1
0
mirror of https://github.com/samba-team/samba.git synced 2024-12-22 13:34:15 +03:00

r15295: Fix some dependencies

Move unistr-specific code to lib/charset/. Remove _m from some places where it's not needed.
This commit is contained in:
Jelmer Vernooij 2006-04-27 16:05:05 +00:00 committed by Gerald (Jerry) Carter
parent e8d69ff8d0
commit 03224e1124
20 changed files with 757 additions and 763 deletions

View File

@ -26,6 +26,7 @@ PRIVATE_PROTO_HEADER = netif/proto.h
OBJ_FILES = \
netif/interface.o \
netif/netif.o
PRIVATE_DEPENDENCIES = LIBSAMBA-UTIL
# End SUBSYSTEM LIBNETIF
##############################
@ -44,19 +45,17 @@ OBJ_FILES = \
################################################
# Start SUBSYSTEM LIBCOMPRESSION
[SUBSYSTEM::LIBCOMPRESSION]
OBJ_FILES = \
compression/mszip.o
OBJ_FILES = compression/mszip.o
# End SUBSYSTEM LIBCOMPRESION
################################################
[SUBSYSTEM::GENCACHE]
PRIVATE_PROTO_HEADER = gencache/gencache.h
OBJ_FILES = \
gencache/gencache.o \
OBJ_FILES = gencache/gencache.o \
[SUBSYSTEM::DB_WRAP]
PUBLIC_PROTO_HEADER = db_wrap_proto.h
PUBLIC_HEADERS = db_wrap.h
OBJ_FILES = db_wrap.o \
gendb.o
PUBLIC_DEPENDENCIES = ldb LIBTDB LDBSAMBA
OBJ_FILES = db_wrap.o gendb.o
PUBLIC_DEPENDENCIES = LIBTDB ldb
PRIVATE_DEPENDENCIES = LDBSAMBA

View File

@ -43,17 +43,16 @@
**/
static const char *charset_name(charset_t ch)
{
const char *ret = NULL;
if (ch == CH_UTF16) ret = "UTF-16LE";
else if (ch == CH_UNIX) ret = lp_unix_charset();
else if (ch == CH_DOS) ret = lp_dos_charset();
else if (ch == CH_DISPLAY) ret = lp_display_charset();
else if (ch == CH_UTF8) ret = "UTF8";
else if (ch == CH_UTF16BE) ret = "UTF-16BE";
if (!ret || !*ret) ret = "ASCII";
return ret;
switch (ch) {
case CH_UTF16: return "UTF-16LE";
case CH_UNIX: return lp_unix_charset();
case CH_DOS: return lp_dos_charset();
case CH_DISPLAY: return lp_display_charset();
case CH_UTF8: return "UTF8";
case CH_UTF16BE: return "UTF-16BE";
default:
return "ASCII";
}
}
static smb_iconv_t conv_handles[NUM_CHARSETS][NUM_CHARSETS];

View File

@ -73,4 +73,9 @@ typedef struct {
#include "lib/charset/charset_proto.h"
/* replace some string functions with multi-byte
versions */
#define strlower(s) strlower_m(s)
#define strupper(s) strupper_m(s)
#endif /* __CHARSET_H__ */

View File

@ -3,7 +3,8 @@
[SUBSYSTEM::CHARSET]
OBJ_FILES = \
iconv.o \
charcnv.o
charcnv.o \
util_unistr.o
PUBLIC_HEADERS = charset.h
PUBLIC_PROTO_HEADER = charset_proto.h
PUBLIC_DEPENDENCIES = ICONV

View File

@ -0,0 +1,615 @@
/*
Unix SMB/CIFS implementation.
Samba utility functions
Copyright (C) Andrew Tridgell 1992-2001
Copyright (C) Simo Sorce 2001
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include "includes.h"
#include "system/iconv.h"
/**
* @file
* @brief Unicode string manipulation
*/
/* these 2 tables define the unicode case handling. They are loaded
at startup either via mmap() or read() from the lib directory */
static void *upcase_table;
static void *lowcase_table;
/*******************************************************************
load the case handling tables
********************************************************************/
static void load_case_tables(void)
{
TALLOC_CTX *mem_ctx;
mem_ctx = talloc_init("load_case_tables");
if (!mem_ctx) {
smb_panic("No memory for case_tables");
}
upcase_table = map_file(data_path(mem_ctx, "upcase.dat"), 0x20000);
lowcase_table = map_file(data_path(mem_ctx, "lowcase.dat"), 0x20000);
talloc_free(mem_ctx);
if (upcase_table == NULL) {
/* try also under codepages for testing purposes */
upcase_table = map_file("codepages/upcase.dat", 0x20000);
if (upcase_table == NULL) {
upcase_table = (void *)-1;
}
}
if (lowcase_table == NULL) {
/* try also under codepages for testing purposes */
lowcase_table = map_file("codepages/lowcase.dat", 0x20000);
if (lowcase_table == NULL) {
lowcase_table = (void *)-1;
}
}
}
/**
Convert a codepoint_t to upper case.
**/
codepoint_t toupper_w(codepoint_t val)
{
if (val < 128) {
return toupper(val);
}
if (upcase_table == NULL) {
load_case_tables();
}
if (upcase_table == (void *)-1) {
return val;
}
if (val & 0xFFFF0000) {
return val;
}
return SVAL(upcase_table, val*2);
}
/**
Convert a codepoint_t to lower case.
**/
codepoint_t tolower_w(codepoint_t val)
{
if (val < 128) {
return tolower(val);
}
if (lowcase_table == NULL) {
load_case_tables();
}
if (lowcase_table == (void *)-1) {
return val;
}
if (val & 0xFFFF0000) {
return val;
}
return SVAL(lowcase_table, val*2);
}
/**
compare two codepoints case insensitively
*/
int codepoint_cmpi(codepoint_t c1, codepoint_t c2)
{
if (c1 == c2 ||
toupper_w(c1) == toupper_w(c2)) {
return 0;
}
return c1 - c2;
}
/**
Case insensitive string compararison
**/
_PUBLIC_ int strcasecmp_m(const char *s1, const char *s2)
{
codepoint_t c1=0, c2=0;
size_t size1, size2;
while (*s1 && *s2) {
c1 = next_codepoint(s1, &size1);
c2 = next_codepoint(s2, &size2);
s1 += size1;
s2 += size2;
if (c1 == c2) {
continue;
}
if (c1 == INVALID_CODEPOINT ||
c2 == INVALID_CODEPOINT) {
/* what else can we do?? */
return strcasecmp(s1, s2);
}
if (toupper_w(c1) != toupper_w(c2)) {
return c1 - c2;
}
}
return *s1 - *s2;
}
/**
* Get the next token from a string, return False if none found.
* Handles double-quotes.
*
* Based on a routine by GJC@VILLAGE.COM.
* Extensively modified by Andrew.Tridgell@anu.edu.au
**/
_PUBLIC_ BOOL next_token(const char **ptr,char *buff, const char *sep, size_t bufsize)
{
const char *s;
BOOL quoted;
size_t len=1;
if (!ptr)
return(False);
s = *ptr;
/* default to simple separators */
if (!sep)
sep = " \t\n\r";
/* find the first non sep char */
while (*s && strchr_m(sep,*s))
s++;
/* nothing left? */
if (! *s)
return(False);
/* copy over the token */
for (quoted = False; len < bufsize && *s && (quoted || !strchr_m(sep,*s)); s++) {
if (*s == '\"') {
quoted = !quoted;
} else {
len++;
*buff++ = *s;
}
}
*ptr = (*s) ? s+1 : s;
*buff = 0;
return(True);
}
/**
Case insensitive string compararison, length limited
**/
_PUBLIC_ int strncasecmp_m(const char *s1, const char *s2, size_t n)
{
codepoint_t c1=0, c2=0;
size_t size1, size2;
while (*s1 && *s2 && n) {
n--;
c1 = next_codepoint(s1, &size1);
c2 = next_codepoint(s2, &size2);
s1 += size1;
s2 += size2;
if (c1 == c2) {
continue;
}
if (c1 == INVALID_CODEPOINT ||
c2 == INVALID_CODEPOINT) {
/* what else can we do?? */
return strcasecmp(s1, s2);
}
if (toupper_w(c1) != toupper_w(c2)) {
return c1 - c2;
}
}
if (n == 0) {
return 0;
}
return *s1 - *s2;
}
/**
* Compare 2 strings.
*
* @note The comparison is case-insensitive.
**/
_PUBLIC_ BOOL strequal_w(const char *s1, const char *s2)
{
if (s1 == s2)
return(True);
if (!s1 || !s2)
return(False);
return strcasecmp_m(s1,s2) == 0;
}
/**
Compare 2 strings (case sensitive).
**/
_PUBLIC_ BOOL strcsequal_w(const char *s1,const char *s2)
{
if (s1 == s2)
return(True);
if (!s1 || !s2)
return(False);
return strcmp(s1,s2) == 0;
}
/**
String replace.
NOTE: oldc and newc must be 7 bit characters
**/
_PUBLIC_ void string_replace_w(char *s, char oldc, char newc)
{
while (*s) {
size_t size;
codepoint_t c = next_codepoint(s, &size);
if (c == oldc) {
*s = newc;
}
s += size;
}
}
/**
Paranoid strcpy into a buffer of given length (includes terminating
zero. Strips out all but 'a-Z0-9' and the character in other_safe_chars
and replaces with '_'. Deliberately does *NOT* check for multibyte
characters. Don't change it !
**/
_PUBLIC_ char *alpha_strcpy(char *dest, const char *src, const char *other_safe_chars, size_t maxlength)
{
size_t len, i;
if (maxlength == 0) {
/* can't fit any bytes at all! */
return NULL;
}
if (!dest) {
DEBUG(0,("ERROR: NULL dest in alpha_strcpy\n"));
return NULL;
}
if (!src) {
*dest = 0;
return dest;
}
len = strlen(src);
if (len >= maxlength)
len = maxlength - 1;
if (!other_safe_chars)
other_safe_chars = "";
for(i = 0; i < len; i++) {
int val = (src[i] & 0xff);
if (isupper(val) || islower(val) || isdigit(val) || strchr_m(other_safe_chars, val))
dest[i] = src[i];
else
dest[i] = '_';
}
dest[i] = '\0';
return dest;
}
/**
Count the number of UCS2 characters in a string. Normally this will
be the same as the number of bytes in a string for single byte strings,
but will be different for multibyte.
**/
_PUBLIC_ size_t strlen_m(const char *s)
{
size_t count = 0;
if (!s) {
return 0;
}
while (*s && !(((uint8_t)*s) & 0x80)) {
s++;
count++;
}
if (!*s) {
return count;
}
while (*s) {
size_t c_size;
codepoint_t c = next_codepoint(s, &c_size);
if (c < 0x10000) {
count += 1;
} else {
count += 2;
}
s += c_size;
}
return count;
}
/**
Work out the number of multibyte chars in a string, including the NULL
terminator.
**/
_PUBLIC_ size_t strlen_m_term(const char *s)
{
if (!s) {
return 0;
}
return strlen_m(s) + 1;
}
/**
Strchr and strrchr_m are a bit complex on general multi-byte strings.
**/
_PUBLIC_ char *strchr_m(const char *s, char c)
{
/* characters below 0x3F are guaranteed to not appear in
non-initial position in multi-byte charsets */
if ((c & 0xC0) == 0) {
return strchr(s, c);
}
while (*s) {
size_t size;
codepoint_t c2 = next_codepoint(s, &size);
if (c2 == c) {
return discard_const(s);
}
s += size;
}
return NULL;
}
/**
* Multibyte-character version of strrchr
*/
_PUBLIC_ char *strrchr_m(const char *s, char c)
{
char *ret = NULL;
/* characters below 0x3F are guaranteed to not appear in
non-initial position in multi-byte charsets */
if ((c & 0xC0) == 0) {
return strrchr(s, c);
}
while (*s) {
size_t size;
codepoint_t c2 = next_codepoint(s, &size);
if (c2 == c) {
ret = discard_const(s);
}
s += size;
}
return ret;
}
/**
return True if any (multi-byte) character is lower case
*/
_PUBLIC_ BOOL strhaslower(const char *string)
{
while (*string) {
size_t c_size;
codepoint_t s;
codepoint_t t;
s = next_codepoint(string, &c_size);
string += c_size;
t = toupper_w(s);
if (s != t) {
return True; /* that means it has lower case chars */
}
}
return False;
}
/**
return True if any (multi-byte) character is upper case
*/
_PUBLIC_ BOOL strhasupper(const char *string)
{
while (*string) {
size_t c_size;
codepoint_t s;
codepoint_t t;
s = next_codepoint(string, &c_size);
string += c_size;
t = tolower_w(s);
if (s != t) {
return True; /* that means it has upper case chars */
}
}
return False;
}
/**
Convert a string to lower case, allocated with talloc
**/
_PUBLIC_ char *strlower_talloc(TALLOC_CTX *ctx, const char *src)
{
size_t size=0;
char *dest;
/* this takes advantage of the fact that upper/lower can't
change the length of a character by more than 1 byte */
dest = talloc_size(ctx, 2*(strlen(src))+1);
if (dest == NULL) {
return NULL;
}
while (*src) {
size_t c_size;
codepoint_t c = next_codepoint(src, &c_size);
src += c_size;
c = tolower_w(c);
c_size = push_codepoint(dest+size, c);
if (c_size == -1) {
talloc_free(dest);
return NULL;
}
size += c_size;
}
dest[size] = 0;
return dest;
}
/**
Convert a string to UPPER case, allocated with talloc
**/
_PUBLIC_ char *strupper_talloc(TALLOC_CTX *ctx, const char *src)
{
size_t size=0;
char *dest;
if (!src) {
return NULL;
}
/* this takes advantage of the fact that upper/lower can't
change the length of a character by more than 1 byte */
dest = talloc_size(ctx, 2*(strlen(src))+1);
if (dest == NULL) {
return NULL;
}
while (*src) {
size_t c_size;
codepoint_t c = next_codepoint(src, &c_size);
src += c_size;
c = toupper_w(c);
c_size = push_codepoint(dest+size, c);
if (c_size == -1) {
talloc_free(dest);
return NULL;
}
size += c_size;
}
dest[size] = 0;
return dest;
}
/**
Convert a string to lower case.
**/
_PUBLIC_ void strlower_m(char *s)
{
char *d;
/* this is quite a common operation, so we want it to be
fast. We optimise for the ascii case, knowing that all our
supported multi-byte character sets are ascii-compatible
(ie. they match for the first 128 chars) */
while (*s && !(((uint8_t)*s) & 0x80)) {
*s = tolower((uint8_t)*s);
s++;
}
if (!*s)
return;
d = s;
while (*s) {
size_t c_size, c_size2;
codepoint_t c = next_codepoint(s, &c_size);
c_size2 = push_codepoint(d, tolower_w(c));
if (c_size2 > c_size) {
DEBUG(0,("FATAL: codepoint 0x%x (0x%x) expanded from %d to %d bytes in strlower_m\n",
c, tolower_w(c), (int)c_size, (int)c_size2));
smb_panic("codepoint expansion in strlower_m\n");
}
s += c_size;
d += c_size2;
}
*d = 0;
}
/**
Convert a string to UPPER case.
**/
_PUBLIC_ void strupper_m(char *s)
{
char *d;
/* this is quite a common operation, so we want it to be
fast. We optimise for the ascii case, knowing that all our
supported multi-byte character sets are ascii-compatible
(ie. they match for the first 128 chars) */
while (*s && !(((uint8_t)*s) & 0x80)) {
*s = toupper((uint8_t)*s);
s++;
}
if (!*s)
return;
d = s;
while (*s) {
size_t c_size, c_size2;
codepoint_t c = next_codepoint(s, &c_size);
c_size2 = push_codepoint(d, toupper_w(c));
if (c_size2 > c_size) {
DEBUG(0,("FATAL: codepoint 0x%x (0x%x) expanded from %d to %d bytes in strupper_m\n",
c, toupper_w(c), (int)c_size, (int)c_size2));
smb_panic("codepoint expansion in strupper_m\n");
}
s += c_size;
d += c_size2;
}
*d = 0;
}

View File

@ -59,11 +59,6 @@ static void ldb_wrap_debug(void *context, enum ldb_debug_level level,
free(s);
}
int wrap_caseless_cmp(void *context, const char *s1, const char *s2)
{
return strcasecmp_m(s1, s2);
}
char *wrap_casefold(void *context, void *mem_ctx, const char *s)
{
return strupper_talloc(mem_ctx, s);

View File

@ -71,6 +71,7 @@ OBJ_FILES = \
################################################
# Start MODULE ldb_ildap
[MODULE::ldb_ildap]
ENABLE = NO
SUBSYSTEM = ldb
INIT_FUNCTION = ldb_ildap_init
ALIASES = ldapi ldaps ldap
@ -164,7 +165,7 @@ PUBLIC_HEADERS = include/ldb.h include/ldb_errors.h
# Start SUBSYSTEM LDBSAMBA
[SUBSYSTEM::LDBSAMBA]
PRIVATE_PROTO_HEADER = samba/ldif_handlers.h
PUBLIC_DEPENDENCIES = LIB_SECURITY SAMDB
PUBLIC_DEPENDENCIES = LIBSECURITY SAMDB
OBJ_FILES = \
samba/ldif_handlers.o
# End SUBSYSTEM LDBSAMBA
@ -176,6 +177,7 @@ OBJ_FILES = \
OBJ_FILES= \
tools/cmdline.o
PUBLIC_DEPENDENCIES = ldb LIBSAMBA-UTIL LIBPOPT POPT_SAMBA POPT_CREDENTIALS
PRIVATE_DEPENDENCIES = gensec
# End SUBSYSTEM LIBLDB_CMDLINE
################################################

View File

@ -5,6 +5,7 @@
OBJ_FILES = \
messaging.o
PUBLIC_DEPENDENCIES = \
DB_WRAP \
NDR_IRPC \
UNIX_PRIVS
# End SUBSYSTEM MESSAGING

View File

@ -98,7 +98,7 @@ OBJ_FILES = \
reg_samba.o \
patchfile.o
PUBLIC_DEPENDENCIES = \
LIBSAMBA-UTIL
LIBSAMBA-UTIL CHARSET
PRIVATE_PROTO_HEADER = registry_proto.h
PUBLIC_HEADERS = registry.h
# End MODULE registry_ldb
@ -132,7 +132,8 @@ MANPAGE = man/regpatch.1
INSTALLDIR = BINDIR
OBJ_FILES = tools/regshell.o
PRIVATE_DEPENDENCIES = \
LIBSAMBA-CONFIG LIBPOPT registry POPT_SAMBA POPT_CREDENTIALS LIBREADLINE
LIBSAMBA-CONFIG LIBPOPT registry POPT_SAMBA POPT_CREDENTIALS \
SMBREADLINE
MANPAGE = man/regshell.1
# End BINARY regshell
################################################

View File

@ -16,7 +16,3 @@ OBJ_FILES = replace.o \
PUBLIC_DEPENDENCIES = REPLACE_READDIR
# End SUBSYSTEM LIBREPLACE
##############################
[SUBSYSTEM::SMBREADLINE]
OBJ_FILES = readline.o
PRIVATE_PROTO_HEADER = readline.h

View File

@ -72,10 +72,10 @@ AC_MSG_CHECKING(whether to use extern readline)
if test x"$EXTERNAL_READLINE" = x"yes"; then
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_LIBREADLINE,1,[Whether the system has readline])
SMB_SUBSYSTEM(LIBREADLINE, [], [SMBREADLINE EXT_READLINE])
SMB_EXT_LIB(EXT_READLINE, [${TERMLIBS}])
SMB_ENABLE(EXT_READLINE,YES)
SMB_SUBSYSTEM(SMBREADLINE, [lib/replace/readline.o], [READLINE])
SMB_EXT_LIB(READLINE, [${TERMLIBS}])
SMB_ENABLE(READLINE,YES)
else
SMB_SUBSYSTEM(LIBREADLINE, [], [SMBREADLINE])
SMB_SUBSYSTEM(SMBREADLINE, [lib/replace/readline.o], [])
AC_MSG_RESULT(no)
fi

View File

@ -7,6 +7,7 @@ OUTPUT_TYPE = MERGEDOBJ
OBJ_FILES = \
socket_ipv4.o
PUBLIC_DEPENDENCIES = EXT_SOCKET
PRIVATE_DEPENDENCIES = LIBSAMBA-ERRORS
# End MODULE socket_ipv4
################################################
@ -41,6 +42,6 @@ OBJ_FILES = \
connect_multi.o \
connect.o
PUBLIC_DEPENDENCIES = LIBTALLOC
PRIVATE_DEPENDENCIES = LIBCLI_RESOLVE SOCKET_WRAPPER
PRIVATE_DEPENDENCIES = SOCKET_WRAPPER LIBCLI_COMPOSITE
# End SUBSYSTEM SOCKET
################################################

View File

@ -2,7 +2,7 @@
################################################
# Start SUBSYSTEM LIBPACKET
[SUBSYSTEM::LIBPACKET]
OBJ_FILES = \
packet.o
OBJ_FILES = packet.o
PRIVATE_DEPENDENCIES = LIBTLS
# End SUBSYSTEM LIBPACKET
################################################

View File

@ -19,7 +19,6 @@ OBJ_FILES = xfile.o \
dprintf.o \
util_str.o \
util_strlist.o \
util_unistr.o \
util_file.o \
data_blob.o \
util.o \
@ -32,10 +31,8 @@ OBJ_FILES = xfile.o \
idtree.o \
module.o
PUBLIC_DEPENDENCIES = \
CHARSET LIBREPLACE LIBCRYPTO DL LIBTALLOC \
SOCKET_WRAPPER \
# for the base64 functions
ldb
LIBREPLACE LIBCRYPTO DL LIBTALLOC \
SOCKET_WRAPPER
[SUBSYSTEM::PIDFILE]
PRIVATE_PROTO_HEADER = pidfile.h

View File

@ -42,9 +42,4 @@
#endif /* !_SPLINT_ */
/* replace some string functions with multi-byte
versions */
#define strlower(s) strlower_m(s)
#define strupper(s) strupper_m(s)
#endif

View File

@ -52,7 +52,7 @@ _PUBLIC_ struct hostent *sys_gethostbyname(const char *name)
/* Does this name have any dots in it? If so, make no change */
if (strchr_m(name, '.'))
if (strchr(name, '.'))
return(gethostbyname(name));
/* Get my hostname, which should have domain name
@ -62,7 +62,7 @@ _PUBLIC_ struct hostent *sys_gethostbyname(const char *name)
gethostname(hostname, sizeof(hostname) - 1);
hostname[sizeof(hostname) - 1] = 0;
if ((domain = strchr_m(hostname, '.')) == NULL)
if ((domain = strchr(hostname, '.')) == NULL)
return(gethostbyname(name));
/* Attach domain name to query and do modified query.

View File

@ -289,7 +289,7 @@ _PUBLIC_ char* get_myname(void)
hostname[host_name_max] = '\0';
/* split off any parts after an initial . */
p = strchr_m(hostname,'.');
p = strchr(hostname,'.');
if (p)
*p = 0;
@ -311,7 +311,7 @@ _PUBLIC_ BOOL is_ipaddress(const char *str)
pure_address = False;
/* Check that a pure number is not misinterpreted as an IP */
pure_address = pure_address && (strchr_m(str, '.') != NULL);
pure_address = pure_address && (strchr(str, '.') != NULL);
return pure_address;
}

View File

@ -90,7 +90,7 @@ _PUBLIC_ void set_socket_options(int fd, const char *options)
char *p;
BOOL got_value = False;
if ((p = strchr_m(tok,'='))) {
if ((p = strchr(tok,'='))) {
*p = 0;
value = atoi(p+1);
got_value = True;

View File

@ -23,208 +23,16 @@
*/
#include "includes.h"
#include "system/iconv.h"
#include "smb.h"
#include "pstring.h"
#include "lib/ldb/include/ldb.h"
#include "system/iconv.h"
/**
* @file
* @brief String utilities.
**/
/**
* Get the next token from a string, return False if none found.
* Handles double-quotes.
*
* Based on a routine by GJC@VILLAGE.COM.
* Extensively modified by Andrew.Tridgell@anu.edu.au
**/
_PUBLIC_ BOOL next_token(const char **ptr,char *buff, const char *sep, size_t bufsize)
{
const char *s;
BOOL quoted;
size_t len=1;
if (!ptr)
return(False);
s = *ptr;
/* default to simple separators */
if (!sep)
sep = " \t\n\r";
/* find the first non sep char */
while (*s && strchr_m(sep,*s))
s++;
/* nothing left? */
if (! *s)
return(False);
/* copy over the token */
for (quoted = False; len < bufsize && *s && (quoted || !strchr_m(sep,*s)); s++) {
if (*s == '\"') {
quoted = !quoted;
} else {
len++;
*buff++ = *s;
}
}
*ptr = (*s) ? s+1 : s;
*buff = 0;
return(True);
}
/**
Case insensitive string compararison
**/
_PUBLIC_ int strcasecmp_m(const char *s1, const char *s2)
{
codepoint_t c1=0, c2=0;
size_t size1, size2;
while (*s1 && *s2) {
c1 = next_codepoint(s1, &size1);
c2 = next_codepoint(s2, &size2);
s1 += size1;
s2 += size2;
if (c1 == c2) {
continue;
}
if (c1 == INVALID_CODEPOINT ||
c2 == INVALID_CODEPOINT) {
/* what else can we do?? */
return strcasecmp(s1, s2);
}
if (toupper_w(c1) != toupper_w(c2)) {
return c1 - c2;
}
}
return *s1 - *s2;
}
/**
Case insensitive string compararison, length limited
**/
_PUBLIC_ int strncasecmp_m(const char *s1, const char *s2, size_t n)
{
codepoint_t c1=0, c2=0;
size_t size1, size2;
while (*s1 && *s2 && n) {
n--;
c1 = next_codepoint(s1, &size1);
c2 = next_codepoint(s2, &size2);
s1 += size1;
s2 += size2;
if (c1 == c2) {
continue;
}
if (c1 == INVALID_CODEPOINT ||
c2 == INVALID_CODEPOINT) {
/* what else can we do?? */
return strcasecmp(s1, s2);
}
if (toupper_w(c1) != toupper_w(c2)) {
return c1 - c2;
}
}
if (n == 0) {
return 0;
}
return *s1 - *s2;
}
/**
* Compare 2 strings.
*
* @note The comparison is case-insensitive.
**/
_PUBLIC_ BOOL strequal(const char *s1, const char *s2)
{
if (s1 == s2)
return(True);
if (!s1 || !s2)
return(False);
return strcasecmp_m(s1,s2) == 0;
}
/**
Compare 2 strings (case sensitive).
**/
_PUBLIC_ BOOL strcsequal(const char *s1,const char *s2)
{
if (s1 == s2)
return(True);
if (!s1 || !s2)
return(False);
return strcmp(s1,s2) == 0;
}
/**
Do a case-insensitive, whitespace-ignoring string compare.
**/
_PUBLIC_ int strwicmp(const char *psz1, const char *psz2)
{
/* if BOTH strings are NULL, return TRUE, if ONE is NULL return */
/* appropriate value. */
if (psz1 == psz2)
return (0);
else if (psz1 == NULL)
return (-1);
else if (psz2 == NULL)
return (1);
/* sync the strings on first non-whitespace */
while (1) {
while (isspace((int)*psz1))
psz1++;
while (isspace((int)*psz2))
psz2++;
if (toupper((unsigned char)*psz1) != toupper((unsigned char)*psz2)
|| *psz1 == '\0'
|| *psz2 == '\0')
break;
psz1++;
psz2++;
}
return (*psz1 - *psz2);
}
/**
String replace.
NOTE: oldc and newc must be 7 bit characters
**/
_PUBLIC_ void string_replace(char *s, char oldc, char newc)
{
while (*s) {
size_t size;
codepoint_t c = next_codepoint(s, &size);
if (c == oldc) {
*s = newc;
}
s += size;
}
}
/**
Trim the specified elements off the front and back of a string.
@ -363,52 +171,6 @@ _PUBLIC_ char *safe_strcat(char *dest, const char *src, size_t maxlength)
return dest;
}
/**
Paranoid strcpy into a buffer of given length (includes terminating
zero. Strips out all but 'a-Z0-9' and the character in other_safe_chars
and replaces with '_'. Deliberately does *NOT* check for multibyte
characters. Don't change it !
**/
_PUBLIC_ char *alpha_strcpy(char *dest, const char *src, const char *other_safe_chars, size_t maxlength)
{
size_t len, i;
if (maxlength == 0) {
/* can't fit any bytes at all! */
return NULL;
}
if (!dest) {
DEBUG(0,("ERROR: NULL dest in alpha_strcpy\n"));
return NULL;
}
if (!src) {
*dest = 0;
return dest;
}
len = strlen(src);
if (len >= maxlength)
len = maxlength - 1;
if (!other_safe_chars)
other_safe_chars = "";
for(i = 0; i < len; i++) {
int val = (src[i] & 0xff);
if (isupper(val) || islower(val) || isdigit(val) || strchr_m(other_safe_chars, val))
dest[i] = src[i];
else
dest[i] = '_';
}
dest[i] = '\0';
return dest;
}
/**
Like strncpy but always null terminates. Make sure there is room!
The variable n should always be one less than the available size.
@ -454,12 +216,12 @@ _PUBLIC_ size_t strhex_to_str(char *p, size_t len, const char *strhex)
continue;
}
if (!(p1 = strchr_m(hexchars, toupper((unsigned char)strhex[i]))))
if (!(p1 = strchr(hexchars, toupper((unsigned char)strhex[i]))))
break;
i++; /* next hex digit */
if (!(p2 = strchr_m(hexchars, toupper((unsigned char)strhex[i]))))
if (!(p2 = strchr(hexchars, toupper((unsigned char)strhex[i]))))
break;
/* get the two nybbles */
@ -662,294 +424,6 @@ _PUBLIC_ void all_string_sub(char *s,const char *pattern,const char *insert, siz
}
/**
Strchr and strrchr_m are a bit complex on general multi-byte strings.
**/
_PUBLIC_ char *strchr_m(const char *s, char c)
{
/* characters below 0x3F are guaranteed to not appear in
non-initial position in multi-byte charsets */
if ((c & 0xC0) == 0) {
return strchr(s, c);
}
while (*s) {
size_t size;
codepoint_t c2 = next_codepoint(s, &size);
if (c2 == c) {
return discard_const(s);
}
s += size;
}
return NULL;
}
/**
* Multibyte-character version of strrchr
*/
_PUBLIC_ char *strrchr_m(const char *s, char c)
{
char *ret = NULL;
/* characters below 0x3F are guaranteed to not appear in
non-initial position in multi-byte charsets */
if ((c & 0xC0) == 0) {
return strrchr(s, c);
}
while (*s) {
size_t size;
codepoint_t c2 = next_codepoint(s, &size);
if (c2 == c) {
ret = discard_const(s);
}
s += size;
}
return ret;
}
/**
return True if any (multi-byte) character is lower case
*/
_PUBLIC_ BOOL strhaslower(const char *string)
{
while (*string) {
size_t c_size;
codepoint_t s;
codepoint_t t;
s = next_codepoint(string, &c_size);
string += c_size;
t = toupper_w(s);
if (s != t) {
return True; /* that means it has lower case chars */
}
}
return False;
}
/**
return True if any (multi-byte) character is upper case
*/
_PUBLIC_ BOOL strhasupper(const char *string)
{
while (*string) {
size_t c_size;
codepoint_t s;
codepoint_t t;
s = next_codepoint(string, &c_size);
string += c_size;
t = tolower_w(s);
if (s != t) {
return True; /* that means it has upper case chars */
}
}
return False;
}
/**
Convert a string to lower case, allocated with talloc
**/
_PUBLIC_ char *strlower_talloc(TALLOC_CTX *ctx, const char *src)
{
size_t size=0;
char *dest;
/* this takes advantage of the fact that upper/lower can't
change the length of a character by more than 1 byte */
dest = talloc_size(ctx, 2*(strlen(src))+1);
if (dest == NULL) {
return NULL;
}
while (*src) {
size_t c_size;
codepoint_t c = next_codepoint(src, &c_size);
src += c_size;
c = tolower_w(c);
c_size = push_codepoint(dest+size, c);
if (c_size == -1) {
talloc_free(dest);
return NULL;
}
size += c_size;
}
dest[size] = 0;
return dest;
}
/**
Convert a string to UPPER case, allocated with talloc
**/
_PUBLIC_ char *strupper_talloc(TALLOC_CTX *ctx, const char *src)
{
size_t size=0;
char *dest;
if (!src) {
return NULL;
}
/* this takes advantage of the fact that upper/lower can't
change the length of a character by more than 1 byte */
dest = talloc_size(ctx, 2*(strlen(src))+1);
if (dest == NULL) {
return NULL;
}
while (*src) {
size_t c_size;
codepoint_t c = next_codepoint(src, &c_size);
src += c_size;
c = toupper_w(c);
c_size = push_codepoint(dest+size, c);
if (c_size == -1) {
talloc_free(dest);
return NULL;
}
size += c_size;
}
dest[size] = 0;
return dest;
}
/**
Convert a string to lower case.
**/
_PUBLIC_ void strlower_m(char *s)
{
char *d;
/* this is quite a common operation, so we want it to be
fast. We optimise for the ascii case, knowing that all our
supported multi-byte character sets are ascii-compatible
(ie. they match for the first 128 chars) */
while (*s && !(((uint8_t)*s) & 0x80)) {
*s = tolower((uint8_t)*s);
s++;
}
if (!*s)
return;
d = s;
while (*s) {
size_t c_size, c_size2;
codepoint_t c = next_codepoint(s, &c_size);
c_size2 = push_codepoint(d, tolower_w(c));
if (c_size2 > c_size) {
DEBUG(0,("FATAL: codepoint 0x%x (0x%x) expanded from %d to %d bytes in strlower_m\n",
c, tolower_w(c), (int)c_size, (int)c_size2));
smb_panic("codepoint expansion in strlower_m\n");
}
s += c_size;
d += c_size2;
}
*d = 0;
}
/**
Convert a string to UPPER case.
**/
_PUBLIC_ void strupper_m(char *s)
{
char *d;
/* this is quite a common operation, so we want it to be
fast. We optimise for the ascii case, knowing that all our
supported multi-byte character sets are ascii-compatible
(ie. they match for the first 128 chars) */
while (*s && !(((uint8_t)*s) & 0x80)) {
*s = toupper((uint8_t)*s);
s++;
}
if (!*s)
return;
d = s;
while (*s) {
size_t c_size, c_size2;
codepoint_t c = next_codepoint(s, &c_size);
c_size2 = push_codepoint(d, toupper_w(c));
if (c_size2 > c_size) {
DEBUG(0,("FATAL: codepoint 0x%x (0x%x) expanded from %d to %d bytes in strupper_m\n",
c, toupper_w(c), (int)c_size, (int)c_size2));
smb_panic("codepoint expansion in strupper_m\n");
}
s += c_size;
d += c_size2;
}
*d = 0;
}
/**
Count the number of UCS2 characters in a string. Normally this will
be the same as the number of bytes in a string for single byte strings,
but will be different for multibyte.
**/
_PUBLIC_ size_t strlen_m(const char *s)
{
size_t count = 0;
if (!s) {
return 0;
}
while (*s && !(((uint8_t)*s) & 0x80)) {
s++;
count++;
}
if (!*s) {
return count;
}
while (*s) {
size_t c_size;
codepoint_t c = next_codepoint(s, &c_size);
if (c < 0x10000) {
count += 1;
} else {
count += 2;
}
s += c_size;
}
return count;
}
/**
Work out the number of multibyte chars in a string, including the NULL
terminator.
**/
_PUBLIC_ size_t strlen_m_term(const char *s)
{
if (!s) {
return 0;
}
return strlen_m(s) + 1;
}
/**
Unescape a URL encoded string, in place.
@ -991,32 +465,6 @@ _PUBLIC_ void rfc1738_unescape(char *buf)
}
}
/**
* Decode a base64 string into a DATA_BLOB - simple and slow algorithm
**/
_PUBLIC_ DATA_BLOB base64_decode_data_blob(TALLOC_CTX *mem_ctx, const char *s)
{
DATA_BLOB ret = data_blob_talloc(mem_ctx, s, strlen(s)+1);
ret.length = ldb_base64_decode((char *)ret.data);
return ret;
}
/**
* Decode a base64 string in-place - wrapper for the above
**/
_PUBLIC_ void base64_decode_inplace(char *s)
{
ldb_base64_decode(s);
}
/**
* Encode a base64 string into a talloc()ed string caller to free.
**/
_PUBLIC_ char *base64_encode_data_blob(TALLOC_CTX *mem_ctx, DATA_BLOB data)
{
return ldb_base64_encode(mem_ctx, (const char *)data.data, data.length);
}
#ifdef VALGRIND
size_t valgrind_strlen(const char *s)
{
@ -1274,3 +722,97 @@ _PUBLIC_ BOOL conv_str_u64(const char * str, uint64_t * val)
*val = (uint64_t)lval;
return True;
}
/**
return the number of bytes occupied by a buffer in CH_UTF16 format
the result includes the null termination
**/
_PUBLIC_ size_t utf16_len(const void *buf)
{
size_t len;
for (len = 0; SVAL(buf,len); len += 2) ;
return len + 2;
}
/**
return the number of bytes occupied by a buffer in CH_UTF16 format
the result includes the null termination
limited by 'n' bytes
**/
_PUBLIC_ size_t utf16_len_n(const void *src, size_t n)
{
size_t len;
for (len = 0; (len+2 < n) && SVAL(src, len); len += 2) ;
if (len+2 <= n) {
len += 2;
}
return len;
}
_PUBLIC_ size_t ucs2_align(const void *base_ptr, const void *p, int flags)
{
if (flags & (STR_NOALIGN|STR_ASCII))
return 0;
return PTR_DIFF(p, base_ptr) & 1;
}
/**
Do a case-insensitive, whitespace-ignoring string compare.
**/
_PUBLIC_ int strwicmp(const char *psz1, const char *psz2)
{
/* if BOTH strings are NULL, return TRUE, if ONE is NULL return */
/* appropriate value. */
if (psz1 == psz2)
return (0);
else if (psz1 == NULL)
return (-1);
else if (psz2 == NULL)
return (1);
/* sync the strings on first non-whitespace */
while (1) {
while (isspace((int)*psz1))
psz1++;
while (isspace((int)*psz2))
psz2++;
if (toupper((unsigned char)*psz1) != toupper((unsigned char)*psz2)
|| *psz1 == '\0'
|| *psz2 == '\0')
break;
psz1++;
psz2++;
}
return (*psz1 - *psz2);
}
/**
String replace.
**/
_PUBLIC_ void string_replace(char *s, char oldc, char newc)
{
while (*s) {
s++;
if (*s == oldc) *s = newc;
}
}
/**
* Compare 2 strings.
*
* @note The comparison is case-insensitive.
**/
_PUBLIC_ BOOL strequal(const char *s1, const char *s2)
{
if (s1 == s2)
return(True);
if (!s1 || !s2)
return(False);
return strcasecmp(s1,s2) == 0;
}

View File

@ -1,155 +0,0 @@
/*
Unix SMB/CIFS implementation.
Samba utility functions
Copyright (C) Andrew Tridgell 1992-2001
Copyright (C) Simo Sorce 2001
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include "includes.h"
#include "system/iconv.h"
/**
* @file
* @brief Unicode string manipulation
*/
/* these 2 tables define the unicode case handling. They are loaded
at startup either via mmap() or read() from the lib directory */
static void *upcase_table;
static void *lowcase_table;
/*******************************************************************
load the case handling tables
********************************************************************/
static void load_case_tables(void)
{
TALLOC_CTX *mem_ctx;
mem_ctx = talloc_init("load_case_tables");
if (!mem_ctx) {
smb_panic("No memory for case_tables");
}
upcase_table = map_file(data_path(mem_ctx, "upcase.dat"), 0x20000);
lowcase_table = map_file(data_path(mem_ctx, "lowcase.dat"), 0x20000);
talloc_free(mem_ctx);
if (upcase_table == NULL) {
/* try also under codepages for testing purposes */
upcase_table = map_file("codepages/upcase.dat", 0x20000);
if (upcase_table == NULL) {
upcase_table = (void *)-1;
}
}
if (lowcase_table == NULL) {
/* try also under codepages for testing purposes */
lowcase_table = map_file("codepages/lowcase.dat", 0x20000);
if (lowcase_table == NULL) {
lowcase_table = (void *)-1;
}
}
}
/**
Convert a codepoint_t to upper case.
**/
codepoint_t toupper_w(codepoint_t val)
{
if (val < 128) {
return toupper(val);
}
if (upcase_table == NULL) {
load_case_tables();
}
if (upcase_table == (void *)-1) {
return val;
}
if (val & 0xFFFF0000) {
return val;
}
return SVAL(upcase_table, val*2);
}
/**
Convert a codepoint_t to lower case.
**/
codepoint_t tolower_w(codepoint_t val)
{
if (val < 128) {
return tolower(val);
}
if (lowcase_table == NULL) {
load_case_tables();
}
if (lowcase_table == (void *)-1) {
return val;
}
if (val & 0xFFFF0000) {
return val;
}
return SVAL(lowcase_table, val*2);
}
/**
return the number of bytes occupied by a buffer in CH_UTF16 format
the result includes the null termination
**/
size_t utf16_len(const void *buf)
{
size_t len;
for (len = 0; SVAL(buf,len); len += 2) ;
return len + 2;
}
/**
return the number of bytes occupied by a buffer in CH_UTF16 format
the result includes the null termination
limited by 'n' bytes
**/
size_t utf16_len_n(const void *src, size_t n)
{
size_t len;
for (len = 0; (len+2 < n) && SVAL(src, len); len += 2) ;
if (len+2 <= n) {
len += 2;
}
return len;
}
size_t ucs2_align(const void *base_ptr, const void *p, int flags)
{
if (flags & (STR_NOALIGN|STR_ASCII))
return 0;
return PTR_DIFF(p, base_ptr) & 1;
}
/**
compare two codepoints case insensitively
*/
int codepoint_cmpi(codepoint_t c1, codepoint_t c2)
{
if (c1 == c2 ||
toupper_w(c1) == toupper_w(c2)) {
return 0;
}
return c1 - c2;
}