/* Convert text in given files from the specified from-set to the to-set.
- Copyright (C) 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
+ Copyright (C) 1998-2014 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1998.
- The GNU C Library is free software; you can redistribute it and/or
- modify it under the terms of the GNU Library 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 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; version 2 of the License, or
+ (at your option) any later version.
- The GNU C Library is distributed in the hope that it will be useful,
+ 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
- Library General Public License for more details.
+ 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 Library General Public
- License along with the GNU C Library; see the file COPYING.LIB. If not,
- write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- Boston, MA 02111-1307, USA. */
+ 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 <argp.h>
#include <assert.h>
#include <langinfo.h>
#include <locale.h>
#include <search.h>
+#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <charmap.h>
#include <gconv_int.h>
#include "iconv_prog.h"
+#include "iconvconfig.h"
/* Get libc version number. */
#include "../version.h"
static const struct argp_option options[] =
{
{ NULL, 0, NULL, 0, N_("Input/Output format specification:") },
- { "from-code", 'f', "NAME", 0, N_("encoding of original text") },
- { "to-code", 't', "NAME", 0, N_("encoding for output") },
+ { "from-code", 'f', N_("NAME"), 0, N_("encoding of original text") },
+ { "to-code", 't', N_("NAME"), 0, N_("encoding for output") },
{ NULL, 0, NULL, 0, N_("Information:") },
{ "list", 'l', NULL, 0, N_("list all known coded character sets") },
{ NULL, 0, NULL, 0, N_("Output control:") },
{ NULL, 'c', NULL, 0, N_("omit invalid characters from output") },
- { "output", 'o', "FILE", 0, N_("output file") },
+ { "output", 'o', N_("FILE"), 0, N_("output file") },
{ "silent", 's', NULL, 0, N_("suppress warnings") },
{ "verbose", OPT_VERBOSE, NULL, 0, N_("print progress information") },
{ NULL, 0, NULL, 0, NULL }
options, parse_opt, args_doc, doc, NULL, more_help
};
-/* Code sets to convert from and to respectively. */
-static const char *from_code;
-static const char *to_code;
+/* Code sets to convert from and to respectively. An empty string as the
+ default causes the 'iconv_open' function to look up the charset of the
+ currently selected locale and use it. */
+static const char *from_code = "";
+static const char *to_code = "";
/* File to write output to. If NULL write to stdout. */
static const char *output_file;
int omit_invalid;
/* Prototypes for the functions doing the actual work. */
-static int process_block (iconv_t cd, char *addr, size_t len, FILE *output);
-static int process_fd (iconv_t cd, int fd, FILE *output);
-static int process_file (iconv_t cd, FILE *input, FILE *output);
+static int process_block (iconv_t cd, char *addr, size_t len, FILE **output,
+ const char *output_file);
+static int process_fd (iconv_t cd, int fd, FILE **output,
+ const char *output_file);
+static int process_file (iconv_t cd, FILE *input, FILE **output,
+ const char *output_file);
static void print_known_names (void) internal_function;
{
int status = EXIT_SUCCESS;
int remaining;
- FILE *output;
iconv_t cd;
const char *orig_to_code;
struct charmap_t *from_charmap = NULL;
print_known_names ();
exit (EXIT_SUCCESS);
}
- if (from_code == NULL)
- {
- /* The Unix standard says that in this case the charset of the current
- locale is used. */
- from_code = nl_langinfo (CODESET);
- assert (from_code != NULL);
- }
- if (to_code == NULL)
- {
- /* The Unix standard says that in this case the charset of the current
- locale is used. */
- to_code = nl_langinfo (CODESET);
- assert (to_code != NULL);
- }
/* If we have to ignore errors make sure we use the appropriate name for
the to-character-set. */
if (*errhand == '/')
{
--nslash;
- errhand = strchrnul (errhand, '/');
+ errhand = strchrnul (errhand + 1, '/');
if (*errhand == '/')
{
--nslash;
- ++errhand;
+ errhand = strchr (errhand, '\0');
}
}
- newp = (char *) alloca (errhand - to_code + nslash + 6 + 1);
+ newp = (char *) alloca (errhand - to_code + nslash + 7 + 1);
cp = mempcpy (newp, to_code, errhand - to_code);
while (nslash-- > 0)
*cp++ = '/';
+ if (cp[-1] != '/')
+ *cp++ = ',';
memcpy (cp, "IGNORE", sizeof ("IGNORE"));
to_code = newp;
if (strchr (from_code, '/') != NULL)
/* The from-name might be a charmap file name. Try reading the
file. */
- from_charmap = charmap_read (from_code, /*0, 1*/1, 0, 0);
+ from_charmap = charmap_read (from_code, /*0, 1*/1, 0, 0, 0);
if (strchr (orig_to_code, '/') != NULL)
/* The to-name might be a charmap file name. Try reading the
file. */
- to_charmap = charmap_read (orig_to_code, /*0, 1,*/1,0, 0);
-
+ to_charmap = charmap_read (orig_to_code, /*0, 1,*/1, 0, 0, 0);
- /* Determine output file. */
- if (output_file != NULL && strcmp (output_file, "-") != 0)
- {
- output = fopen (output_file, "w");
- if (output == NULL)
- error (EXIT_FAILURE, errno, _("cannot open output file"));
- }
- else
- output = stdout;
/* At this point we have to handle two cases. The first one is
where a charmap is used for the from- or to-charset, or both. We
if (from_charmap != NULL || to_charmap != NULL)
/* Construct the conversion table and do the conversion. */
status = charmap_conversion (from_code, from_charmap, to_code, to_charmap,
- argc, remaining, argv, output);
+ argc, remaining, argv, output_file);
else
{
/* Let's see whether we have these coded character sets. */
if (cd == (iconv_t) -1)
{
if (errno == EINVAL)
- error (EXIT_FAILURE, 0,
- _("conversion from `%s' to `%s' not supported"),
- from_code, orig_to_code);
+ {
+ /* Try to be nice with the user and tell her which of the
+ two encoding names is wrong. This is possible because
+ all supported encodings can be converted from/to Unicode,
+ in other words, because the graph of encodings is
+ connected. */
+ bool from_wrong =
+ (iconv_open ("UTF-8", from_code) == (iconv_t) -1
+ && errno == EINVAL);
+ bool to_wrong =
+ (iconv_open (to_code, "UTF-8") == (iconv_t) -1
+ && errno == EINVAL);
+ const char *from_pretty =
+ (from_code[0] ? from_code : nl_langinfo (CODESET));
+ const char *to_pretty =
+ (orig_to_code[0] ? orig_to_code : nl_langinfo (CODESET));
+
+ if (from_wrong)
+ {
+ if (to_wrong)
+ error (0, 0,
+ _("\
+conversions from `%s' and to `%s' are not supported"),
+ from_pretty, to_pretty);
+ else
+ error (0, 0,
+ _("conversion from `%s' is not supported"),
+ from_pretty);
+ }
+ else
+ {
+ if (to_wrong)
+ error (0, 0,
+ _("conversion to `%s' is not supported"),
+ to_pretty);
+ else
+ error (0, 0,
+ _("conversion from `%s' to `%s' is not supported"),
+ from_pretty, to_pretty);
+ }
+
+ argp_help (&argp, stderr, ARGP_HELP_SEE,
+ program_invocation_short_name);
+ exit (1);
+ }
else
error (EXIT_FAILURE, errno,
_("failed to start conversion processing"));
}
+ /* The output file. Will be opened when we are ready to produce
+ output. */
+ FILE *output = NULL;
+
/* Now process the remaining files. Write them to stdout or the file
specified with the `-o' parameter. If we have no file given as
the parameter process all from stdin. */
if (remaining == argc)
{
- if (process_file (cd, stdin, output) != 0)
+ if (process_file (cd, stdin, &output, output_file) != 0)
status = EXIT_FAILURE;
}
else
do
{
- struct stat st;
+#ifdef _POSIX_MAPPED_FILES
+ struct stat64 st;
char *addr;
- int fd;
+#endif
+ int fd, ret;
if (verbose)
- printf ("%s:\n", argv[remaining]);
+ fprintf (stderr, "%s:\n", argv[remaining]);
if (strcmp (argv[remaining], "-") == 0)
fd = 0;
else
#ifdef _POSIX_MAPPED_FILES
/* We have possibilities for reading the input file. First try
to mmap() it since this will provide the fastest solution. */
- if (fstat (fd, &st) == 0
+ if (fstat64 (fd, &st) == 0
&& ((addr = mmap (NULL, st.st_size, PROT_READ, MAP_PRIVATE,
fd, 0)) != MAP_FAILED))
{
_("error while closing input `%s'"),
argv[remaining]);
- if (process_block (cd, addr, st.st_size, output) < 0)
+ ret = process_block (cd, addr, st.st_size, &output,
+ output_file);
+
+ /* We don't need the input data anymore. */
+ munmap ((void *) addr, st.st_size);
+
+ if (ret != 0)
{
- /* Something went wrong. */
status = EXIT_FAILURE;
- /* We don't need the input data anymore. */
- munmap ((void *) addr, st.st_size);
-
- /* We cannot go on with producing output since it might
- lead to problem because the last output might leave
- the output stream in an undefined state. */
- break;
+ if (ret < 0)
+ /* We cannot go on with producing output since it might
+ lead to problem because the last output might leave
+ the output stream in an undefined state. */
+ break;
}
-
- /* We don't need the input data anymore. */
- munmap ((void *) addr, st.st_size);
}
else
#endif /* _POSIX_MAPPED_FILES */
{
/* Read the file in pieces. */
- if (process_fd (cd, fd, output) != 0)
+ ret = process_fd (cd, fd, &output, output_file);
+
+ /* Now close the file. */
+ close (fd);
+
+ if (ret != 0)
{
/* Something went wrong. */
status = EXIT_FAILURE;
- /* We don't need the input file anymore. */
- close (fd);
-
- /* We cannot go on with producing output since it might
- lead to problem because the last output might leave
- the output stream in an undefined state. */
- break;
+ if (ret < 0)
+ /* We cannot go on with producing output since it might
+ lead to problem because the last output might leave
+ the output stream in an undefined state. */
+ break;
}
-
- /* Now close the file. */
- close (fd);
}
}
while (++remaining < argc);
- }
- /* Close the output file now. */
- if (fclose (output))
- error (EXIT_FAILURE, errno, _("error while closing output file"));
+ /* Close the output file now. */
+ if (output != NULL && fclose (output))
+ error (EXIT_FAILURE, errno, _("error while closing output file"));
+ }
return status;
}
static char *
more_help (int key, const char *text, void *input)
{
+ char *tp = NULL;
switch (key)
{
case ARGP_KEY_HELP_EXTRA:
/* We print some extra information. */
- return strdup (gettext ("\
-Report bugs using the `glibcbug' script to <bugs@gnu.org>.\n"));
+ if (asprintf (&tp, gettext ("\
+For bug reporting instructions, please see:\n\
+%s.\n"), REPORT_BUGS_TO) < 0)
+ return NULL;
+ return tp;
default:
break;
}
static void
print_version (FILE *stream, struct argp_state *state)
{
- fprintf (stream, "iconv (GNU %s) %s\n", PACKAGE, VERSION);
+ fprintf (stream, "iconv %s%s\n", PKGVERSION, VERSION);
fprintf (stream, gettext ("\
Copyright (C) %s Free Software Foundation, Inc.\n\
This is free software; see the source for copying conditions. There is NO\n\
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.\n\
-"), "2001");
+"), "2013");
fprintf (stream, gettext ("Written by %s.\n"), "Ulrich Drepper");
}
static int
-process_block (iconv_t cd, char *addr, size_t len, FILE *output)
+write_output (const char *outbuf, const char *outptr, FILE **output,
+ const char *output_file)
+{
+ /* We have something to write out. */
+ int errno_save = errno;
+
+ if (*output == NULL)
+ {
+ /* Determine output file. */
+ if (output_file != NULL && strcmp (output_file, "-") != 0)
+ {
+ *output = fopen (output_file, "w");
+ if (*output == NULL)
+ error (EXIT_FAILURE, errno, _("cannot open output file"));
+ }
+ else
+ *output = stdout;
+ }
+
+ if (fwrite (outbuf, 1, outptr - outbuf, *output) < (size_t) (outptr - outbuf)
+ || ferror (*output))
+ {
+ /* Error occurred while printing the result. */
+ error (0, 0, _("\
+conversion stopped due to problem in writing the output"));
+ return -1;
+ }
+
+ errno = errno_save;
+
+ return 0;
+}
+
+
+static int
+process_block (iconv_t cd, char *addr, size_t len, FILE **output,
+ const char *output_file)
{
#define OUTBUF_SIZE 32768
const char *start = addr;
char *outptr;
size_t outlen;
size_t n;
+ int ret = 0;
while (len > 0)
{
outlen = OUTBUF_SIZE;
n = iconv (cd, &addr, &len, &outptr, &outlen);
- if (outptr != outbuf)
+ if (n == (size_t) -1 && omit_invalid && errno == EILSEQ)
{
- /* We have something to write out. */
- int errno_save = errno;
-
- if (fwrite (outbuf, 1, outptr - outbuf, output) < outptr - outbuf
- || ferror (output))
- {
- /* Error occurred while printing the result. */
- error (0, 0, _("\
-conversion stopped due to problem in writing the output"));
- return -1;
- }
+ ret = 1;
+ if (len == 0)
+ n = 0;
+ else
+ errno = E2BIG;
+ }
- errno = errno_save;
+ if (outptr != outbuf)
+ {
+ ret = write_output (outbuf, outptr, output, output_file);
+ if (ret != 0)
+ break;
}
if (n != (size_t) -1)
{
/* All the input test is processed. For state-dependent
- character sets we have to flush the state now. */
+ character sets we have to flush the state now. */
outptr = outbuf;
outlen = OUTBUF_SIZE;
- (void) iconv (cd, NULL, NULL, &outptr, &outlen);
+ n = iconv (cd, NULL, NULL, &outptr, &outlen);
if (outptr != outbuf)
{
- /* We have something to write out. */
- int errno_save = errno;
+ ret = write_output (outbuf, outptr, output, output_file);
+ if (ret != 0)
+ break;
+ }
- if (fwrite (outbuf, 1, outptr - outbuf, output) < outptr - outbuf
- || ferror (output))
- {
- /* Error occurred while printing the result. */
- error (0, 0, _("\
-conversion stopped due to problem in writing the output"));
- return -1;
- }
+ if (n != (size_t) -1)
+ break;
- errno = errno_save;
+ if (omit_invalid && errno == EILSEQ)
+ {
+ ret = 1;
+ break;
}
-
- break;
}
if (errno != E2BIG)
switch (errno)
{
case EILSEQ:
- error (0, 0, _("illegal input sequence at position %ld"),
- (long) (addr - start));
+ if (! omit_invalid)
+ error (0, 0, _("illegal input sequence at position %ld"),
+ (long int) (addr - start));
break;
case EINVAL:
error (0, 0, _("\
}
}
- return 0;
+ return ret;
}
static int
-process_fd (iconv_t cd, int fd, FILE *output)
+process_fd (iconv_t cd, int fd, FILE **output, const char *output_file)
{
/* we have a problem with reading from a desriptor since we must not
provide the iconv() function an incomplete character or shift
while (1)
{
ssize_t n;
+ char *new_inbuf;
/* Increase the buffer. */
+ new_inbuf = (char *) realloc (inbuf, maxlen + 32768);
+ if (new_inbuf == NULL)
+ {
+ error (0, errno, _("unable to allocate buffer for input"));
+ return -1;
+ }
+ inbuf = new_inbuf;
maxlen += 32768;
- inbuf = realloc (inbuf, maxlen);
- if (inbuf == NULL)
- error (0, errno, _("unable to allocate buffer for input"));
inptr = inbuf + actlen;
do
}
/* Now we have all the input in the buffer. Process it in one run. */
- return process_block (cd, inbuf, actlen, output);
+ return process_block (cd, inbuf, actlen, output, output_file);
}
static int
-process_file (iconv_t cd, FILE *input, FILE *output)
+process_file (iconv_t cd, FILE *input, FILE **output, const char *output_file)
{
/* This should be safe since we use this function only for `stdin' and
we haven't read anything so far. */
- return process_fd (cd, fileno (input), output);
+ return process_fd (cd, fileno (input), output, output_file);
}
add_known_names (node->right);
do
{
- if (strcmp (node->from_string, "INTERNAL"))
- tsearch (node->from_string, &printlist,
- (__compar_fn_t) strverscmp);
- if (strcmp (node->to_string, "INTERNAL"))
+ if (strcmp (node->from_string, "INTERNAL") != 0)
+ tsearch (node->from_string, &printlist, (__compar_fn_t) strverscmp);
+ if (strcmp (node->to_string, "INTERNAL") != 0)
tsearch (node->to_string, &printlist, (__compar_fn_t) strverscmp);
node = node->same;
while (node != NULL);
}
+
+static void
+insert_cache (void)
+{
+ const struct gconvcache_header *header;
+ const char *strtab;
+ const struct hash_entry *hashtab;
+ size_t cnt;
+
+ header = (const struct gconvcache_header *) __gconv_get_cache ();
+ strtab = (char *) header + header->string_offset;
+ hashtab = (struct hash_entry *) ((char *) header + header->hash_offset);
+
+ for (cnt = 0; cnt < header->hash_size; ++cnt)
+ if (hashtab[cnt].string_offset != 0)
+ {
+ const char *str = strtab + hashtab[cnt].string_offset;
+
+ if (strcmp (str, "INTERNAL") != 0)
+ tsearch (str, &printlist, (__compar_fn_t) strverscmp);
+ }
+}
+
+
static void
internal_function
print_known_names (void)
{
iconv_t h;
+ void *cache;
/* We must initialize the internal databases first. */
h = iconv_open ("L1", "L1");
iconv_close (h);
- /* First add the aliases. */
- twalk (__gconv_alias_db, insert_print_list);
+ /* See whether we have a cache. */
+ cache = __gconv_get_cache ();
+ if (cache != NULL)
+ /* Yep, use only this information. */
+ insert_cache ();
+ else
+ {
+ struct gconv_module *modules;
- /* Add the from- and to-names from the known modules. */
- add_known_names (__gconv_modules_db);
+ /* No, then use the information read from the gconv-modules file.
+ First add the aliases. */
+ twalk (__gconv_get_alias_db (), insert_print_list);
- fputs (_("\
-The following list contain all the coded character sets known. This does\n\
+ /* Add the from- and to-names from the known modules. */
+ modules = __gconv_get_modules_db ();
+ if (modules != NULL)
+ add_known_names (modules);
+ }
+
+ bool human_readable = isatty (fileno (stdout));
+
+ if (human_readable)
+ fputs (_("\
+The following list contains all the coded character sets known. This does\n\
not necessarily mean that all combinations of these names can be used for\n\
the FROM and TO command line parameters. One coded character set can be\n\
listed with several different names (aliases).\n\n "), stdout);
/* Now print the collected names. */
column = 2;
- if (isatty (fileno (stdout)))
- {
- twalk (printlist, do_print_human);
+ twalk (printlist, human_readable ? do_print_human : do_print);
- if (column != 0)
- puts ("");
- }
- else
- twalk (printlist, do_print);
+ if (human_readable && column != 0)
+ puts ("");
}
projects / thirdparty / glibc.git / blobdiff