wordexp.c revision 108865
1/*-
2 * Copyright (c) 2002 Tim J. Robbins.
3 * All rights reserved.
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 * 2. Redistributions in binary form must reproduce the above copyright
11 *    notice, this list of conditions and the following disclaimer in the
12 *    documentation and/or other materials provided with the distribution.
13 *
14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
24 * SUCH DAMAGE.
25 */
26
27#include "namespace.h"
28#include <sys/cdefs.h>
29#include <sys/types.h>
30#include <sys/wait.h>
31#include <fcntl.h>
32#include <paths.h>
33#include <stdio.h>
34#include <stdlib.h>
35#include <string.h>
36#include <unistd.h>
37#include <wordexp.h>
38#include "un-namespace.h"
39
40__FBSDID("$FreeBSD: head/lib/libc/gen/wordexp.c 108865 2003-01-07 06:07:56Z tjr $");
41
42static int	we_askshell(const char *, wordexp_t *, int);
43static int	we_check(const char *, int);
44
45/*
46 * wordexp --
47 *	Perform shell word expansion on `words' and place the resulting list
48 *	of words in `we'. See wordexp(3).
49 *
50 *	Specified by IEEE Std. 1003.1-2001.
51 */
52int
53wordexp(const char * __restrict words, wordexp_t * __restrict we, int flags)
54{
55	int error;
56
57	if (flags & WRDE_REUSE)
58		wordfree(we);
59	if ((flags & WRDE_APPEND) == 0) {
60		we->we_wordc = 0;
61		we->we_wordv = NULL;
62		we->we_strings = NULL;
63		we->we_nbytes = 0;
64	}
65	if ((error = we_check(words, flags)) != 0) {
66		wordfree(we);
67		return (error);
68	}
69	if ((error = we_askshell(words, we, flags)) != 0) {
70		wordfree(we);
71		return (error);
72	}
73	return (0);
74}
75
76/*
77 * we_askshell --
78 *	Use the `wordexp' /bin/sh builtin function to do most of the work
79 *	in expanding the word string. This function is complicated by
80 *	memory management.
81 */
82static int
83we_askshell(const char *words, wordexp_t *we, int flags)
84{
85	int pdes[2];			/* Pipe to child */
86	char bbuf[9];			/* Buffer for byte count */
87	char wbuf[9];			/* Buffer for word count */
88	long nwords, nbytes;		/* Number of words, bytes from child */
89	long i;				/* Handy integer */
90	size_t sofs;			/* Offset into we->we_strings */
91	size_t vofs;			/* Offset into we->we_wordv */
92	pid_t pid;			/* Process ID of child */
93	int status;			/* Child exit status */
94	char *ifs;			/* IFS env. var. */
95	char *np, *p;			/* Handy pointers */
96	char *nstrings;			/* Temporary for realloc() */
97	char **nwv;			/* Temporary for realloc() */
98
99	if ((ifs = getenv("IFS")) == NULL)
100		ifs = " \t\n";
101
102	if (pipe(pdes) < 0)
103		return (WRDE_NOSPACE);	/* XXX */
104	if ((pid = fork()) < 0) {
105		_close(pdes[0]);
106		_close(pdes[1]);
107		return (WRDE_NOSPACE);	/* XXX */
108	}
109	else if (pid == 0) {
110		/*
111		 * We are the child; just get /bin/sh to run the wordexp
112		 * builtin on `words'.
113		 */
114		int devnull;
115		char *cmd;
116
117		_close(pdes[0]);
118		if (_dup2(pdes[1], STDOUT_FILENO) < 0)
119			_exit(1);
120		_close(pdes[1]);
121		if (asprintf(&cmd, "wordexp%c%s\n", *ifs, words) < 0)
122			_exit(1);
123		if ((flags & WRDE_SHOWERR) == 0) {
124			if ((devnull = _open(_PATH_DEVNULL, O_RDWR, 0666)) < 0)
125				_exit(1);
126			if (_dup2(devnull, STDERR_FILENO) < 0)
127				_exit(1);
128			_close(devnull);
129		}
130		execl(_PATH_BSHELL, "sh", flags & WRDE_UNDEF ? "-u" : "+u",
131		    "-c", cmd, NULL);
132		_exit(1);
133	}
134
135	/*
136	 * We are the parent; read the output of the shell wordexp function,
137	 * which is a 32-bit hexadecimal word count, a 32-bit hexadecimal
138	 * byte count (not including terminating null bytes), followed by
139	 * the expanded words separated by nulls.
140	 */
141	_close(pdes[1]);
142	if (_read(pdes[0], wbuf, 8) != 8 || _read(pdes[0], bbuf, 8) != 8) {
143		_close(pdes[0]);
144		_waitpid(pid, &status, 0);
145		return (flags & WRDE_UNDEF ? WRDE_BADVAL : WRDE_SYNTAX);
146	}
147	wbuf[8] = bbuf[8] = '\0';
148	nwords = strtol(wbuf, NULL, 16);
149	nbytes = strtol(bbuf, NULL, 16) + nwords;
150
151	/*
152	 * Allocate or reallocate (when flags & WRDE_APPEND) the word vector
153	 * and string storage buffers for the expanded words we're about to
154	 * read from the child.
155	 */
156	sofs = we->we_nbytes;
157	vofs = we->we_wordc;
158	if ((flags & (WRDE_DOOFS|WRDE_APPEND)) == (WRDE_DOOFS|WRDE_APPEND))
159		vofs += we->we_offs;
160	we->we_wordc += nwords;
161	we->we_nbytes += nbytes;
162	if ((nwv = realloc(we->we_wordv, (we->we_wordc + 1 +
163	    (flags & WRDE_DOOFS ?  we->we_offs : 0)) *
164	    sizeof(char *))) == NULL) {
165		_close(pdes[0]);
166		_waitpid(pid, &status, 0);
167		return (WRDE_NOSPACE);
168	}
169	we->we_wordv = nwv;
170	if ((nstrings = realloc(we->we_strings, we->we_nbytes)) == NULL) {
171		_close(pdes[0]);
172		_waitpid(pid, &status, 0);
173		return (WRDE_NOSPACE);
174	}
175	for (i = 0; i < vofs; i++)
176		if (we->we_wordv[i] != NULL)
177			we->we_wordv[i] += nstrings - we->we_strings;
178	we->we_strings = nstrings;
179
180	if (_read(pdes[0], we->we_strings + sofs, nbytes) != nbytes) {
181		_close(pdes[0]);
182		_waitpid(pid, &status, 0);
183		return (flags & WRDE_UNDEF ? WRDE_BADVAL : WRDE_SYNTAX);
184	}
185
186	if (_waitpid(pid, &status, 0) < 0 || !WIFEXITED(status) ||
187	    WEXITSTATUS(status) != 0) {
188		_close(pdes[0]);
189		return (flags & WRDE_UNDEF ? WRDE_BADVAL : WRDE_SYNTAX);
190	}
191	_close(pdes[0]);
192
193	/*
194	 * Break the null-terminated expanded word strings out into
195	 * the vector.
196	 */
197	if (vofs == 0 && flags & WRDE_DOOFS)
198		while (vofs < we->we_offs)
199			we->we_wordv[vofs++] = NULL;
200	p = we->we_strings + sofs;
201	while (nwords-- != 0) {
202		we->we_wordv[vofs++] = p;
203		if ((np = memchr(p, '\0', nbytes)) == NULL)
204			return (WRDE_NOSPACE);	/* XXX */
205		nbytes -= np - p + 1;
206		p = np + 1;
207	}
208	we->we_wordv[vofs] = NULL;
209
210	return (0);
211}
212
213/*
214 * we_check --
215 *	Check that the string contains none of the following unquoted
216 *	special characters: <newline> |&;<>(){}
217 *	or command substitutions when WRDE_NOCMD is set in flags.
218 */
219static int
220we_check(const char *words, int flags)
221{
222	char c;
223	int dquote, level, quote, squote;
224
225	quote = squote = dquote = 0;
226	while ((c = *words++) != '\0') {
227		switch (c) {
228		case '\\':
229			quote ^= 1;
230			continue;
231		case '\'':
232			if (quote + dquote == 0)
233				squote ^= 1;
234			break;
235		case '"':
236			if (quote + squote == 0)
237				dquote ^= 1;
238			break;
239		case '`':
240			if (quote + squote == 0 && flags & WRDE_NOCMD)
241				return (WRDE_CMDSUB);
242			while ((c = *words++) != '\0' && c != '`')
243				if (c == '\\' && (c = *words++) == '\0')
244					break;
245			if (c == '\0')
246				return (WRDE_SYNTAX);
247			break;
248		case '|': case '&': case ';': case '<': case '>':
249		case '{': case '}': case '(': case ')': case '\n':
250			if (quote + squote + dquote == 0)
251				return (WRDE_BADCHAR);
252			break;
253		case '$':
254			if ((c = *words++) == '\0')
255				break;
256			else if (quote + squote == 0 && c == '(') {
257				if (flags & WRDE_NOCMD && *words != '(')
258					return (WRDE_CMDSUB);
259				level = 1;
260				while ((c = *words++) != '\0') {
261					if (c == '\\') {
262						if ((c = *words++) == '\0')
263							break;
264					} else if (c == '(')
265						level++;
266					else if (c == ')' && --level == 0)
267						break;
268				}
269				if (c == '\0' || level != 0)
270					return (WRDE_SYNTAX);
271			} else if (quote + squote == 0 && c == '{') {
272				level = 1;
273				while ((c = *words++) != '\0') {
274					if (c == '\\') {
275						if ((c = *words++) == '\0')
276							break;
277					} else if (c == '{')
278						level++;
279					else if (c == '}' && --level == 0)
280						break;
281				}
282				if (c == '\0' || level != 0)
283					return (WRDE_SYNTAX);
284			} else
285				c = *--words;
286			break;
287		default:
288			break;
289		}
290		quote = 0;
291	}
292	if (quote + squote + dquote != 0)
293		return (WRDE_SYNTAX);
294
295	return (0);
296}
297
298/*
299 * wordfree --
300 *	Free the result of wordexp(). See wordexp(3).
301 *
302 *	Specified by IEEE Std. 1003.1-2001.
303 */
304void
305wordfree(wordexp_t *we)
306{
307
308	if (we == NULL)
309		return;
310	free(we->we_wordv);
311	free(we->we_strings);
312	we->we_wordv = NULL;
313	we->we_strings = NULL;
314	we->we_nbytes = 0;
315	we->we_wordc = 0;
316}
317