blob: 4bd41214346f13582f73185097b5551df889fb9d [file] [log] [blame]
/* Copyright (C) 2000-2009 Free Software Foundation, Inc.
This file is part of the GNU LIBICONV Library.
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 3 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, see <http://www.gnu.org/licenses/>. */
#include "config.h"
#ifndef ICONV_CONST
# define ICONV_CONST
#endif
#include <limits.h>
#include <stddef.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <iconv.h>
#include <errno.h>
#include <locale.h>
#include <fcntl.h>
/* Ensure that iconv_no_i18n does not depend on libintl. */
#ifdef NO_I18N
# undef ENABLE_NLS
# undef ENABLE_RELOCATABLE
#endif
#include "binary-io.h"
#include "progname.h"
#include "relocatable.h"
#include "xalloc.h"
#include "uniwidth.h"
#include "uniwidth/cjk.h"
/* Ensure that iconv_no_i18n does not depend on libintl. */
#ifdef NO_I18N
#include <stdarg.h>
static void
error (int status, int errnum, const char *message, ...)
{
va_list args;
fflush(stdout);
fprintf(stderr,"%s: ",program_name);
va_start(args,message);
vfprintf(stderr,message,args);
va_end(args);
if (errnum) {
const char *s = strerror(errnum);
if (s == NULL)
s = "Unknown system error";
}
putc('\n',stderr);
fflush(stderr);
if (status)
exit(status);
}
#else
# include "error.h"
#endif
#include "gettext.h"
#define _(str) gettext(str)
/* Ensure that iconv_no_i18n does not depend on libintl. */
#ifdef NO_I18N
# define xmalloc malloc
# define xalloc_die abort
#endif
/* Locale independent test for a decimal digit.
Argument can be 'char' or 'unsigned char'. (Whereas the argument of
<ctype.h> isdigit must be an 'unsigned char'.) */
#undef isdigit
#define isdigit(c) ((unsigned int) ((c) - '0') < 10)
/* Locale independent test for a printable character.
Argument can be 'char' or 'unsigned char'. (Whereas the argument of
<ctype.h> isdigit must be an 'unsigned char'.) */
#define c_isprint(c) ((c) >= ' ' && (c) <= '~')
/* ========================================================================= */
static int discard_unconvertible = 0;
static int silent = 0;
static void usage (int exitcode)
{
if (exitcode != 0) {
const char* helpstring1 =
/* TRANSLATORS: The first line of the short usage message. */
_("Usage: iconv [-c] [-s] [-f fromcode] [-t tocode] [file ...]");
const char* helpstring2 =
/* TRANSLATORS: The second line of the short usage message.
Align it correctly against the first line. */
_("or: iconv -l");
fprintf(stderr, "%s\n%s\n", helpstring1, helpstring2);
fprintf(stderr, _("Try `%s --help' for more information.\n"), program_name);
} else {
/* xgettext: no-wrap */
/* TRANSLATORS: The first line of the long usage message.
The %s placeholder expands to the program name. */
printf(_("\
Usage: %s [OPTION...] [-f ENCODING] [-t ENCODING] [INPUTFILE...]\n"),
program_name);
/* xgettext: no-wrap */
/* TRANSLATORS: The second line of the long usage message.
Align it correctly against the first line.
The %s placeholder expands to the program name. */
printf(_("\
or: %s -l\n"),
program_name);
printf("\n");
/* xgettext: no-wrap */
/* TRANSLATORS: Description of the iconv program. */
printf(_("\
Converts text from one encoding to another encoding.\n"));
printf("\n");
/* xgettext: no-wrap */
printf(_("\
Options controlling the input and output format:\n"));
/* xgettext: no-wrap */
printf(_("\
-f ENCODING, --from-code=ENCODING\n\
the encoding of the input\n"));
/* xgettext: no-wrap */
printf(_("\
-t ENCODING, --to-code=ENCODING\n\
the encoding of the output\n"));
printf("\n");
/* xgettext: no-wrap */
printf(_("\
Options controlling conversion problems:\n"));
/* xgettext: no-wrap */
printf(_("\
-c discard unconvertible characters\n"));
/* xgettext: no-wrap */
printf(_("\
--unicode-subst=FORMATSTRING\n\
substitution for unconvertible Unicode characters\n"));
/* xgettext: no-wrap */
printf(_("\
--byte-subst=FORMATSTRING substitution for unconvertible bytes\n"));
/* xgettext: no-wrap */
printf(_("\
--widechar-subst=FORMATSTRING\n\
substitution for unconvertible wide characters\n"));
printf("\n");
/* xgettext: no-wrap */
printf(_("\
Options controlling error output:\n"));
/* xgettext: no-wrap */
printf(_("\
-s, --silent suppress error messages about conversion problems\n"));
printf("\n");
/* xgettext: no-wrap */
printf(_("\
Informative output:\n"));
/* xgettext: no-wrap */
printf(_("\
-l, --list list the supported encodings\n"));
/* xgettext: no-wrap */
printf(_("\
--help display this help and exit\n"));
/* xgettext: no-wrap */
printf(_("\
--version output version information and exit\n"));
printf("\n");
/* TRANSLATORS: The placeholder indicates the bug-reporting address
for this package. Please add _another line_ saying
"Report translation bugs to <...>\n" with the address for translation
bugs (typically your translation team's web or email address). */
fputs(_("Report bugs to <bug-gnu-libiconv@gnu.org>.\n"),stdout);
}
exit(exitcode);
}
static void print_version (void)
{
printf("iconv (GNU libiconv %d.%d)\n",
_libiconv_version >> 8, _libiconv_version & 0xff);
printf("Copyright (C) %s Free Software Foundation, Inc.\n", "2000-2009");
/* xgettext: no-wrap */
fputs (_("\
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>\n\
This is free software: you are free to change and redistribute it.\n\
There is NO WARRANTY, to the extent permitted by law.\n\
"),stdout);
/* TRANSLATORS: The %s placeholder expands to an author's name. */
printf(_("Written by %s.\n"),"Bruno Haible");
exit(EXIT_SUCCESS);
}
static int print_one (unsigned int namescount, const char * const * names,
void* data)
{
unsigned int i;
(void)data;
for (i = 0; i < namescount; i++) {
if (i > 0)
putc(' ',stdout);
fputs(names[i],stdout);
}
putc('\n',stdout);
return 0;
}
/* ========================================================================= */
/* Line number and column position. */
static unsigned int line;
static unsigned int column;
static const char* cjkcode;
/* Update the line number and column position after a character was
successfully converted. */
static void update_line_column (unsigned int uc, void* data)
{
if (uc == 0x000A) {
line++;
column = 0;
} else {
int width = uc_width(uc, cjkcode);
if (width >= 0)
column += width;
else if (uc == 0x0009)
column += 8 - (column % 8);
}
}
/* ========================================================================= */
/* Production of placeholder strings as fallback for unconvertible
characters. */
/* Check that the argument is a format string taking either no argument
or exactly one unsigned integer argument. Returns the maximum output
size of the format string. */
static size_t check_subst_formatstring (const char *format, const char *param_name)
{
/* C format strings are described in POSIX (IEEE P1003.1 2001), section
XSH 3 fprintf(). See also Linux fprintf(3) manual page.
For simplicity, we don't accept
- the '%m$' reordering syntax,
- the 'I' flag,
- width specifications referring to an argument,
- precision specifications referring to an argument,
- size specifiers,
- format specifiers other than 'o', 'u', 'x', 'X'.
What remains?
A directive
- starts with '%',
- is optionally followed by any of the characters '#', '0', '-', ' ',
'+', "'", each of which acts as a flag,
- is optionally followed by a width specification: a nonempty digit
sequence,
- is optionally followed by '.' and a precision specification: a
nonempty digit sequence,
- is finished by a specifier
- '%', that needs no argument,
- 'o', 'u', 'x', 'X', that need an unsigned integer argument.
*/
size_t maxsize = 0;
unsigned int unnumbered_arg_count = 0;
for (; *format != '\0';) {
if (*format++ == '%') {
/* A directive. */
unsigned int width = 0;
unsigned int precision = 0;
unsigned int length;
/* Parse flags. */
for (;;) {
if (*format == ' ' || *format == '+' || *format == '-'
|| *format == '#' || *format == '0' || *format == '\'')
format++;
else
break;
}
/* Parse width. */
if (*format == '*')
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option. */
_("%s argument: A format directive with a variable width is not allowed here."),
param_name);
if (isdigit (*format)) {
do {
width = 10*width + (*format - '0');
format++;
} while (isdigit (*format));
}
/* Parse precision. */
if (*format == '.') {
format++;
if (*format == '*')
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option. */
_("%s argument: A format directive with a variable precision is not allowed here."),
param_name);
if (isdigit (*format)) {
do {
precision = 10*precision + (*format - '0');
format++;
} while (isdigit (*format));
}
}
/* Parse size. */
switch (*format) {
case 'h': case 'l': case 'L': case 'q':
case 'j': case 'z': case 'Z': case 't':
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option. */
_("%s argument: A format directive with a size is not allowed here."),
param_name);
}
/* Parse end of directive. */
switch (*format) {
case '%':
length = 1;
break;
case 'u': case 'o': case 'x': case 'X':
if (*format == 'u') {
length = (unsigned int) (sizeof (unsigned int) * CHAR_BIT
* 0.30103 /* binary -> decimal */
)
+ 1; /* turn floor into ceil */
if (length < precision)
length = precision;
length *= 2; /* estimate for FLAG_GROUP */
length += 1; /* account for leading sign */
} else if (*format == 'o') {
length = (unsigned int) (sizeof (unsigned int) * CHAR_BIT
* 0.333334 /* binary -> octal */
)
+ 1; /* turn floor into ceil */
if (length < precision)
length = precision;
length += 1; /* account for leading sign */
} else { /* 'x', 'X' */
length = (unsigned int) (sizeof (unsigned int) * CHAR_BIT
* 0.25 /* binary -> hexadecimal */
)
+ 1; /* turn floor into ceil */
if (length < precision)
length = precision;
length += 2; /* account for leading sign or alternate form */
}
unnumbered_arg_count++;
break;
default:
if (*format == '\0')
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option. */
_("%s argument: The string ends in the middle of a directive."),
param_name);
else if (c_isprint(*format))
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option.
The %c placeholder expands to an unknown format directive. */
_("%s argument: The character '%c' is not a valid conversion specifier."),
param_name,*format);
else
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option. */
_("%s argument: The character that terminates the format directive is not a valid conversion specifier."),
param_name);
abort(); /*NOTREACHED*/
}
format++;
if (length < width)
length = width;
maxsize += length;
} else
maxsize++;
}
if (unnumbered_arg_count > 1)
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a command-line option.
The %u placeholder expands to the number of arguments consumed by the format string. */
ngettext("%s argument: The format string consumes more than one argument: %u argument.",
"%s argument: The format string consumes more than one argument: %u arguments.",
unnumbered_arg_count),
param_name,unnumbered_arg_count);
return maxsize;
}
/* Format strings. */
static const char* ilseq_byte_subst;
static const char* ilseq_wchar_subst;
static const char* ilseq_unicode_subst;
/* Maximum result size for each format string. */
static size_t ilseq_byte_subst_size;
static size_t ilseq_wchar_subst_size;
static size_t ilseq_unicode_subst_size;
/* Buffer of size ilseq_byte_subst_size+1. */
static char* ilseq_byte_subst_buffer;
#if HAVE_WCHAR_T
/* Buffer of size ilseq_wchar_subst_size+1. */
static char* ilseq_wchar_subst_buffer;
#endif
/* Buffer of size ilseq_unicode_subst_size+1. */
static char* ilseq_unicode_subst_buffer;
/* Auxiliary variables for subst_mb_to_uc_fallback. */
/* Converter from locale encoding to UCS-4. */
static iconv_t subst_mb_to_uc_cd;
/* Buffer of size ilseq_byte_subst_size. */
static unsigned int* subst_mb_to_uc_temp_buffer;
static void subst_mb_to_uc_fallback
(const char* inbuf, size_t inbufsize,
void (*write_replacement) (const unsigned int *buf, size_t buflen,
void* callback_arg),
void* callback_arg,
void* data)
{
for (; inbufsize > 0; inbuf++, inbufsize--) {
const char* inptr;
size_t inbytesleft;
char* outptr;
size_t outbytesleft;
sprintf(ilseq_byte_subst_buffer,
ilseq_byte_subst, (unsigned int)(unsigned char)*inbuf);
inptr = ilseq_byte_subst_buffer;
inbytesleft = strlen(ilseq_byte_subst_buffer);
outptr = (char*)subst_mb_to_uc_temp_buffer;
outbytesleft = ilseq_byte_subst_size*sizeof(unsigned int);
iconv(subst_mb_to_uc_cd,NULL,NULL,NULL,NULL);
if (iconv(subst_mb_to_uc_cd, (ICONV_CONST char**)&inptr,&inbytesleft, &outptr,&outbytesleft)
== (size_t)(-1)
|| iconv(subst_mb_to_uc_cd, NULL,NULL, &outptr,&outbytesleft)
== (size_t)(-1))
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a piece of text, specified through --byte-subst. */
_("cannot convert byte substitution to Unicode: %s"),
ilseq_byte_subst_buffer);
if (!(outbytesleft%sizeof(unsigned int) == 0))
abort();
write_replacement(subst_mb_to_uc_temp_buffer,
ilseq_byte_subst_size-(outbytesleft/sizeof(unsigned int)),
callback_arg);
}
}
/* Auxiliary variables for subst_uc_to_mb_fallback. */
/* Converter from locale encoding to target encoding. */
static iconv_t subst_uc_to_mb_cd;
/* Buffer of size ilseq_unicode_subst_size*4. */
static char* subst_uc_to_mb_temp_buffer;
static void subst_uc_to_mb_fallback
(unsigned int code,
void (*write_replacement) (const char *buf, size_t buflen,
void* callback_arg),
void* callback_arg,
void* data)
{
const char* inptr;
size_t inbytesleft;
char* outptr;
size_t outbytesleft;
sprintf(ilseq_unicode_subst_buffer, ilseq_unicode_subst, code);
inptr = ilseq_unicode_subst_buffer;
inbytesleft = strlen(ilseq_unicode_subst_buffer);
outptr = subst_uc_to_mb_temp_buffer;
outbytesleft = ilseq_unicode_subst_size*4;
iconv(subst_uc_to_mb_cd,NULL,NULL,NULL,NULL);
if (iconv(subst_uc_to_mb_cd, (ICONV_CONST char**)&inptr,&inbytesleft, &outptr,&outbytesleft)
== (size_t)(-1)
|| iconv(subst_uc_to_mb_cd, NULL,NULL, &outptr,&outbytesleft)
== (size_t)(-1))
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a piece of text, specified through --unicode-subst. */
_("cannot convert unicode substitution to target encoding: %s"),
ilseq_unicode_subst_buffer);
write_replacement(subst_uc_to_mb_temp_buffer,
ilseq_unicode_subst_size*4-outbytesleft,
callback_arg);
}
#if HAVE_WCHAR_T
/* Auxiliary variables for subst_mb_to_wc_fallback. */
/* Converter from locale encoding to wchar_t. */
static iconv_t subst_mb_to_wc_cd;
/* Buffer of size ilseq_byte_subst_size. */
static wchar_t* subst_mb_to_wc_temp_buffer;
static void subst_mb_to_wc_fallback
(const char* inbuf, size_t inbufsize,
void (*write_replacement) (const wchar_t *buf, size_t buflen,
void* callback_arg),
void* callback_arg,
void* data)
{
for (; inbufsize > 0; inbuf++, inbufsize--) {
const char* inptr;
size_t inbytesleft;
char* outptr;
size_t outbytesleft;
sprintf(ilseq_byte_subst_buffer,
ilseq_byte_subst, (unsigned int)(unsigned char)*inbuf);
inptr = ilseq_byte_subst_buffer;
inbytesleft = strlen(ilseq_byte_subst_buffer);
outptr = (char*)subst_mb_to_wc_temp_buffer;
outbytesleft = ilseq_byte_subst_size*sizeof(wchar_t);
iconv(subst_mb_to_wc_cd,NULL,NULL,NULL,NULL);
if (iconv(subst_mb_to_wc_cd, (ICONV_CONST char**)&inptr,&inbytesleft, &outptr,&outbytesleft)
== (size_t)(-1)
|| iconv(subst_mb_to_wc_cd, NULL,NULL, &outptr,&outbytesleft)
== (size_t)(-1))
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a piece of text, specified through --byte-subst. */
_("cannot convert byte substitution to wide string: %s"),
ilseq_byte_subst_buffer);
if (!(outbytesleft%sizeof(wchar_t) == 0))
abort();
write_replacement(subst_mb_to_wc_temp_buffer,
ilseq_byte_subst_size-(outbytesleft/sizeof(wchar_t)),
callback_arg);
}
}
/* Auxiliary variables for subst_wc_to_mb_fallback. */
/* Converter from locale encoding to target encoding. */
static iconv_t subst_wc_to_mb_cd;
/* Buffer of size ilseq_wchar_subst_size*4.
Hardcode factor 4, because MB_LEN_MAX is not reliable on some platforms. */
static char* subst_wc_to_mb_temp_buffer;
static void subst_wc_to_mb_fallback
(wchar_t code,
void (*write_replacement) (const char *buf, size_t buflen,
void* callback_arg),
void* callback_arg,
void* data)
{
const char* inptr;
size_t inbytesleft;
char* outptr;
size_t outbytesleft;
sprintf(ilseq_wchar_subst_buffer, ilseq_wchar_subst, (unsigned int) code);
inptr = ilseq_wchar_subst_buffer;
inbytesleft = strlen(ilseq_wchar_subst_buffer);
outptr = subst_wc_to_mb_temp_buffer;
outbytesleft = ilseq_wchar_subst_size*4;
iconv(subst_wc_to_mb_cd,NULL,NULL,NULL,NULL);
if (iconv(subst_wc_to_mb_cd, (ICONV_CONST char**)&inptr,&inbytesleft, &outptr,&outbytesleft)
== (size_t)(-1)
|| iconv(subst_wc_to_mb_cd, NULL,NULL, &outptr,&outbytesleft)
== (size_t)(-1))
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a piece of text, specified through --widechar-subst. */
_("cannot convert widechar substitution to target encoding: %s"),
ilseq_wchar_subst_buffer);
write_replacement(subst_wc_to_mb_temp_buffer,
ilseq_wchar_subst_size*4-outbytesleft,
callback_arg);
}
#else
#define subst_mb_to_wc_fallback NULL
#define subst_wc_to_mb_fallback NULL
#endif
/* Auxiliary variables for subst_mb_to_mb_fallback. */
/* Converter from locale encoding to target encoding. */
static iconv_t subst_mb_to_mb_cd;
/* Buffer of size ilseq_byte_subst_size*4. */
static char* subst_mb_to_mb_temp_buffer;
static void subst_mb_to_mb_fallback (const char* inbuf, size_t inbufsize)
{
for (; inbufsize > 0; inbuf++, inbufsize--) {
const char* inptr;
size_t inbytesleft;
char* outptr;
size_t outbytesleft;
sprintf(ilseq_byte_subst_buffer,
ilseq_byte_subst, (unsigned int)(unsigned char)*inbuf);
inptr = ilseq_byte_subst_buffer;
inbytesleft = strlen(ilseq_byte_subst_buffer);
outptr = subst_mb_to_mb_temp_buffer;
outbytesleft = ilseq_byte_subst_size*4;
iconv(subst_mb_to_mb_cd,NULL,NULL,NULL,NULL);
if (iconv(subst_mb_to_mb_cd, (ICONV_CONST char**)&inptr,&inbytesleft, &outptr,&outbytesleft)
== (size_t)(-1)
|| iconv(subst_mb_to_mb_cd, NULL,NULL, &outptr,&outbytesleft)
== (size_t)(-1))
error(EXIT_FAILURE,0,
/* TRANSLATORS: An error message.
The %s placeholder expands to a piece of text, specified through --byte-subst. */
_("cannot convert byte substitution to target encoding: %s"),
ilseq_byte_subst_buffer);
fwrite(subst_mb_to_mb_temp_buffer,1,ilseq_byte_subst_size*4-outbytesleft,
stdout);
}
}
/* ========================================================================= */
/* Error messages during conversion. */
static void conversion_error_EILSEQ (const char* infilename)
{
fflush(stdout);
if (column > 0)
putc('\n',stderr);
error(0,0,
/* TRANSLATORS: An error message.
The placeholders expand to the input file name, a line number, and a column number. */
_("%s:%u:%u: cannot convert"),
infilename,line,column);
}
static void conversion_error_EINVAL (const char* infilename)
{
fflush(stdout);
if (column > 0)
putc('\n',stderr);
error(0,0,
/* TRANSLATORS: An error message.
The placeholders expand to the input file name, a line number, and a column number.
A "shift sequence" is a sequence of bytes that changes the state of the converter;
this concept exists only for "stateful" encodings like ISO-2022-JP. */
_("%s:%u:%u: incomplete character or shift sequence"),
infilename,line,column);
}
static void conversion_error_other (int errnum, const char* infilename)
{
fflush(stdout);
if (column > 0)
putc('\n',stderr);
error(0,errnum,
/* TRANSLATORS: The first part of an error message.
It is followed by a colon and a detail message.
The placeholders expand to the input file name, a line number, and a column number. */
_("%s:%u:%u"),
infilename,line,column);
}
/* Convert the input given in infile. */
static int convert (iconv_t cd, FILE* infile, const char* infilename)
{
char inbuf[4096+4096];
size_t inbufrest = 0;
char initial_outbuf[4096];
char *outbuf = initial_outbuf;
size_t outbufsize = sizeof(initial_outbuf);
int status = 0;
#if O_BINARY
SET_BINARY(fileno(infile));
#endif
line = 1; column = 0;
iconv(cd,NULL,NULL,NULL,NULL);
for (;;) {
size_t inbufsize = fread(inbuf+4096,1,4096,infile);
if (inbufsize == 0) {
if (inbufrest == 0)
break;
else {
if (ilseq_byte_subst != NULL)
subst_mb_to_mb_fallback(inbuf+4096-inbufrest, inbufrest);
if (!silent)
conversion_error_EINVAL(infilename);
status = 1;
goto done;
}
} else {
const char* inptr = inbuf+4096-inbufrest;
size_t insize = inbufrest+inbufsize;
inbufrest = 0;
while (insize > 0) {
char* outptr = outbuf;
size_t outsize = outbufsize;
size_t res = iconv(cd,(ICONV_CONST char**)&inptr,&insize,&outptr,&outsize);
if (outptr != outbuf) {
int saved_errno = errno;
if (fwrite(outbuf,1,outptr-outbuf,stdout) < outptr-outbuf) {
status = 1;
goto done;
}
errno = saved_errno;
}
if (res == (size_t)(-1)) {
if (errno == EILSEQ) {
if (discard_unconvertible == 1) {
int one = 1;
iconvctl(cd,ICONV_SET_DISCARD_ILSEQ,&one);
discard_unconvertible = 2;
status = 1;
} else {
if (!silent)
conversion_error_EILSEQ(infilename);
status = 1;
goto done;
}
} else if (errno == EINVAL) {
if (inbufsize == 0 || insize > 4096) {
if (!silent)
conversion_error_EINVAL(infilename);
status = 1;
goto done;
} else {
inbufrest = insize;
if (insize > 0) {
/* Like memcpy(inbuf+4096-insize,inptr,insize), except that
we cannot use memcpy here, because source and destination
regions may overlap. */
char* restptr = inbuf+4096-insize;
do { *restptr++ = *inptr++; } while (--insize > 0);
}
break;
}
} else if (errno == E2BIG) {
if (outptr==outbuf) {
/* outbuf is too small. Double its size. */
if (outbuf != initial_outbuf)
free(outbuf);
outbufsize = 2*outbufsize;
if (outbufsize==0) /* integer overflow? */
xalloc_die();
outbuf = (char*)xmalloc(outbufsize);
}
} else {
if (!silent)
conversion_error_other(errno,infilename);
status = 1;
goto done;
}
}
}
}
}
for (;;) {
char* outptr = outbuf;
size_t outsize = outbufsize;
size_t res = iconv(cd,NULL,NULL,&outptr,&outsize);
if (outptr != outbuf) {
int saved_errno = errno;
if (fwrite(outbuf,1,outptr-outbuf,stdout) < outptr-outbuf) {
status = 1;
goto done;
}
errno = saved_errno;
}
if (res == (size_t)(-1)) {
if (errno == EILSEQ) {
if (discard_unconvertible == 1) {
int one = 1;
iconvctl(cd,ICONV_SET_DISCARD_ILSEQ,&one);
discard_unconvertible = 2;
status = 1;
} else {
if (!silent)
conversion_error_EILSEQ(infilename);
status = 1;
goto done;
}
} else if (errno == EINVAL) {
if (!silent)
conversion_error_EINVAL(infilename);
status = 1;
goto done;
} else if (errno == E2BIG) {
if (outptr==outbuf) {
/* outbuf is too small. Double its size. */
if (outbuf != initial_outbuf)
free(outbuf);
outbufsize = 2*outbufsize;
if (outbufsize==0) /* integer overflow? */
xalloc_die();
outbuf = (char*)xmalloc(outbufsize);
}
} else {
if (!silent)
conversion_error_other(errno,infilename);
status = 1;
goto done;
}
} else
break;
}
if (ferror(infile)) {
fflush(stdout);
if (column > 0)
putc('\n',stderr);
error(0,0,
/* TRANSLATORS: An error message.
The placeholder expands to the input file name. */
_("%s: I/O error"),
infilename);
status = 1;
goto done;
}
done:
if (outbuf != initial_outbuf)
free(outbuf);
return status;
}
/* ========================================================================= */
int main (int argc, char* argv[])
{
const char* fromcode = NULL;
const char* tocode = NULL;
int do_list = 0;
iconv_t cd;
struct iconv_fallbacks fallbacks;
struct iconv_hooks hooks;
int i;
int status;
set_program_name (argv[0]);
#if HAVE_SETLOCALE
/* Needed for the locale dependent encodings, "char" and "wchar_t",
and for gettext. */
setlocale(LC_CTYPE,"");
#if ENABLE_NLS
/* Needed for gettext. */
setlocale(LC_MESSAGES,"");
#endif
#endif
#if ENABLE_NLS
bindtextdomain("libiconv",relocate(LOCALEDIR));
#endif
textdomain("libiconv");
for (i = 1; i < argc;) {
size_t len = strlen(argv[i]);
if (!strcmp(argv[i],"--")) {
i++;
break;
}
if (!strcmp(argv[i],"-f")
/* --f ... --from-code */
|| (len >= 3 && len <= 11 && !strncmp(argv[i],"--from-code",len))
/* --from-code=... */
|| (len >= 12 && !strncmp(argv[i],"--from-code=",12))) {
if (len < 12)
if (i == argc-1) usage(1);
if (fromcode != NULL) usage(1);
if (len < 12) {
fromcode = argv[i+1];
i += 2;
} else {
fromcode = argv[i]+12;
i++;
}
continue;
}
if (!strcmp(argv[i],"-t")
/* --t ... --to-code */
|| (len >= 3 && len <= 9 && !strncmp(argv[i],"--to-code",len))
/* --from-code=... */
|| (len >= 10 && !strncmp(argv[i],"--to-code=",10))) {
if (len < 10)
if (i == argc-1) usage(1);
if (tocode != NULL) usage(1);
if (len < 10) {
tocode = argv[i+1];
i += 2;
} else {
tocode = argv[i]+10;
i++;
}
continue;
}
if (!strcmp(argv[i],"-l")
/* --l ... --list */
|| (len >= 3 && len <= 6 && !strncmp(argv[i],"--list",len))) {
do_list = 1;
i++;
continue;
}
if (/* --by ... --byte-subst */
(len >= 4 && len <= 12 && !strncmp(argv[i],"--byte-subst",len))
/* --byte-subst=... */
|| (len >= 13 && !strncmp(argv[i],"--byte-subst=",13))) {
if (len < 13) {
if (i == argc-1) usage(1);
ilseq_byte_subst = argv[i+1];
i += 2;
} else {
ilseq_byte_subst = argv[i]+13;
i++;
}
ilseq_byte_subst_size =
check_subst_formatstring(ilseq_byte_subst, "--byte-subst");
continue;
}
if (/* --w ... --widechar-subst */
(len >= 3 && len <= 16 && !strncmp(argv[i],"--widechar-subst",len))
/* --widechar-subst=... */
|| (len >= 17 && !strncmp(argv[i],"--widechar-subst=",17))) {
if (len < 17) {
if (i == argc-1) usage(1);
ilseq_wchar_subst = argv[i+1];
i += 2;
} else {
ilseq_wchar_subst = argv[i]+17;
i++;
}
ilseq_wchar_subst_size =
check_subst_formatstring(ilseq_wchar_subst, "--widechar-subst");
continue;
}
if (/* --u ... --unicode-subst */
(len >= 3 && len <= 15 && !strncmp(argv[i],"--unicode-subst",len))
/* --unicode-subst=... */
|| (len >= 16 && !strncmp(argv[i],"--unicode-subst=",16))) {
if (len < 16) {
if (i == argc-1) usage(1);
ilseq_unicode_subst = argv[i+1];
i += 2;
} else {
ilseq_unicode_subst = argv[i]+16;
i++;
}
ilseq_unicode_subst_size =
check_subst_formatstring(ilseq_unicode_subst, "--unicode-subst");
continue;
}
if /* --s ... --silent */
(len >= 3 && len <= 8 && !strncmp(argv[i],"--silent",len)) {
silent = 1;
continue;
}
if /* --h ... --help */
(len >= 3 && len <= 6 && !strncmp(argv[i],"--help",len)) {
usage(0);
}
if /* --v ... --version */
(len >= 3 && len <= 9 && !strncmp(argv[i],"--version",len)) {
print_version();
}
#if O_BINARY
/* Backward compatibility with iconv <= 1.9.1. */
if /* --bi ... --binary */
(len >= 4 && len <= 8 && !strncmp(argv[i],"--binary",len)) {
i++;
continue;
}
#endif
if (argv[i][0] == '-') {
const char *option = argv[i] + 1;
if (*option == '\0')
usage(1);
for (; *option; option++)
switch (*option) {
case 'c': discard_unconvertible = 1; break;
case 's': silent = 1; break;
default: usage(1);
}
i++;
continue;
}
break;
}
if (do_list) {
if (i != 2 || i != argc)
usage(1);
iconvlist(print_one,NULL);
status = 0;
} else {
#if O_BINARY
SET_BINARY(fileno(stdout));
#endif
if (fromcode == NULL)
fromcode = "char";
if (tocode == NULL)
tocode = "char";
cd = iconv_open(tocode,fromcode);
if (cd == (iconv_t)(-1)) {
if (iconv_open("UCS-4",fromcode) == (iconv_t)(-1))
error(0,0,
/* TRANSLATORS: An error message.
The placeholder expands to the encoding name, specified through --from-code. */
_("conversion from %s unsupported"),
fromcode);
else if (iconv_open(tocode,"UCS-4") == (iconv_t)(-1))
error(0,0,
/* TRANSLATORS: An error message.
The placeholder expands to the encoding name, specified through --to-code. */
_("conversion to %s unsupported"),
tocode);
else
error(0,0,
/* TRANSLATORS: An error message.
The placeholders expand to the encoding names, specified through --from-code and --to-code, respectively. */
_("conversion from %s to %s unsupported"),
fromcode,tocode);
error(EXIT_FAILURE,0,
/* TRANSLATORS: Additional advice after an error message.
The %s placeholder expands to the program name. */
_("try '%s -l' to get the list of supported encodings"),
program_name);
}
/* Look at fromcode and tocode, to determine whether character widths
should be determined according to legacy CJK conventions. */
cjkcode = iconv_canonicalize(tocode);
if (!is_cjk_encoding(cjkcode))
cjkcode = iconv_canonicalize(fromcode);
/* Set up fallback routines for handling impossible conversions. */
if (ilseq_byte_subst != NULL)
ilseq_byte_subst_buffer = (char*)xmalloc((ilseq_byte_subst_size+1)*sizeof(char));
if (!discard_unconvertible) {
#if HAVE_WCHAR_T
if (ilseq_wchar_subst != NULL)
ilseq_wchar_subst_buffer = (char*)xmalloc((ilseq_wchar_subst_size+1)*sizeof(char));
#endif
if (ilseq_unicode_subst != NULL)
ilseq_unicode_subst_buffer = (char*)xmalloc((ilseq_unicode_subst_size+1)*sizeof(char));
if (ilseq_byte_subst != NULL) {
subst_mb_to_uc_cd = iconv_open("UCS-4-INTERNAL","char");
subst_mb_to_uc_temp_buffer = (unsigned int*)xmalloc(ilseq_byte_subst_size*sizeof(unsigned int));
#if HAVE_WCHAR_T
subst_mb_to_wc_cd = iconv_open("wchar_t","char");
subst_mb_to_wc_temp_buffer = (wchar_t*)xmalloc(ilseq_byte_subst_size*sizeof(wchar_t));
#endif
subst_mb_to_mb_cd = iconv_open(tocode,"char");
subst_mb_to_mb_temp_buffer = (char*)xmalloc(ilseq_byte_subst_size*4);
}
#if HAVE_WCHAR_T
if (ilseq_wchar_subst != NULL) {
subst_wc_to_mb_cd = iconv_open(tocode,"char");
subst_wc_to_mb_temp_buffer = (char*)xmalloc(ilseq_wchar_subst_size*4);
}
#endif
if (ilseq_unicode_subst != NULL) {
subst_uc_to_mb_cd = iconv_open(tocode,"char");
subst_uc_to_mb_temp_buffer = (char*)xmalloc(ilseq_unicode_subst_size*4);
}
fallbacks.mb_to_uc_fallback =
(ilseq_byte_subst != NULL ? subst_mb_to_uc_fallback : NULL);
fallbacks.uc_to_mb_fallback =
(ilseq_unicode_subst != NULL ? subst_uc_to_mb_fallback : NULL);
fallbacks.mb_to_wc_fallback =
(ilseq_byte_subst != NULL ? subst_mb_to_wc_fallback : NULL);
fallbacks.wc_to_mb_fallback =
(ilseq_wchar_subst != NULL ? subst_wc_to_mb_fallback : NULL);
fallbacks.data = NULL;
iconvctl(cd, ICONV_SET_FALLBACKS, &fallbacks);
}
/* Set up hooks for updating the line and column position. */
hooks.uc_hook = update_line_column;
hooks.wc_hook = NULL;
hooks.data = NULL;
iconvctl(cd, ICONV_SET_HOOKS, &hooks);
if (i == argc)
status = convert(cd,stdin,
/* TRANSLATORS: A filename substitute denoting standard input. */
_("(stdin)"));
else {
status = 0;
for (; i < argc; i++) {
const char* infilename = argv[i];
FILE* infile = fopen(infilename,"r");
if (infile == NULL) {
int saved_errno = errno;
error(0,saved_errno,
/* TRANSLATORS: The first part of an error message.
It is followed by a colon and a detail message.
The %s placeholder expands to the input file name. */
_("%s"),
infilename);
status = 1;
} else {
status |= convert(cd,infile,infilename);
fclose(infile);
}
}
}
iconv_close(cd);
}
if (ferror(stdout) || fclose(stdout)) {
error(0,0,
/* TRANSLATORS: An error message. */
_("I/O error"));
status = 1;
}
exit(status);
}