charcnv.c File Reference

#include <arpa/inet.h>
#include <ctype.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/param.h>
#include <sys/stat.h>
#include <unistd.h>
#include <iconv.h>
#include <atalk/logger.h>
#include <atalk/unicode.h>
#include <atalk/util.h>
#include <atalk/compat.h>
#include <atalk/byteorder.h>

Macros
#define	MAX_CHARSETS   20
#define	CHECK_FLAGS(a, b)
#define	hextoint(c)

Functions
static const char *	charset_name (charset_t ch)
int	set_charset_name (charset_t ch, const char *name)
void	free_charset_names (void)
static struct charset_functions *	get_charset_functions (charset_t ch)
static void	lazy_initialize_conv (void)
charset_t	add_charset (const char *name)
void	init_iconv (void)
	Initialize iconv conversion descriptors.
static size_t	add_null (charset_t to, char *buf, size_t bytesleft, size_t len)
static size_t	convert_string_internal (charset_t from, charset_t to, void const *src, size_t srclen, void *dest, size_t destlen)
	Convert string from one encoding to another, making error checking etc.
size_t	convert_string (charset_t from, charset_t to, void const *src, size_t srclen, void *dest, size_t destlen)
static size_t	convert_string_allocate_internal (charset_t from, charset_t to, void const *src, size_t srclen, char **dest)
	Convert between character sets, allocating a new buffer for the result.
size_t	convert_string_allocate (charset_t from, charset_t to, void const *src, size_t srclen, char **dest)
size_t	charset_strupper (charset_t ch, const char *src, size_t srclen, char *dest, size_t destlen)
size_t	charset_strlower (charset_t ch, const char *src, size_t srclen, char *dest, size_t destlen)
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.
size_t	ucs2_to_charset_allocate (charset_t ch, char **dest, const ucs2_t *src)
size_t	charset_precompose (charset_t ch, char *src, size_t inlen, char *dst, size_t outlen)
size_t	charset_decompose (charset_t ch, char *src, size_t inlen, char *dst, size_t outlen)
static size_t	pull_charset_flags (charset_t from_set, charset_t to_set, charset_t cap_set, const char *src, size_t srclen, char *dest, size_t destlen, uint16_t *flags)
	Convert from MB to UCS2 charset.
static size_t	push_charset_flags (charset_t to_set, charset_t cap_set, char *src, size_t srclen, char *dest, size_t destlen, uint16_t *flags)
	Convert from UCS2 to MB 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)

Variables
static atalk_iconv_t	conv_handles [MAX_CHARSETS][MAX_CHARSETS]
static char *	charset_names [MAX_CHARSETS]
static struct charset_functions *	charsets [MAX_CHARSETS]
static char	hexdig [] = "0123456789abcdef"

Detailed Description

Character-set conversion routines based on samba iconv wrappers.

Macro Definition Documentation

CHECK_FLAGS

#define CHECK_FLAGS	(		a,
			b )

Value:

(((a)!=NULL) ? (*(a) & (b)) : 0 )

hextoint

#define hextoint

(

c

)

Value:

( isdigit( c ) ? c - '0' : c + 10 - 'a' )

MAX_CHARSETS

#define MAX_CHARSETS   20

Function Documentation

add_charset()

charset_t add_charset

(

const char *

name

)

add_null()

size_t add_null ( charset_t to, char * buf, size_t bytesleft, size_t len )

static

charset_decompose()

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

charset_name()

const char * charset_name ( charset_t ch )

static

Return the name of a charset to give to iconv().

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 size must be dest_len +2

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 )

convert_string_allocate_internal()

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

static

Convert between character sets, allocating a new buffer for the result.

Parameters

from	source character set
to	destination character set
src	pointer to source string (multibyte or singlebyte)
srclen	length of source buffer
dest	always set at least to NULL

Note

-1 is not accepted for srclen

Returns

Size in bytes of the converted string; or -1 in case of error

convert_string_internal()

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

static

Convert string from one encoding to another, making error checking etc.

Parameters

from	source character set
to	destination character set
src	pointer to source string (multibyte or singlebyte)
srclen	length of the source string in bytes
dest	pointer to destination string (multibyte or singlebyte)
destlen	maximal length allowed for string

Returns

the number of bytes occupied in the destination

free_charset_names()

void free_charset_names

(

void

)

get_charset_functions()

struct charset_functions * get_charset_functions ( charset_t ch )

static

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.

lazy_initialize_conv()

void lazy_initialize_conv ( void )

static

pull_charset_flags()

size_t pull_charset_flags ( charset_t from_set, charset_t to_set, charset_t cap_set, const char * src, size_t srclen, char * dest, size_t destlen, uint16_t * flags )

static

Convert from MB to UCS2 charset.

Flags:

• CONV_UNESCAPEHEX: ':XX' will be converted to an UCS2 character
• CONV_IGNORE: return the first convertable characters.
• CONV_FORCE: force convertion

Bug

This will not work if the destination charset is not multibyte, i.e. UCS2->UCS2 will fail The (un)escape scheme is not compatible to the old cap style escape. This is bad, we need it for e.g. HFS cdroms.

push_charset_flags()

size_t push_charset_flags ( charset_t to_set, charset_t cap_set, char * src, size_t srclen, char * dest, size_t destlen, uint16_t * flags )

static

Convert from UCS2 to MB charset.

Flags:

• CONV_ESCAPEDOTS: escape leading dots
• CONV_ESCAPEHEX: unconvertable characters and '/' will be escaped to :XX
• CONV_IGNORE: return the first convertable characters.
• CONV__EILSEQ: unconvertable characters will be replaced with '_'
• CONV_FORCE: force convertion

  Bug

  CONV_IGNORE and CONV_ESCAPEHEX can't work together. Should we check this ? This will not work if the destination charset is not multibyte, i.e. UCS2->UCS2 will fail The escape scheme is not compatible to the old cap style escape. This is bad, we need it for e.g. HFS cdroms.

set_charset_name()

int set_charset_name	(	charset_t	ch,
		const char *	name )

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 )

Variable Documentation

charset_names

char* charset_names[MAX_CHARSETS]

static

charsets

struct charset_functions* charsets[MAX_CHARSETS]

static

conv_handles

atalk_iconv_t conv_handles[MAX_CHARSETS][MAX_CHARSETS]

static

hexdig

char hexdig[] = "0123456789abcdef"

static