1 /* $FreeBSD$ */ 2 /* $NetBSD: iconv.c,v 1.16 2009/02/20 15:28:21 yamt 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/cdefs.h> 31 32 #include <err.h> 33 #include <errno.h> 34 #include <getopt.h> 35 #include <iconv.h> 36 #include <limits.h> 37 #include <locale.h> 38 #include <stdbool.h> 39 #include <stdio.h> 40 #include <stdlib.h> 41 #include <string.h> 42 #include <unistd.h> 43 44 static unsigned long long invalids; 45 46 static void do_conv(FILE *, const char *, const char *, bool, bool); 47 static int do_list(unsigned int, const char * const *, void *); 48 static void usage(void); 49 50 struct option long_options[] = 51 { 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 void 73 do_conv(FILE *fp, const char *from, const char *to, bool silent, 74 bool hide_invalid) 75 { 76 iconv_t cd; 77 char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *out; 78 char *in; 79 size_t inbytes, outbytes, ret; 80 81 if ((cd = iconv_open(to, from)) == (iconv_t)-1) 82 err(EXIT_FAILURE, "iconv_open(%s, %s)", to, from); 83 84 if (hide_invalid) { 85 int arg = 1; 86 87 if (iconvctl(cd, ICONV_SET_DISCARD_ILSEQ, (void *)&arg) == -1) 88 err(1, NULL); 89 } 90 while ((inbytes = fread(inbuf, 1, INBUFSIZE, fp)) > 0) { 91 in = inbuf; 92 while (inbytes > 0) { 93 size_t inval; 94 95 out = outbuf; 96 outbytes = OUTBUFSIZE; 97 ret = __iconv(cd, &in, &inbytes, &out, &outbytes, 98 0, &inval); 99 invalids += inval; 100 if (outbytes < OUTBUFSIZE) 101 (void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, 102 stdout); 103 if (ret == (size_t)-1 && errno != E2BIG) { 104 if (errno != EINVAL || in == inbuf) 105 err(EXIT_FAILURE, "iconv()"); 106 107 /* incomplete input character */ 108 (void)memmove(inbuf, in, inbytes); 109 ret = fread(inbuf + inbytes, 1, 110 INBUFSIZE - inbytes, fp); 111 if (ret == 0) { 112 fflush(stdout); 113 if (feof(fp)) 114 errx(EXIT_FAILURE, 115 "unexpected end of file; " 116 "the last character is " 117 "incomplete."); 118 else 119 err(EXIT_FAILURE, "fread()"); 120 } 121 in = inbuf; 122 inbytes += ret; 123 } 124 } 125 } 126 /* reset the shift state of the output buffer */ 127 outbytes = OUTBUFSIZE; 128 out = outbuf; 129 ret = iconv(cd, NULL, NULL, &out, &outbytes); 130 if (ret == (size_t)-1) 131 err(EXIT_FAILURE, "iconv()"); 132 if (outbytes < OUTBUFSIZE) 133 (void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, stdout); 134 135 if (invalids > 0 && !silent) 136 warnx("warning: invalid characters: %llu", invalids); 137 138 iconv_close(cd); 139 } 140 141 static int 142 do_list(unsigned int n, const char * const *list, void *data __unused) 143 { 144 unsigned int i; 145 146 for(i = 0; i < n; i++) { 147 printf("%s", list[i]); 148 if (i < n - 1) 149 printf(" "); 150 } 151 printf("\n"); 152 153 return (1); 154 } 155 156 int 157 main(int argc, char **argv) 158 { 159 FILE *fp; 160 char *opt_f, *opt_t; 161 int ch, i; 162 bool opt_c = false, opt_s = false; 163 164 opt_f = opt_t = strdup(""); 165 166 setlocale(LC_ALL, ""); 167 setprogname(argv[0]); 168 169 while ((ch = getopt_long(argc, argv, "csLlf:t:", 170 long_options, NULL)) != -1) { 171 switch (ch) { 172 case 'c': 173 opt_c = true; 174 break; 175 case 's': 176 opt_s = true; 177 break; 178 case 'l': 179 /* list */ 180 if (opt_s || opt_c || strcmp(opt_f, "") != 0 || 181 strcmp(opt_t, "") != 0) { 182 warnx("-l is not allowed with other flags."); 183 usage(); 184 } 185 iconvlist(do_list, NULL); 186 return (EXIT_SUCCESS); 187 case 'f': 188 /* from */ 189 if (optarg != NULL) 190 opt_f = strdup(optarg); 191 break; 192 case 't': 193 /* to */ 194 if (optarg != NULL) 195 opt_t = strdup(optarg); 196 break; 197 default: 198 usage(); 199 } 200 } 201 argc -= optind; 202 argv += optind; 203 if ((strcmp(opt_f, "") == 0) && (strcmp(opt_t, "") == 0)) 204 usage(); 205 if (argc == 0) 206 do_conv(stdin, opt_f, opt_t, opt_s, opt_c); 207 else { 208 for (i = 0; i < argc; i++) { 209 fp = (strcmp(argv[i], "-") != 0) ? 210 fopen(argv[i], "r") : stdin; 211 if (fp == NULL) 212 err(EXIT_FAILURE, "Cannot open `%s'", 213 argv[i]); 214 do_conv(fp, opt_f, opt_t, opt_s, 215 opt_c); 216 (void)fclose(fp); 217 } 218 } 219 return (EXIT_SUCCESS); 220 } 221