usr.bin/iconv/iconv.c

   1 /*      $NetBSD: iconv.c,v 1.4 2003/10/20 12:56:18 yamt Exp $   */
   2 /*      $DragonFly: src/usr.bin/iconv/iconv.c,v 1.2 2008/07/10 18:29:51 swildner Exp $ */
   3
   4 /*-
   5  * Copyright (c)2003 Citrus Project,
   6  * All rights reserved.
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  *
  17  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  27  * SUCH DAMAGE.
  28  */
  29
  30 #include <sys/endian.h>
  31 #include <err.h>
  32 #include <errno.h>
  33 #include <iconv.h>
  34 #include <stdio.h>
  35 #include <stdlib.h>
  36 #include <string.h>
  37 #include <unistd.h>
  38
  39 static void
  40 usage(void)
  41 {
  42         fprintf(stderr,
  43                 "usage:\n"
  44                 "\t%s [-cs] -f <from> -t <to> [file ...]\n"
  45                 "\t%s -l\n",
  46                 getprogname(), getprogname());
  47         exit(1);
  48 }
  49
  50 /*
  51  * qsort() helper function
  52  */
  53 static int
  54 scmp(const void *v1, const void *v2)
  55 {
  56         const char * const *s1 = v1;
  57         const char * const *s2 = v2;
  58
  59         return(strcasecmp(*s1, *s2));
  60 }
  61
  62 static void
  63 show_codesets(void)
  64 {
  65         char **list;
  66         size_t sz, i;
  67
  68         if (__iconv_get_list(&list, &sz))
  69                 err(EXIT_FAILURE, "__iconv_get_list()");
  70
  71         qsort(list, sz, sizeof(char *), scmp);
  72
  73         for (i=0; i<sz; i++) {
  74                 printf("%s\n", list[i]);
  75         }
  76
  77         __iconv_free_list(list, sz);
  78 }
  79
  80 #define INBUFSIZE 1024
  81 #define OUTBUFSIZE (INBUFSIZE*2)
  82
  83 static void
  84 do_conv(FILE *fp, const char *from, const char *to, int silent,
  85         int hide_invalid)
  86 {
  87         char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *out;
  88         const char *in;
  89         size_t inbytes, outbytes, invalids;
  90         ssize_t ret;
  91         iconv_t cd;
  92         u_int32_t flags = 0;
  93
  94         if (hide_invalid)
  95                 flags |= __ICONV_F_HIDE_INVALID;
  96         cd = iconv_open(to, from);
  97         if (cd == (iconv_t)-1)
  98                 err(EXIT_FAILURE, "iconv_open(%s, %s)", to, from);
  99
 100         invalids = 0;
 101         while ((inbytes = fread(inbuf, 1, INBUFSIZE, fp)) > 0) {
 102                 in = inbuf;
 103                 while (inbytes>0) {
 104                         size_t inval;
 105
 106                         out = outbuf;
 107                         outbytes = OUTBUFSIZE;
 108                         ret = __iconv(cd, &in, &inbytes, &out, &outbytes,
 109                                       flags, &inval);
 110                         invalids += inval;
 111                         if (ret == -1 && errno != E2BIG) {
 112                                 /*
 113                                  * XXX: iconv(3) is bad interface.
 114                                  *   invalid character count is lost here.
 115                                  *   instead, we just provide __iconv function.
 116                                  */
 117                                 if (errno != EINVAL || in == inbuf)
 118                                         err(EXIT_FAILURE, "iconv()");
 119
 120                                 /* incomplete input character */
 121                                 memmove(inbuf, in, inbytes);
 122                                 ret = fread(inbuf+inbytes, 1,
 123                                             INBUFSIZE-inbytes, fp);
 124                                 if (ret == 0) {
 125                                         if (feof(fp))
 126                                                 errx(EXIT_FAILURE,
 127                                                      "iconv(): %s",
 128                                                      strerror(EINVAL));
 129                                         else
 130                                                 err(EXIT_FAILURE, "fread()");
 131                                 }
 132                                 in = inbuf;
 133                                 inbytes += ret;
 134                         }
 135                         if (outbytes < OUTBUFSIZE)
 136                                 fwrite(outbuf, 1, OUTBUFSIZE-outbytes, stdout);
 137                 }
 138         }
 139         /* reset the shift state of the output buffer */
 140         outbytes = OUTBUFSIZE;
 141         out = outbuf;
 142         ret = iconv(cd, NULL, NULL, &out, &outbytes);
 143         if (ret == -1)
 144                 err(EXIT_FAILURE, "iconv()");
 145         if (outbytes < OUTBUFSIZE)
 146                 fwrite(outbuf, 1, OUTBUFSIZE-outbytes, stdout);
 147
 148         if (invalids > 0 && !silent)
 149                 warnx("warning: invalid characters: %lu",
 150                       (unsigned long)invalids);
 151
 152         iconv_close(cd);
 153 }
 154
 155 int
 156 main(int argc, char **argv)
 157 {
 158         int ch, i;
 159         int opt_l = 0, opt_s = 0, opt_c = 0;
 160         char *opt_f = NULL, *opt_t = NULL;
 161         FILE *fp;
 162
 163         while ((ch=getopt(argc, argv, "cslf:t:")) != -1) {
 164                 switch (ch) {
 165                 case 'c':
 166                         opt_c = 1;
 167                         break;
 168                 case 's':
 169                         opt_s = 1;
 170                         break;
 171                 case 'l':
 172                         /* list */
 173                         opt_l = 1;
 174                         break;
 175                 case 'f':
 176                         /* from */
 177                         opt_f = strdup(optarg);
 178                         break;
 179                 case 't':
 180                         /* to */
 181                         opt_t = strdup(optarg);
 182                         break;
 183                 default:
 184                         usage();
 185                 }
 186         }
 187         argc-=optind;
 188         argv+=optind;
 189         if (opt_l) {
 190                 if (argc>0 || opt_s || opt_f != NULL || opt_t != NULL) {
 191                         warnx("%s: -l should be specified solely.",
 192                               getprogname());
 193                         usage();
 194                 }
 195                 show_codesets();
 196         } else {
 197                 if (opt_f == NULL || opt_t == NULL)
 198                         usage();
 199
 200                 if (argc == 0)
 201                         do_conv(stdin, opt_f, opt_t, opt_s, opt_c);
 202                 else {
 203                         for (i=0; i<argc; i++) {
 204                                 fp = fopen(argv[i], "r");
 205                                 if (fp == NULL)
 206                                         errx(EXIT_FAILURE, "%s: %s:%s",
 207                                              getprogname(), argv[i],
 208                                              strerror(errno));
 209                                 do_conv(fp, opt_f, opt_t, opt_s, opt_c);
 210                                 fclose(fp);
 211                         }
 212                 }
 213         }
 214
 215         return(EXIT_SUCCESS);
 216 }