#include <talloc.h>
/* this defines the charset types used in samba */
-typedef enum {CH_UTF16LE=0, CH_UTF16=0, CH_UNIX, CH_DISPLAY, CH_DOS, CH_UTF8, CH_UTF16BE, CH_UTF16MUNGED} charset_t;
+typedef enum {CH_UTF16LE=0, CH_UTF16=0, CH_UNIX, CH_DOS, CH_UTF8, CH_UTF16BE, CH_UTF16MUNGED} charset_t;
#define NUM_CHARSETS 7
typedef uint16_t smb_ucs2_t;
-/*
- * SMB UCS2 (16-bit unicode) internal type.
- * smb_ucs2_t is *always* in little endian format.
- */
-
#ifdef WORDS_BIGENDIAN
#define UCS2_SHIFT 8
#else
/* turn a 7 bit character into a ucs2 character */
#define UCS2_CHAR(c) ((c) << UCS2_SHIFT)
-/* return an ascii version of a ucs2 character */
-#define UCS2_TO_CHAR(c) (((c) >> UCS2_SHIFT) & 0xff)
-
-/* Copy into a smb_ucs2_t from a possibly unaligned buffer. Return the copied smb_ucs2_t */
-#define COPY_UCS2_CHAR(dest,src) (((unsigned char *)(dest))[0] = ((unsigned char *)(src))[0],\
- ((unsigned char *)(dest))[1] = ((unsigned char *)(src))[1], (dest))
-
-
-
/*
* for each charset we have a function that pulls from that charset to
* a ucs2 buffer, and a function that pushes to a ucs2 buffer
char **outbuf, size_t *outbytesleft);
size_t (*push)(void *, const char **inbuf, size_t *inbytesleft,
char **outbuf, size_t *outbytesleft);
- struct charset_functions *prev, *next;
+ bool samba_internal_charset;
};
/* this type is used for manipulating unicode codepoints */
#define INVALID_CODEPOINT ((codepoint_t)-1)
-/*
- * This is auxiliary struct used by source/script/gen-8-bit-gap.sh script
- * during generation of an encoding table for charset module
- * */
-
-struct charset_gap_table {
- uint16_t start;
- uint16_t end;
- int32_t idx;
-};
-
-
/* generic iconv conversion structure */
typedef struct smb_iconv_s {
size_t (*direct)(void *cd, const char **inbuf, size_t *inbytesleft,
#define STR_LEN_NOTERM 256 /* the length field is the unterminated length */
struct loadparm_context;
-struct smb_iconv_convenience;
-
-/* replace some string functions with multi-byte
- versions */
-#define strlower(s) strlower_m(s)
-#define strupper(s) strupper_m(s)
+struct smb_iconv_handle;
char *strchr_m(const char *s, char c);
+/**
+ * Calculate the number of units (8 or 16-bit, depending on the
+ * destination charset), that would be needed to convert the input
+ * string which is expected to be in in src_charset encoding to the
+ * destination charset (which should be a unicode charset).
+ */
+size_t strlen_m_ext_handle(struct smb_iconv_handle *ic,
+ const char *s, charset_t src_charset, charset_t dst_charset);
+size_t strlen_m_ext(const char *s, charset_t src_charset, charset_t dst_charset);
+size_t strlen_m_ext_term(const char *s, charset_t src_charset,
+ charset_t dst_charset);
size_t strlen_m_term(const char *s);
size_t strlen_m_term_null(const char *s);
size_t strlen_m(const char *s);
char *alpha_strcpy(char *dest, const char *src, const char *other_safe_chars, size_t maxlength);
void string_replace_m(char *s, char oldc, char newc);
-bool strcsequal_m(const char *s1,const char *s2);
+bool strcsequal(const char *s1,const char *s2);
bool strequal_m(const char *s1, const char *s2);
int strncasecmp_m(const char *s1, const char *s2, size_t n);
+int strncasecmp_m_handle(struct smb_iconv_handle *iconv_handle,
+ const char *s1, const char *s2, size_t n);
bool next_token(const char **ptr,char *buff, const char *sep, size_t bufsize);
+int strcasecmp_m_handle(struct smb_iconv_handle *iconv_handle,
+ const char *s1, const char *s2);
int strcasecmp_m(const char *s1, const char *s2);
size_t count_chars_m(const char *s, char c);
-void strupper_m(char *s);
-void strlower_m(char *s);
char *strupper_talloc(TALLOC_CTX *ctx, const char *src);
char *talloc_strdup_upper(TALLOC_CTX *ctx, const char *src);
+char *strupper_talloc_n_handle(struct smb_iconv_handle *iconv_handle,
+ TALLOC_CTX *ctx, const char *src, size_t n);
char *strupper_talloc_n(TALLOC_CTX *ctx, const char *src, size_t n);
+ char *strlower_talloc_handle(struct smb_iconv_handle *iconv_handle,
+ TALLOC_CTX *ctx, const char *src);
char *strlower_talloc(TALLOC_CTX *ctx, const char *src);
bool strhasupper(const char *string);
+bool strhaslower_handle(struct smb_iconv_handle *ic,
+ const char *string);
bool strhaslower(const char *string);
+bool strhasupper_handle(struct smb_iconv_handle *ic,
+ const char *string);
char *strrchr_m(const char *s, char c);
char *strchr_m(const char *s, char c);
+char *strstr_m(const char *src, const char *findstr);
bool push_ascii_talloc(TALLOC_CTX *ctx, char **dest, const char *src, size_t *converted_size);
bool push_ucs2_talloc(TALLOC_CTX *ctx, smb_ucs2_t **dest, const char *src, size_t *converted_size);
ssize_t pull_string(char *dest, const void *src, size_t dest_len, size_t src_len, int flags);
bool convert_string_talloc(TALLOC_CTX *ctx,
- charset_t from, charset_t to,
- void const *src, size_t srclen,
- void *dest, size_t *converted_size,
- bool allow_badcharcnv);
-
-size_t convert_string(charset_t from, charset_t to,
- void const *src, size_t srclen,
- void *dest, size_t destlen, bool allow_badcharcnv);
-
-ssize_t iconv_talloc(TALLOC_CTX *mem_ctx,
- smb_iconv_t cd,
- void const *src, size_t srclen,
- void *dest);
-
-extern struct smb_iconv_convenience *global_iconv_convenience;
-
+ charset_t from, charset_t to,
+ void const *src, size_t srclen,
+ void *dest, size_t *converted_size);
+
+bool convert_string(charset_t from, charset_t to,
+ void const *src, size_t srclen,
+ void *dest, size_t destlen,
+ size_t *converted_size);
+bool convert_string_error(charset_t from, charset_t to,
+ void const *src, size_t srclen,
+ void *dest, size_t destlen,
+ size_t *converted_size);
+
+extern struct smb_iconv_handle *global_iconv_handle;
+struct smb_iconv_handle *get_iconv_handle(void);
+struct smb_iconv_handle *get_iconv_testing_handle(TALLOC_CTX *mem_ctx,
+ const char *dos_charset,
+ const char *unix_charset,
+ bool use_builtin_handlers);
+smb_iconv_t get_conv_handle(struct smb_iconv_handle *ic,
+ charset_t from, charset_t to);
+const char *charset_name(struct smb_iconv_handle *ic, charset_t ch);
+
+codepoint_t next_codepoint_ext(const char *str, size_t len,
+ charset_t src_charset, size_t *size);
codepoint_t next_codepoint(const char *str, size_t *size);
ssize_t push_codepoint(char *str, codepoint_t c);
/* codepoints */
-codepoint_t next_codepoint_convenience(struct smb_iconv_convenience *ic,
+codepoint_t next_codepoint_handle_ext(struct smb_iconv_handle *ic,
+ const char *str, size_t len,
+ charset_t src_charset,
+ size_t *size);
+codepoint_t next_codepoint_handle(struct smb_iconv_handle *ic,
const char *str, size_t *size);
-ssize_t push_codepoint_convenience(struct smb_iconv_convenience *ic,
+ssize_t push_codepoint_handle(struct smb_iconv_handle *ic,
char *str, codepoint_t c);
+
codepoint_t toupper_m(codepoint_t val);
codepoint_t tolower_m(codepoint_t val);
+bool islower_m(codepoint_t val);
+bool isupper_m(codepoint_t val);
int codepoint_cmpi(codepoint_t c1, codepoint_t c2);
/* Iconv convenience functions */
-struct smb_iconv_convenience *smb_iconv_convenience_reinit(TALLOC_CTX *mem_ctx,
+struct smb_iconv_handle *smb_iconv_handle_reinit(TALLOC_CTX *mem_ctx,
const char *dos_charset,
const char *unix_charset,
- bool native_iconv,
- struct smb_iconv_convenience *old_ic);
+ bool use_builtin_handlers,
+ struct smb_iconv_handle *old_ic);
-bool convert_string_convenience(struct smb_iconv_convenience *ic,
+bool convert_string_handle(struct smb_iconv_handle *ic,
charset_t from, charset_t to,
void const *src, size_t srclen,
- void *dest, size_t destlen, size_t *converted_size,
- bool allow_badcharcnv);
-bool convert_string_talloc_convenience(TALLOC_CTX *ctx,
- struct smb_iconv_convenience *ic,
+ void *dest, size_t destlen, size_t *converted_size);
+bool convert_string_error_handle(struct smb_iconv_handle *ic,
+ charset_t from, charset_t to,
+ void const *src, size_t srclen,
+ void *dest, size_t destlen,
+ size_t *converted_size);
+
+bool convert_string_talloc_handle(TALLOC_CTX *ctx,
+ struct smb_iconv_handle *ic,
charset_t from, charset_t to,
void const *src, size_t srclen,
- void *dest, size_t *converted_size, bool allow_badcharcnv);
+ void *dest, size_t *converted_size);
/* iconv */
smb_iconv_t smb_iconv_open(const char *tocode, const char *fromcode);
int smb_iconv_close(smb_iconv_t cd);
const char **inbuf, size_t *inbytesleft,
char **outbuf, size_t *outbytesleft);
smb_iconv_t smb_iconv_open_ex(TALLOC_CTX *mem_ctx, const char *tocode,
- const char *fromcode, bool native_iconv);
+ const char *fromcode, bool use_builtin_handlers);
void load_case_tables(void);
-bool charset_register_backend(const void *_funcs);
-
-/*
- * Define stub for charset module which implements 8-bit encoding with gaps.
- * Encoding tables for such module should be produced from glibc's CHARMAPs
- * using script source/script/gen-8bit-gap.sh
- * CHARSETNAME is CAPITALIZED charset name
- *
- * */
-#define SMB_GENERATE_CHARSET_MODULE_8_BIT_GAP(CHARSETNAME) \
-static size_t CHARSETNAME ## _push(void *cd, const char **inbuf, size_t *inbytesleft, \
- char **outbuf, size_t *outbytesleft) \
-{ \
- while (*inbytesleft >= 2 && *outbytesleft >= 1) { \
- int i; \
- int done = 0; \
- \
- uint16 ch = SVAL(*inbuf,0); \
- \
- for (i=0; from_idx[i].start != 0xffff; i++) { \
- if ((from_idx[i].start <= ch) && (from_idx[i].end >= ch)) { \
- ((unsigned char*)(*outbuf))[0] = from_ucs2[from_idx[i].idx+ch]; \
- (*inbytesleft) -= 2; \
- (*outbytesleft) -= 1; \
- (*inbuf) += 2; \
- (*outbuf) += 1; \
- done = 1; \
- break; \
- } \
- } \
- if (!done) { \
- errno = EINVAL; \
- return -1; \
- } \
- \
- } \
- \
- if (*inbytesleft == 1) { \
- errno = EINVAL; \
- return -1; \
- } \
- \
- if (*inbytesleft > 1) { \
- errno = E2BIG; \
- return -1; \
- } \
- \
- return 0; \
-} \
- \
-static size_t CHARSETNAME ## _pull(void *cd, const char **inbuf, size_t *inbytesleft, \
- char **outbuf, size_t *outbytesleft) \
-{ \
- while (*inbytesleft >= 1 && *outbytesleft >= 2) { \
- *(uint16*)(*outbuf) = to_ucs2[((unsigned char*)(*inbuf))[0]]; \
- (*inbytesleft) -= 1; \
- (*outbytesleft) -= 2; \
- (*inbuf) += 1; \
- (*outbuf) += 2; \
- } \
- \
- if (*inbytesleft > 0) { \
- errno = E2BIG; \
- return -1; \
- } \
- \
- return 0; \
-} \
- \
-struct charset_functions CHARSETNAME ## _functions = \
- {#CHARSETNAME, CHARSETNAME ## _pull, CHARSETNAME ## _push}; \
- \
-NTSTATUS charset_ ## CHARSETNAME ## _init(void); \
-NTSTATUS charset_ ## CHARSETNAME ## _init(void) \
-{ \
- return smb_register_charset(& CHARSETNAME ## _functions); \
-} \
-
+void load_case_tables_library(void);
+
+/* The following definitions come from util_unistr_w.c */
+
+size_t strlen_w(const smb_ucs2_t *src);
+size_t strnlen_w(const smb_ucs2_t *src, size_t max);
+smb_ucs2_t *strchr_w(const smb_ucs2_t *s, smb_ucs2_t c);
+smb_ucs2_t *strchr_wa(const smb_ucs2_t *s, char c);
+smb_ucs2_t *strrchr_w(const smb_ucs2_t *s, smb_ucs2_t c);
+smb_ucs2_t *strnrchr_w(const smb_ucs2_t *s, smb_ucs2_t c, unsigned int n);
+smb_ucs2_t *strstr_w(const smb_ucs2_t *s, const smb_ucs2_t *ins);
+bool strlower_w(smb_ucs2_t *s);
+bool strupper_w(smb_ucs2_t *s);
+int strcasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b);
+int strncasecmp_w(const smb_ucs2_t *a, const smb_ucs2_t *b, size_t len);
+int strcmp_wa(const smb_ucs2_t *a, const char *b);
+smb_ucs2_t toupper_w(smb_ucs2_t v);
#endif /* __CHARSET_H__ */