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/capsicum.h>
32
33#include <capsicum_helpers.h>
34#include <err.h>
35#include <errno.h>
36#include <getopt.h>
37#include <iconv.h>
38#include <limits.h>
39#include <locale.h>
40#include <stdbool.h>
41#include <stdio.h>
42#include <stdlib.h>
43#include <string.h>
44#include <unistd.h>
45
46static int		do_conv(FILE *, iconv_t, bool, bool);
47static int		do_list(unsigned int, const char * const *, void *);
48static void		usage(void) __dead2;
49
50static const struct option long_options[] = {
51	{"from-code",		required_argument,	NULL, 'f'},
52	{"list",		no_argument,		NULL, 'l'},
53	{"silent",		no_argument,		NULL, 's'},
54        {"to-code",		required_argument,	NULL, 't'},
55        {NULL,                  no_argument,            NULL, 0}
56};
57
58static void
59usage(void)
60{
61	(void)fprintf(stderr,
62	    "Usage:\t%1$s [-cs] -f <from_code> -t <to_code> [file ...]\n"
63	    "\t%1$s -f <from_code> [-cs] [-t <to_code>] [file ...]\n"
64	    "\t%1$s -t <to_code> [-cs] [-f <from_code>] [file ...]\n"
65	    "\t%1$s -l\n", getprogname());
66	exit(1);
67}
68
69#define INBUFSIZE 1024
70#define OUTBUFSIZE (INBUFSIZE * 2)
71static int
72do_conv(FILE *fp, iconv_t cd, bool silent, bool hide_invalid)
73{
74	char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *in, *out;
75	unsigned long long invalids;
76	size_t inbytes, outbytes, ret;
77
78	/*
79	 * Don't touch ICONV_SET_DISCARD_ILSEQ if -c wasn't specified.  It may
80	 * be that the user has specified //IGNORE in the -t specification, and
81	 * we don't want to clobber that.
82	 */
83	if (hide_invalid) {
84		int arg = (int)hide_invalid;
85		if (iconvctl(cd, ICONV_SET_DISCARD_ILSEQ, (void *)&arg) == -1)
86			err(EXIT_FAILURE, "iconvctl(DISCARD_ILSEQ, %d)", arg);
87	}
88
89	invalids = 0;
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	return (invalids > 0);
139}
140
141static int
142do_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
156int
157main(int argc, char **argv)
158{
159	iconv_t cd;
160	FILE *fp;
161	const char *opt_f, *opt_t;
162	int ch, i, res;
163	bool opt_c = false, opt_s = false;
164
165	opt_f = opt_t = "";
166
167	setlocale(LC_ALL, "");
168	setprogname(argv[0]);
169
170	while ((ch = getopt_long(argc, argv, "csLlf:t:",
171	    long_options, NULL)) != -1) {
172		switch (ch) {
173		case 'c':
174			opt_c = true;
175			break;
176		case 's':
177			opt_s = true;
178			break;
179		case 'l':
180			/* list */
181			if (opt_s || opt_c || strcmp(opt_f, "") != 0 ||
182			    strcmp(opt_t, "") != 0) {
183				warnx("-l is not allowed with other flags.");
184				usage();
185			}
186			iconvlist(do_list, NULL);
187			return (EXIT_SUCCESS);
188		case 'f':
189			/* from */
190			if (optarg != NULL)
191				opt_f = optarg;
192			break;
193		case 't':
194			/* to */
195			if (optarg != NULL)
196				opt_t = optarg;
197			break;
198		default:
199			usage();
200		}
201	}
202	argc -= optind;
203	argv += optind;
204	if ((strcmp(opt_f, "") == 0) && (strcmp(opt_t, "") == 0))
205		usage();
206
207	if (caph_limit_stdio() < 0)
208		err(EXIT_FAILURE, "capsicum");
209
210	/*
211	 * Cache NLS data, for strerror, for err(3), before entering capability
212	 * mode.
213	 */
214	caph_cache_catpages();
215
216	/*
217	 * Cache iconv conversion handle before entering sandbox.
218	 */
219	cd = iconv_open(opt_t, opt_f);
220	if (cd == (iconv_t)-1)
221		err(EXIT_FAILURE, "iconv_open(%s, %s)", opt_t, opt_f);
222
223	if (argc == 0) {
224		if (caph_enter() < 0)
225			err(EXIT_FAILURE, "unable to enter capability mode");
226		res = do_conv(stdin, cd, opt_s, opt_c);
227	} else {
228		res = 0;
229		for (i = 0; i < argc; i++) {
230			fp = (strcmp(argv[i], "-") != 0) ?
231			    fopen(argv[i], "r") : stdin;
232			if (fp == NULL)
233				err(EXIT_FAILURE, "Cannot open `%s'",
234				    argv[i]);
235			/* Enter Capsicum sandbox for final input file. */
236			if (i + 1 == argc && caph_enter() < 0)
237				err(EXIT_FAILURE,
238				    "unable to enter capability mode");
239			res |= do_conv(fp, cd, opt_s, opt_c);
240			(void)fclose(fp);
241
242			/* Reset iconv descriptor state. */
243			(void)iconv(cd, NULL, NULL, NULL, NULL);
244		}
245	}
246	iconv_close(cd);
247	return (res == 0 ? EXIT_SUCCESS : EXIT_FAILURE);
248}
249