source/lib/util_unistr.c

   1 /*
   2    Unix SMB/Netbios implementation.
   3    Version 3.0
   4    Samba utility functions
   5    Copyright (C) Andrew Tridgell 1992-2001
   6
   7    This program is free software; you can redistribute it and/or modify
   8    it under the terms of the GNU General Public License as published by
   9    the Free Software Foundation; either version 2 of the License, or
  10    (at your option) any later version.
  11
  12    This program is distributed in the hope that it will be useful,
  13    but WITHOUT ANY WARRANTY; without even the implied warranty of
  14    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15    GNU General Public License for more details.
  16
  17    You should have received a copy of the GNU General Public License
  18    along with this program; if not, write to the Free Software
  19    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  20 */
  21
  22 #include "includes.h"
  23
  24 #ifndef MAXUNI
  25 #define MAXUNI 1024
  26 #endif
  27
  28 /* these 3 tables define the unicode case handling.  They are loaded
  29    at startup either via mmap() or read() from the lib directory */
  30 static smb_ucs2_t *upcase_table;
  31 static smb_ucs2_t *lowcase_table;
  32 static uint8 *valid_table;
  33
  34 /*******************************************************************
  35 load the case handling tables
  36 ********************************************************************/
  37 void load_case_tables(void)
  38 {
  39         static int initialised;
  40         int i;
  41
  42         if (initialised) return;
  43         initialised = 1;
  44
  45         upcase_table = map_file(lib_path("upcase.dat"), 0x20000);
  46         lowcase_table = map_file(lib_path("lowcase.dat"), 0x20000);
  47         valid_table = map_file(lib_path("valid.dat"), 0x10000);
  48
  49         /* we would like Samba to limp along even if these tables are
  50            not available */
  51         if (!upcase_table) {
  52                 DEBUG(1,("creating lame upcase table\n"));
  53                 upcase_table = malloc(0x20000);
  54                 for (i=0;i<256;i++) upcase_table[i] = islower(i)?toupper(i):i;
  55                 for (;i<0x10000;i++) upcase_table[i] = i;
  56         }
  57
  58         if (!lowcase_table) {
  59                 DEBUG(1,("creating lame lowcase table\n"));
  60                 lowcase_table = malloc(0x20000);
  61                 for (i=0;i<256;i++) lowcase_table[i] = isupper(i)?tolower(i):i;
  62                 for (;i<0x10000;i++) lowcase_table[i] = i;
  63         }
  64
  65         if (!valid_table) {
  66                 const char *allowed = "!#$%&'()_-@^`~";
  67                 DEBUG(1,("creating lame valid table\n"));
  68                 valid_table = malloc(0x10000);
  69                 for (i=0;i<256;i++) valid_table[i] = isalnum(i) || strchr(allowed,i);
  70                 for (;i<0x10000;i++) valid_table[i] = 0;
  71         }
  72 }
  73
  74
  75 /*******************************************************************
  76  Write a string in (little-endian) unicode format. src is in
  77  the current DOS codepage. len is the length in bytes of the
  78  string pointed to by dst.
  79
  80  if null_terminate is True then null terminate the packet (adds 2 bytes)
  81
  82  the return value is the length in bytes consumed by the string, including the
  83  null termination if applied
  84 ********************************************************************/
  85
  86 size_t dos_PutUniCode(char *dst,const char *src, ssize_t len, BOOL null_terminate)
  87 {
  88         return push_ucs2(NULL, dst, src, len,
  89                          STR_UNICODE|STR_NOALIGN | (null_terminate?STR_TERMINATE:0));
  90 }
  91
  92
  93 /*******************************************************************
  94  Skip past a unicode string, but not more than len. Always move
  95  past a terminating zero if found.
  96 ********************************************************************/
  97
  98 char *skip_unibuf(char *src, size_t len)
  99 {
 100     char *srcend = src + len;
 101
 102     while (src < srcend && SVAL(src,0))
 103         src += 2;
 104
 105     if(!SVAL(src,0))
 106         src += 2;
 107
 108     return src;
 109 }
 110
 111 /* Copy a string from little-endian or big-endian unicode source (depending
 112  * on flags) to internal samba format destination
 113  */
 114 int rpcstr_pull(char* dest, void *src, int dest_len, int src_len, int flags)
 115 {
 116         if(dest_len==-1) dest_len=MAXUNI-3;
 117         return pull_ucs2(NULL, dest, src, dest_len, src_len, flags|STR_UNICODE|STR_NOALIGN);
 118 }
 119
 120 /* Copy a string from a unistr2 source to internal samba format
 121    destination.  Use this instead of direct calls to rpcstr_pull() to avoid
 122    having to determine whether the source string is null terminated. */
 123
 124 int rpcstr_pull_unistr2_fstring(char *dest, UNISTR2 *src)
 125 {
 126         return pull_ucs2(NULL, dest, src->buffer, sizeof(fstring),
 127                          src->uni_str_len * 2, 0);
 128 }
 129
 130 /* Converts a string from internal samba format to unicode
 131  */
 132 int rpcstr_push(void* dest, const char *src, int dest_len, int flags)
 133 {
 134         return push_ucs2(NULL, dest, src, dest_len, flags|STR_UNICODE|STR_NOALIGN);
 135 }
 136
 137 /*******************************************************************
 138  Return a DOS codepage version of a little-endian unicode string.
 139  len is the filename length (ignoring any terminating zero) in uin16
 140  units. Always null terminates.
 141  Hack alert: uses fixed buffer(s).
 142 ********************************************************************/
 143 char *dos_unistrn2(const uint16 *src, int len)
 144 {
 145         static char lbufs[8][MAXUNI];
 146         static int nexti;
 147         char *lbuf = lbufs[nexti];
 148         nexti = (nexti+1)%8;
 149         pull_ucs2(NULL, lbuf, src, MAXUNI-3, len*2, STR_NOALIGN);
 150         return lbuf;
 151 }
 152
 153 /*******************************************************************
 154  Convert a (little-endian) UNISTR2 structure to an ASCII string
 155 ********************************************************************/
 156 void unistr2_to_ascii(char *dest, const UNISTR2 *str, size_t maxlen)
 157 {
 158         if (str == NULL) {
 159                 *dest='\0';
 160                 return;
 161         }
 162         pull_ucs2(NULL, dest, str->buffer, maxlen, str->uni_str_len*2, STR_NOALIGN);
 163 }
 164
 165
 166 /*******************************************************************
 167 Return a number stored in a buffer
 168 ********************************************************************/
 169
 170 uint32 buffer2_to_uint32(BUFFER2 *str)
 171 {
 172         if (str->buf_len == 4)
 173                 return IVAL(str->buffer, 0);
 174         else
 175                 return 0;
 176 }
 177
 178 /*******************************************************************
 179  Convert a wchar to upper case.
 180 ********************************************************************/
 181
 182 smb_ucs2_t toupper_w(smb_ucs2_t val)
 183 {
 184         return upcase_table[SVAL(&val,0)];
 185 }
 186
 187 /*******************************************************************
 188  Convert a wchar to lower case.
 189 ********************************************************************/
 190
 191 smb_ucs2_t tolower_w( smb_ucs2_t val )
 192 {
 193         return lowcase_table[SVAL(&val,0)];
 194 }
 195
 196 /*******************************************************************
 197 determine if a character is lowercase
 198 ********************************************************************/
 199 BOOL islower_w(smb_ucs2_t c)
 200 {
 201         return upcase_table[SVAL(&c,0)] != c;
 202 }
 203
 204 /*******************************************************************
 205 determine if a character is uppercase
 206 ********************************************************************/
 207 BOOL isupper_w(smb_ucs2_t c)
 208 {
 209         return lowcase_table[SVAL(&c,0)] != c;
 210 }
 211
 212
 213 /*******************************************************************
 214 determine if a character is valid in a 8.3 name
 215 ********************************************************************/
 216 BOOL isvalid83_w(smb_ucs2_t c)
 217 {
 218         return valid_table[SVAL(&c,0)] != 0;
 219 }
 220
 221 /*******************************************************************
 222  Count the number of characters in a smb_ucs2_t string.
 223 ********************************************************************/
 224 size_t strlen_w(const smb_ucs2_t *src)
 225 {
 226         size_t len;
 227
 228         for(len = 0; *src++; len++) ;
 229
 230         return len;
 231 }
 232
 233 /*******************************************************************
 234 wide strchr()
 235 ********************************************************************/
 236 smb_ucs2_t *strchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
 237 {
 238         while (*s != 0) {
 239                 if (c == *s) return (smb_ucs2_t *)s;
 240                 s++;
 241         }
 242         return NULL;
 243 }
 244
 245 smb_ucs2_t *strchr_wa(const smb_ucs2_t *s, char c)
 246 {
 247         return strchr_w(s, UCS2_CHAR(c));
 248 }
 249
 250 smb_ucs2_t *strrchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
 251 {
 252         const smb_ucs2_t *p = s;
 253         int len = strlen_w(s);
 254         if (len == 0) return NULL;
 255         p += (len - 1);
 256         do {
 257                 if (c == *p) return (smb_ucs2_t *)p;
 258         } while (p-- != s);
 259         return NULL;
 260 }
 261
 262 /*******************************************************************
 263 wide strstr()
 264 ********************************************************************/
 265 smb_ucs2_t *strstr_w(const smb_ucs2_t *s, const smb_ucs2_t *ins)
 266 {
 267         smb_ucs2_t *r;
 268         size_t slen, inslen;
 269
 270         if (!s || !*s || !ins || !*ins) return NULL;
 271         slen = strlen_w(s);
 272         inslen = strlen_w(ins);
 273         r = (smb_ucs2_t *)s;
 274         while (r = strchr_w(r, *ins)) {
 275                 if (strncmp_w(r, ins, inslen) == 0) return r;
 276                 r++;
 277         }
 278         return NULL;
 279 }
 280
 281 smb_ucs2_t *strstr_wa(const smb_ucs2_t *s, const char *ins)
 282 {
 283         smb_ucs2_t *r;
 284         size_t slen, inslen;
 285
 286         if (!s || !*s || !ins || !*ins) return NULL;
 287         slen = strlen_w(s);
 288         inslen = strlen(ins);
 289         r = (smb_ucs2_t *)s;
 290         while (r = strchr_w(r, UCS2_CHAR(*ins))) {
 291                 if (strncmp_wa(r, ins, inslen) == 0) return r;
 292                 r++;
 293         }
 294         return NULL;
 295 }
 296
 297 /*******************************************************************
 298  Convert a string to lower case.
 299  return True if any char is converted
 300 ********************************************************************/
 301 BOOL strlower_w(smb_ucs2_t *s)
 302 {
 303         BOOL ret = False;
 304         while (*s) {
 305                 smb_ucs2_t v = tolower_w(*s);
 306                 if (v != *s) {
 307                         *s = v;
 308                         ret = True;
 309                 }
 310                 s++;
 311         }
 312         return ret;
 313 }
 314
 315 /*******************************************************************
 316  Convert a string to upper case.
 317  return True if any char is converted
 318 ********************************************************************/
 319 BOOL strupper_w(smb_ucs2_t *s)
 320 {
 321         BOOL ret = False;
 322         while (*s) {
 323                 smb_ucs2_t v = toupper_w(*s);
 324                 if (v != *s) {
 325                         *s = v;
 326                         ret = True;
 327                 }
 328                 s++;
 329         }
 330         return ret;
 331 }
 332
 333 /*******************************************************************
 334   convert a string to "normal" form
 335 ********************************************************************/
 336 void strnorm_w(smb_ucs2_t *s)
 337 {
 338   extern int case_default;
 339   if (case_default == CASE_UPPER)
 340     strupper_w(s);
 341   else
 342     strlower_w(s);
 343 }
 344
 345 /*******************************************************************
 346 case insensitive string comparison
 347 ********************************************************************/
 348 int strcasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
 349 {
 350         while (*b && toupper_w(*a) == toupper_w(*b)) { a++; b++; }
 351         return (tolower_w(*a) - tolower_w(*b));
 352 }
 353
 354 /*******************************************************************
 355 case insensitive string comparison, lenght limited
 356 ********************************************************************/
 357 int strncasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
 358 {
 359         size_t n = 0;
 360         while ((n < len) && *b && (toupper_w(*a) == toupper_w(*b))) { a++; b++; n++; }
 361         return (len - n)?(tolower_w(*a) - tolower_w(*b)):0;
 362 }
 363
 364 /*******************************************************************
 365   compare 2 strings
 366 ********************************************************************/
 367 BOOL strequal_w(const smb_ucs2_t *s1, const smb_ucs2_t *s2)
 368 {
 369         if (s1 == s2) return(True);
 370         if (!s1 || !s2) return(False);
 371
 372         return(strcasecmp_w(s1,s2)==0);
 373 }
 374
 375 /*******************************************************************
 376   compare 2 strings up to and including the nth char.
 377   ******************************************************************/
 378 BOOL strnequal_w(const smb_ucs2_t *s1,const smb_ucs2_t *s2,size_t n)
 379 {
 380   if (s1 == s2) return(True);
 381   if (!s1 || !s2 || !n) return(False);
 382
 383   return(strncasecmp_w(s1,s2,n)==0);
 384 }
 385
 386 /*******************************************************************
 387 duplicate string
 388 ********************************************************************/
 389 smb_ucs2_t *strdup_w(const smb_ucs2_t *src)
 390 {
 391         return strndup_w(src, 0);
 392 }
 393
 394 /* if len == 0 then duplicate the whole string */
 395 smb_ucs2_t *strndup_w(const smb_ucs2_t *src, size_t len)
 396 {
 397         smb_ucs2_t *dest;
 398
 399         if (!len) len = strlen_w(src);
 400         dest = (smb_ucs2_t *)malloc((len + 1) * sizeof(smb_ucs2_t));
 401         if (!dest) {
 402                 DEBUG(0,("strdup_w: out of memory!\n"));
 403                 return NULL;
 404         }
 405
 406         memcpy(dest, src, len * sizeof(smb_ucs2_t));
 407         dest[len] = 0;
 408
 409         return dest;
 410 }
 411
 412 /*******************************************************************
 413 copy a string with max len
 414 ********************************************************************/
 415
 416 smb_ucs2_t *strncpy_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
 417 {
 418         size_t len;
 419
 420         if (!dest || !src) return NULL;
 421
 422         for (len = 0; (src[len] != 0) && (len < max); len++)
 423                 dest[len] = src[len];
 424         while (len < max)
 425                 dest[len++] = 0;
 426
 427         return dest;
 428 }
 429
 430
 431 /*******************************************************************
 432 append a string of len bytes and add a terminator
 433 ********************************************************************/
 434
 435 smb_ucs2_t *strncat_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
 436 {
 437         size_t start;
 438         size_t len;
 439
 440         if (!dest || !src) return NULL;
 441
 442         start = strlen_w(dest);
 443         len = strlen_w(src);
 444         if (len > max) len = max;
 445
 446         memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
 447         dest[start+len] = 0;
 448
 449         return dest;
 450 }
 451
 452
 453 /*
 454   The *_wa() functions take a combination of 7 bit ascii
 455   and wide characters They are used so that you can use string
 456   functions combining C string constants with ucs2 strings
 457
 458   The char* arguments must NOT be multibyte - to be completely sure
 459   of this only pass string constants */
 460
 461
 462 void pstrcpy_wa(smb_ucs2_t *dest, const char *src)
 463 {
 464         int i;
 465         for (i=0;i<PSTRING_LEN;i++) {
 466                 dest[i] = UCS2_CHAR(src[i]);
 467                 if (src[i] == 0) return;
 468         }
 469 }
 470
 471 int strcmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
 472 {
 473         while (*b && *a == *b) { a++; b++; }
 474         return (*a - *b);
 475         /* warning: if *a != *b and both are not 0 we retrun a random
 476                 greater or lesser than 0 number not realted to which
 477                 string is longer */
 478 }
 479
 480 int strncmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
 481 {
 482         size_t n = 0;
 483         while ((n < len) && *b && *a == *b) { a++; b++; n++;}
 484         return (len - n)?(*a - *b):0;
 485 }
 486
 487 int strcmp_wa(const smb_ucs2_t *a, const char *b)
 488 {
 489         while (*b && *a == UCS2_CHAR(*b)) { a++; b++; }
 490         return (*a - UCS2_CHAR(*b));
 491 }
 492
 493 int strncmp_wa(const smb_ucs2_t *a, const char *b, size_t len)
 494 {
 495         size_t n = 0;
 496         while ((n < len) && *b && *a == UCS2_CHAR(*b)) { a++; b++; n++;}
 497         return (len - n)?(*a - UCS2_CHAR(*b)):0;
 498 }
 499
 500 smb_ucs2_t *strpbrk_wa(const smb_ucs2_t *s, const char *p)
 501 {
 502         while (*s != 0) {
 503                 int i;
 504                 for (i=0; p[i] && *s != UCS2_CHAR(p[i]); i++)
 505                         ;
 506                 if (p[i]) return (smb_ucs2_t *)s;
 507                 s++;
 508         }
 509         return NULL;
 510 }
 511
 512
 513 /*******************************************************************
 514 copy a string with max len
 515 ********************************************************************/
 516
 517 smb_ucs2_t *strncpy_wa(smb_ucs2_t *dest, const char *src, const size_t max)
 518 {
 519         smb_ucs2_t *ucs2_src;
 520
 521         if (!dest || !src) return NULL;
 522         ucs2_src = (smb_ucs2_t *)malloc((strlen(src)+1)*sizeof(smb_ucs2_t));
 523         if (!ucs2_src) {
 524                 DEBUG(0,("strncpy_wa: out of memory!\n"));
 525                 return NULL;
 526         }
 527         push_ucs2(NULL, ucs2_src, src, -1, STR_TERMINATE|STR_NOALIGN);
 528
 529         strncpy_w(dest, ucs2_src, max);
 530         SAFE_FREE(ucs2_src);
 531         return dest;
 532 }
 533
 534
 535 /*******************************************************************
 536 append a string of len bytes and add a terminator
 537 ********************************************************************/
 538
 539 smb_ucs2_t *strncat_wa(smb_ucs2_t *dest, const char *src, const size_t max)
 540 {
 541         smb_ucs2_t *ucs2_src;
 542
 543         if (!dest || !src) return NULL;
 544         ucs2_src = (smb_ucs2_t *)malloc((strlen(src)+1)*sizeof(smb_ucs2_t));
 545         if (!ucs2_src) {
 546                 DEBUG(0,("strncat_wa: out of memory!\n"));
 547                 return NULL;
 548         }
 549         push_ucs2(NULL, ucs2_src, src, -1, STR_TERMINATE|STR_NOALIGN);
 550
 551         strncat_w(dest, ucs2_src, max);
 552         SAFE_FREE(ucs2_src);
 553         return dest;
 554 }
 555
 556 /*******************************************************************
 557 replace any occurence of oldc with newc in unicode string
 558 ********************************************************************/
 559
 560 void string_replace_w(smb_ucs2_t *s, smb_ucs2_t oldc, smb_ucs2_t newc)
 561 {
 562         for(;*s;s++) {
 563                 if(*s==oldc) *s=newc;
 564         }
 565 }
 566
 567 /*******************************************************************
 568 trim unicode string
 569 ********************************************************************/
 570
 571 BOOL trim_string_w(smb_ucs2_t *s, const smb_ucs2_t *front,
 572                                   const smb_ucs2_t *back)
 573 {
 574         BOOL ret = False;
 575         size_t len, lw, front_len, flw, back_len, blw;
 576
 577         if (!s || !*s) return False;
 578
 579         len = strlen_w(s);
 580
 581         if (front && *front) {
 582                 front_len = strlen_w(front);
 583                 flw = front_len * sizeof(smb_ucs2_t);
 584                 lw = (len + 1) * sizeof(smb_ucs2_t);
 585                 while (len && strncmp_w(s, front, front_len) == 0) {
 586                         memcpy(s, s + flw, lw - flw);
 587                         len -= front_len;
 588                         lw -= flw;
 589                         ret = True;
 590                 }
 591         }
 592
 593         if (back && *back) {
 594                 back_len = strlen_w(back);
 595                 blw = back_len * sizeof(smb_ucs2_t);
 596                 lw = len * sizeof(smb_ucs2_t);
 597                 while (len && strncmp_w(s + lw - blw, back, back_len) == 0) {
 598                         s[len - back_len] = 0;
 599                         len -= back_len;
 600                         lw -= blw;
 601                         ret = True;
 602                 }
 603         }
 604
 605         return ret;
 606 }
 607
 608 BOOL trim_string_wa(smb_ucs2_t *s, const char *front,
 609                                   const char *back)
 610 {
 611         wpstring f, b;
 612
 613         if (front) push_ucs2(NULL, f, front, sizeof(wpstring) - 1, STR_TERMINATE);
 614         else *f = 0;
 615         if (back) push_ucs2(NULL, b, back, sizeof(wpstring) - 1, STR_TERMINATE);
 616         else *b = 0;
 617         return trim_string_w(s, f, b);
 618 }