322 if (!*ucs2) /* Match found */
323 return (wchar_t *) anchor1;
324 ucs1 = ++anchor1; /* No match */
325 ucs2 = anchor2;
326 }
327 }
328
329 if (!*ucs2) /* Both end together */
330 return (wchar_t *) anchor1; /* Match found */
331 return NULL; /* No match */
332 }
333
334 #ifndef UNIUPR_NOUPPER
335 /*
336 * UniToupper: Convert a unicode character to upper case
337 */
338 static inline wchar_t
UniToupper(register wchar_t uc)339 UniToupper(register wchar_t uc)
340 {
341 register const struct UniCaseRange *rp;
342
343 if (uc < sizeof(CifsUniUpperTable)) {
344 /* Latin characters */
345 return uc + CifsUniUpperTable[uc]; /* Use base tables */
346 } else {
347 rp = CifsUniUpperRange; /* Use range tables */
348 while (rp->start) {
349 if (uc < rp->start) /* Before start of range */
350 return uc; /* Uppercase = input */
351 if (uc <= rp->end) /* In range */
352 return uc + rp->table[uc - rp->start];
353 rp++; /* Try next range */
354 }
355 }
356 return uc; /* Past last range */
357 }
358
359 /*
360 * UniStrupr: Upper case a unicode string
361 */
362 static inline __le16 *
UniStrupr(register __le16 * upin)363 UniStrupr(register __le16 *upin)
364 {
365 register __le16 *up;
366
367 up = upin;
368 while (*up) { /* For all characters */
369 *up = cpu_to_le16(UniToupper(le16_to_cpu(*up)));
370 up++;
371 }
372 return upin; /* Return input pointer */
373 }
374 #endif /* UNIUPR_NOUPPER */
375
376 #ifndef UNIUPR_NOLOWER
377 /*
378 * UniTolower: Convert a unicode character to lower case
379 */
380 static inline wchar_t
UniTolower(register wchar_t uc)381 UniTolower(register wchar_t uc)
382 {
383 register const struct UniCaseRange *rp;
384
385 if (uc < sizeof(CifsUniLowerTable)) {
386 /* Latin characters */
387 return uc + CifsUniLowerTable[uc]; /* Use base tables */
388 } else {
389 rp = CifsUniLowerRange; /* Use range tables */
390 while (rp->start) {
391 if (uc < rp->start) /* Before start of range */
392 return uc; /* Uppercase = input */
393 if (uc <= rp->end) /* In range */
394 return uc + rp->table[uc - rp->start];
395 rp++; /* Try next range */
396 }
397 }
398 return uc; /* Past last range */
399 }
400
401 /*
402 * UniStrlwr: Lower case a unicode string
403 */
404 static inline wchar_t *
UniStrlwr(register wchar_t * upin)405 UniStrlwr(register wchar_t *upin)
406 {
407 register wchar_t *up;
408
409 up = upin;
410 while (*up) { /* For all characters */
411 *up = UniTolower(*up);
412 up++;
413 }
414 return upin; /* Return input pointer */
415 }
416
417 #endif
418
419 #endif /* _CIFS_UNICODE_H */
420