summaryrefslogtreecommitdiff
path: root/utf8.c
blob: b3d30047827ac8c384b6dc6487523da4e30f06c9 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
/* $OpenBSD: utf8.c,v 1.9 2020/03/06 18:12:55 markus Exp $ */
/*
 * Copyright (c) 2016 Ingo Schwarze <schwarze@openbsd.org>
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 */

/*
 * Utility functions for multibyte-character handling,
 * in particular to sanitize untrusted strings for terminal output.
 */

#include "includes.h"

#include <sys/types.h>
#ifdef HAVE_LANGINFO_H
# include <langinfo.h>
#endif
#include <limits.h>
#include <locale.h>
#include <stdarg.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#if defined(HAVE_STRNVIS) && defined(HAVE_VIS_H) && !defined(BROKEN_STRNVIS)
# include <vis.h>
#endif
#ifdef HAVE_WCHAR_H
# include <wchar.h>
#endif

#include "utf8.h"

static int	 dangerous_locale(void);
static int	 grow_dst(char **, size_t *, size_t, char **, size_t);
static int	 vasnmprintf(char **, size_t, int *, const char *, va_list);


/*
 * For US-ASCII and UTF-8 encodings, we can safely recover from
 * encoding errors and from non-printable characters.  For any
 * other encodings, err to the side of caution and abort parsing:
 * For state-dependent encodings, recovery is impossible.
 * For arbitrary encodings, replacement of non-printable
 * characters would be non-trivial and too fragile.
 * The comments indicate what nl_langinfo(CODESET)
 * returns for US-ASCII on various operating systems.
 */

static int
dangerous_locale(void) {
	char	*loc;

	loc = nl_langinfo(CODESET);
	return strcmp(loc, "UTF-8") != 0 &&
	    strcmp(loc, "US-ASCII") != 0 &&		/* OpenBSD */
	    strcmp(loc, "ANSI_X3.4-1968") != 0 &&	/* Linux */
	    strcmp(loc, "ISO8859-1") != 0 &&		/* AIX */
	    strcmp(loc, "646") != 0 &&			/* Solaris, NetBSD */
	    strcmp(loc, "") != 0;			/* Solaris 6 */
}

static int
grow_dst(char **dst, size_t *sz, size_t maxsz, char **dp, size_t need)
{
	char	*tp;
	size_t	 tsz;

	if (*dp + need < *dst + *sz)
		return 0;
	tsz = *sz + 128;
	if (tsz > maxsz)
		tsz = maxsz;
	if ((tp = recallocarray(*dst, *sz, tsz, 1)) == NULL)
		return -1;
	*dp = tp + (*dp - *dst);
	*dst = tp;
	*sz = tsz;
	return 0;
}

/*
 * The following two functions limit the number of bytes written,
 * including the terminating '\0', to sz.  Unless wp is NULL,
 * they limit the number of display columns occupied to *wp.
 * Whichever is reached first terminates the output string.
 * To stay close to the standard interfaces, they return the number of
 * non-NUL bytes that would have been written if both were unlimited.
 * If wp is NULL, newline, carriage return, and tab are allowed;
 * otherwise, the actual number of columns occupied by what was
 * written is returned in *wp.
 */

static int
vasnmprintf(char **str, size_t maxsz, int *wp, const char *fmt, va_list ap)
{
	char	*src;	/* Source string returned from vasprintf. */
	char	*sp;	/* Pointer into src. */
	char	*dst;	/* Destination string to be returned. */
	char	*dp;	/* Pointer into dst. */
	char	*tp;	/* Temporary pointer for dst. */
	size_t	 sz;	/* Number of bytes allocated for dst. */
	wchar_t	 wc;	/* Wide character at sp. */
	int	 len;	/* Number of bytes in the character at sp. */
	int	 ret;	/* Number of bytes needed to format src. */
	int	 width;	/* Display width of the character wc. */
	int	 total_width, max_width, print;

	src = NULL;
	if ((ret = vasprintf(&src, fmt, ap)) <= 0)
		goto fail;

	sz = strlen(src) + 1;
	if ((dst = malloc(sz)) == NULL) {
		free(src);
		ret = -1;
		goto fail;
	}

	if (maxsz > INT_MAX)
		maxsz = INT_MAX;

	sp = src;
	dp = dst;
	ret = 0;
	print = 1;
	total_width = 0;
	max_width = wp == NULL ? INT_MAX : *wp;
	while (*sp != '\0') {
		if ((len = mbtowc(&wc, sp, MB_CUR_MAX)) == -1) {
			(void)mbtowc(NULL, NULL, MB_CUR_MAX);
			if (dangerous_locale()) {
				ret = -1;
				break;
			}
			len = 1;
			width = -1;
		} else if (wp == NULL &&
		    (wc == L'\n' || wc == L'\r' || wc == L'\t')) {
			/*
			 * Don't use width uninitialized; the actual
			 * value doesn't matter because total_width
			 * is only returned for wp != NULL.
			 */
			width = 0;
		} else if ((width = wcwidth(wc)) == -1 &&
		    dangerous_locale()) {
			ret = -1;
			break;
		}

		/* Valid, printable character. */

		if (width >= 0) {
			if (print && (dp - dst >= (int)maxsz - len ||
			    total_width > max_width - width))
				print = 0;
			if (print) {
				if (grow_dst(&dst, &sz, maxsz,
				    &dp, len) == -1) {
					ret = -1;
					break;
				}
				total_width += width;
				memcpy(dp, sp, len);
				dp += len;
			}
			sp += len;
			if (ret >= 0)
				ret += len;
			continue;
		}

		/* Escaping required. */

		while (len > 0) {
			if (print && (dp - dst >= (int)maxsz - 4 ||
			    total_width > max_width - 4))
				print = 0;
			if (print) {
				if (grow_dst(&dst, &sz, maxsz,
				    &dp, 4) == -1) {
					ret = -1;
					break;
				}
				tp = vis(dp, *sp, VIS_OCTAL | VIS_ALL, 0);
				width = tp - dp;
				total_width += width;
				dp = tp;
			} else
				width = 4;
			len--;
			sp++;
			if (ret >= 0)
				ret += width;
		}
		if (len > 0)
			break;
	}
	free(src);
	*dp = '\0';
	*str = dst;
	if (wp != NULL)
		*wp = total_width;

	/*
	 * If the string was truncated by the width limit but
	 * would have fit into the size limit, the only sane way
	 * to report the problem is using the return value, such
	 * that the usual idiom "if (ret < 0 || ret >= sz) error"
	 * works as expected.
	 */

	if (ret < (int)maxsz && !print)
		ret = -1;
	return ret;

fail:
	if (wp != NULL)
		*wp = 0;
	if (ret == 0) {
		*str = src;
		return 0;
	} else {
		*str = NULL;
		return -1;
	}
}

int
snmprintf(char *str, size_t sz, int *wp, const char *fmt, ...)
{
	va_list	 ap;
	char	*cp = NULL;
	int	 ret;

	va_start(ap, fmt);
	ret = vasnmprintf(&cp, sz, wp, fmt, ap);
	va_end(ap);
	if (cp != NULL) {
		(void)strlcpy(str, cp, sz);
		free(cp);
	} else
		*str = '\0';
	return ret;
}

/*
 * To stay close to the standard interfaces, the following functions
 * return the number of non-NUL bytes written.
 */

int
vfmprintf(FILE *stream, const char *fmt, va_list ap)
{
	char	*str = NULL;
	int	 ret;

	if ((ret = vasnmprintf(&str, INT_MAX, NULL, fmt, ap)) < 0) {
		free(str);
		return -1;
	}
	if (fputs(str, stream) == EOF)
		ret = -1;
	free(str);
	return ret;
}

int
fmprintf(FILE *stream, const char *fmt, ...)
{
	va_list	 ap;
	int	 ret;

	va_start(ap, fmt);
	ret = vfmprintf(stream, fmt, ap);
	va_end(ap);
	return ret;
}

int
mprintf(const char *fmt, ...)
{
	va_list	 ap;
	int	 ret;

	va_start(ap, fmt);
	ret = vfmprintf(stdout, fmt, ap);
	va_end(ap);
	return ret;
}

/*
 * Set up libc for multibyte output in the user's chosen locale.
 *
 * XXX: we are known to have problems with Turkish (i/I confusion) so we
 *      deliberately fall back to the C locale for now. Longer term we should
 *      always prefer to select C.[encoding] if possible, but there's no
 *      standardisation in locales between systems, so we'll need to survey
 *      what's out there first.
 */
void
msetlocale(void)
{
	const char *vars[] = { "LC_ALL", "LC_CTYPE", "LANG", NULL };
	char *cp;
	int i;

	/*
	 * We can't yet cope with dotless/dotted I in Turkish locales,
	 * so fall back to the C locale for these.
	 */
	for (i = 0; vars[i] != NULL; i++) {
		if ((cp = getenv(vars[i])) == NULL)
			continue;
		if (strncasecmp(cp, "TR", 2) != 0)
			break;
		/*
		 * If we're in a UTF-8 locale then prefer to use
		 * the C.UTF-8 locale (or equivalent) if it exists.
		 */
		if ((strcasestr(cp, "UTF-8") != NULL ||
		    strcasestr(cp, "UTF8") != NULL) &&
		    (setlocale(LC_CTYPE, "C.UTF-8") != NULL ||
		    setlocale(LC_CTYPE, "POSIX.UTF-8") != NULL))
			return;
		setlocale(LC_CTYPE, "C");
		return;
	}
	/* We can handle this locale */
	setlocale(LC_CTYPE, "");
}