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;
36 * This table says which Unicode characters are valid dos
39 * Each value is just a single bit.
41 static uint8 doschar_table[8192]; /* 65536 characters / 8 bits/byte */
45 * Load or generate the case handling tables.
47 * The case tables are defined in UCS2 and don't depend on any
48 * configured parameters, so they never need to be reloaded.
51 void load_case_tables(void)
53 static int initialised;
61 upcase_table = map_file(lib_path("upcase.dat"), 0x20000);
62 lowcase_table = map_file(lib_path("lowcase.dat"), 0x20000);
64 /* we would like Samba to limp along even if these tables are
67 DEBUG(1,("creating lame upcase table\n"));
68 upcase_table = SMB_MALLOC(0x20000);
69 for (i=0;i<0x10000;i++) {
76 SSVAL(&v, 0, UCS2_CHAR(i));
77 upcase_table[v] = UCS2_CHAR(islower(i)?toupper(i):i);
82 DEBUG(1,("creating lame lowcase table\n"));
83 lowcase_table = SMB_MALLOC(0x20000);
84 for (i=0;i<0x10000;i++) {
91 SSVAL(&v, 0, UCS2_CHAR(i));
92 lowcase_table[v] = UCS2_CHAR(isupper(i)?tolower(i):i);
98 see if a ucs2 character can be mapped correctly to a dos character
99 and mapped back to the same character in ucs2
102 int check_dos_char(smb_ucs2_t c)
104 lazy_initialize_conv();
106 /* Find the right byte, and right bit within the byte; return
108 return (doschar_table[(c & 0xffff) / 8] & (1 << (c & 7))) != 0;
112 static int check_dos_char_slowly(smb_ucs2_t c)
118 len1 = convert_string(CH_UCS2, CH_DOS, &c, 2, buf, sizeof(buf),False);
122 len2 = convert_string(CH_DOS, CH_UCS2, buf, len1, &c2, 2,False);
131 * Fill out doschar table the hard way, by examining each character
134 void init_doschar_table(void)
138 /* For each byte of packed table */
140 for (i = 0; i <= 0xffff; i += 8) {
142 for (j = 0; j <= 7; j++) {
147 if (check_dos_char_slowly(c)) {
151 doschar_table[i/8] = byteval;
157 * Load the valid character map table from <tt>valid.dat</tt> or
158 * create from the configured codepage.
160 * This function is called whenever the configuration is reloaded.
161 * However, the valid character table is not changed if it's loaded
162 * from a file, because we can't unmap files.
165 void init_valid_table(void)
167 static int mapped_file;
169 const char *allowed = ".!#$%&'()_-@^`~";
173 /* Can't unmap files, so stick with what we have */
177 valid_file = map_file(lib_path("valid.dat"), 0x10000);
179 valid_table = valid_file;
184 /* Otherwise, we're using a dynamically created valid_table.
185 * It might need to be regenerated if the code page changed.
186 * We know that we're not using a mapped file, so we can
187 * free() the old one. */
188 if (valid_table) free(valid_table);
190 DEBUG(2,("creating default valid table\n"));
191 valid_table = SMB_MALLOC(0x10000);
192 for (i=0;i<128;i++) {
193 valid_table[i] = isalnum(i) || strchr(allowed,i);
196 for (;i<0x10000;i++) {
199 valid_table[i] = check_dos_char(c);
203 /*******************************************************************
204 Write a string in (little-endian) unicode format. src is in
205 the current DOS codepage. len is the length in bytes of the
206 string pointed to by dst.
208 if null_terminate is True then null terminate the packet (adds 2 bytes)
210 the return value is the length in bytes consumed by the string, including the
211 null termination if applied
212 ********************************************************************/
214 size_t dos_PutUniCode(char *dst,const char *src, ssize_t len, BOOL null_terminate)
216 return push_ucs2(NULL, dst, src, len,
217 STR_UNICODE|STR_NOALIGN | (null_terminate?STR_TERMINATE:0));
221 /*******************************************************************
222 Skip past a unicode string, but not more than len. Always move
223 past a terminating zero if found.
224 ********************************************************************/
226 char *skip_unibuf(char *src, size_t len)
228 char *srcend = src + len;
230 while (src < srcend && SVAL(src,0)) {
241 /* Copy a string from little-endian or big-endian unicode source (depending
242 * on flags) to internal samba format destination
245 int rpcstr_pull(char* dest, void *src, int dest_len, int src_len, int flags)
254 return pull_ucs2(NULL, dest, src, dest_len, src_len, flags|STR_UNICODE|STR_NOALIGN);
257 /* Copy a string from a unistr2 source to internal samba format
258 destination. Use this instead of direct calls to rpcstr_pull() to avoid
259 having to determine whether the source string is null terminated. */
261 int rpcstr_pull_unistr2_fstring(char *dest, UNISTR2 *src)
263 return pull_ucs2(NULL, dest, src->buffer, sizeof(fstring),
264 src->uni_str_len * 2, 0);
267 /* Converts a string from internal samba format to unicode
270 int rpcstr_push(void* dest, const char *src, int dest_len, int flags)
272 return push_ucs2(NULL, dest, src, dest_len, flags|STR_UNICODE|STR_NOALIGN);
275 /*******************************************************************
276 Convert a (little-endian) UNISTR2 structure to an ASCII string.
277 ********************************************************************/
279 void unistr2_to_ascii(char *dest, const UNISTR2 *str, size_t maxlen)
285 pull_ucs2(NULL, dest, str->buffer, maxlen, str->uni_str_len*2, STR_NOALIGN);
288 /*******************************************************************
289 Convert a (little-endian) UNISTR3 structure to an ASCII string.
290 ********************************************************************/
292 void unistr3_to_ascii(char *dest, const UNISTR3 *str, size_t maxlen)
298 pull_ucs2(NULL, dest, str->str.buffer, maxlen, str->uni_str_len*2,
302 /*******************************************************************
303 Give a static string for displaying a UNISTR2.
304 ********************************************************************/
306 const char *unistr2_static(const UNISTR2 *str)
309 unistr2_to_ascii(ret, str, sizeof(ret));
313 /*******************************************************************
314 Duplicate a UNISTR2 string into a null terminated char*
315 using a talloc context.
316 ********************************************************************/
318 char *unistr2_tdup(TALLOC_CTX *ctx, const UNISTR2 *str)
321 int maxlen = (str->uni_str_len+1)*4;
325 s = (char *)TALLOC(ctx, maxlen); /* convervative */
329 pull_ucs2(NULL, s, str->buffer, maxlen, str->uni_str_len*2, STR_NOALIGN);
333 /*******************************************************************
334 Convert a wchar to upper case.
335 ********************************************************************/
337 smb_ucs2_t toupper_w(smb_ucs2_t val)
339 return upcase_table[SVAL(&val,0)];
342 /*******************************************************************
343 Convert a wchar to lower case.
344 ********************************************************************/
346 smb_ucs2_t tolower_w( smb_ucs2_t val )
348 return lowcase_table[SVAL(&val,0)];
351 /*******************************************************************
352 Determine if a character is lowercase.
353 ********************************************************************/
355 BOOL islower_w(smb_ucs2_t c)
357 return upcase_table[SVAL(&c,0)] != c;
360 /*******************************************************************
361 Determine if a character is uppercase.
362 ********************************************************************/
364 BOOL isupper_w(smb_ucs2_t c)
366 return lowcase_table[SVAL(&c,0)] != c;
369 /*******************************************************************
370 Determine if a character is valid in a 8.3 name.
371 ********************************************************************/
373 BOOL isvalid83_w(smb_ucs2_t c)
375 return valid_table[SVAL(&c,0)] != 0;
378 /*******************************************************************
379 Count the number of characters in a smb_ucs2_t string.
380 ********************************************************************/
382 size_t strlen_w(const smb_ucs2_t *src)
387 for(len = 0; *(COPY_UCS2_CHAR(&c,src)); src++, len++) {
394 /*******************************************************************
395 Count up to max number of characters in a smb_ucs2_t string.
396 ********************************************************************/
398 size_t strnlen_w(const smb_ucs2_t *src, size_t max)
403 for(len = 0; *(COPY_UCS2_CHAR(&c,src)) && (len < max); src++, len++) {
410 /*******************************************************************
412 ********************************************************************/
414 smb_ucs2_t *strchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
417 while (*(COPY_UCS2_CHAR(&cp,s))) {
419 return (smb_ucs2_t *)s;
424 return (smb_ucs2_t *)s;
430 smb_ucs2_t *strchr_wa(const smb_ucs2_t *s, char c)
432 return strchr_w(s, UCS2_CHAR(c));
435 /*******************************************************************
437 ********************************************************************/
439 smb_ucs2_t *strrchr_w(const smb_ucs2_t *s, smb_ucs2_t c)
442 const smb_ucs2_t *p = s;
443 int len = strlen_w(s);
450 if (c == *(COPY_UCS2_CHAR(&cp,p))) {
451 return (smb_ucs2_t *)p;
457 /*******************************************************************
458 Wide version of strrchr that returns after doing strrchr 'n' times.
459 ********************************************************************/
461 smb_ucs2_t *strnrchr_w(const smb_ucs2_t *s, smb_ucs2_t c, unsigned int n)
464 const smb_ucs2_t *p = s;
465 int len = strlen_w(s);
467 if (len == 0 || !n) {
472 if (c == *(COPY_UCS2_CHAR(&cp,p))) {
477 return (smb_ucs2_t *)p;
483 /*******************************************************************
485 ********************************************************************/
487 smb_ucs2_t *strstr_w(const smb_ucs2_t *s, const smb_ucs2_t *ins)
492 if (!s || !*s || !ins || !*ins) {
496 inslen = strlen_w(ins);
499 while ((r = strchr_w(r, *ins))) {
500 if (strncmp_w(r, ins, inslen) == 0) {
509 /*******************************************************************
510 Convert a string to lower case.
511 return True if any char is converted
512 ********************************************************************/
514 BOOL strlower_w(smb_ucs2_t *s)
519 while (*(COPY_UCS2_CHAR(&cp,s))) {
520 smb_ucs2_t v = tolower_w(cp);
522 COPY_UCS2_CHAR(s,&v);
530 /*******************************************************************
531 Convert a string to upper case.
532 return True if any char is converted
533 ********************************************************************/
535 BOOL strupper_w(smb_ucs2_t *s)
539 while (*(COPY_UCS2_CHAR(&cp,s))) {
540 smb_ucs2_t v = toupper_w(cp);
542 COPY_UCS2_CHAR(s,&v);
550 /*******************************************************************
551 Convert a string to "normal" form.
552 ********************************************************************/
554 void strnorm_w(smb_ucs2_t *s, int case_default)
556 if (case_default == CASE_UPPER) {
563 int strcmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
567 while ((*(COPY_UCS2_CHAR(&cpb,b))) && (*(COPY_UCS2_CHAR(&cpa,a)) == cpb)) {
572 /* warning: if *a != *b and both are not 0 we return a random
573 greater or lesser than 0 number not realted to which
577 int strncmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
582 while ((n < len) && (*(COPY_UCS2_CHAR(&cpb,b))) && (*(COPY_UCS2_CHAR(&cpa,a)) == cpb)) {
587 return (len - n)?(cpa - cpb):0;
590 /*******************************************************************
591 Case insensitive string comparison.
592 ********************************************************************/
594 int strcasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b)
598 while ((*COPY_UCS2_CHAR(&cpb,b)) && toupper_w(*(COPY_UCS2_CHAR(&cpa,a))) == toupper_w(cpb)) {
602 return (tolower_w(cpa) - tolower_w(cpb));
605 /*******************************************************************
606 Case insensitive string comparison, length limited.
607 ********************************************************************/
609 int strncasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len)
614 while ((n < len) && *COPY_UCS2_CHAR(&cpb,b) && (toupper_w(*(COPY_UCS2_CHAR(&cpa,a))) == toupper_w(cpb))) {
619 return (len - n)?(tolower_w(cpa) - tolower_w(cpb)):0;
622 /*******************************************************************
624 ********************************************************************/
626 BOOL strequal_w(const smb_ucs2_t *s1, const smb_ucs2_t *s2)
635 return(strcasecmp_w(s1,s2)==0);
638 /*******************************************************************
639 Compare 2 strings up to and including the nth char.
640 ******************************************************************/
642 BOOL strnequal_w(const smb_ucs2_t *s1,const smb_ucs2_t *s2,size_t n)
647 if (!s1 || !s2 || !n) {
651 return(strncasecmp_w(s1,s2,n)==0);
654 /*******************************************************************
656 ********************************************************************/
658 smb_ucs2_t *strdup_w(const smb_ucs2_t *src)
660 return strndup_w(src, 0);
663 /* if len == 0 then duplicate the whole string */
665 smb_ucs2_t *strndup_w(const smb_ucs2_t *src, size_t len)
672 dest = SMB_MALLOC_ARRAY(smb_ucs2_t, len + 1);
674 DEBUG(0,("strdup_w: out of memory!\n"));
678 memcpy(dest, src, len * sizeof(smb_ucs2_t));
683 /*******************************************************************
684 Copy a string with max len.
685 ********************************************************************/
687 smb_ucs2_t *strncpy_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
696 for (len = 0; (*COPY_UCS2_CHAR(&cp,(src+len))) && (len < max); len++) {
697 cp = *COPY_UCS2_CHAR(dest+len,src+len);
700 for ( /*nothing*/ ; len < max; len++ ) {
701 cp = *COPY_UCS2_CHAR(dest+len,&cp);
707 /*******************************************************************
708 Append a string of len bytes and add a terminator.
709 ********************************************************************/
711 smb_ucs2_t *strncat_w(smb_ucs2_t *dest, const smb_ucs2_t *src, const size_t max)
721 start = strlen_w(dest);
722 len = strnlen_w(src, max);
724 memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
725 z = *COPY_UCS2_CHAR(dest+start+len,&z);
730 smb_ucs2_t *strcat_w(smb_ucs2_t *dest, const smb_ucs2_t *src)
740 start = strlen_w(dest);
743 memcpy(&dest[start], src, len*sizeof(smb_ucs2_t));
744 z = *COPY_UCS2_CHAR(dest+start+len,&z);
750 /*******************************************************************
751 Replace any occurence of oldc with newc in unicode string.
752 ********************************************************************/
754 void string_replace_w(smb_ucs2_t *s, smb_ucs2_t oldc, smb_ucs2_t newc)
758 for(;*(COPY_UCS2_CHAR(&cp,s));s++) {
760 COPY_UCS2_CHAR(s,&newc);
765 /*******************************************************************
767 ********************************************************************/
769 BOOL trim_string_w(smb_ucs2_t *s, const smb_ucs2_t *front,
770 const smb_ucs2_t *back)
773 size_t len, front_len, back_len;
781 if (front && *front) {
782 front_len = strlen_w(front);
783 while (len && strncmp_w(s, front, front_len) == 0) {
784 memmove(s, (s + front_len), (len - front_len + 1) * sizeof(smb_ucs2_t));
791 back_len = strlen_w(back);
792 while (len && strncmp_w((s + (len - back_len)), back, back_len) == 0) {
793 s[len - back_len] = 0;
803 The *_wa() functions take a combination of 7 bit ascii
804 and wide characters They are used so that you can use string
805 functions combining C string constants with ucs2 strings
807 The char* arguments must NOT be multibyte - to be completely sure
808 of this only pass string constants */
810 int strcmp_wa(const smb_ucs2_t *a, const char *b)
814 while (*b && *(COPY_UCS2_CHAR(&cp,a)) == UCS2_CHAR(*b)) {
818 return (cp - UCS2_CHAR(*b));
821 int strncmp_wa(const smb_ucs2_t *a, const char *b, size_t len)
826 while ((n < len) && *b && *(COPY_UCS2_CHAR(&cp,a)) == UCS2_CHAR(*b)) {
831 return (len - n)?(cp - UCS2_CHAR(*b)):0;
834 smb_ucs2_t *strpbrk_wa(const smb_ucs2_t *s, const char *p)
838 while (*(COPY_UCS2_CHAR(&cp,s))) {
840 for (i=0; p[i] && cp != UCS2_CHAR(p[i]); i++)
843 return (smb_ucs2_t *)s;
850 smb_ucs2_t *strstr_wa(const smb_ucs2_t *s, const char *ins)
859 inslen = strlen(ins);
862 while ((r = strchr_w(r, UCS2_CHAR(*ins)))) {
863 if (strncmp_wa(r, ins, inslen) == 0)
871 BOOL trim_string_wa(smb_ucs2_t *s, const char *front,
877 push_ucs2(NULL, f, front, sizeof(wpstring) - 1, STR_TERMINATE);
882 push_ucs2(NULL, b, back, sizeof(wpstring) - 1, STR_TERMINATE);
886 return trim_string_w(s, f, b);
889 /*******************************************************************
890 Returns the length in number of wide characters.
891 ******************************************************************/
893 int unistrlen(uint16 *s)
901 for (len=0; SVAL(s,0); s++,len++) {
908 /*******************************************************************
909 Strcpy for unicode strings. Returns length (in num of wide chars).
911 ********************************************************************/
913 int unistrcpy(uint16 *dst, uint16 *src)
917 while (SVAL(src,0)) {
927 * Samba ucs2 type to UNISTR2 conversion
929 * @param ctx Talloc context to create the dst strcture (if null) and the
930 * contents of the unicode string.
931 * @param dst UNISTR2 destination. If equals null, then it's allocated.
932 * @param src smb_ucs2_t source.
933 * @param max_len maximum number of unicode characters to copy. If equals
934 * null, then null-termination of src is taken
936 * @return copied UNISTR2 destination
939 UNISTR2* ucs2_to_unistr2(TALLOC_CTX *ctx, UNISTR2* dst, smb_ucs2_t* src)
949 /* allocate UNISTR2 destination if not given */
951 dst = TALLOC_P(ctx, UNISTR2);
956 dst->buffer = TALLOC_ARRAY(ctx, uint16, len + 1);
961 /* set UNISTR2 parameters */
962 dst->uni_max_len = len + 1;
964 dst->uni_str_len = len;
966 /* copy the actual unicode string */
967 strncpy_w(dst->buffer, src, dst->uni_max_len);