1/*	$NetBSD: iconv.c,v 1.17 2011/09/16 15:39:26 joerg 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 1.17 2011/09/16 15:39:26 joerg Exp $");
32#endif /* LIBC_SCCS and not lint */
33
34#include <err.h>
35#include <errno.h>
36#include <iconv.h>
37#include <langinfo.h>
38#include <locale.h>
39#include <stdio.h>
40#include <stdlib.h>
41#include <string.h>
42#include <unistd.h>
43#include <util.h>
44
45static void usage(void) __dead;
46static int scmp(const void *, const void *);
47static void show_codesets(void);
48static void do_conv(const char *, FILE *, const char *, const char *, int, int);
49
50static void
51usage(void)
52{
53	(void)fprintf(stderr,
54	    "Usage:\t%1$s [-cs] -f <from_code> -t <to_code> [file ...]\n"
55	    "\t%1$s -f <from_code> [-cs] [-t <to_code>] [file ...]\n"
56	    "\t%1$s -t <to_code> [-cs] [-f <from_code>] [file ...]\n"
57	    "\t%1$s -l\n", getprogname());
58	exit(1);
59}
60
61/*
62 * qsort() helper function
63 */
64static int
65scmp(const void *v1, const void *v2)
66{
67	const char * const *s1 = v1;
68	const char * const *s2 = v2;
69
70	return strcasecmp(*s1, *s2);
71}
72
73static void
74show_codesets(void)
75{
76	char **list;
77	size_t sz, i;
78
79	if (__iconv_get_list(&list, &sz))
80		err(EXIT_FAILURE, "__iconv_get_list()");
81
82	qsort(list, sz, sizeof(char *), scmp);
83
84	for (i = 0; i < sz; i++)
85		(void)printf("%s\n", list[i]);
86
87	__iconv_free_list(list, sz);
88}
89
90#define INBUFSIZE 1024
91#define OUTBUFSIZE (INBUFSIZE * 2)
92/*ARGSUSED*/
93static void
94do_conv(const char *fn, FILE *fp, const char *from, const char *to, int silent,
95    int hide_invalid)
96{
97	char inbuf[INBUFSIZE], outbuf[OUTBUFSIZE], *out;
98	const char *in;
99	size_t inbytes, outbytes, ret, invalids;
100	iconv_t cd;
101	uint32_t flags = 0;
102
103	if (hide_invalid)
104		flags |= __ICONV_F_HIDE_INVALID;
105	cd = iconv_open(to, from);
106	if (cd == (iconv_t)-1)
107		err(EXIT_FAILURE, "iconv_open(%s, %s)", to, from);
108
109	invalids = 0;
110	while ((inbytes = fread(inbuf, 1, INBUFSIZE, fp)) > 0) {
111		in = inbuf;
112		while (inbytes > 0) {
113			size_t inval;
114
115			out = outbuf;
116			outbytes = OUTBUFSIZE;
117			ret = __iconv(cd, &in, &inbytes, &out, &outbytes,
118			    flags, &inval);
119			invalids += inval;
120			if (outbytes < OUTBUFSIZE)
121				(void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes,
122				    stdout);
123			if (ret == (size_t)-1 && errno != E2BIG) {
124				/*
125				 * XXX: iconv(3) is bad interface.
126				 *   invalid character count is lost here.
127				 *   instead, we just provide __iconv function.
128				 */
129				if (errno != EINVAL || in == inbuf)
130					err(EXIT_FAILURE, "iconv()");
131
132				/* incomplete input character */
133				(void)memmove(inbuf, in, inbytes);
134				ret = fread(inbuf + inbytes, 1,
135				    INBUFSIZE - inbytes, fp);
136				if (ret == 0) {
137					fflush(stdout);
138					if (feof(fp))
139						errx(EXIT_FAILURE,
140						     "unexpected end of file; "
141						     "the last character is "
142						     "incomplete.");
143					else
144						err(EXIT_FAILURE, "fread()");
145				}
146				in = inbuf;
147				inbytes += ret;
148			}
149		}
150	}
151	/* reset the shift state of the output buffer */
152	outbytes = OUTBUFSIZE;
153	out = outbuf;
154	ret = iconv(cd, NULL, NULL, &out, &outbytes);
155	if (ret == (size_t)-1)
156		err(EXIT_FAILURE, "iconv()");
157	if (outbytes < OUTBUFSIZE)
158		(void)fwrite(outbuf, 1, OUTBUFSIZE - outbytes, stdout);
159
160	if (invalids > 0 && !silent)
161		warnx("warning: invalid characters: %lu",
162		    (unsigned long)invalids);
163
164	iconv_close(cd);
165}
166
167int
168main(int argc, char **argv)
169{
170	int ch, i;
171	int opt_l = 0, opt_s = 0, opt_c = 0;
172	char *opt_f = NULL, *opt_t = NULL;
173	FILE *fp;
174
175	setlocale(LC_ALL, "");
176	setprogname(argv[0]);
177
178	while ((ch = getopt(argc, argv, "cslf:t:")) != EOF) {
179		switch (ch) {
180		case 'c':
181			opt_c = 1;
182			break;
183		case 's':
184			opt_s = 1;
185			break;
186		case 'l':
187			/* list */
188			opt_l = 1;
189			break;
190		case 'f':
191			/* from */
192			opt_f = estrdup(optarg);
193			break;
194		case 't':
195			/* to */
196			opt_t = estrdup(optarg);
197			break;
198		default:
199			usage();
200		}
201	}
202	argc -= optind;
203	argv += optind;
204	if (opt_l) {
205		if (argc > 0 || opt_s || opt_f != NULL || opt_t != NULL) {
206			warnx("-l is not allowed with other flags.");
207			usage();
208		}
209		show_codesets();
210	} else {
211		if (opt_f == NULL) {
212			if (opt_t == NULL)
213				usage();
214			opt_f = nl_langinfo(CODESET);
215		} else if (opt_t == NULL)
216			opt_t = nl_langinfo(CODESET);
217
218		if (argc == 0)
219			do_conv("<stdin>", stdin, opt_f, opt_t, opt_s, opt_c);
220		else {
221			for (i = 0; i < argc; i++) {
222				fp = fopen(argv[i], "r");
223				if (fp == NULL)
224					err(EXIT_FAILURE, "Cannot open `%s'",
225					    argv[i]);
226				do_conv(argv[i], fp, opt_f, opt_t, opt_s,
227				    opt_c);
228				(void)fclose(fp);
229			}
230		}
231	}
232	return EXIT_SUCCESS;
233}
234