fs/cifs/cifs_unicode.h

   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_ASTERISK    ((__u16) 0xF021)
  61 #define SFM_QUESTION    ((__u16) 0xF025)
  62 #define SFM_COLON       ((__u16) 0xF022)
  63 #define SFM_GRTRTHAN    ((__u16) 0xF024)
  64 #define SFM_LESSTHAN    ((__u16) 0xF023)
  65 #define SFM_PIPE        ((__u16) 0xF027)
  66 #define SFM_SLASH       ((__u16) 0xF026)
  67 #define SFM_PERIOD      ((__u16) 0xF028)
  68 #define SFM_SPACE       ((__u16) 0xF029)
  69
  70 /*
  71  * Mapping mechanism to use when one of the seven reserved characters is
  72  * encountered.  We can only map using one of the mechanisms at a time
  73  * since otherwise readdir could return directory entries which we would
  74  * not be able to open
  75  *
  76  * NO_MAP_UNI_RSVD  = do not perform any remapping of the character
  77  * SFM_MAP_UNI_RSVD = map reserved characters using SFM scheme (MAC compatible)
  78  * SFU_MAP_UNI_RSVD = map reserved characters ala SFU ("mapchars" option)
  79  *
  80  */
  81 #define NO_MAP_UNI_RSVD         0
  82 #define SFM_MAP_UNI_RSVD        1
  83 #define SFU_MAP_UNI_RSVD        2
  84
  85 /* Just define what we want from uniupr.h.  We don't want to define the tables
  86  * in each source file.
  87  */
  88 #ifndef UNICASERANGE_DEFINED
  89 struct UniCaseRange {
  90         wchar_t start;
  91         wchar_t end;
  92         signed char *table;
  93 };
  94 #endif                          /* UNICASERANGE_DEFINED */
  95
  96 #ifndef UNIUPR_NOUPPER
  97 extern signed char CifsUniUpperTable[512];
  98 extern const struct UniCaseRange CifsUniUpperRange[];
  99 #endif                          /* UNIUPR_NOUPPER */
 100
 101 #ifndef UNIUPR_NOLOWER
 102 extern signed char CifsUniLowerTable[512];
 103 extern const struct UniCaseRange CifsUniLowerRange[];
 104 #endif                          /* UNIUPR_NOLOWER */
 105
 106 #ifdef __KERNEL__
 107 int cifs_from_utf16(char *to, const __le16 *from, int tolen, int fromlen,
 108                     const struct nls_table *cp, int map_type);
 109 int cifs_utf16_bytes(const __le16 *from, int maxbytes,
 110                      const struct nls_table *codepage);
 111 int cifs_strtoUTF16(__le16 *, const char *, int, const struct nls_table *);
 112 char *cifs_strndup_from_utf16(const char *src, const int maxlen,
 113                               const bool is_unicode,
 114                               const struct nls_table *codepage);
 115 extern int cifsConvertToUTF16(__le16 *target, const char *source, int maxlen,
 116                               const struct nls_table *cp, int mapChars);
 117 extern int cifs_remap(struct cifs_sb_info *cifs_sb);
 118 #ifdef CONFIG_CIFS_SMB2
 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 /* CONFIG_CIFS_SMB2 */
 123 #endif
 124
 125 wchar_t cifs_toupper(wchar_t in);
 126
 127 /*
 128  * UniStrcat:  Concatenate the second string to the first
 129  *
 130  * Returns:
 131  *     Address of the first string
 132  */
 133 static inline wchar_t *
 134 UniStrcat(wchar_t *ucs1, const wchar_t *ucs2)
 135 {
 136         wchar_t *anchor = ucs1; /* save a pointer to start of ucs1 */
 137
 138         while (*ucs1++) ;       /* To end of first string */
 139         ucs1--;                 /* Return to the null */
 140         while ((*ucs1++ = *ucs2++)) ;   /* copy string 2 over */
 141         return anchor;
 142 }
 143
 144 /*
 145  * UniStrchr:  Find a character in a string
 146  *
 147  * Returns:
 148  *     Address of first occurrence of character in string
 149  *     or NULL if the character is not in the string
 150  */
 151 static inline wchar_t *
 152 UniStrchr(const wchar_t *ucs, wchar_t uc)
 153 {
 154         while ((*ucs != uc) && *ucs)
 155                 ucs++;
 156
 157         if (*ucs == uc)
 158                 return (wchar_t *) ucs;
 159         return NULL;
 160 }
 161
 162 /*
 163  * UniStrcmp:  Compare two strings
 164  *
 165  * Returns:
 166  *     < 0:  First string is less than second
 167  *     = 0:  Strings are equal
 168  *     > 0:  First string is greater than second
 169  */
 170 static inline int
 171 UniStrcmp(const wchar_t *ucs1, const wchar_t *ucs2)
 172 {
 173         while ((*ucs1 == *ucs2) && *ucs1) {
 174                 ucs1++;
 175                 ucs2++;
 176         }
 177         return (int) *ucs1 - (int) *ucs2;
 178 }
 179
 180 /*
 181  * UniStrcpy:  Copy a string
 182  */
 183 static inline wchar_t *
 184 UniStrcpy(wchar_t *ucs1, const wchar_t *ucs2)
 185 {
 186         wchar_t *anchor = ucs1; /* save the start of result string */
 187
 188         while ((*ucs1++ = *ucs2++)) ;
 189         return anchor;
 190 }
 191
 192 /*
 193  * UniStrlen:  Return the length of a string (in 16 bit Unicode chars not bytes)
 194  */
 195 static inline size_t
 196 UniStrlen(const wchar_t *ucs1)
 197 {
 198         int i = 0;
 199
 200         while (*ucs1++)
 201                 i++;
 202         return i;
 203 }
 204
 205 /*
 206  * UniStrnlen:  Return the length (in 16 bit Unicode chars not bytes) of a
 207  *              string (length limited)
 208  */
 209 static inline size_t
 210 UniStrnlen(const wchar_t *ucs1, int maxlen)
 211 {
 212         int i = 0;
 213
 214         while (*ucs1++) {
 215                 i++;
 216                 if (i >= maxlen)
 217                         break;
 218         }
 219         return i;
 220 }
 221
 222 /*
 223  * UniStrncat:  Concatenate length limited string
 224  */
 225 static inline wchar_t *
 226 UniStrncat(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
 227 {
 228         wchar_t *anchor = ucs1; /* save pointer to string 1 */
 229
 230         while (*ucs1++) ;
 231         ucs1--;                 /* point to null terminator of s1 */
 232         while (n-- && (*ucs1 = *ucs2)) {        /* copy s2 after s1 */
 233                 ucs1++;
 234                 ucs2++;
 235         }
 236         *ucs1 = 0;              /* Null terminate the result */
 237         return (anchor);
 238 }
 239
 240 /*
 241  * UniStrncmp:  Compare length limited string
 242  */
 243 static inline int
 244 UniStrncmp(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
 245 {
 246         if (!n)
 247                 return 0;       /* Null strings are equal */
 248         while ((*ucs1 == *ucs2) && *ucs1 && --n) {
 249                 ucs1++;
 250                 ucs2++;
 251         }
 252         return (int) *ucs1 - (int) *ucs2;
 253 }
 254
 255 /*
 256  * UniStrncmp_le:  Compare length limited string - native to little-endian
 257  */
 258 static inline int
 259 UniStrncmp_le(const wchar_t *ucs1, const wchar_t *ucs2, size_t n)
 260 {
 261         if (!n)
 262                 return 0;       /* Null strings are equal */
 263         while ((*ucs1 == __le16_to_cpu(*ucs2)) && *ucs1 && --n) {
 264                 ucs1++;
 265                 ucs2++;
 266         }
 267         return (int) *ucs1 - (int) __le16_to_cpu(*ucs2);
 268 }
 269
 270 /*
 271  * UniStrncpy:  Copy length limited string with pad
 272  */
 273 static inline wchar_t *
 274 UniStrncpy(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
 275 {
 276         wchar_t *anchor = ucs1;
 277
 278         while (n-- && *ucs2)    /* Copy the strings */
 279                 *ucs1++ = *ucs2++;
 280
 281         n++;
 282         while (n--)             /* Pad with nulls */
 283                 *ucs1++ = 0;
 284         return anchor;
 285 }
 286
 287 /*
 288  * UniStrncpy_le:  Copy length limited string with pad to little-endian
 289  */
 290 static inline wchar_t *
 291 UniStrncpy_le(wchar_t *ucs1, const wchar_t *ucs2, size_t n)
 292 {
 293         wchar_t *anchor = ucs1;
 294
 295         while (n-- && *ucs2)    /* Copy the strings */
 296                 *ucs1++ = __le16_to_cpu(*ucs2++);
 297
 298         n++;
 299         while (n--)             /* Pad with nulls */
 300                 *ucs1++ = 0;
 301         return anchor;
 302 }
 303
 304 /*
 305  * UniStrstr:  Find a string in a string
 306  *
 307  * Returns:
 308  *     Address of first match found
 309  *     NULL if no matching string is found
 310  */
 311 static inline wchar_t *
 312 UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2)
 313 {
 314         const wchar_t *anchor1 = ucs1;
 315         const wchar_t *anchor2 = ucs2;
 316
 317         while (*ucs1) {
 318                 if (*ucs1 == *ucs2) {
 319                         /* Partial match found */
 320                         ucs1++;
 321                         ucs2++;
 322                 } else {
 323                         if (!*ucs2)     /* Match found */
 324                                 return (wchar_t *) anchor1;
 325                         ucs1 = ++anchor1;       /* No match */
 326                         ucs2 = anchor2;
 327                 }
 328         }
 329
 330         if (!*ucs2)             /* Both end together */
 331                 return (wchar_t *) anchor1;     /* Match found */
 332         return NULL;            /* No match */
 333 }
 334
 335 #ifndef UNIUPR_NOUPPER
 336 /*
 337  * UniToupper:  Convert a unicode character to upper case
 338  */
 339 static inline wchar_t
 340 UniToupper(register wchar_t uc)
 341 {
 342         register const struct UniCaseRange *rp;
 343
 344         if (uc < sizeof(CifsUniUpperTable)) {
 345                 /* Latin characters */
 346                 return uc + CifsUniUpperTable[uc];      /* Use base tables */
 347         } else {
 348                 rp = CifsUniUpperRange; /* Use range tables */
 349                 while (rp->start) {
 350                         if (uc < rp->start)     /* Before start of range */
 351                                 return uc;      /* Uppercase = input */
 352                         if (uc <= rp->end)      /* In range */
 353                                 return uc + rp->table[uc - rp->start];
 354                         rp++;   /* Try next range */
 355                 }
 356         }
 357         return uc;              /* Past last range */
 358 }
 359
 360 /*
 361  * UniStrupr:  Upper case a unicode string
 362  */
 363 static inline __le16 *
 364 UniStrupr(register __le16 *upin)
 365 {
 366         register __le16 *up;
 367
 368         up = upin;
 369         while (*up) {           /* For all characters */
 370                 *up = cpu_to_le16(UniToupper(le16_to_cpu(*up)));
 371                 up++;
 372         }
 373         return upin;            /* Return input pointer */
 374 }
 375 #endif                          /* UNIUPR_NOUPPER */
 376
 377 #ifndef UNIUPR_NOLOWER
 378 /*
 379  * UniTolower:  Convert a unicode character to lower case
 380  */
 381 static inline wchar_t
 382 UniTolower(register wchar_t uc)
 383 {
 384         register const struct UniCaseRange *rp;
 385
 386         if (uc < sizeof(CifsUniLowerTable)) {
 387                 /* Latin characters */
 388                 return uc + CifsUniLowerTable[uc];      /* Use base tables */
 389         } else {
 390                 rp = CifsUniLowerRange; /* Use range tables */
 391                 while (rp->start) {
 392                         if (uc < rp->start)     /* Before start of range */
 393                                 return uc;      /* Uppercase = input */
 394                         if (uc <= rp->end)      /* In range */
 395                                 return uc + rp->table[uc - rp->start];
 396                         rp++;   /* Try next range */
 397                 }
 398         }
 399         return uc;              /* Past last range */
 400 }
 401
 402 /*
 403  * UniStrlwr:  Lower case a unicode string
 404  */
 405 static inline wchar_t *
 406 UniStrlwr(register wchar_t *upin)
 407 {
 408         register wchar_t *up;
 409
 410         up = upin;
 411         while (*up) {           /* For all characters */
 412                 *up = UniTolower(*up);
 413                 up++;
 414         }
 415         return upin;            /* Return input pointer */
 416 }
 417
 418 #endif
 419
 420 #endif /* _CIFS_UNICODE_H */