1 /* $NetBSD: iconv.c,v 1.16 2009/02/20 15:28:21 yamt Exp $ */ 2 3 /*- 4 * SPDX-License-Identifier: BSD-2-Clause 5 * 6 * Copyright (c)2003 Citrus Project, 7 * All rights reserved. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 18 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 19 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 20 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 21 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 22 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 23 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 24 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 25 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 26 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 27 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 28 * SUCH DAMAGE. 29 */ 30 31 #include <sys/cdefs.h> 32 #include <sys/capsicum.h> 33 34 #include <capsicum_helpers.h> 35 #include <err.h> 36 #include <errno.h> 37 #include <getopt.h> 38 #include <iconv.h> 39 #include <limits.h> 40 #include <locale.h> 41 #include <stdbool.h> 42 #include <stdio.h> 43 #include <stdlib.h> 44 #include <string.h> 45 #include <unistd.h> 46 47 static int do_conv(FILE *, iconv_t, bool, bool); 48 static int do_list(unsigned int, const char * const *, void *); 49 static void usage(void) __dead2; 50 51 static const struct option long_options[] = { 52 {"from-code", required_argument, NULL, 'f'}, 53 {"list", no_argument, NULL, 'l'}, 54 {"silent", no_argument, NULL, 's'}, 55 {"to-code", required_argument, NULL, 't'}, 56 {NULL, no_argument, NULL, 0} 57 }; 58 59 static void 60 usage(void) 61 { 62 (void)fprintf(stderr, 63 "Usage:\t%1$s [-cs] -f <from_code> -t <to_code> [file ...]\n" 64 "\t%1$s -f <from_code> [-cs] [-t <to_code>] [file ...]\n" 65 "\t%1$s -t <to_code> [-cs] [-f <from_code>] [file ...]\n" 66 "\t%1$s -l\n", getprogname()); 67 exit(1); 68 } 69 70 #define INBUFSIZE 1024 71 #define OUTBUFSIZE (INBUFSIZE * 2) 72 static int 73 do_conv(FILE *fp, iconv_t cd, bool silent, bool hide_invalid) 74 { 75 char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *in, *out; 76 unsigned long long invalids; 77 size_t inbytes, outbytes, ret; 78 79 /* 80 * Don't touch ICONV_SET_DISCARD_ILSEQ if -c wasn't specified. It may 81 * be that the user has specified //IGNORE in the -t specification, and 82 * we don't want to clobber that. 83 */ 84 if (hide_invalid) { 85 int arg = (int)hide_invalid; 86 if (iconvctl(cd, ICONV_SET_DISCARD_ILSEQ, (void *)&arg) == -1) 87 err(EXIT_FAILURE, "iconvctl(DISCARD_ILSEQ, %d)", arg); 88 } 89 90 invalids = 0; 91 while ((inbytes = fread(inbuf, 1, INBUFSIZE, fp)) > 0) { 92 in = inbuf; 93 while (inbytes > 0) { 94 size_t inval; 95 96 out = outbuf; 97 outbytes = OUTBUFSIZE; 98 ret = __iconv(cd, &in, &inbytes, &out, &outbytes, 99 0, &inval); 100 invalids += inval; 101 if (outbytes < OUTBUFSIZE) 102 (void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, 103 stdout); 104 if (ret == (size_t)-1 && errno != E2BIG) { 105 if (errno != EINVAL || in == inbuf) 106 err(EXIT_FAILURE, "iconv()"); 107 108 /* incomplete input character */ 109 (void)memmove(inbuf, in, inbytes); 110 ret = fread(inbuf + inbytes, 1, 111 INBUFSIZE - inbytes, fp); 112 if (ret == 0) { 113 fflush(stdout); 114 if (feof(fp)) 115 errx(EXIT_FAILURE, 116 "unexpected end of file; " 117 "the last character is " 118 "incomplete."); 119 else 120 err(EXIT_FAILURE, "fread()"); 121 } 122 in = inbuf; 123 inbytes += ret; 124 } 125 } 126 } 127 /* reset the shift state of the output buffer */ 128 outbytes = OUTBUFSIZE; 129 out = outbuf; 130 ret = iconv(cd, NULL, NULL, &out, &outbytes); 131 if (ret == (size_t)-1) 132 err(EXIT_FAILURE, "iconv()"); 133 if (outbytes < OUTBUFSIZE) 134 (void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, stdout); 135 136 if (invalids > 0 && !silent) 137 warnx("warning: invalid characters: %llu", invalids); 138 139 return (invalids > 0); 140 } 141 142 static int 143 do_list(unsigned int n, const char * const *list, void *data __unused) 144 { 145 unsigned int i; 146 147 for(i = 0; i < n; i++) { 148 printf("%s", list[i]); 149 if (i < n - 1) 150 printf(" "); 151 } 152 printf("\n"); 153 154 return (1); 155 } 156 157 int 158 main(int argc, char **argv) 159 { 160 iconv_t cd; 161 FILE *fp; 162 const char *opt_f, *opt_t; 163 int ch, i, res; 164 bool opt_c = false, opt_s = false; 165 166 opt_f = opt_t = ""; 167 168 setlocale(LC_ALL, ""); 169 setprogname(argv[0]); 170 171 while ((ch = getopt_long(argc, argv, "csLlf:t:", 172 long_options, NULL)) != -1) { 173 switch (ch) { 174 case 'c': 175 opt_c = true; 176 break; 177 case 's': 178 opt_s = true; 179 break; 180 case 'l': 181 /* list */ 182 if (opt_s || opt_c || strcmp(opt_f, "") != 0 || 183 strcmp(opt_t, "") != 0) { 184 warnx("-l is not allowed with other flags."); 185 usage(); 186 } 187 iconvlist(do_list, NULL); 188 return (EXIT_SUCCESS); 189 case 'f': 190 /* from */ 191 if (optarg != NULL) 192 opt_f = optarg; 193 break; 194 case 't': 195 /* to */ 196 if (optarg != NULL) 197 opt_t = optarg; 198 break; 199 default: 200 usage(); 201 } 202 } 203 argc -= optind; 204 argv += optind; 205 if ((strcmp(opt_f, "") == 0) && (strcmp(opt_t, "") == 0)) 206 usage(); 207 208 if (caph_limit_stdio() < 0) 209 err(EXIT_FAILURE, "capsicum"); 210 211 /* 212 * Cache NLS data, for strerror, for err(3), before entering capability 213 * mode. 214 */ 215 caph_cache_catpages(); 216 217 /* 218 * Cache iconv conversion handle before entering sandbox. 219 */ 220 cd = iconv_open(opt_t, opt_f); 221 if (cd == (iconv_t)-1) 222 err(EXIT_FAILURE, "iconv_open(%s, %s)", opt_t, opt_f); 223 224 if (argc == 0) { 225 if (caph_enter() < 0) 226 err(EXIT_FAILURE, "unable to enter capability mode"); 227 res = do_conv(stdin, cd, opt_s, opt_c); 228 } else { 229 res = 0; 230 for (i = 0; i < argc; i++) { 231 fp = (strcmp(argv[i], "-") != 0) ? 232 fopen(argv[i], "r") : stdin; 233 if (fp == NULL) 234 err(EXIT_FAILURE, "Cannot open `%s'", 235 argv[i]); 236 /* Enter Capsicum sandbox for final input file. */ 237 if (i + 1 == argc && caph_enter() < 0) 238 err(EXIT_FAILURE, 239 "unable to enter capability mode"); 240 res |= do_conv(fp, cd, opt_s, opt_c); 241 (void)fclose(fp); 242 243 /* Reset iconv descriptor state. */ 244 (void)iconv(cd, NULL, NULL, NULL, NULL); 245 } 246 } 247 iconv_close(cd); 248 return (res == 0 ? EXIT_SUCCESS : EXIT_FAILURE); 249 } 250