unicode.h File Reference

#include <errno.h>
#include <stdint.h>
#include <sys/param.h>

Go to the source code of this file.

Data Structures
struct	atalk_iconv_t
struct	charset_functions

Macros
#define	ucs2_t   uint16_t
#define	SAFE_FREE(x)   do { if ((x) != NULL) {free(x); x=NULL;} } while(0)
#define	EILSEQ   84
#define	CHARSET_CLIENT   1
#define	CHARSET_VOLUME   2
#define	CHARSET_PRECOMPOSED   4
#define	CHARSET_DECOMPOSED   8
#define	CHARSET_MULTIBYTE   16
#define	CHARSET_WIDECHAR   32
#define	CHARSET_ICONV   64
#define	IGNORE_CHAR   '_'
#define	CONV_IGNORE   (1<<0)
#define	CONV_ESCAPEHEX   (1<<1)
#define	CONV_ESCAPEDOTS   (1<<2)
#define	CONV_UNESCAPEHEX   (1<<3)
#define	CONV_TOUPPER   (1<<4)
#define	CONV_TOLOWER   (1<<5)
#define	CONV_PRECOMPOSE   (1<<6)
#define	CONV_DECOMPOSE   (1<<7)
#define	CONV_FORCE   (1<<8)
#define	CONV__EILSEQ   (1<<9)
#define	CONV_REQMANGLE   (1<<14)
#define	CONV_REQESCAPE   (1<<15)
#define	NUM_CHARSETS   5

Enumerations
enum	charset_t {   CH_UCS2 = 0 , CH_UTF8 = 1 , CH_MAC = 2 , CH_UNIX = 3 ,   CH_UTF8_MAC = 4 }

Functions
atalk_iconv_t	atalk_iconv_open (const char *, const char *)
size_t	atalk_iconv (atalk_iconv_t, const char **, size_t *, char **, size_t *)
	This is a simple portable iconv() implementaion. More...
int	atalk_iconv_close (atalk_iconv_t)
struct charset_functions *	find_charset_functions (const char *)
int	atalk_register_charset (struct charset_functions *)
ucs2_t	toupper_w (ucs2_t)
uint32_t	toupper_sp (uint32_t)
ucs2_t	tolower_w (ucs2_t)
uint32_t	tolower_sp (uint32_t)
int	strupper_w (ucs2_t *)
	Convert a string to upper case. More...
int	strlower_w (ucs2_t *)
	Convert a string to lower case. More...
size_t	strlen_w (const ucs2_t *)
	wide strlen() More...
size_t	strnlen_w (const ucs2_t *, size_t)
	wide strnlen() More...
ucs2_t *	strchr_w (const ucs2_t *, ucs2_t)
	wide strchr() More...
ucs2_t *	strcasechr_w (const ucs2_t *s, ucs2_t c)
	wide strcasechr() More...
int	strcmp_w (const ucs2_t *, const ucs2_t *)
	wide strcmp() More...
int	strncmp_w (const ucs2_t *, const ucs2_t *, size_t)
	wide strncmp() More...
int	strcasecmp_w (const ucs2_t *, const ucs2_t *)
	wide strcasecmp() More...
int	strncasecmp_w (const ucs2_t *, const ucs2_t *, size_t)
	wide strncasecmp() More...
ucs2_t *	strstr_w (const ucs2_t *s, const ucs2_t *ins)
	wide strstr() More...
ucs2_t *	strcasestr_w (const ucs2_t *, const ucs2_t *)
	wide strcasestr() More...
ucs2_t *	strndup_w (const ucs2_t *, size_t)
	wide strndup() More...
ucs2_t *	strdup_w (const ucs2_t *)
	wide strdup() More...
size_t	precompose_w (ucs2_t *, size_t, ucs2_t *, size_t *)
	pre|decomposition More...
size_t	decompose_w (ucs2_t *, size_t, ucs2_t *, size_t *)
int	set_charset_name (charset_t, const char *)
void	free_charset_names (void)
void	init_iconv (void)
	Initialize iconv conversion descriptors. More...
size_t	convert_string (charset_t, charset_t, void const *, size_t, void *, size_t)
size_t	convert_string_allocate (charset_t, charset_t, void const *, size_t, char **)
size_t	charset_strupper (charset_t, const char *, size_t, char *, size_t)
size_t	charset_strlower (charset_t, const char *, size_t, char *, size_t)
size_t	ucs2_to_charset_allocate (charset_t, char **dest, const ucs2_t *src)
size_t	ucs2_to_charset (charset_t, const ucs2_t *src, char *dest, size_t)
	Copy a string from a UCS2 src to a unix char * destination, allocating a buffer. More...
size_t	convert_charset (charset_t, charset_t, charset_t, const char *, size_t, char *, size_t, uint16_t *)
size_t	charset_precompose (charset_t, char *, size_t, char *, size_t)
size_t	charset_decompose (charset_t, char *, size_t, char *, size_t)
charset_t	add_charset (const char *name)

Macro Definition Documentation

CHARSET_CLIENT

#define CHARSET_CLIENT   1

CHARSET_DECOMPOSED

#define CHARSET_DECOMPOSED   8

CHARSET_ICONV

#define CHARSET_ICONV   64

CHARSET_MULTIBYTE

#define CHARSET_MULTIBYTE   16

CHARSET_PRECOMPOSED

#define CHARSET_PRECOMPOSED   4

CHARSET_VOLUME

#define CHARSET_VOLUME   2

CHARSET_WIDECHAR

#define CHARSET_WIDECHAR   32

CONV__EILSEQ

#define CONV__EILSEQ   (1<<9)

ignore EILSEQ, replace with IGNORE_CHAR (try USC2)

CONV_DECOMPOSE

#define CONV_DECOMPOSE   (1<<7)

decompose

CONV_ESCAPEDOTS

#define CONV_ESCAPEDOTS   (1<<2)

escape leading dots with :2600

CONV_ESCAPEHEX

#define CONV_ESCAPEHEX   (1<<1)

escape unconvertable chars with :[UCS2HEX], also escape '/'. Escape ':' if also CONV_ALLOW_COLON, else ':' raises EILSEQ

CONV_FORCE

#define CONV_FORCE   (1<<8)

force convertion

CONV_IGNORE

#define CONV_IGNORE   (1<<0)

return the first convertable characters.

CONV_PRECOMPOSE

#define CONV_PRECOMPOSE   (1<<6)

precompose

CONV_REQESCAPE

#define CONV_REQESCAPE   (1<<15)

espace unconvertable chars with :[UCS2HEX]

CONV_REQMANGLE

#define CONV_REQMANGLE   (1<<14)

mangling of returned name is required

CONV_TOLOWER

#define CONV_TOLOWER   (1<<5)

convert to lowercase

CONV_TOUPPER

#define CONV_TOUPPER   (1<<4)

convert to UPPERcase

CONV_UNESCAPEHEX

#define CONV_UNESCAPEHEX   (1<<3)

EILSEQ

#define EILSEQ   84

Illegal byte sequence.

IGNORE_CHAR

#define IGNORE_CHAR   '_'

NUM_CHARSETS

#define NUM_CHARSETS   5

SAFE_FREE

#define SAFE_FREE

(

x

)

do { if ((x) != NULL) {free(x); x=NULL;} } while(0)

ucs2_t

#define ucs2_t   uint16_t

Enumeration Type Documentation

charset_t

enum charset_t

this defines the charset types used in netatalk

Enumerator
CH_UCS2
CH_UTF8
CH_MAC
CH_UNIX
CH_UTF8_MAC

Function Documentation

add_charset()

charset_t add_charset

(

const char *

name

)

atalk_iconv()

size_t atalk_iconv	(	atalk_iconv_t	cd,
		const char **	inbuf,
		size_t *	inbytesleft,
		char **	outbuf,
		size_t *	outbytesleft
	)

This is a simple portable iconv() implementaion.

It only knows about a very small number of character sets - just enough that netatalk works on systems that don't have iconv.

atalk_iconv_close()

int atalk_iconv_close

(

atalk_iconv_t

cd

)

simple iconv_close() wrapper

atalk_iconv_open()

atalk_iconv_t atalk_iconv_open	(	const char *	tocode,
		const char *	fromcode
	)

simple iconv_open() wrapper

atalk_register_charset()

int atalk_register_charset

(

struct charset_functions *

funcs

)

charset_decompose()

size_t charset_decompose	(	charset_t	ch,
		char *	src,
		size_t	inlen,
		char *	dst,
		size_t	outlen
	)

charset_precompose()

size_t charset_precompose	(	charset_t	ch,
		char *	src,
		size_t	inlen,
		char *	dst,
		size_t	outlen
	)

charset_strlower()

size_t charset_strlower	(	charset_t	ch,
		const char *	src,
		size_t	srclen,
		char *	dest,
		size_t	destlen
	)

charset_strupper()

size_t charset_strupper	(	charset_t	ch,
		const char *	src,
		size_t	srclen,
		char *	dest,
		size_t	destlen
	)

convert_charset()

size_t convert_charset	(	charset_t	from_set,
		charset_t	to_set,
		charset_t	cap_charset,
		const char *	src,
		size_t	src_len,
		char *	dest,
		size_t	dest_len,
		uint16_t *	flags
	)

Bug:

the size is a mess we really need a malloc/free logic

Note

dest_len must include space for the two-byte terminator

convert_string()

size_t convert_string	(	charset_t	from,
		charset_t	to,
		void const *	src,
		size_t	srclen,
		void *	dest,
		size_t	destlen
	)

convert_string_allocate()

size_t convert_string_allocate	(	charset_t	from,
		charset_t	to,
		void const *	src,
		size_t	srclen,
		char **	dest
	)

decompose_w()

size_t decompose_w	(	ucs2_t *	name,
		size_t	inplen,
		ucs2_t *	comp,
		size_t *	outlen
	)

find_charset_functions()

struct charset_functions * find_charset_functions

(

const char *

name

)

free_charset_names()

void free_charset_names

(

void

)

init_iconv()

void init_iconv

(

void

)

Initialize iconv conversion descriptors.

This is called the first time it is needed, and also called again every time the configuration is reloaded, because the charset or codepage might have changed.

precompose_w()

size_t precompose_w	(	ucs2_t *	name,
		size_t	inplen,
		ucs2_t *	comp,
		size_t *	outlen
	)

pre|decomposition

we can't use static, this stuff needs to be reentrant static char comp[MAXPATHLEN +1];

We don't implement Singleton and Canonical Ordering. We ignore CompositionExclusions.txt because they cause the problem of the roundtrip such as Dancing Icon.

exclude U2000-U2FFF, UFE30-UFE4F and U2F800-U2FA1F ranges in precompose.h from composition according to AFP 3.x spec

set_charset_name()

int set_charset_name	(	charset_t	ch,
		const char *	name
	)

strcasechr_w()

ucs2_t * strcasechr_w	(	const ucs2_t *	s,
		ucs2_t	c
	)

wide strcasechr()

Note

separately process BMP and surrogate pair

strcasecmp_w()

int strcasecmp_w	(	const ucs2_t *	a,
		const ucs2_t *	b
	)

wide strcasecmp()

case insensitive string comparison

Note

surrogate pair support

strcasestr_w()

ucs2_t * strcasestr_w	(	const ucs2_t *	s,
		const ucs2_t *	ins
	)

wide strcasestr()

Note

surrogate pair support

strchr_w()

ucs2_t * strchr_w	(	const ucs2_t *	s,
		ucs2_t	c
	)

wide strchr()

Note

hi and lo of surrogate pair are separately processed.

strcmp_w()

int strcmp_w	(	const ucs2_t *	a,
		const ucs2_t *	b
	)

wide strcmp()

Note

no problem of surrogate pair

strdup_w()

ucs2_t * strdup_w

(

const ucs2_t *

src

)

wide strdup()

duplicate string

Note

no problem of surrogate pair

strlen_w()

size_t strlen_w

(

const ucs2_t *

src

)

wide strlen()

Count the number of characters in a UTF-16 string.

Note

one surrogate pair is two characters.

strlower_w()

int strlower_w

(

ucs2_t *

s

)

Convert a string to lower case.

Returns

True if any char is converted

Note

surrogate pair support

strncasecmp_w()

int strncasecmp_w	(	const ucs2_t *	a,
		const ucs2_t *	b,
		size_t	len
	)

wide strncasecmp()

case insensitive string comparison, length limited

Note

compare up to 'len+1' if 'len' isolate surrogate pair

strncmp_w()

int strncmp_w	(	const ucs2_t *	a,
		const ucs2_t *	b,
		size_t	len
	)

wide strncmp()

Note

no problem of surrogate pair

strndup_w()

ucs2_t * strndup_w	(	const ucs2_t *	src,
		size_t	len
	)

wide strndup()

duplicate string

Note

not check isolation of surrogate pair

if len == 0 then duplicate the whole string

strnlen_w()

size_t strnlen_w	(	const ucs2_t *	src,
		size_t	max
	)

wide strnlen()

Count up to max number of characters in a UTF-16 string.

Note

one surrogate pair is two characters.

strstr_w()

ucs2_t * strstr_w	(	const ucs2_t *	s,
		const ucs2_t *	ins
	)

wide strstr()

Note

no problem of surrogate pair

strupper_w()

int strupper_w

(

ucs2_t *

s

)

Convert a string to upper case.

Returns

True if any char is converted

Note

surrogate pair support

tolower_sp()

uint32_t tolower_sp

(

uint32_t

val

)

tolower_w()

ucs2_t tolower_w

(

ucs2_t

val

)

toupper_sp()

uint32_t toupper_sp

(

uint32_t

val

)

toupper_w()

ucs2_t toupper_w

(

ucs2_t

val

)

ucs2_to_charset()

size_t ucs2_to_charset	(	charset_t	ch,
		const ucs2_t *	src,
		char *	dest,
		size_t	destlen
	)

Copy a string from a UCS2 src to a unix char * destination, allocating a buffer.

Parameters

ch	destination character set
src	source UCS2 string
dest	always set at least to NULL
destlen	maximum length of destination buffer

Returns

The number of bytes occupied by the string in the destination

ucs2_to_charset_allocate()

size_t ucs2_to_charset_allocate	(	charset_t	ch,
		char **	dest,
		const ucs2_t *	src
	)