1 /* $FreeBSD$ */ 2 /* $NetBSD: iconv.c,v 1.16 2009/02/20 15:28:21 yamt Exp $ */ 3 4 /*- 5 * SPDX-License-Identifier: BSD-2-Clause 6 * 7 * Copyright (c)2003 Citrus Project, 8 * All rights reserved. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 22 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 29 * SUCH DAMAGE. 30 */ 31 32 #include <sys/cdefs.h> 33 #include <sys/capsicum.h> 34 35 #include <capsicum_helpers.h> 36 #include <err.h> 37 #include <errno.h> 38 #include <getopt.h> 39 #include <iconv.h> 40 #include <limits.h> 41 #include <locale.h> 42 #include <stdbool.h> 43 #include <stdio.h> 44 #include <stdlib.h> 45 #include <string.h> 46 #include <unistd.h> 47 48 static int do_conv(FILE *, iconv_t, bool, bool); 49 static int do_list(unsigned int, const char * const *, void *); 50 static void usage(void) __dead2; 51 52 static const struct option long_options[] = { 53 {"from-code", required_argument, NULL, 'f'}, 54 {"list", no_argument, NULL, 'l'}, 55 {"silent", no_argument, NULL, 's'}, 56 {"to-code", required_argument, NULL, 't'}, 57 {NULL, no_argument, NULL, 0} 58 }; 59 60 static void 61 usage(void) 62 { 63 (void)fprintf(stderr, 64 "Usage:\t%1$s [-cs] -f <from_code> -t <to_code> [file ...]\n" 65 "\t%1$s -f <from_code> [-cs] [-t <to_code>] [file ...]\n" 66 "\t%1$s -t <to_code> [-cs] [-f <from_code>] [file ...]\n" 67 "\t%1$s -l\n", getprogname()); 68 exit(1); 69 } 70 71 #define INBUFSIZE 1024 72 #define OUTBUFSIZE (INBUFSIZE * 2) 73 static int 74 do_conv(FILE *fp, iconv_t cd, bool silent, bool hide_invalid) 75 { 76 char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *in, *out; 77 unsigned long long invalids; 78 size_t inbytes, outbytes, ret; 79 80 /* 81 * Don't touch ICONV_SET_DISCARD_ILSEQ if -c wasn't specified. It may 82 * be that the user has specified //IGNORE in the -t specification, and 83 * we don't want to clobber that. 84 */ 85 if (hide_invalid) { 86 int arg = (int)hide_invalid; 87 if (iconvctl(cd, ICONV_SET_DISCARD_ILSEQ, (void *)&arg) == -1) 88 err(EXIT_FAILURE, "iconvctl(DISCARD_ILSEQ, %d)", arg); 89 } 90 91 invalids = 0; 92 while ((inbytes = fread(inbuf, 1, INBUFSIZE, fp)) > 0) { 93 in = inbuf; 94 while (inbytes > 0) { 95 size_t inval; 96 97 out = outbuf; 98 outbytes = OUTBUFSIZE; 99 ret = __iconv(cd, &in, &inbytes, &out, &outbytes, 100 0, &inval); 101 invalids += inval; 102 if (outbytes < OUTBUFSIZE) 103 (void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, 104 stdout); 105 if (ret == (size_t)-1 && errno != E2BIG) { 106 if (errno != EINVAL || in == inbuf) 107 err(EXIT_FAILURE, "iconv()"); 108 109 /* incomplete input character */ 110 (void)memmove(inbuf, in, inbytes); 111 ret = fread(inbuf + inbytes, 1, 112 INBUFSIZE - inbytes, fp); 113 if (ret == 0) { 114 fflush(stdout); 115 if (feof(fp)) 116 errx(EXIT_FAILURE, 117 "unexpected end of file; " 118 "the last character is " 119 "incomplete."); 120 else 121 err(EXIT_FAILURE, "fread()"); 122 } 123 in = inbuf; 124 inbytes += ret; 125 } 126 } 127 } 128 /* reset the shift state of the output buffer */ 129 outbytes = OUTBUFSIZE; 130 out = outbuf; 131 ret = iconv(cd, NULL, NULL, &out, &outbytes); 132 if (ret == (size_t)-1) 133 err(EXIT_FAILURE, "iconv()"); 134 if (outbytes < OUTBUFSIZE) 135 (void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, stdout); 136 137 if (invalids > 0 && !silent) 138 warnx("warning: invalid characters: %llu", invalids); 139 140 return (invalids > 0); 141 } 142 143 static int 144 do_list(unsigned int n, const char * const *list, void *data __unused) 145 { 146 unsigned int i; 147 148 for(i = 0; i < n; i++) { 149 printf("%s", list[i]); 150 if (i < n - 1) 151 printf(" "); 152 } 153 printf("\n"); 154 155 return (1); 156 } 157 158 int 159 main(int argc, char **argv) 160 { 161 iconv_t cd; 162 FILE *fp; 163 const char *opt_f, *opt_t; 164 int ch, i, res; 165 bool opt_c = false, opt_s = false; 166 167 opt_f = opt_t = ""; 168 169 setlocale(LC_ALL, ""); 170 setprogname(argv[0]); 171 172 while ((ch = getopt_long(argc, argv, "csLlf:t:", 173 long_options, NULL)) != -1) { 174 switch (ch) { 175 case 'c': 176 opt_c = true; 177 break; 178 case 's': 179 opt_s = true; 180 break; 181 case 'l': 182 /* list */ 183 if (opt_s || opt_c || strcmp(opt_f, "") != 0 || 184 strcmp(opt_t, "") != 0) { 185 warnx("-l is not allowed with other flags."); 186 usage(); 187 } 188 iconvlist(do_list, NULL); 189 return (EXIT_SUCCESS); 190 case 'f': 191 /* from */ 192 if (optarg != NULL) 193 opt_f = optarg; 194 break; 195 case 't': 196 /* to */ 197 if (optarg != NULL) 198 opt_t = optarg; 199 break; 200 default: 201 usage(); 202 } 203 } 204 argc -= optind; 205 argv += optind; 206 if ((strcmp(opt_f, "") == 0) && (strcmp(opt_t, "") == 0)) 207 usage(); 208 209 if (caph_limit_stdio() < 0) 210 err(EXIT_FAILURE, "capsicum"); 211 212 /* 213 * Cache NLS data, for strerror, for err(3), before entering capability 214 * mode. 215 */ 216 caph_cache_catpages(); 217 218 /* 219 * Cache iconv conversion handle before entering sandbox. 220 */ 221 cd = iconv_open(opt_t, opt_f); 222 if (cd == (iconv_t)-1) 223 err(EXIT_FAILURE, "iconv_open(%s, %s)", opt_t, opt_f); 224 225 if (argc == 0) { 226 if (caph_enter() < 0) 227 err(EXIT_FAILURE, "unable to enter capability mode"); 228 res = do_conv(stdin, cd, opt_s, opt_c); 229 } else { 230 res = 0; 231 for (i = 0; i < argc; i++) { 232 fp = (strcmp(argv[i], "-") != 0) ? 233 fopen(argv[i], "r") : stdin; 234 if (fp == NULL) 235 err(EXIT_FAILURE, "Cannot open `%s'", 236 argv[i]); 237 /* Enter Capsicum sandbox for final input file. */ 238 if (i + 1 == argc && caph_enter() < 0) 239 err(EXIT_FAILURE, 240 "unable to enter capability mode"); 241 res |= do_conv(fp, cd, opt_s, opt_c); 242 (void)fclose(fp); 243 244 /* Reset iconv descriptor state. */ 245 (void)iconv(cd, NULL, NULL, NULL, NULL); 246 } 247 } 248 iconv_close(cd); 249 return (res == 0 ? EXIT_SUCCESS : EXIT_FAILURE); 250 } 251