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 /* Helper function to return a talloc'ed string. I have implemented it with a
269 * copy because I don't really know how pull_ucs2 and friends calculate the
270 * target size. If this turns out to be a major bottleneck someone with deeper
271 * multi-byte knowledge needs to revisit this.
272 * My (VL) use is dsr_getdcname, which returns 6 strings, the alternative would
273 * have been to manually talloc_strdup them in rpc_client/cli_netlogon.c.
276 size_t rpcstr_pull_unistr2_talloc(TALLOC_CTX
*mem_ctx
, char **dest
,
282 result
= pull_ucs2(NULL
, tmp
, src
->buffer
, sizeof(tmp
),
283 src
->uni_str_len
* 2, 0);
288 *dest
= talloc_strdup(mem_ctx
, tmp
);
296 /* Converts a string from internal samba format to unicode
299 int rpcstr_push(void* dest
, const char *src
, int dest_len
, int flags
)
301 return push_ucs2(NULL
, dest
, src
, dest_len
, flags
|STR_UNICODE
|STR_NOALIGN
);
304 /*******************************************************************
305 Convert a (little-endian) UNISTR2 structure to an ASCII string.
306 ********************************************************************/
308 void unistr2_to_ascii(char *dest
, const UNISTR2
*str
, size_t maxlen
)
314 pull_ucs2(NULL
, dest
, str
->buffer
, maxlen
, str
->uni_str_len
*2, STR_NOALIGN
);
317 /*******************************************************************
318 Convert a (little-endian) UNISTR3 structure to an ASCII string.
319 ********************************************************************/
321 void unistr3_to_ascii(char *dest
, const UNISTR3
*str
, size_t maxlen
)
327 pull_ucs2(NULL
, dest
, str
->str
.buffer
, maxlen
, str
->uni_str_len
*2,
331 /*******************************************************************
332 Give a static string for displaying a UNISTR2.
333 ********************************************************************/
335 const char *unistr2_static(const UNISTR2
*str
)
338 unistr2_to_ascii(ret
, str
, sizeof(ret
));
342 /*******************************************************************
343 Duplicate a UNISTR2 string into a null terminated char*
344 using a talloc context.
345 ********************************************************************/
347 char *unistr2_tdup(TALLOC_CTX
*ctx
, const UNISTR2
*str
)
350 int maxlen
= (str
->uni_str_len
+1)*4;
354 s
= (char *)TALLOC(ctx
, maxlen
); /* convervative */
358 pull_ucs2(NULL
, s
, str
->buffer
, maxlen
, str
->uni_str_len
*2, STR_NOALIGN
);
362 /*******************************************************************
363 Convert a wchar to upper case.
364 ********************************************************************/
366 smb_ucs2_t
toupper_w(smb_ucs2_t val
)
368 return upcase_table
[SVAL(&val
,0)];
371 /*******************************************************************
372 Convert a wchar to lower case.
373 ********************************************************************/
375 smb_ucs2_t
tolower_w( smb_ucs2_t val
)
377 return lowcase_table
[SVAL(&val
,0)];
380 /*******************************************************************
381 Determine if a character is lowercase.
382 ********************************************************************/
384 BOOL
islower_w(smb_ucs2_t c
)
386 return upcase_table
[SVAL(&c
,0)] != c
;
389 /*******************************************************************
390 Determine if a character is uppercase.
391 ********************************************************************/
393 BOOL
isupper_w(smb_ucs2_t c
)
395 return lowcase_table
[SVAL(&c
,0)] != c
;
398 /*******************************************************************
399 Determine if a character is valid in a 8.3 name.
400 ********************************************************************/
402 BOOL
isvalid83_w(smb_ucs2_t c
)
404 return valid_table
[SVAL(&c
,0)] != 0;
407 /*******************************************************************
408 Count the number of characters in a smb_ucs2_t string.
409 ********************************************************************/
411 size_t strlen_w(const smb_ucs2_t
*src
)
416 for(len
= 0; *(COPY_UCS2_CHAR(&c
,src
)); src
++, len
++) {
423 /*******************************************************************
424 Count up to max number of characters in a smb_ucs2_t string.
425 ********************************************************************/
427 size_t strnlen_w(const smb_ucs2_t
*src
, size_t max
)
432 for(len
= 0; *(COPY_UCS2_CHAR(&c
,src
)) && (len
< max
); src
++, len
++) {
439 /*******************************************************************
441 ********************************************************************/
443 smb_ucs2_t
*strchr_w(const smb_ucs2_t
*s
, smb_ucs2_t c
)
446 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
448 return (smb_ucs2_t
*)s
;
453 return (smb_ucs2_t
*)s
;
459 smb_ucs2_t
*strchr_wa(const smb_ucs2_t
*s
, char c
)
461 return strchr_w(s
, UCS2_CHAR(c
));
464 /*******************************************************************
466 ********************************************************************/
468 smb_ucs2_t
*strrchr_w(const smb_ucs2_t
*s
, smb_ucs2_t c
)
471 const smb_ucs2_t
*p
= s
;
472 int len
= strlen_w(s
);
479 if (c
== *(COPY_UCS2_CHAR(&cp
,p
))) {
480 return (smb_ucs2_t
*)p
;
486 /*******************************************************************
487 Wide version of strrchr that returns after doing strrchr 'n' times.
488 ********************************************************************/
490 smb_ucs2_t
*strnrchr_w(const smb_ucs2_t
*s
, smb_ucs2_t c
, unsigned int n
)
493 const smb_ucs2_t
*p
= s
;
494 int len
= strlen_w(s
);
496 if (len
== 0 || !n
) {
501 if (c
== *(COPY_UCS2_CHAR(&cp
,p
))) {
506 return (smb_ucs2_t
*)p
;
512 /*******************************************************************
514 ********************************************************************/
516 smb_ucs2_t
*strstr_w(const smb_ucs2_t
*s
, const smb_ucs2_t
*ins
)
521 if (!s
|| !*s
|| !ins
|| !*ins
) {
525 inslen
= strlen_w(ins
);
528 while ((r
= strchr_w(r
, *ins
))) {
529 if (strncmp_w(r
, ins
, inslen
) == 0) {
538 /*******************************************************************
539 Convert a string to lower case.
540 return True if any char is converted
541 ********************************************************************/
543 BOOL
strlower_w(smb_ucs2_t
*s
)
548 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
549 smb_ucs2_t v
= tolower_w(cp
);
551 COPY_UCS2_CHAR(s
,&v
);
559 /*******************************************************************
560 Convert a string to upper case.
561 return True if any char is converted
562 ********************************************************************/
564 BOOL
strupper_w(smb_ucs2_t
*s
)
568 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
569 smb_ucs2_t v
= toupper_w(cp
);
571 COPY_UCS2_CHAR(s
,&v
);
579 /*******************************************************************
580 Convert a string to "normal" form.
581 ********************************************************************/
583 void strnorm_w(smb_ucs2_t
*s
, int case_default
)
585 if (case_default
== CASE_UPPER
) {
592 int strcmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
)
596 while ((*(COPY_UCS2_CHAR(&cpb
,b
))) && (*(COPY_UCS2_CHAR(&cpa
,a
)) == cpb
)) {
600 return (*(COPY_UCS2_CHAR(&cpa
,a
)) - *(COPY_UCS2_CHAR(&cpb
,b
)));
601 /* warning: if *a != *b and both are not 0 we return a random
602 greater or lesser than 0 number not realted to which
606 int strncmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
, size_t len
)
611 while ((n
< len
) && (*(COPY_UCS2_CHAR(&cpb
,b
))) && (*(COPY_UCS2_CHAR(&cpa
,a
)) == cpb
)) {
616 return (len
- n
)?(*(COPY_UCS2_CHAR(&cpa
,a
)) - *(COPY_UCS2_CHAR(&cpb
,b
))):0;
619 /*******************************************************************
620 Case insensitive string comparison.
621 ********************************************************************/
623 int strcasecmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
)
627 while ((*COPY_UCS2_CHAR(&cpb
,b
)) && toupper_w(*(COPY_UCS2_CHAR(&cpa
,a
))) == toupper_w(cpb
)) {
631 return (tolower_w(*(COPY_UCS2_CHAR(&cpa
,a
))) - tolower_w(*(COPY_UCS2_CHAR(&cpb
,b
))));
634 /*******************************************************************
635 Case insensitive string comparison, length limited.
636 ********************************************************************/
638 int strncasecmp_w(const smb_ucs2_t
*a
, const smb_ucs2_t
*b
, size_t len
)
643 while ((n
< len
) && *COPY_UCS2_CHAR(&cpb
,b
) && (toupper_w(*(COPY_UCS2_CHAR(&cpa
,a
))) == toupper_w(cpb
))) {
648 return (len
- n
)?(tolower_w(*(COPY_UCS2_CHAR(&cpa
,a
))) - tolower_w(*(COPY_UCS2_CHAR(&cpb
,b
)))):0;
651 /*******************************************************************
653 ********************************************************************/
655 BOOL
strequal_w(const smb_ucs2_t
*s1
, const smb_ucs2_t
*s2
)
664 return(strcasecmp_w(s1
,s2
)==0);
667 /*******************************************************************
668 Compare 2 strings up to and including the nth char.
669 ******************************************************************/
671 BOOL
strnequal_w(const smb_ucs2_t
*s1
,const smb_ucs2_t
*s2
,size_t n
)
676 if (!s1
|| !s2
|| !n
) {
680 return(strncasecmp_w(s1
,s2
,n
)==0);
683 /*******************************************************************
685 ********************************************************************/
687 smb_ucs2_t
*strdup_w(const smb_ucs2_t
*src
)
689 return strndup_w(src
, 0);
692 /* if len == 0 then duplicate the whole string */
694 smb_ucs2_t
*strndup_w(const smb_ucs2_t
*src
, size_t len
)
701 dest
= SMB_MALLOC_ARRAY(smb_ucs2_t
, len
+ 1);
703 DEBUG(0,("strdup_w: out of memory!\n"));
707 memcpy(dest
, src
, len
* sizeof(smb_ucs2_t
));
712 /*******************************************************************
713 Copy a string with max len.
714 ********************************************************************/
716 smb_ucs2_t
*strncpy_w(smb_ucs2_t
*dest
, const smb_ucs2_t
*src
, const size_t max
)
725 for (len
= 0; (*COPY_UCS2_CHAR(&cp
,(src
+len
))) && (len
< max
); len
++) {
726 cp
= *COPY_UCS2_CHAR(dest
+len
,src
+len
);
729 for ( /*nothing*/ ; len
< max
; len
++ ) {
730 cp
= *COPY_UCS2_CHAR(dest
+len
,&cp
);
736 /*******************************************************************
737 Append a string of len bytes and add a terminator.
738 ********************************************************************/
740 smb_ucs2_t
*strncat_w(smb_ucs2_t
*dest
, const smb_ucs2_t
*src
, const size_t max
)
750 start
= strlen_w(dest
);
751 len
= strnlen_w(src
, max
);
753 memcpy(&dest
[start
], src
, len
*sizeof(smb_ucs2_t
));
754 z
= *COPY_UCS2_CHAR(dest
+start
+len
,&z
);
759 smb_ucs2_t
*strcat_w(smb_ucs2_t
*dest
, const smb_ucs2_t
*src
)
769 start
= strlen_w(dest
);
772 memcpy(&dest
[start
], src
, len
*sizeof(smb_ucs2_t
));
773 z
= *COPY_UCS2_CHAR(dest
+start
+len
,&z
);
779 /*******************************************************************
780 Replace any occurence of oldc with newc in unicode string.
781 ********************************************************************/
783 void string_replace_w(smb_ucs2_t
*s
, smb_ucs2_t oldc
, smb_ucs2_t newc
)
787 for(;*(COPY_UCS2_CHAR(&cp
,s
));s
++) {
789 COPY_UCS2_CHAR(s
,&newc
);
794 /*******************************************************************
796 ********************************************************************/
798 BOOL
trim_string_w(smb_ucs2_t
*s
, const smb_ucs2_t
*front
,
799 const smb_ucs2_t
*back
)
802 size_t len
, front_len
, back_len
;
810 if (front
&& *front
) {
811 front_len
= strlen_w(front
);
812 while (len
&& strncmp_w(s
, front
, front_len
) == 0) {
813 memmove(s
, (s
+ front_len
), (len
- front_len
+ 1) * sizeof(smb_ucs2_t
));
820 back_len
= strlen_w(back
);
821 while (len
&& strncmp_w((s
+ (len
- back_len
)), back
, back_len
) == 0) {
822 s
[len
- back_len
] = 0;
832 The *_wa() functions take a combination of 7 bit ascii
833 and wide characters They are used so that you can use string
834 functions combining C string constants with ucs2 strings
836 The char* arguments must NOT be multibyte - to be completely sure
837 of this only pass string constants */
839 int strcmp_wa(const smb_ucs2_t
*a
, const char *b
)
843 while (*b
&& *(COPY_UCS2_CHAR(&cp
,a
)) == UCS2_CHAR(*b
)) {
847 return (*(COPY_UCS2_CHAR(&cp
,a
)) - UCS2_CHAR(*b
));
850 int strncmp_wa(const smb_ucs2_t
*a
, const char *b
, size_t len
)
855 while ((n
< len
) && *b
&& *(COPY_UCS2_CHAR(&cp
,a
)) == UCS2_CHAR(*b
)) {
860 return (len
- n
)?(*(COPY_UCS2_CHAR(&cp
,a
)) - UCS2_CHAR(*b
)):0;
863 smb_ucs2_t
*strpbrk_wa(const smb_ucs2_t
*s
, const char *p
)
867 while (*(COPY_UCS2_CHAR(&cp
,s
))) {
869 for (i
=0; p
[i
] && cp
!= UCS2_CHAR(p
[i
]); i
++)
872 return (smb_ucs2_t
*)s
;
879 smb_ucs2_t
*strstr_wa(const smb_ucs2_t
*s
, const char *ins
)
888 inslen
= strlen(ins
);
891 while ((r
= strchr_w(r
, UCS2_CHAR(*ins
)))) {
892 if (strncmp_wa(r
, ins
, inslen
) == 0)
900 BOOL
trim_string_wa(smb_ucs2_t
*s
, const char *front
,
906 push_ucs2(NULL
, f
, front
, sizeof(wpstring
) - 1, STR_TERMINATE
);
911 push_ucs2(NULL
, b
, back
, sizeof(wpstring
) - 1, STR_TERMINATE
);
915 return trim_string_w(s
, f
, b
);
918 /*******************************************************************
919 Returns the length in number of wide characters.
920 ******************************************************************/
922 int unistrlen(uint16
*s
)
930 for (len
=0; SVAL(s
,0); s
++,len
++) {
937 /*******************************************************************
938 Strcpy for unicode strings. Returns length (in num of wide chars).
940 ********************************************************************/
942 int unistrcpy(uint16
*dst
, uint16
*src
)
946 while (SVAL(src
,0)) {
956 * Samba ucs2 type to UNISTR2 conversion
958 * @param ctx Talloc context to create the dst strcture (if null) and the
959 * contents of the unicode string.
960 * @param dst UNISTR2 destination. If equals null, then it's allocated.
961 * @param src smb_ucs2_t source.
962 * @param max_len maximum number of unicode characters to copy. If equals
963 * null, then null-termination of src is taken
965 * @return copied UNISTR2 destination
968 UNISTR2
* ucs2_to_unistr2(TALLOC_CTX
*ctx
, UNISTR2
* dst
, smb_ucs2_t
* src
)
978 /* allocate UNISTR2 destination if not given */
980 dst
= TALLOC_P(ctx
, UNISTR2
);
985 dst
->buffer
= TALLOC_ARRAY(ctx
, uint16
, len
+ 1);
990 /* set UNISTR2 parameters */
991 dst
->uni_max_len
= len
+ 1;
993 dst
->uni_str_len
= len
;
995 /* copy the actual unicode string */
996 strncpy_w(dst
->buffer
, src
, dst
->uni_max_len
);