2 Unix SMB/CIFS implementation.
3 Samba utility functions
4 Copyright (C) Andrew Tridgell 1992-2001
5 Copyright (C) Simo Sorce 2001
6 Copyright (C) Jeremy Allison 2005
8 This program is free software; you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation; either version 3 of the License, or
11 (at your option) any later version.
13 This program is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
18 You should have received a copy of the GNU General Public License
19 along with this program. If not, see <http://www.gnu.org/licenses/>.
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 static bool upcase_table_use_unmap;
34 static bool lowcase_table_use_unmap;
35 static bool valid_table_use_unmap;
36 static bool initialized;
39 * Destroy global objects allocated by load_case_tables()
41 void gfree_case_tables(void)
44 if ( upcase_table_use_unmap )
45 unmap_file(upcase_table, 0x20000);
47 SAFE_FREE(upcase_table);
51 if ( lowcase_table ) {
52 if ( lowcase_table_use_unmap )
53 unmap_file(lowcase_table, 0x20000);
55 SAFE_FREE(lowcase_table);
60 if ( valid_table_use_unmap )
61 unmap_file(valid_table, 0x10000);
63 SAFE_FREE(valid_table);
70 * Load or generate the case handling tables.
72 * The case tables are defined in UCS2 and don't depend on any
73 * configured parameters, so they never need to be reloaded.
76 void load_case_tables(void)
78 char *old_locale = NULL, *saved_locale = NULL;
80 TALLOC_CTX *frame = NULL;
87 frame = talloc_stackframe();
89 upcase_table = (smb_ucs2_t *)map_file(data_path("upcase.dat"),
91 upcase_table_use_unmap = ( upcase_table != NULL );
93 lowcase_table = (smb_ucs2_t *)map_file(data_path("lowcase.dat"),
95 lowcase_table_use_unmap = ( lowcase_table != NULL );
98 /* Get the name of the current locale. */
99 old_locale = setlocale(LC_ALL, NULL);
102 /* Save it as it is in static storage. */
103 saved_locale = SMB_STRDUP(old_locale);
106 /* We set back the locale to C to get ASCII-compatible toupper/lower functions. */
107 setlocale(LC_ALL, "C");
110 /* we would like Samba to limp along even if these tables are
113 DEBUG(1,("creating lame upcase table\n"));
114 upcase_table = (smb_ucs2_t *)SMB_MALLOC(0x20000);
115 for (i=0;i<0x10000;i++) {
120 for (i=0;i<256;i++) {
122 SSVAL(&v, 0, UCS2_CHAR(i));
123 upcase_table[v] = UCS2_CHAR(islower(i)?toupper(i):i);
127 if (!lowcase_table) {
128 DEBUG(1,("creating lame lowcase table\n"));
129 lowcase_table = (smb_ucs2_t *)SMB_MALLOC(0x20000);
130 for (i=0;i<0x10000;i++) {
133 lowcase_table[v] = i;
135 for (i=0;i<256;i++) {
137 SSVAL(&v, 0, UCS2_CHAR(i));
138 lowcase_table[v] = UCS2_CHAR(isupper(i)?tolower(i):i);
142 #ifdef HAVE_SETLOCALE
143 /* Restore the old locale. */
145 setlocale (LC_ALL, saved_locale);
146 SAFE_FREE(saved_locale);
152 static int check_dos_char_slowly(smb_ucs2_t c)
158 len1 = convert_string(CH_UTF16LE, CH_DOS, &c, 2, buf, sizeof(buf),False);
162 len2 = convert_string(CH_DOS, CH_UTF16LE, buf, len1, &c2, 2,False);
170 * Load the valid character map table from <tt>valid.dat</tt> or
171 * create from the configured codepage.
173 * This function is called whenever the configuration is reloaded.
174 * However, the valid character table is not changed if it's loaded
175 * from a file, because we can't unmap files.
178 void init_valid_table(void)
180 static int mapped_file;
182 const char *allowed = ".!#$%&'()_-@^`~";
186 /* Can't unmap files, so stick with what we have */
190 valid_file = (uint8 *)map_file(data_path("valid.dat"), 0x10000);
192 valid_table = valid_file;
194 valid_table_use_unmap = True;
198 /* Otherwise, we're using a dynamically created valid_table.
199 * It might need to be regenerated if the code page changed.
200 * We know that we're not using a mapped file, so we can
201 * free() the old one. */
202 SAFE_FREE(valid_table);
204 /* use free rather than unmap */
205 valid_table_use_unmap = False;
207 DEBUG(2,("creating default valid table\n"));
208 valid_table = (uint8 *)SMB_MALLOC(0x10000);
209 SMB_ASSERT(valid_table != NULL);
210 for (i=0;i<128;i++) {
211 valid_table[i] = isalnum(i) || strchr(allowed,i);
214 lazy_initialize_conv();
216 for (;i<0x10000;i++) {
219 valid_table[i] = check_dos_char_slowly(c);
223 /*******************************************************************
224 Write a string in (little-endian) unicode format. src is in
225 the current DOS codepage. len is the length in bytes of the
226 string pointed to by dst.
228 if null_terminate is True then null terminate the packet (adds 2 bytes)
230 the return value is the length in bytes consumed by the string, including the
231 null termination if applied
232 ********************************************************************/
234 size_t dos_PutUniCode(char *dst,const char *src, size_t len, bool null_terminate)
236 int flags = null_terminate ? STR_UNICODE|STR_NOALIGN|STR_TERMINATE
237 : STR_UNICODE|STR_NOALIGN;
238 return push_ucs2(NULL, dst, src, len, flags);
242 /*******************************************************************
243 Skip past a unicode string, but not more than len. Always move
244 past a terminating zero if found.
245 ********************************************************************/
247 char *skip_unibuf(char *src, size_t len)
249 char *srcend = src + len;
251 while (src < srcend && SVAL(src,0)) {
262 /* Converts a string from internal samba format to unicode
265 int rpcstr_push(void *dest, const char *src, size_t dest_len, int flags)
267 return push_ucs2(NULL, dest, src, dest_len, flags|STR_UNICODE|STR_NOALIGN);
270 /* Converts a string from internal samba format to unicode. Always terminates.
271 * Actually just a wrapper round push_ucs2_talloc().
274 int rpcstr_push_talloc(TALLOC_CTX *ctx, smb_ucs2_t **dest, const char *src)
277 if (push_ucs2_talloc(ctx, dest, src, &size))
283 /*******************************************************************
284 Convert a wchar to upper case.
285 ********************************************************************/
287 smb_ucs2_t toupper_w(smb_ucs2_t val)
289 return upcase_table[SVAL(&val,0)];
292 /*******************************************************************
293 Convert a wchar to lower case.
294 ********************************************************************/
296 smb_ucs2_t tolower_w( smb_ucs2_t val )
298 return lowcase_table[SVAL(&val,0)];
301 /*******************************************************************
302 Determine if a character is lowercase.
303 ********************************************************************/
305 bool islower_w(smb_ucs2_t c)
307 return upcase_table[SVAL(&c,0)] != c;
310 /*******************************************************************
311 Determine if a character is uppercase.
312 ********************************************************************/
314 bool isupper_w(smb_ucs2_t c)
316 return lowcase_table[SVAL(&c,0)] != c;
319 /*******************************************************************
320 Determine if a character is valid in a 8.3 name.
321 ********************************************************************/
323 bool isvalid83_w(smb_ucs2_t c)
325 return valid_table[SVAL(&c,0)] != 0;
328 /*******************************************************************
329 Count the number of characters in a smb_ucs2_t string.
330 ********************************************************************/
332 size_t strlen_w(const smb_ucs2_t *src)
337 for(len = 0; *(COPY_UCS2_CHAR(&c,src)); src++, len++) {
344 /*******************************************************************
345 Count up to max number of characters in a smb_ucs2_t string.
346 ********************************************************************/
348 size_t strnlen_w(const smb_ucs2_t *src, size_t max)
353 for(len = 0; (len < max) && *(COPY_UCS2_CHAR(&c,src)); src++, len++) {
360 /*******************************************************************
362 ********************************************************************/
364 smb_ucs2_t *strchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
367 while (*(COPY_UCS2_CHAR(&cp,s))) {
369 return (smb_ucs2_t *)s;
374 return (smb_ucs2_t *)s;
380 smb_ucs2_t *strchr_wa(const smb_ucs2_t *s, char c)
382 return strchr_w(s, UCS2_CHAR(c));
385 /*******************************************************************
387 ********************************************************************/
389 smb_ucs2_t *strrchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
392 const smb_ucs2_t *p = s;
393 int len = strlen_w(s);
400 if (c == *(COPY_UCS2_CHAR(&cp,p))) {
401 return (smb_ucs2_t *)p;
407 /*******************************************************************
408 Wide version of strrchr that returns after doing strrchr 'n' times.
409 ********************************************************************/
411 smb_ucs2_t *strnrchr_w(const smb_ucs2_t *s, smb_ucs2_t c, unsigned int n)
414 const smb_ucs2_t *p = s;
415 int len = strlen_w(s);
417 if (len == 0 || !n) {
422 if (c == *(COPY_UCS2_CHAR(&cp,p))) {
427 return (smb_ucs2_t *)p;
433 /*******************************************************************
435 ********************************************************************/
437 smb_ucs2_t *strstr_w(const smb_ucs2_t *s, const smb_ucs2_t *ins)
442 if (!s || !*s || !ins || !*ins) {
446 inslen = strlen_w(ins);
449 while ((r = strchr_w(r, *ins))) {
450 if (strncmp_w(r, ins, inslen) == 0) {
459 /*******************************************************************
460 Convert a string to lower case.
461 return True if any char is converted
462 ********************************************************************/
464 bool strlower_w(smb_ucs2_t *s)
469 while (*(COPY_UCS2_CHAR(&cp,s))) {
470 smb_ucs2_t v = tolower_w(cp);
472 COPY_UCS2_CHAR(s,&v);
480 /*******************************************************************
481 Convert a string to upper case.
482 return True if any char is converted
483 ********************************************************************/
485 bool strupper_w(smb_ucs2_t *s)
489 while (*(COPY_UCS2_CHAR(&cp,s))) {
490 smb_ucs2_t v = toupper_w(cp);
492 COPY_UCS2_CHAR(s,&v);
500 /*******************************************************************
501 Convert a string to "normal" form.
502 ********************************************************************/
504 void strnorm_w(smb_ucs2_t *s, int case_default)
506 if (case_default == CASE_UPPER) {
513 int strcmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
517 while ((*(COPY_UCS2_CHAR(&cpb,b))) && (*(COPY_UCS2_CHAR(&cpa,a)) == cpb)) {
521 return (*(COPY_UCS2_CHAR(&cpa,a)) - *(COPY_UCS2_CHAR(&cpb,b)));
522 /* warning: if *a != *b and both are not 0 we return a random
523 greater or lesser than 0 number not realted to which
527 int strncmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
532 while ((n < len) && (*(COPY_UCS2_CHAR(&cpb,b))) && (*(COPY_UCS2_CHAR(&cpa,a)) == cpb)) {
537 return (len - n)?(*(COPY_UCS2_CHAR(&cpa,a)) - *(COPY_UCS2_CHAR(&cpb,b))):0;
540 /*******************************************************************
541 Case insensitive string comparison.
542 ********************************************************************/
544 int strcasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
548 while ((*COPY_UCS2_CHAR(&cpb,b)) && toupper_w(*(COPY_UCS2_CHAR(&cpa,a))) == toupper_w(cpb)) {
552 return (tolower_w(*(COPY_UCS2_CHAR(&cpa,a))) - tolower_w(*(COPY_UCS2_CHAR(&cpb,b))));
555 /*******************************************************************
556 Case insensitive string comparison, length limited.
557 ********************************************************************/
559 int strncasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
564 while ((n < len) && *COPY_UCS2_CHAR(&cpb,b) && (toupper_w(*(COPY_UCS2_CHAR(&cpa,a))) == toupper_w(cpb))) {
569 return (len - n)?(tolower_w(*(COPY_UCS2_CHAR(&cpa,a))) - tolower_w(*(COPY_UCS2_CHAR(&cpb,b)))):0;
572 /*******************************************************************
574 ********************************************************************/
576 bool strequal_w(const smb_ucs2_t *s1, const smb_ucs2_t *s2)
585 return(strcasecmp_w(s1,s2)==0);
588 /*******************************************************************
589 Compare 2 strings up to and including the nth char.
590 ******************************************************************/
592 bool strnequal_w(const smb_ucs2_t *s1,const smb_ucs2_t *s2,size_t n)
597 if (!s1 || !s2 || !n) {
601 return(strncasecmp_w(s1,s2,n)==0);
604 /*******************************************************************
606 ********************************************************************/
608 smb_ucs2_t *strdup_w(const smb_ucs2_t *src)
610 return strndup_w(src, 0);
613 /* if len == 0 then duplicate the whole string */
615 smb_ucs2_t *strndup_w(const smb_ucs2_t *src, size_t len)
622 dest = SMB_MALLOC_ARRAY(smb_ucs2_t, len + 1);
624 DEBUG(0,("strdup_w: out of memory!\n"));
628 memcpy(dest, src, len * sizeof(smb_ucs2_t));
633 /*******************************************************************
634 Copy a string with max len.
635 ********************************************************************/
637 smb_ucs2_t *strncpy_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
646 for (len = 0; (*COPY_UCS2_CHAR(&cp,(src+len))) && (len < max); len++) {
647 cp = *COPY_UCS2_CHAR(dest+len,src+len);
650 for ( /*nothing*/ ; len < max; len++ ) {
651 cp = *COPY_UCS2_CHAR(dest+len,&cp);
657 /*******************************************************************
658 Append a string of len bytes and add a terminator.
659 ********************************************************************/
661 smb_ucs2_t *strncat_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
671 start = strlen_w(dest);
672 len = strnlen_w(src, max);
674 memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
675 z = *COPY_UCS2_CHAR(dest+start+len,&z);
680 smb_ucs2_t *strcat_w(smb_ucs2_t *dest, const smb_ucs2_t *src)
690 start = strlen_w(dest);
693 memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
694 z = *COPY_UCS2_CHAR(dest+start+len,&z);
700 /*******************************************************************
701 Replace any occurence of oldc with newc in unicode string.
702 ********************************************************************/
704 void string_replace_w(smb_ucs2_t *s, smb_ucs2_t oldc, smb_ucs2_t newc)
708 for(;*(COPY_UCS2_CHAR(&cp,s));s++) {
710 COPY_UCS2_CHAR(s,&newc);
715 /*******************************************************************
717 ********************************************************************/
719 bool trim_string_w(smb_ucs2_t *s, const smb_ucs2_t *front,
720 const smb_ucs2_t *back)
723 size_t len, front_len, back_len;
731 if (front && *front) {
732 front_len = strlen_w(front);
733 while (len && strncmp_w(s, front, front_len) == 0) {
734 memmove(s, (s + front_len), (len - front_len + 1) * sizeof(smb_ucs2_t));
741 back_len = strlen_w(back);
742 while (len && strncmp_w((s + (len - back_len)), back, back_len) == 0) {
743 s[len - back_len] = 0;
753 The *_wa() functions take a combination of 7 bit ascii
754 and wide characters They are used so that you can use string
755 functions combining C string constants with ucs2 strings
757 The char* arguments must NOT be multibyte - to be completely sure
758 of this only pass string constants */
760 int strcmp_wa(const smb_ucs2_t *a, const char *b)
764 while (*b && *(COPY_UCS2_CHAR(&cp,a)) == UCS2_CHAR(*b)) {
768 return (*(COPY_UCS2_CHAR(&cp,a)) - UCS2_CHAR(*b));
771 int strncmp_wa(const smb_ucs2_t *a, const char *b, size_t len)
776 while ((n < len) && *b && *(COPY_UCS2_CHAR(&cp,a)) == UCS2_CHAR(*b)) {
781 return (len - n)?(*(COPY_UCS2_CHAR(&cp,a)) - UCS2_CHAR(*b)):0;
784 smb_ucs2_t *strpbrk_wa(const smb_ucs2_t *s, const char *p)
788 while (*(COPY_UCS2_CHAR(&cp,s))) {
790 for (i=0; p[i] && cp != UCS2_CHAR(p[i]); i++)
793 return (smb_ucs2_t *)s;
800 smb_ucs2_t *strstr_wa(const smb_ucs2_t *s, const char *ins)
809 inslen = strlen(ins);
812 while ((r = strchr_w(r, UCS2_CHAR(*ins)))) {
813 if (strncmp_wa(r, ins, inslen) == 0)
821 /*************************************************************
822 ascii only toupper - saves the need for smbd to be in C locale.
823 *************************************************************/
825 int toupper_ascii(int c)
827 smb_ucs2_t uc = toupper_w(UCS2_CHAR(c));
828 return UCS2_TO_CHAR(uc);
831 /*************************************************************
832 ascii only tolower - saves the need for smbd to be in C locale.
833 *************************************************************/
835 int tolower_ascii(int c)
837 smb_ucs2_t uc = tolower_w(UCS2_CHAR(c));
838 return UCS2_TO_CHAR(uc);
841 /*************************************************************
842 ascii only isupper - saves the need for smbd to be in C locale.
843 *************************************************************/
845 int isupper_ascii(int c)
847 return isupper_w(UCS2_CHAR(c));
850 /*************************************************************
851 ascii only islower - saves the need for smbd to be in C locale.
852 *************************************************************/
854 int islower_ascii(int c)
856 return islower_w(UCS2_CHAR(c));