xref: /freebsd/usr.bin/ident/ident.c (revision 4ffeb3b88ed228e72e13abe1be888c77c6ede95d)
1 /*-
2  * Copyright (c) 2015-2021 Baptiste Daroussin <bapt@FreeBSD.org>
3  * Copyright (c) 2015 Xin LI <delphij@FreeBSD.org>
4  *
5  * Redistribution and use in source and binary forms, with or without
6  * modification, are permitted provided that the following conditions
7  * are met:
8  * 1. Redistributions of source code must retain the above copyright
9  *    notice, this list of conditions and the following disclaimer
10  *    in this position and unchanged.
11  * 2. Redistributions in binary form must reproduce the above copyright
12  *    notice, this list of conditions and the following disclaimer in the
13  *    documentation and/or other materials provided with the distribution.
14  *
15  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
16  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
17  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
18  * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
19  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
20  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
21  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
22  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
23  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
24  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25  */
26 
27 #include <sys/cdefs.h>
28 __FBSDID("$FreeBSD$");
29 
30 #include <sys/capsicum.h>
31 #include <sys/types.h>
32 #include <sys/sbuf.h>
33 
34 #include <capsicum_helpers.h>
35 #include <ctype.h>
36 #include <err.h>
37 #include <errno.h>
38 #include <stdbool.h>
39 #include <stdio.h>
40 #include <stdlib.h>
41 #include <string.h>
42 #include <unistd.h>
43 #include <xlocale.h>
44 
45 typedef enum {
46 	/* state	condition to transit to next state */
47 	INIT,		/* '$' */
48 	DELIM_SEEN,	/* letter */
49 	KEYWORD,	/* punctuation mark */
50 	PUNC_SEEN,	/* ':' -> _SVN; space -> TEXT */
51 	PUNC_SEEN_SVN,	/* space */
52 	TEXT
53 } analyzer_states;
54 
55 static int
56 scan(FILE *fp, const char *name, bool quiet)
57 {
58 	int c;
59 	bool hasid = false;
60 	bool subversion = false;
61 	analyzer_states state = INIT;
62 	FILE* buffp;
63 	char *buf;
64 	size_t sz;
65 	locale_t l;
66 
67 	l = newlocale(LC_ALL_MASK, "C", NULL);
68 	sz = 0;
69 	buf = NULL;
70 	buffp = open_memstream(&buf, &sz);
71 	if (buffp == NULL)
72 		err(EXIT_FAILURE, "open_memstream()");
73 
74 	if (name != NULL)
75 		printf("%s:\n", name);
76 
77 	while ((c = fgetc(fp)) != EOF) {
78 		switch (state) {
79 		case INIT:
80 			if (c == '$') {
81 				/* Transit to DELIM_SEEN if we see $ */
82 				state = DELIM_SEEN;
83 			} else {
84 				/* Otherwise, stay in INIT state */
85 				continue;
86 			}
87 			break;
88 		case DELIM_SEEN:
89 			if (isalpha_l(c, l)) {
90 				/* Transit to KEYWORD if we see letter */
91 				if (buf != NULL)
92 					memset(buf, 0, sz);
93 				rewind(buffp);
94 				fputc('$', buffp);
95 				fputc(c, buffp);
96 				state = KEYWORD;
97 
98 				continue;
99 			} else if (c == '$') {
100 				/* Or, stay in DELIM_SEEN if more $ */
101 				continue;
102 			} else {
103 				/* Otherwise, transit back to INIT */
104 				state = INIT;
105 			}
106 			break;
107 		case KEYWORD:
108 			fputc(c, buffp);
109 
110 			if (isalpha_l(c, l)) {
111 				/*
112 				 * Stay in KEYWORD if additional letter is seen
113 				 */
114 				continue;
115 			} else if (c == ':') {
116 				/*
117 				 * See ':' for the first time, transit to
118 				 * PUNC_SEEN.
119 				 */
120 				state = PUNC_SEEN;
121 				subversion = false;
122 			} else if (c == '$') {
123 				/*
124 				 * Incomplete ident.  Go back to DELIM_SEEN
125 				 * state because we see a '$' which could be
126 				 * the beginning of a keyword.
127 				 */
128 				state = DELIM_SEEN;
129 			} else {
130 				/*
131 				 * Go back to INIT state otherwise.
132 				 */
133 				state = INIT;
134 			}
135 			break;
136 		case PUNC_SEEN:
137 		case PUNC_SEEN_SVN:
138 			fputc(c, buffp);
139 
140 			switch (c) {
141 			case ':':
142 				/*
143 				 * If we see '::' (seen : in PUNC_SEEN),
144 				 * activate subversion treatment and transit
145 				 * to PUNC_SEEN_SVN state.
146 				 *
147 				 * If more than two :'s were seen, the ident
148 				 * is invalid and we would therefore go back
149 				 * to INIT state.
150 				 */
151 				if (state == PUNC_SEEN) {
152 					state = PUNC_SEEN_SVN;
153 					subversion = true;
154 				} else {
155 					state = INIT;
156 				}
157 				break;
158 			case ' ':
159 				/*
160 				 * A space after ':' or '::' indicates we are at the
161 				 * last component of potential ident.
162 				 */
163 				state = TEXT;
164 				break;
165 			default:
166 				/* All other characters are invalid */
167 				state = INIT;
168 				break;
169 			}
170 			break;
171 		case TEXT:
172 			fputc(c, buffp);
173 
174 			if (iscntrl_l(c, l)) {
175 				/* Control characters are not allowed in this state */
176 				state = INIT;
177 			} else if (c == '$') {
178 				fflush(buffp);
179 				/*
180 				 * valid ident should end with a space.
181 				 *
182 				 * subversion extension uses '#' to indicate that
183 				 * the keyword expansion have exceeded the fixed
184 				 * width, so it is also permitted if we are in
185 				 * subversion mode.  No length check is enforced
186 				 * because GNU RCS ident(1) does not do it either.
187 				 */
188 				c = buf[strlen(buf) -2 ];
189 				if (c == ' ' || (subversion && c == '#')) {
190 					printf("     %s\n", buf);
191 					hasid = true;
192 				}
193 				state = INIT;
194 			}
195 			/* Other characters: stay in the state */
196 			break;
197 		}
198 	}
199 	fclose(buffp);
200 	free(buf);
201 	freelocale(l);
202 
203 	if (!hasid) {
204 		if (!quiet)
205 			fprintf(stderr, "%s warning: no id keywords in %s\n",
206 			    getprogname(), name ? name : "standard input");
207 
208 		return (EXIT_FAILURE);
209 	}
210 
211 	return (EXIT_SUCCESS);
212 }
213 
214 int
215 main(int argc, char **argv)
216 {
217 	bool quiet = false;
218 	int ch, i, *fds, fd;
219 	int ret = EXIT_SUCCESS;
220 	size_t nfds;
221 	FILE *fp;
222 
223 	while ((ch = getopt(argc, argv, "qV")) != -1) {
224 		switch (ch) {
225 		case 'q':
226 			quiet = true;
227 			break;
228 		case 'V':
229 			/* Do nothing, compat with GNU rcs's ident */
230 			return (EXIT_SUCCESS);
231 		default:
232 			errx(EXIT_FAILURE, "usage: %s [-q] [-V] [file...]",
233 			    getprogname());
234 		}
235 	}
236 
237 	argc -= optind;
238 	argv += optind;
239 
240 	if (caph_limit_stdio() < 0)
241 		err(EXIT_FAILURE, "unable to limit stdio");
242 
243 	if (argc == 0) {
244 		nfds = 1;
245 		fds = malloc(sizeof(*fds));
246 		if (fds == NULL)
247 			err(EXIT_FAILURE, "unable to allocate fds array");
248 		fds[0] = STDIN_FILENO;
249 	} else {
250 		nfds = argc;
251 		fds = malloc(sizeof(*fds) * nfds);
252 		if (fds == NULL)
253 			err(EXIT_FAILURE, "unable to allocate fds array");
254 
255 		for (i = 0; i < argc; i++) {
256 			fds[i] = fd = open(argv[i], O_RDONLY);
257 			if (fd < 0) {
258 				warn("%s", argv[i]);
259 				ret = EXIT_FAILURE;
260 				continue;
261 			}
262 			if (caph_limit_stream(fd, CAPH_READ) < 0)
263 				err(EXIT_FAILURE,
264 				    "unable to limit fcntls/rights for %s",
265 				    argv[i]);
266 		}
267 	}
268 
269 	/* Enter Capsicum sandbox. */
270 	if (caph_enter() < 0)
271 		err(EXIT_FAILURE, "unable to enter capability mode");
272 
273 	for (i = 0; i < (int)nfds; i++) {
274 		if (fds[i] < 0)
275 			continue;
276 
277 		fp = fdopen(fds[i], "r");
278 		if (fp == NULL) {
279 			warn("%s", argv[i]);
280 			ret = EXIT_FAILURE;
281 			continue;
282 		}
283 		if (scan(fp, argc == 0 ? NULL : argv[i], quiet) != EXIT_SUCCESS)
284 			ret = EXIT_FAILURE;
285 		fclose(fp);
286 	}
287 
288 	return (ret);
289 }
290