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) Andrew Bartlett 2011
7 Copyright (C) Jeremy Allison 1992-2007
8 Copyright (C) Martin Pool 2003
9 Copyright (C) James Peach 2006
11 This program is free software; you can redistribute it and/or modify
12 it under the terms of the GNU General Public License as published by
13 the Free Software Foundation; either version 3 of the License, or
14 (at your option) any later version.
16 This program is distributed in the hope that it will be useful,
17 but WITHOUT ANY WARRANTY; without even the implied warranty of
18 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 GNU General Public License for more details.
21 You should have received a copy of the GNU General Public License
22 along with this program. If not, see <http://www.gnu.org/licenses/>.
26 #include "system/locale.h"
33 Case insensitive string compararison, handle specified for testing
35 _PUBLIC_
int strcasecmp_m_handle(struct smb_iconv_handle
*iconv_handle
,
36 const char *s1
, const char *s2
)
38 codepoint_t c1
=0, c2
=0;
41 /* handle null ptr comparisons to simplify the use in qsort */
42 if (s1
== s2
) return 0;
43 if (s1
== NULL
) return -1;
44 if (s2
== NULL
) return 1;
47 c1
= next_codepoint_handle(iconv_handle
, s1
, &size1
);
48 c2
= next_codepoint_handle(iconv_handle
, s2
, &size2
);
50 if (c1
== INVALID_CODEPOINT
||
51 c2
== INVALID_CODEPOINT
) {
52 return strcasecmp(s1
, s2
);
62 if (toupper_m(c1
) != toupper_m(c2
)) {
71 Case insensitive string compararison
73 _PUBLIC_
int strcasecmp_m(const char *s1
, const char *s2
)
75 struct smb_iconv_handle
*iconv_handle
= get_iconv_handle();
76 return strcasecmp_m_handle(iconv_handle
, s1
, s2
);
80 Case insensitive string compararison, length limited, handle specified for testing
82 _PUBLIC_
int strncasecmp_m_handle(struct smb_iconv_handle
*iconv_handle
,
83 const char *s1
, const char *s2
, size_t n
)
85 codepoint_t c1
=0, c2
=0;
88 /* handle null ptr comparisons to simplify the use in qsort */
89 if (s1
== s2
) return 0;
90 if (s1
== NULL
) return -1;
91 if (s2
== NULL
) return 1;
93 while (*s1
&& *s2
&& n
) {
96 c1
= next_codepoint_handle(iconv_handle
, s1
, &size1
);
97 c2
= next_codepoint_handle(iconv_handle
, s2
, &size2
);
99 if (c1
== INVALID_CODEPOINT
||
100 c2
== INVALID_CODEPOINT
) {
102 * n was specified in characters,
103 * now we must convert it to bytes.
104 * As bytes are the smallest
105 * character unit, the following
106 * increment and strncasecmp is always
109 * The source string was already known
110 * to be n characters long, so we are
111 * guaranteed to be able to look at the
112 * (n remaining + size1) bytes from the
116 return strncasecmp(s1
, s2
, n
);
126 if (toupper_m(c1
) != toupper_m(c2
)) {
139 Case insensitive string compararison, length limited
141 _PUBLIC_
int strncasecmp_m(const char *s1
, const char *s2
, size_t n
)
143 struct smb_iconv_handle
*iconv_handle
= get_iconv_handle();
144 return strncasecmp_m_handle(iconv_handle
, s1
, s2
, n
);
150 * @note The comparison is case-insensitive.
152 _PUBLIC_
bool strequal_m(const char *s1
, const char *s2
)
154 return strcasecmp_m(s1
,s2
) == 0;
158 Compare 2 strings (case sensitive).
160 _PUBLIC_
bool strcsequal(const char *s1
,const char *s2
)
167 return strcmp(s1
,s2
) == 0;
171 * Calculate the number of units (8 or 16-bit, depending on the
172 * destination charset), that would be needed to convert the input
173 * string which is expected to be in in src_charset encoding to the
174 * destination charset (which should be a unicode charset).
176 _PUBLIC_
size_t strlen_m_ext_handle(struct smb_iconv_handle
*ic
,
177 const char *s
, charset_t src_charset
, charset_t dst_charset
)
182 switch (dst_charset
) {
185 smb_panic("cannot call strlen_m_ext() with a variable dest charset (must be UTF16* or UTF8)");
190 switch (src_charset
) {
193 smb_panic("cannot call strlen_m_ext() with a UTF16 src charset (must be DOS, UNIX, DISPLAY or UTF8)");
202 while (*s
&& !(((uint8_t)*s
) & 0x80)) {
213 codepoint_t c
= next_codepoint_handle_ext(ic
, s
, src_charset
, &c_size
);
216 switch (dst_charset
) {
221 /* Unicode char fits into 16 bits. */
224 /* Double-width unicode char - 32 bits. */
230 * this only checks ranges, and does not
231 * check for invalid codepoints
235 } else if (c
< 0x800) {
237 } else if (c
< 0x10000) {
245 * non-unicode encoding:
246 * assume that each codepoint fits into
247 * one unit in the destination encoding.
257 * Calculate the number of units (8 or 16-bit, depending on the
258 * destination charset), that would be needed to convert the input
259 * string which is expected to be in in src_charset encoding to the
260 * destination charset (which should be a unicode charset).
262 _PUBLIC_
size_t strlen_m_ext(const char *s
, charset_t src_charset
, charset_t dst_charset
)
264 struct smb_iconv_handle
*ic
= get_iconv_handle();
265 return strlen_m_ext_handle(ic
, s
, src_charset
, dst_charset
);
268 _PUBLIC_
size_t strlen_m_ext_term(const char *s
, const charset_t src_charset
,
269 const charset_t dst_charset
)
274 return strlen_m_ext(s
, src_charset
, dst_charset
) + 1;
278 * Calculate the number of 16-bit units that would be needed to convert
279 * the input string which is expected to be in CH_UNIX encoding to UTF16.
281 * This will be the same as the number of bytes in a string for single
282 * byte strings, but will be different for multibyte.
284 _PUBLIC_
size_t strlen_m(const char *s
)
286 return strlen_m_ext(s
, CH_UNIX
, CH_UTF16LE
);
290 Work out the number of multibyte chars in a string, including the NULL
293 _PUBLIC_
size_t strlen_m_term(const char *s
)
299 return strlen_m(s
) + 1;
303 * Weird helper routine for the winreg pipe: If nothing is around, return 0,
304 * if a string is there, include the terminator.
307 _PUBLIC_
size_t strlen_m_term_null(const char *s
)
322 Strchr and strrchr_m are a bit complex on general multi-byte strings.
324 _PUBLIC_
char *strchr_m(const char *src
, char c
)
327 struct smb_iconv_handle
*ic
= get_iconv_handle();
331 /* characters below 0x3F are guaranteed to not appear in
332 non-initial position in multi-byte charsets */
333 if ((c
& 0xC0) == 0) {
334 return strchr(src
, c
);
337 /* this is quite a common operation, so we want it to be
338 fast. We optimise for the ascii case, knowing that all our
339 supported multi-byte character sets are ascii-compatible
340 (ie. they match for the first 128 chars) */
342 for (s
= src
; *s
&& !(((unsigned char)s
[0]) & 0x80); s
++) {
344 return discard_const_p(char, s
);
350 #ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
351 /* With compose characters we must restart from the beginning. JRA. */
357 codepoint_t c2
= next_codepoint_handle(ic
, s
, &size
);
359 return discard_const_p(char, s
);
368 * Multibyte-character version of strrchr
370 _PUBLIC_
char *strrchr_m(const char *s
, char c
)
372 struct smb_iconv_handle
*ic
= get_iconv_handle();
379 /* characters below 0x3F are guaranteed to not appear in
380 non-initial position in multi-byte charsets */
381 if ((c
& 0xC0) == 0) {
382 return strrchr(s
, c
);
385 /* this is quite a common operation, so we want it to be
386 fast. We optimise for the ascii case, knowing that all our
387 supported multi-byte character sets are ascii-compatible
388 (ie. they match for the first 128 chars). Also, in Samba
389 we only search for ascii characters in 'c' and that
390 in all mb character sets with a compound character
391 containing c, if 'c' is not a match at position
392 p, then p[-1] > 0x7f. JRA. */
395 size_t len
= strlen(s
);
404 /* Could be a match. Part of a multibyte ? */
406 (((unsigned char)cp
[-1]) & 0x80)) {
407 /* Yep - go slow :-( */
411 /* No - we have a match ! */
412 return discard_const_p(char , cp
);
421 codepoint_t c2
= next_codepoint_handle(ic
, s
, &size
);
423 ret
= discard_const_p(char, s
);
432 return True if any (multi-byte) character is lower case
434 _PUBLIC_
bool strhaslower_handle(struct smb_iconv_handle
*ic
,
442 s
= next_codepoint_handle(ic
, string
, &c_size
);
448 return true; /* that means it has lower case chars */
455 _PUBLIC_
bool strhaslower(const char *string
)
457 struct smb_iconv_handle
*ic
= get_iconv_handle();
458 return strhaslower_handle(ic
, string
);
462 return True if any (multi-byte) character is upper case
464 _PUBLIC_
bool strhasupper_handle(struct smb_iconv_handle
*ic
,
472 s
= next_codepoint_handle(ic
, string
, &c_size
);
478 return true; /* that means it has upper case chars */
485 _PUBLIC_
bool strhasupper(const char *string
)
487 struct smb_iconv_handle
*ic
= get_iconv_handle();
488 return strhasupper_handle(ic
, string
);
491 /***********************************************************************
492 strstr_m - We convert via ucs2 for now.
493 ***********************************************************************/
495 char *strstr_m(const char *src
, const char *findstr
)
498 smb_ucs2_t
*src_w
, *find_w
;
502 size_t converted_size
, findstr_len
= 0;
504 TALLOC_CTX
*frame
; /* Only set up in the iconv case */
506 /* for correctness */
508 return discard_const_p(char, src
);
511 /* Samba does single character findstr calls a *lot*. */
512 if (findstr
[1] == '\0')
513 return strchr_m(src
, *findstr
);
515 /* We optimise for the ascii case, knowing that all our
516 supported multi-byte character sets are ascii-compatible
517 (ie. they match for the first 128 chars) */
519 for (s
= src
; *s
&& !(((unsigned char)s
[0]) & 0x80); s
++) {
520 if (*s
== *findstr
) {
522 findstr_len
= strlen(findstr
);
524 if (strncmp(s
, findstr
, findstr_len
) == 0) {
525 return discard_const_p(char, s
);
533 #if 1 /* def BROKEN_UNICODE_COMPOSE_CHARACTERS */
534 /* 'make check' fails unless we do this */
536 /* With compose characters we must restart from the beginning. JRA. */
540 frame
= talloc_stackframe();
542 if (!push_ucs2_talloc(frame
, &src_w
, src
, &converted_size
)) {
543 DEBUG(0,("strstr_m: src malloc fail\n"));
548 if (!push_ucs2_talloc(frame
, &find_w
, findstr
, &converted_size
)) {
549 DEBUG(0,("strstr_m: find malloc fail\n"));
554 p
= strstr_w(src_w
, find_w
);
562 if (!pull_ucs2_talloc(frame
, &s2
, src_w
, &converted_size
)) {
564 DEBUG(0,("strstr_m: dest malloc fail\n"));
567 retp
= discard_const_p(char, (s
+strlen(s2
)));