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 int flags
= null_terminate
? STR_UNICODE
|STR_NOALIGN
|STR_TERMINATE
217 : STR_UNICODE
|STR_NOALIGN
;
218 return push_ucs2(NULL
, dst
, src
, len
, flags
);
222 /*******************************************************************
223 Skip past a unicode string, but not more than len. Always move
224 past a terminating zero if found.
225 ********************************************************************/
227 char *skip_unibuf(char *src
, size_t len
)
229 char *srcend
= src
+ len
;
231 while (src
< srcend
&& SVAL(src
,0)) {
242 /* Copy a string from little-endian or big-endian unicode source (depending
243 * on flags) to internal samba format destination
246 int rpcstr_pull(char* dest
, void *src
, int dest_len
, int src_len
, int flags
)
255 return pull_ucs2(NULL
, dest
, src
, dest_len
, src_len
, flags
|STR_UNICODE
|STR_NOALIGN
);
258 /* Copy a string from a unistr2 source to internal samba format
259 destination. Use this instead of direct calls to rpcstr_pull() to avoid
260 having to determine whether the source string is null terminated. */
262 int rpcstr_pull_unistr2_fstring(char *dest
, UNISTR2
*src
)
264 return pull_ucs2(NULL
, dest
, src
->buffer
, sizeof(fstring
),
265 src
->uni_str_len
* 2, 0);
268 /* Converts a string from internal samba format to unicode
271 int rpcstr_push(void* dest
, const char *src
, int dest_len
, int flags
)
273 return push_ucs2(NULL
, dest
, src
, dest_len
, flags
|STR_UNICODE
|STR_NOALIGN
);
276 /*******************************************************************
277 Convert a (little-endian) UNISTR2 structure to an ASCII string.
278 ********************************************************************/
280 void unistr2_to_ascii(char *dest
, const UNISTR2
*str
, size_t maxlen
)
286 pull_ucs2(NULL
, dest
, str
->buffer
, maxlen
, str
->uni_str_len
*2, STR_NOALIGN
);
289 /*******************************************************************
290 Convert a (little-endian) UNISTR3 structure to an ASCII string.
291 ********************************************************************/
293 void unistr3_to_ascii(char *dest
, const UNISTR3
*str
, size_t maxlen
)
299 pull_ucs2(NULL
, dest
, str
->str
.buffer
, maxlen
, str
->uni_str_len
*2,
303 /*******************************************************************
304 Give a static string for displaying a UNISTR2.
305 ********************************************************************/
307 const char *unistr2_static(const UNISTR2
*str
)
310 unistr2_to_ascii(ret
, str
, sizeof(ret
));
314 /*******************************************************************
315 Duplicate a UNISTR2 string into a null terminated char*
316 using a talloc context.
317 ********************************************************************/
319 char *unistr2_tdup(TALLOC_CTX
*ctx
, const UNISTR2
*str
)
322 int maxlen
= (str
->uni_str_len
+1)*4;
326 s
= (char *)TALLOC(ctx
, maxlen
); /* convervative */
330 pull_ucs2(NULL
, s
, str
->buffer
, maxlen
, str
->uni_str_len
*2, STR_NOALIGN
);
334 /*******************************************************************
335 Convert a wchar to upper case.
336 ********************************************************************/
338 smb_ucs2_t
toupper_w(smb_ucs2_t val
)
340 return upcase_table
[SVAL(&val
,0)];
343 /*******************************************************************
344 Convert a wchar to lower case.
345 ********************************************************************/
347 smb_ucs2_t
tolower_w( smb_ucs2_t val
)
349 return lowcase_table
[SVAL(&val
,0)];
352 /*******************************************************************
353 Determine if a character is lowercase.
354 ********************************************************************/
356 BOOL
islower_w(smb_ucs2_t c
)
358 return upcase_table
[SVAL(&c
,0)] != c
;
361 /*******************************************************************
362 Determine if a character is uppercase.
363 ********************************************************************/
365 BOOL
isupper_w(smb_ucs2_t c
)
367 return lowcase_table
[SVAL(&c
,0)] != c
;
370 /*******************************************************************
371 Determine if a character is valid in a 8.3 name.
372 ********************************************************************/
374 BOOL
isvalid83_w(smb_ucs2_t c
)
376 return valid_table
[SVAL(&c
,0)] != 0;
379 /*******************************************************************
380 Count the number of characters in a smb_ucs2_t string.
381 ********************************************************************/
383 size_t strlen_w(const smb_ucs2_t
*src
)
388 for(len
= 0; *(COPY_UCS2_CHAR(&c
,src
)); src
++, len
++) {
395 /*******************************************************************
396 Count up to max number of characters in a smb_ucs2_t string.
397 ********************************************************************/
399 size_t strnlen_w(const smb_ucs2_t
*src
, size_t max
)
404 for(len
= 0; *(COPY_UCS2_CHAR(&c
,src
)) && (len
< max
); src
++, len
++) {
411 /*******************************************************************
413 ********************************************************************/
415 smb_ucs2_t
*strchr_w(const smb_ucs2_t
*s
, smb_ucs2_t c
)
418 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
420 return (smb_ucs2_t
*)s
;
425 return (smb_ucs2_t
*)s
;
431 smb_ucs2_t
*strchr_wa(const smb_ucs2_t
*s
, char c
)
433 return strchr_w(s
, UCS2_CHAR(c
));
436 /*******************************************************************
438 ********************************************************************/
440 smb_ucs2_t
*strrchr_w(const smb_ucs2_t
*s
, smb_ucs2_t c
)
443 const smb_ucs2_t
*p
= s
;
444 int len
= strlen_w(s
);
451 if (c
== *(COPY_UCS2_CHAR(&cp
,p
))) {
452 return (smb_ucs2_t
*)p
;
458 /*******************************************************************
459 Wide version of strrchr that returns after doing strrchr 'n' times.
460 ********************************************************************/
462 smb_ucs2_t
*strnrchr_w(const smb_ucs2_t
*s
, smb_ucs2_t c
, unsigned int n
)
465 const smb_ucs2_t
*p
= s
;
466 int len
= strlen_w(s
);
468 if (len
== 0 || !n
) {
473 if (c
== *(COPY_UCS2_CHAR(&cp
,p
))) {
478 return (smb_ucs2_t
*)p
;
484 /*******************************************************************
486 ********************************************************************/
488 smb_ucs2_t
*strstr_w(const smb_ucs2_t
*s
, const smb_ucs2_t
*ins
)
493 if (!s
|| !*s
|| !ins
|| !*ins
) {
497 inslen
= strlen_w(ins
);
500 while ((r
= strchr_w(r
, *ins
))) {
501 if (strncmp_w(r
, ins
, inslen
) == 0) {
510 /*******************************************************************
511 Convert a string to lower case.
512 return True if any char is converted
513 ********************************************************************/
515 BOOL
strlower_w(smb_ucs2_t
*s
)
520 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
521 smb_ucs2_t v
= tolower_w(cp
);
523 COPY_UCS2_CHAR(s
,&v
);
531 /*******************************************************************
532 Convert a string to upper case.
533 return True if any char is converted
534 ********************************************************************/
536 BOOL
strupper_w(smb_ucs2_t
*s
)
540 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
541 smb_ucs2_t v
= toupper_w(cp
);
543 COPY_UCS2_CHAR(s
,&v
);
551 /*******************************************************************
552 Convert a string to "normal" form.
553 ********************************************************************/
555 void strnorm_w(smb_ucs2_t
*s
, int case_default
)
557 if (case_default
== CASE_UPPER
) {
564 int strcmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
)
568 while ((*(COPY_UCS2_CHAR(&cpb
,b
))) && (*(COPY_UCS2_CHAR(&cpa
,a
)) == cpb
)) {
572 return (*(COPY_UCS2_CHAR(&cpa
,a
)) - *(COPY_UCS2_CHAR(&cpb
,b
)));
573 /* warning: if *a != *b and both are not 0 we return a random
574 greater or lesser than 0 number not realted to which
578 int strncmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
, size_t len
)
583 while ((n
< len
) && (*(COPY_UCS2_CHAR(&cpb
,b
))) && (*(COPY_UCS2_CHAR(&cpa
,a
)) == cpb
)) {
588 return (len
- n
)?(*(COPY_UCS2_CHAR(&cpa
,a
)) - *(COPY_UCS2_CHAR(&cpb
,b
))):0;
591 /*******************************************************************
592 Case insensitive string comparison.
593 ********************************************************************/
595 int strcasecmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
)
599 while ((*COPY_UCS2_CHAR(&cpb
,b
)) && toupper_w(*(COPY_UCS2_CHAR(&cpa
,a
))) == toupper_w(cpb
)) {
603 return (tolower_w(*(COPY_UCS2_CHAR(&cpa
,a
))) - tolower_w(*(COPY_UCS2_CHAR(&cpb
,b
))));
606 /*******************************************************************
607 Case insensitive string comparison, length limited.
608 ********************************************************************/
610 int strncasecmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
, size_t len
)
615 while ((n
< len
) && *COPY_UCS2_CHAR(&cpb
,b
) && (toupper_w(*(COPY_UCS2_CHAR(&cpa
,a
))) == toupper_w(cpb
))) {
620 return (len
- n
)?(tolower_w(*(COPY_UCS2_CHAR(&cpa
,a
))) - tolower_w(*(COPY_UCS2_CHAR(&cpb
,b
)))):0;
623 /*******************************************************************
625 ********************************************************************/
627 BOOL
strequal_w(const smb_ucs2_t
*s1
, const smb_ucs2_t
*s2
)
636 return(strcasecmp_w(s1
,s2
)==0);
639 /*******************************************************************
640 Compare 2 strings up to and including the nth char.
641 ******************************************************************/
643 BOOL
strnequal_w(const smb_ucs2_t
*s1
,const smb_ucs2_t
*s2
,size_t n
)
648 if (!s1
|| !s2
|| !n
) {
652 return(strncasecmp_w(s1
,s2
,n
)==0);
655 /*******************************************************************
657 ********************************************************************/
659 smb_ucs2_t
*strdup_w(const smb_ucs2_t
*src
)
661 return strndup_w(src
, 0);
664 /* if len == 0 then duplicate the whole string */
666 smb_ucs2_t
*strndup_w(const smb_ucs2_t
*src
, size_t len
)
673 dest
= SMB_MALLOC_ARRAY(smb_ucs2_t
, len
+ 1);
675 DEBUG(0,("strdup_w: out of memory!\n"));
679 memcpy(dest
, src
, len
* sizeof(smb_ucs2_t
));
684 /*******************************************************************
685 Copy a string with max len.
686 ********************************************************************/
688 smb_ucs2_t
*strncpy_w(smb_ucs2_t
*dest
, const smb_ucs2_t
*src
, const size_t max
)
697 for (len
= 0; (*COPY_UCS2_CHAR(&cp
,(src
+len
))) && (len
< max
); len
++) {
698 cp
= *COPY_UCS2_CHAR(dest
+len
,src
+len
);
701 for ( /*nothing*/ ; len
< max
; len
++ ) {
702 cp
= *COPY_UCS2_CHAR(dest
+len
,&cp
);
708 /*******************************************************************
709 Append a string of len bytes and add a terminator.
710 ********************************************************************/
712 smb_ucs2_t
*strncat_w(smb_ucs2_t
*dest
, const smb_ucs2_t
*src
, const size_t max
)
722 start
= strlen_w(dest
);
723 len
= strnlen_w(src
, max
);
725 memcpy(&dest
[start
], src
, len
*sizeof(smb_ucs2_t
));
726 z
= *COPY_UCS2_CHAR(dest
+start
+len
,&z
);
731 smb_ucs2_t
*strcat_w(smb_ucs2_t
*dest
, const smb_ucs2_t
*src
)
741 start
= strlen_w(dest
);
744 memcpy(&dest
[start
], src
, len
*sizeof(smb_ucs2_t
));
745 z
= *COPY_UCS2_CHAR(dest
+start
+len
,&z
);
751 /*******************************************************************
752 Replace any occurence of oldc with newc in unicode string.
753 ********************************************************************/
755 void string_replace_w(smb_ucs2_t
*s
, smb_ucs2_t oldc
, smb_ucs2_t newc
)
759 for(;*(COPY_UCS2_CHAR(&cp
,s
));s
++) {
761 COPY_UCS2_CHAR(s
,&newc
);
766 /*******************************************************************
768 ********************************************************************/
770 BOOL
trim_string_w(smb_ucs2_t
*s
, const smb_ucs2_t
*front
,
771 const smb_ucs2_t
*back
)
774 size_t len
, front_len
, back_len
;
782 if (front
&& *front
) {
783 front_len
= strlen_w(front
);
784 while (len
&& strncmp_w(s
, front
, front_len
) == 0) {
785 memmove(s
, (s
+ front_len
), (len
- front_len
+ 1) * sizeof(smb_ucs2_t
));
792 back_len
= strlen_w(back
);
793 while (len
&& strncmp_w((s
+ (len
- back_len
)), back
, back_len
) == 0) {
794 s
[len
- back_len
] = 0;
804 The *_wa() functions take a combination of 7 bit ascii
805 and wide characters They are used so that you can use string
806 functions combining C string constants with ucs2 strings
808 The char* arguments must NOT be multibyte - to be completely sure
809 of this only pass string constants */
811 int strcmp_wa(const smb_ucs2_t
*a
, const char *b
)
815 while (*b
&& *(COPY_UCS2_CHAR(&cp
,a
)) == UCS2_CHAR(*b
)) {
819 return (*(COPY_UCS2_CHAR(&cp
,a
)) - UCS2_CHAR(*b
));
822 int strncmp_wa(const smb_ucs2_t
*a
, const char *b
, size_t len
)
827 while ((n
< len
) && *b
&& *(COPY_UCS2_CHAR(&cp
,a
)) == UCS2_CHAR(*b
)) {
832 return (len
- n
)?(*(COPY_UCS2_CHAR(&cp
,a
)) - UCS2_CHAR(*b
)):0;
835 smb_ucs2_t
*strpbrk_wa(const smb_ucs2_t
*s
, const char *p
)
839 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
841 for (i
=0; p
[i
] && cp
!= UCS2_CHAR(p
[i
]); i
++)
844 return (smb_ucs2_t
*)s
;
851 smb_ucs2_t
*strstr_wa(const smb_ucs2_t
*s
, const char *ins
)
860 inslen
= strlen(ins
);
863 while ((r
= strchr_w(r
, UCS2_CHAR(*ins
)))) {
864 if (strncmp_wa(r
, ins
, inslen
) == 0)
872 BOOL
trim_string_wa(smb_ucs2_t
*s
, const char *front
,
878 push_ucs2(NULL
, f
, front
, sizeof(wpstring
) - 1, STR_TERMINATE
);
883 push_ucs2(NULL
, b
, back
, sizeof(wpstring
) - 1, STR_TERMINATE
);
887 return trim_string_w(s
, f
, b
);
890 /*******************************************************************
891 Returns the length in number of wide characters.
892 ******************************************************************/
894 int unistrlen(uint16
*s
)
902 for (len
=0; SVAL(s
,0); s
++,len
++) {
909 /*******************************************************************
910 Strcpy for unicode strings. Returns length (in num of wide chars).
912 ********************************************************************/
914 int unistrcpy(uint16
*dst
, uint16
*src
)
918 while (SVAL(src
,0)) {
928 * Samba ucs2 type to UNISTR2 conversion
930 * @param ctx Talloc context to create the dst strcture (if null) and the
931 * contents of the unicode string.
932 * @param dst UNISTR2 destination. If equals null, then it's allocated.
933 * @param src smb_ucs2_t source.
934 * @param max_len maximum number of unicode characters to copy. If equals
935 * null, then null-termination of src is taken
937 * @return copied UNISTR2 destination
940 UNISTR2
* ucs2_to_unistr2(TALLOC_CTX
*ctx
, UNISTR2
* dst
, smb_ucs2_t
* src
)
950 /* allocate UNISTR2 destination if not given */
952 dst
= TALLOC_P(ctx
, UNISTR2
);
957 dst
->buffer
= TALLOC_ARRAY(ctx
, uint16
, len
+ 1);
962 /* set UNISTR2 parameters */
963 dst
->uni_max_len
= len
+ 1;
965 dst
->uni_str_len
= len
;
967 /* copy the actual unicode string */
968 strncpy_w(dst
->buffer
, src
, dst
->uni_max_len
);