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 2 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, write to the Free Software
20 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
29 /* these 3 tables define the unicode case handling. They are loaded
30 at startup either via mmap() or read() from the lib directory */
31 static smb_ucs2_t *upcase_table;
32 static smb_ucs2_t *lowcase_table;
33 static uint8 *valid_table;
34 static BOOL upcase_table_use_unmap;
35 static BOOL lowcase_table_use_unmap;
36 static BOOL valid_table_use_unmap;
39 * This table says which Unicode characters are valid dos
42 * Each value is just a single bit.
44 static uint8 doschar_table[8192]; /* 65536 characters / 8 bits/byte */
47 * Destroy global objects allocated by load_case_tables()
49 void gfree_case_tables(void)
52 if ( upcase_table_use_unmap )
53 unmap_file(upcase_table, 0x20000);
55 SAFE_FREE(upcase_table);
58 if ( lowcase_table ) {
59 if ( lowcase_table_use_unmap )
60 unmap_file(lowcase_table, 0x20000);
62 SAFE_FREE(lowcase_table);
66 if ( valid_table_use_unmap )
67 unmap_file(valid_table, 0x10000);
69 SAFE_FREE(valid_table);
74 * Load or generate the case handling tables.
76 * The case tables are defined in UCS2 and don't depend on any
77 * configured parameters, so they never need to be reloaded.
80 void load_case_tables(void)
82 static int initialised;
83 char *old_locale = NULL, *saved_locale = NULL;
91 upcase_table = (smb_ucs2_t *)map_file(lib_path("upcase.dat"),
93 upcase_table_use_unmap = ( upcase_table != NULL );
95 lowcase_table = (smb_ucs2_t *)map_file(lib_path("lowcase.dat"),
97 lowcase_table_use_unmap = ( lowcase_table != NULL );
100 /* Get the name of the current locale. */
101 old_locale = setlocale(LC_ALL, NULL);
104 /* Save it as it is in static storage. */
105 saved_locale = SMB_STRDUP(old_locale);
108 /* We set back the locale to C to get ASCII-compatible toupper/lower functions. */
109 setlocale(LC_ALL, "C");
112 /* we would like Samba to limp along even if these tables are
115 DEBUG(1,("creating lame upcase table\n"));
116 upcase_table = (smb_ucs2_t *)SMB_MALLOC(0x20000);
117 for (i=0;i<0x10000;i++) {
122 for (i=0;i<256;i++) {
124 SSVAL(&v, 0, UCS2_CHAR(i));
125 upcase_table[v] = UCS2_CHAR(islower(i)?toupper(i):i);
129 if (!lowcase_table) {
130 DEBUG(1,("creating lame lowcase table\n"));
131 lowcase_table = (smb_ucs2_t *)SMB_MALLOC(0x20000);
132 for (i=0;i<0x10000;i++) {
135 lowcase_table[v] = i;
137 for (i=0;i<256;i++) {
139 SSVAL(&v, 0, UCS2_CHAR(i));
140 lowcase_table[v] = UCS2_CHAR(isupper(i)?tolower(i):i);
144 #ifdef HAVE_SETLOCALE
145 /* Restore the old locale. */
147 setlocale (LC_ALL, saved_locale);
148 SAFE_FREE(saved_locale);
154 see if a ucs2 character can be mapped correctly to a dos character
155 and mapped back to the same character in ucs2
158 int check_dos_char(smb_ucs2_t c)
160 lazy_initialize_conv();
162 /* Find the right byte, and right bit within the byte; return
164 return (doschar_table[(c & 0xffff) / 8] & (1 << (c & 7))) != 0;
168 static int check_dos_char_slowly(smb_ucs2_t c)
174 len1 = convert_string(CH_UTF16LE, CH_DOS, &c, 2, buf, sizeof(buf),False);
178 len2 = convert_string(CH_DOS, CH_UTF16LE, buf, len1, &c2, 2,False);
187 * Fill out doschar table the hard way, by examining each character
190 void init_doschar_table(void)
194 /* For each byte of packed table */
196 for (i = 0; i <= 0xffff; i += 8) {
198 for (j = 0; j <= 7; j++) {
203 if (check_dos_char_slowly(c)) {
207 doschar_table[i/8] = byteval;
213 * Load the valid character map table from <tt>valid.dat</tt> or
214 * create from the configured codepage.
216 * This function is called whenever the configuration is reloaded.
217 * However, the valid character table is not changed if it's loaded
218 * from a file, because we can't unmap files.
221 void init_valid_table(void)
223 static int mapped_file;
225 const char *allowed = ".!#$%&'()_-@^`~";
229 /* Can't unmap files, so stick with what we have */
233 valid_file = (uint8 *)map_file(lib_path("valid.dat"), 0x10000);
235 valid_table = valid_file;
237 valid_table_use_unmap = True;
241 /* Otherwise, we're using a dynamically created valid_table.
242 * It might need to be regenerated if the code page changed.
243 * We know that we're not using a mapped file, so we can
244 * free() the old one. */
246 SAFE_FREE(valid_table);
248 /* use free rather than unmap */
249 valid_table_use_unmap = False;
251 DEBUG(2,("creating default valid table\n"));
252 valid_table = (uint8 *)SMB_MALLOC(0x10000);
253 for (i=0;i<128;i++) {
254 valid_table[i] = isalnum(i) || strchr(allowed,i);
257 for (;i<0x10000;i++) {
260 valid_table[i] = check_dos_char(c);
264 /*******************************************************************
265 Write a string in (little-endian) unicode format. src is in
266 the current DOS codepage. len is the length in bytes of the
267 string pointed to by dst.
269 if null_terminate is True then null terminate the packet (adds 2 bytes)
271 the return value is the length in bytes consumed by the string, including the
272 null termination if applied
273 ********************************************************************/
275 size_t dos_PutUniCode(char *dst,const char *src, size_t len, BOOL null_terminate)
277 int flags = null_terminate ? STR_UNICODE|STR_NOALIGN|STR_TERMINATE
278 : STR_UNICODE|STR_NOALIGN;
279 return push_ucs2(NULL, dst, src, len, flags);
283 /*******************************************************************
284 Skip past a unicode string, but not more than len. Always move
285 past a terminating zero if found.
286 ********************************************************************/
288 char *skip_unibuf(char *src, size_t len)
290 char *srcend = src + len;
292 while (src < srcend && SVAL(src,0)) {
303 /* Copy a string from little-endian or big-endian unicode source (depending
304 * on flags) to internal samba format destination
307 int rpcstr_pull(char* dest, void *src, int dest_len, int src_len, int flags)
316 return pull_ucs2(NULL, dest, src, dest_len, src_len, flags|STR_UNICODE|STR_NOALIGN);
319 /* Copy a string from a unistr2 source to internal samba format
320 destination. Use this instead of direct calls to rpcstr_pull() to avoid
321 having to determine whether the source string is null terminated. */
323 int rpcstr_pull_unistr2_fstring(char *dest, UNISTR2 *src)
325 return pull_ucs2(NULL, dest, src->buffer, sizeof(fstring),
326 src->uni_str_len * 2, 0);
329 /* Helper function to return a talloc'ed string. I have implemented it with a
330 * copy because I don't really know how pull_ucs2 and friends calculate the
331 * target size. If this turns out to be a major bottleneck someone with deeper
332 * multi-byte knowledge needs to revisit this.
333 * My (VL) use is dsr_getdcname, which returns 6 strings, the alternative would
334 * have been to manually talloc_strdup them in rpc_client/cli_netlogon.c.
337 char *rpcstr_pull_unistr2_talloc(TALLOC_CTX *mem_ctx, const UNISTR2 *src)
342 result = pull_ucs2(NULL, tmp, src->buffer, sizeof(tmp),
343 src->uni_str_len * 2, 0);
344 if (result == (size_t)-1) {
348 return talloc_strdup(mem_ctx, tmp);
351 /* Converts a string from internal samba format to unicode
354 int rpcstr_push(void* dest, const char *src, size_t dest_len, int flags)
356 return push_ucs2(NULL, dest, src, dest_len, flags|STR_UNICODE|STR_NOALIGN);
359 /*******************************************************************
360 Convert a (little-endian) UNISTR2 structure to an ASCII string.
361 ********************************************************************/
363 void unistr2_to_ascii(char *dest, const UNISTR2 *str, size_t maxlen)
369 pull_ucs2(NULL, dest, str->buffer, maxlen, str->uni_str_len*2, STR_NOALIGN);
372 /*******************************************************************
373 Convert a (little-endian) UNISTR3 structure to an ASCII string.
374 ********************************************************************/
376 void unistr3_to_ascii(char *dest, const UNISTR3 *str, size_t maxlen)
382 pull_ucs2(NULL, dest, str->str.buffer, maxlen, str->uni_str_len*2,
386 /*******************************************************************
387 Give a static string for displaying a UNISTR2.
388 ********************************************************************/
390 const char *unistr2_static(const UNISTR2 *str)
393 unistr2_to_ascii(ret, str, sizeof(ret));
397 /*******************************************************************
398 Duplicate a UNISTR2 string into a null terminated char*
399 using a talloc context.
400 ********************************************************************/
402 char *unistr2_tdup(TALLOC_CTX *ctx, const UNISTR2 *str)
405 int maxlen = (str->uni_str_len+1)*4;
409 s = (char *)TALLOC(ctx, maxlen); /* convervative */
413 pull_ucs2(NULL, s, str->buffer, maxlen, str->uni_str_len*2, STR_NOALIGN);
417 /*******************************************************************
418 Convert a wchar to upper case.
419 ********************************************************************/
421 smb_ucs2_t toupper_w(smb_ucs2_t val)
423 return upcase_table[SVAL(&val,0)];
426 /*******************************************************************
427 Convert a wchar to lower case.
428 ********************************************************************/
430 smb_ucs2_t tolower_w( smb_ucs2_t val )
432 return lowcase_table[SVAL(&val,0)];
435 /*******************************************************************
436 Determine if a character is lowercase.
437 ********************************************************************/
439 BOOL islower_w(smb_ucs2_t c)
441 return upcase_table[SVAL(&c,0)] != c;
444 /*******************************************************************
445 Determine if a character is uppercase.
446 ********************************************************************/
448 BOOL isupper_w(smb_ucs2_t c)
450 return lowcase_table[SVAL(&c,0)] != c;
453 /*******************************************************************
454 Determine if a character is valid in a 8.3 name.
455 ********************************************************************/
457 BOOL isvalid83_w(smb_ucs2_t c)
459 return valid_table[SVAL(&c,0)] != 0;
462 /*******************************************************************
463 Count the number of characters in a smb_ucs2_t string.
464 ********************************************************************/
466 size_t strlen_w(const smb_ucs2_t *src)
471 for(len = 0; *(COPY_UCS2_CHAR(&c,src)); src++, len++) {
478 /*******************************************************************
479 Count up to max number of characters in a smb_ucs2_t string.
480 ********************************************************************/
482 size_t strnlen_w(const smb_ucs2_t *src, size_t max)
487 for(len = 0; (len < max) && *(COPY_UCS2_CHAR(&c,src)); src++, len++) {
494 /*******************************************************************
496 ********************************************************************/
498 smb_ucs2_t *strchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
501 while (*(COPY_UCS2_CHAR(&cp,s))) {
503 return (smb_ucs2_t *)s;
508 return (smb_ucs2_t *)s;
514 smb_ucs2_t *strchr_wa(const smb_ucs2_t *s, char c)
516 return strchr_w(s, UCS2_CHAR(c));
519 /*******************************************************************
521 ********************************************************************/
523 smb_ucs2_t *strrchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
526 const smb_ucs2_t *p = s;
527 int len = strlen_w(s);
534 if (c == *(COPY_UCS2_CHAR(&cp,p))) {
535 return (smb_ucs2_t *)p;
541 /*******************************************************************
542 Wide version of strrchr that returns after doing strrchr 'n' times.
543 ********************************************************************/
545 smb_ucs2_t *strnrchr_w(const smb_ucs2_t *s, smb_ucs2_t c, unsigned int n)
548 const smb_ucs2_t *p = s;
549 int len = strlen_w(s);
551 if (len == 0 || !n) {
556 if (c == *(COPY_UCS2_CHAR(&cp,p))) {
561 return (smb_ucs2_t *)p;
567 /*******************************************************************
569 ********************************************************************/
571 smb_ucs2_t *strstr_w(const smb_ucs2_t *s, const smb_ucs2_t *ins)
576 if (!s || !*s || !ins || !*ins) {
580 inslen = strlen_w(ins);
583 while ((r = strchr_w(r, *ins))) {
584 if (strncmp_w(r, ins, inslen) == 0) {
593 /*******************************************************************
594 Convert a string to lower case.
595 return True if any char is converted
596 ********************************************************************/
598 BOOL strlower_w(smb_ucs2_t *s)
603 while (*(COPY_UCS2_CHAR(&cp,s))) {
604 smb_ucs2_t v = tolower_w(cp);
606 COPY_UCS2_CHAR(s,&v);
614 /*******************************************************************
615 Convert a string to upper case.
616 return True if any char is converted
617 ********************************************************************/
619 BOOL strupper_w(smb_ucs2_t *s)
623 while (*(COPY_UCS2_CHAR(&cp,s))) {
624 smb_ucs2_t v = toupper_w(cp);
626 COPY_UCS2_CHAR(s,&v);
634 /*******************************************************************
635 Convert a string to "normal" form.
636 ********************************************************************/
638 void strnorm_w(smb_ucs2_t *s, int case_default)
640 if (case_default == CASE_UPPER) {
647 int strcmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
651 while ((*(COPY_UCS2_CHAR(&cpb,b))) && (*(COPY_UCS2_CHAR(&cpa,a)) == cpb)) {
655 return (*(COPY_UCS2_CHAR(&cpa,a)) - *(COPY_UCS2_CHAR(&cpb,b)));
656 /* warning: if *a != *b and both are not 0 we return a random
657 greater or lesser than 0 number not realted to which
661 int strncmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
666 while ((n < len) && (*(COPY_UCS2_CHAR(&cpb,b))) && (*(COPY_UCS2_CHAR(&cpa,a)) == cpb)) {
671 return (len - n)?(*(COPY_UCS2_CHAR(&cpa,a)) - *(COPY_UCS2_CHAR(&cpb,b))):0;
674 /*******************************************************************
675 Case insensitive string comparison.
676 ********************************************************************/
678 int strcasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
682 while ((*COPY_UCS2_CHAR(&cpb,b)) && toupper_w(*(COPY_UCS2_CHAR(&cpa,a))) == toupper_w(cpb)) {
686 return (tolower_w(*(COPY_UCS2_CHAR(&cpa,a))) - tolower_w(*(COPY_UCS2_CHAR(&cpb,b))));
689 /*******************************************************************
690 Case insensitive string comparison, length limited.
691 ********************************************************************/
693 int strncasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
698 while ((n < len) && *COPY_UCS2_CHAR(&cpb,b) && (toupper_w(*(COPY_UCS2_CHAR(&cpa,a))) == toupper_w(cpb))) {
703 return (len - n)?(tolower_w(*(COPY_UCS2_CHAR(&cpa,a))) - tolower_w(*(COPY_UCS2_CHAR(&cpb,b)))):0;
706 /*******************************************************************
708 ********************************************************************/
710 BOOL strequal_w(const smb_ucs2_t *s1, const smb_ucs2_t *s2)
719 return(strcasecmp_w(s1,s2)==0);
722 /*******************************************************************
723 Compare 2 strings up to and including the nth char.
724 ******************************************************************/
726 BOOL strnequal_w(const smb_ucs2_t *s1,const smb_ucs2_t *s2,size_t n)
731 if (!s1 || !s2 || !n) {
735 return(strncasecmp_w(s1,s2,n)==0);
738 /*******************************************************************
740 ********************************************************************/
742 smb_ucs2_t *strdup_w(const smb_ucs2_t *src)
744 return strndup_w(src, 0);
747 /* if len == 0 then duplicate the whole string */
749 smb_ucs2_t *strndup_w(const smb_ucs2_t *src, size_t len)
756 dest = SMB_MALLOC_ARRAY(smb_ucs2_t, len + 1);
758 DEBUG(0,("strdup_w: out of memory!\n"));
762 memcpy(dest, src, len * sizeof(smb_ucs2_t));
767 /*******************************************************************
768 Copy a string with max len.
769 ********************************************************************/
771 smb_ucs2_t *strncpy_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
780 for (len = 0; (*COPY_UCS2_CHAR(&cp,(src+len))) && (len < max); len++) {
781 cp = *COPY_UCS2_CHAR(dest+len,src+len);
784 for ( /*nothing*/ ; len < max; len++ ) {
785 cp = *COPY_UCS2_CHAR(dest+len,&cp);
791 /*******************************************************************
792 Append a string of len bytes and add a terminator.
793 ********************************************************************/
795 smb_ucs2_t *strncat_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
805 start = strlen_w(dest);
806 len = strnlen_w(src, max);
808 memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
809 z = *COPY_UCS2_CHAR(dest+start+len,&z);
814 smb_ucs2_t *strcat_w(smb_ucs2_t *dest, const smb_ucs2_t *src)
824 start = strlen_w(dest);
827 memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
828 z = *COPY_UCS2_CHAR(dest+start+len,&z);
834 /*******************************************************************
835 Replace any occurence of oldc with newc in unicode string.
836 ********************************************************************/
838 void string_replace_w(smb_ucs2_t *s, smb_ucs2_t oldc, smb_ucs2_t newc)
842 for(;*(COPY_UCS2_CHAR(&cp,s));s++) {
844 COPY_UCS2_CHAR(s,&newc);
849 /*******************************************************************
851 ********************************************************************/
853 BOOL trim_string_w(smb_ucs2_t *s, const smb_ucs2_t *front,
854 const smb_ucs2_t *back)
857 size_t len, front_len, back_len;
865 if (front && *front) {
866 front_len = strlen_w(front);
867 while (len && strncmp_w(s, front, front_len) == 0) {
868 memmove(s, (s + front_len), (len - front_len + 1) * sizeof(smb_ucs2_t));
875 back_len = strlen_w(back);
876 while (len && strncmp_w((s + (len - back_len)), back, back_len) == 0) {
877 s[len - back_len] = 0;
887 The *_wa() functions take a combination of 7 bit ascii
888 and wide characters They are used so that you can use string
889 functions combining C string constants with ucs2 strings
891 The char* arguments must NOT be multibyte - to be completely sure
892 of this only pass string constants */
894 int strcmp_wa(const smb_ucs2_t *a, const char *b)
898 while (*b && *(COPY_UCS2_CHAR(&cp,a)) == UCS2_CHAR(*b)) {
902 return (*(COPY_UCS2_CHAR(&cp,a)) - UCS2_CHAR(*b));
905 int strncmp_wa(const smb_ucs2_t *a, const char *b, size_t len)
910 while ((n < len) && *b && *(COPY_UCS2_CHAR(&cp,a)) == UCS2_CHAR(*b)) {
915 return (len - n)?(*(COPY_UCS2_CHAR(&cp,a)) - UCS2_CHAR(*b)):0;
918 smb_ucs2_t *strpbrk_wa(const smb_ucs2_t *s, const char *p)
922 while (*(COPY_UCS2_CHAR(&cp,s))) {
924 for (i=0; p[i] && cp != UCS2_CHAR(p[i]); i++)
927 return (smb_ucs2_t *)s;
934 smb_ucs2_t *strstr_wa(const smb_ucs2_t *s, const char *ins)
943 inslen = strlen(ins);
946 while ((r = strchr_w(r, UCS2_CHAR(*ins)))) {
947 if (strncmp_wa(r, ins, inslen) == 0)
955 BOOL trim_string_wa(smb_ucs2_t *s, const char *front,
961 push_ucs2(NULL, f, front, sizeof(wpstring) - 1, STR_TERMINATE);
966 push_ucs2(NULL, b, back, sizeof(wpstring) - 1, STR_TERMINATE);
970 return trim_string_w(s, f, b);
973 /*******************************************************************
974 Returns the length in number of wide characters.
975 ******************************************************************/
977 int unistrlen(uint16 *s)
985 for (len=0; SVAL(s,0); s++,len++) {
992 /*******************************************************************
993 Strcpy for unicode strings. Returns length (in num of wide chars).
995 ********************************************************************/
997 int unistrcpy(uint16 *dst, uint16 *src)
1001 while (SVAL(src,0)) {
1011 * Samba ucs2 type to UNISTR2 conversion
1013 * @param ctx Talloc context to create the dst strcture (if null) and the
1014 * contents of the unicode string.
1015 * @param dst UNISTR2 destination. If equals null, then it's allocated.
1016 * @param src smb_ucs2_t source.
1017 * @param max_len maximum number of unicode characters to copy. If equals
1018 * null, then null-termination of src is taken
1020 * @return copied UNISTR2 destination
1023 UNISTR2* ucs2_to_unistr2(TALLOC_CTX *ctx, UNISTR2* dst, smb_ucs2_t* src)
1031 len = strlen_w(src);
1033 /* allocate UNISTR2 destination if not given */
1035 dst = TALLOC_P(ctx, UNISTR2);
1040 dst->buffer = TALLOC_ARRAY(ctx, uint16, len + 1);
1045 /* set UNISTR2 parameters */
1046 dst->uni_max_len = len + 1;
1048 dst->uni_str_len = len;
1050 /* copy the actual unicode string */
1051 strncpy_w(dst->buffer, src, dst->uni_max_len);
1056 /*************************************************************
1057 ascii only toupper - saves the need for smbd to be in C locale.
1058 *************************************************************/
1060 int toupper_ascii(int c)
1062 smb_ucs2_t uc = toupper_w(UCS2_CHAR(c));
1063 return UCS2_TO_CHAR(uc);
1066 /*************************************************************
1067 ascii only tolower - saves the need for smbd to be in C locale.
1068 *************************************************************/
1070 int tolower_ascii(int c)
1072 smb_ucs2_t uc = tolower_w(UCS2_CHAR(c));
1073 return UCS2_TO_CHAR(uc);
1076 /*************************************************************
1077 ascii only isupper - saves the need for smbd to be in C locale.
1078 *************************************************************/
1080 int isupper_ascii(int c)
1082 return isupper_w(UCS2_CHAR(c));
1085 /*************************************************************
1086 ascii only islower - saves the need for smbd to be in C locale.
1087 *************************************************************/
1089 int islower_ascii(int c)
1091 return islower_w(UCS2_CHAR(c));