1 /* $NetBSD: iconv.c,v 1.21 2025/11/21 15:20:54 christos Exp $ */ 2 3 /*- 4 * Copyright (c)2003 Citrus Project, 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 26 * SUCH DAMAGE. 27 */ 28 29 #include <sys/cdefs.h> 30 #if defined(LIBC_SCCS) && !defined(lint) 31 __RCSID("$NetBSD: iconv.c,v 1.21 2025/11/21 15:20:54 christos Exp $"); 32 #endif /* LIBC_SCCS and not lint */ 33 34 #include <err.h> 35 #include <errno.h> 36 #include <iconv.h> 37 #include <langinfo.h> 38 #include <locale.h> 39 #include <stdio.h> 40 #include <stdlib.h> 41 #include <string.h> 42 #include <unistd.h> 43 #include <util.h> 44 45 static void usage(void) __dead; 46 static int scmp(const void *, const void *); 47 static void show_codesets(void); 48 static void do_conv(const char *, const char *, FILE *, FILE *, 49 const char *, const char *, int, int); 50 51 static void 52 usage(void) 53 { 54 (void)fprintf(stderr, 55 "Usage:\t%1$s [-cs] [-o <outfile>] -f <from_code> -t <to_code> [file ...]\n" 56 "\t%1$s [-cs] [-o <outfile>] -f <from_code> [-t <to_code>] [file ...]\n" 57 "\t%1$s [-cs] [-o <outfile>] -t <to_code> [-f <from_code>] [file ...]\n" 58 "\t%1$s -l\n", getprogname()); 59 exit(EXIT_FAILURE); 60 } 61 62 /* 63 * qsort() helper function 64 */ 65 static int 66 scmp(const void *v1, const void *v2) 67 { 68 const char * const *s1 = v1; 69 const char * const *s2 = v2; 70 71 return strcasecmp(*s1, *s2); 72 } 73 74 static void 75 show_codesets(void) 76 { 77 char **list; 78 size_t sz, i; 79 80 if (__iconv_get_list(&list, &sz)) 81 err(EXIT_FAILURE, "__iconv_get_list()"); 82 83 qsort(list, sz, sizeof(char *), scmp); 84 85 for (i = 0; i < sz; i++) 86 (void)printf("%s\n", list[i]); 87 88 __iconv_free_list(list, sz); 89 } 90 91 #define INBUFSIZE 1024 92 #define OUTBUFSIZE (INBUFSIZE * 2) 93 /*ARGSUSED*/ 94 static void 95 do_conv(const char *fni, const char *fno, FILE *fi, FILE *fo, 96 const char *from, const char *to, int silent, int hide_invalid) 97 { 98 char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *out; 99 const char *in; 100 size_t inbytes, outbytes, ret, invalids; 101 iconv_t cd; 102 uint32_t flags = 0; 103 int serrno; 104 105 if (hide_invalid) 106 flags |= __ICONV_F_HIDE_INVALID; 107 cd = iconv_open(to, from); 108 if (cd == (iconv_t)-1) 109 err(EXIT_FAILURE, "iconv_open(%s, %s)", to, from); 110 111 invalids = 0; 112 while ((inbytes = fread(inbuf, 1, INBUFSIZE, fi)) > 0) { 113 in = inbuf; 114 while (inbytes > 0) { 115 size_t inval; 116 117 out = outbuf; 118 outbytes = OUTBUFSIZE; 119 ret = __iconv(cd, __UNCONST(&in), &inbytes, &out, 120 &outbytes, flags, &inval); 121 serrno = errno; 122 invalids += inval; 123 if (outbytes < OUTBUFSIZE) { 124 if (fwrite(outbuf, 1, OUTBUFSIZE - outbytes, fo) 125 != OUTBUFSIZE - outbytes) 126 err(EXIT_FAILURE, 127 "failed writing to `%s'", fno); 128 } 129 errno = serrno; 130 if (ret == (size_t)-1 && errno != E2BIG) { 131 /* 132 * XXX: iconv(3) is bad interface. 133 * invalid character count is lost here. 134 * instead, we just provide __iconv function. 135 */ 136 if (errno != EINVAL || in == inbuf) 137 err(EXIT_FAILURE, "iconv()"); 138 139 /* incomplete input character */ 140 (void)memmove(inbuf, in, inbytes); 141 ret = fread(inbuf + inbytes, 1, 142 INBUFSIZE - inbytes, fi); 143 if (ret == 0) { 144 fflush(fo); 145 if (feof(fi)) 146 errx(EXIT_FAILURE, 147 "unexpected end of file; " 148 "the last character is " 149 "incomplete."); 150 else 151 err(EXIT_FAILURE, "fread()"); 152 } 153 in = inbuf; 154 inbytes += ret; 155 } 156 } 157 } 158 /* reset the shift state of the output buffer */ 159 outbytes = OUTBUFSIZE; 160 out = outbuf; 161 ret = iconv(cd, NULL, NULL, &out, &outbytes); 162 if (ret == (size_t)-1) 163 err(EXIT_FAILURE, "iconv()"); 164 if (outbytes < OUTBUFSIZE) { 165 if (fwrite(outbuf, 1, OUTBUFSIZE - outbytes, fo) 166 != OUTBUFSIZE - outbytes) 167 err(EXIT_FAILURE, "failed writing to `%s'", fno); 168 } 169 170 if (invalids > 0 && !silent) 171 warnx("warning: invalid characters: %lu", 172 (unsigned long)invalids); 173 174 iconv_close(cd); 175 } 176 177 int 178 main(int argc, char **argv) 179 { 180 int ch, i; 181 int opt_l = 0, opt_s = 0, opt_c = 0; 182 const char *opt_f = NULL, *opt_t = NULL, *opt_o = NULL; 183 FILE *fi, *fo; 184 185 setlocale(LC_ALL, ""); 186 setprogname(argv[0]); 187 188 while ((ch = getopt(argc, argv, "cslf:o:t:")) != EOF) { 189 switch (ch) { 190 case 'c': 191 opt_c = 1; 192 break; 193 case 's': 194 opt_s = 1; 195 break; 196 case 'l': 197 /* list */ 198 opt_l = 1; 199 break; 200 case 'f': 201 /* from */ 202 opt_f = estrdup(optarg); 203 break; 204 case 't': 205 /* to */ 206 opt_t = estrdup(optarg); 207 break; 208 case 'o': 209 /* outfile */ 210 opt_o = estrdup(optarg); 211 break; 212 default: 213 usage(); 214 } 215 } 216 argc -= optind; 217 argv += optind; 218 if (opt_l) { 219 if (argc > 0 || opt_s || opt_f != NULL || opt_t != NULL) { 220 warnx("-l is not allowed with other flags."); 221 usage(); 222 } 223 show_codesets(); 224 } else { 225 if (opt_f == NULL) { 226 if (opt_t == NULL) 227 usage(); 228 opt_f = nl_langinfo(CODESET); 229 } else if (opt_t == NULL) 230 opt_t = nl_langinfo(CODESET); 231 232 if (opt_o == NULL) { 233 fo = stdout; 234 opt_o = "<stdout>"; 235 } else { 236 fo = fopen(opt_o, "w"); 237 if (fo == NULL) 238 err(EXIT_FAILURE, "Cannot open `%s'", opt_o); 239 } 240 241 if (argc == 0) 242 do_conv("<stdin>", opt_o, stdin, fo, opt_f, opt_t, 243 opt_s, opt_c); 244 else { 245 for (i = 0; i < argc; i++) { 246 fi = fopen(argv[i], "r"); 247 if (fi == NULL) 248 err(EXIT_FAILURE, "Cannot open `%s'", 249 argv[i]); 250 do_conv(argv[i], opt_o, fi, fo, opt_f, opt_t, 251 opt_s, opt_c); 252 (void)fclose(fi); 253 } 254 } 255 if (fo != stdout) 256 (void)fclose(fo); 257 } 258 return EXIT_SUCCESS; 259 } 260
Indexes created Sun Oct 25 00:26:45 UTC 2026