xref: /freebsd/usr.sbin/jail/jaillex.l (revision 1d386b48a555f61cb7325543adbbb5c3f3407a66)
1 %{
2 /*-
3  * SPDX-License-Identifier: BSD-2-Clause
4  *
5  * Copyright (c) 2011 James Gritton
6  * All rights reserved.
7  *
8  * Redistribution and use in source and binary forms, with or without
9  * modification, are permitted provided that the following conditions
10  * are met:
11  * 1. Redistributions of source code must retain the above copyright
12  *    notice, this list of conditions and the following disclaimer.
13  * 2. Redistributions in binary form must reproduce the above copyright
14  *    notice, this list of conditions and the following disclaimer in the
15  *    documentation and/or other materials provided with the distribution.
16  *
17  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27  * SUCH DAMAGE.
28  */
29 
30 #include <sys/cdefs.h>
31 #include <err.h>
32 #include <stddef.h>
33 #include <stdlib.h>
34 #include <string.h>
35 
36 #include "jailp.h"
37 #include "y.tab.h"
38 
39 #define YY_DECL int yylex(YYSTYPE *yylval, yyscan_t yyscanner)
40 #define YY_EXTRA_TYPE struct cflex*
41 
42 extern YY_DECL;
43 
44 static ssize_t text2lval(size_t triml, size_t trimr, int tovar,
45     YYSTYPE *yylval, yyscan_t scanner);
46 
47 static int instr;
48 %}
49 
50 %option noyywrap
51 %option noinput
52 %option nounput
53 %option reentrant
54 %option yylineno
55 
56 %start _ DQ
57 
58 %%
59 
60 			/* Whitespace or equivalent */
61 <_>[ \t\r\n]+		instr = 0;
62 <_>#.*			instr = 0;
63 <_>\/\/.*		instr = 0;
64 <_>\/\*([^*]|(\*+([^*\/])))*\*+\/ instr = 0;
65 
66 			/* Reserved tokens */
67 <_>\+=			{
68 				instr = 0;
69 				return PLEQ;
70 			}
71 <_>[,;={}]		{
72 				instr = 0;
73 				return yytext[0];
74 			}
75 
76 			/* Atomic (unquoted) strings */
77 <_,DQ>[A-Za-z0-9_!%&()\-.:<>?@\[\]^`|~]+ |
78 <_,DQ>\\(.|\n|[0-7]{1,3}|x[0-9A-Fa-f]{1,2}) |
79 <_,DQ>[$*+/\\]		{
80 				(void)text2lval(0, 0, 0, yylval, yyscanner);
81 				return instr ? STR1 : (instr = 1, STR);
82 			}
83 
84 			/* Single and double quoted strings */
85 <_>'([^\'\\]|\\(.|\n))*' {
86 				(void)text2lval(1, 1, 0, yylval, yyscanner);
87 				return instr ? STR1 : (instr = 1, STR);
88 			}
89 <_>\"([^"\\]|\\(.|\n))*\" |
90 <DQ>[^\"$\\]([^"\\]|\\(.|\n))*\" {
91 				size_t skip;
92 				ssize_t atvar;
93 
94 				skip = yytext[0] == '"' ? 1 : 0;
95 				atvar = text2lval(skip, 1, 1, yylval,
96 					yyscanner);
97 				if (atvar < 0)
98 					BEGIN _;
99 				else {
100 					/*
101 					 * The string has a variable inside it.
102 					 * Go into DQ mode to get the variable
103 					 * and then the rest of the string.
104 					 */
105 					BEGIN DQ;
106 					yyless(atvar);
107 				}
108 				return instr ? STR1 : (instr = 1, STR);
109 			}
110 <DQ>\"			BEGIN _;
111 
112 			/* Variables, single-word or bracketed */
113 <_,DQ>$[A-Za-z_][A-Za-z_0-9]* {
114 				(void)text2lval(1, 0, 0, yylval, yyscanner);
115 				return instr ? VAR1 : (instr = 1, VAR);
116 			}
117 <_>$\{([^\n{}]|\\(.|\n))*\} |
118 <DQ>$\{([^\n\"{}]|\\(.|\n))*\} {
119 				(void)text2lval(2, 1, 0, yylval, yyscanner);
120 				return instr ? VAR1 : (instr = 1, VAR);
121 			}
122 
123 			/* Partially formed bits worth complaining about */
124 <_>\/\*([^*]|(\*+([^*\/])))*\** {
125 				warnx("%s line %d: unterminated comment",
126 				    yyextra->cfname, yylineno);
127 				yyextra->error = 1;
128 			}
129 <_>'([^\n'\\]|\\.)*	|
130 <_>\"([^\n\"\\]|\\.)*	{
131 				warnx("%s line %d: unterminated string",
132 				    yyextra->cfname, yylineno);
133 				yyextra->error = 1;
134 			}
135 <_>$\{([^\n{}]|\\.)*	|
136 <DQ>$\{([^\n\"{}]|\\.)*	{
137 				warnx("%s line %d: unterminated variable",
138 				    yyextra->cfname, yylineno);
139 				yyextra->error = 1;
140 			}
141 
142 			/* A hack because "<0>" rules aren't allowed */
143 <_>.			return yytext[0];
144 .|\n			{
145 				BEGIN _;
146 				yyless(0);
147 			}
148 
149 %%
150 
151 /*
152  * Copy string from yytext to yylval, handling backslash escapes,
153  * and optionally stopping at the beginning of a variable.
154  */
155 static ssize_t
156 text2lval(size_t triml, size_t trimr, int tovar, YYSTYPE *yylval,
157     yyscan_t scanner)
158 {
159 	char *d;
160 	const char *s, *se;
161 
162 	struct yyguts_t *yyg = scanner;
163 	yylval->cs = d = emalloc(yyleng - trimr - triml + 1);
164 	se = yytext + (yyleng - trimr);
165 	for (s = yytext + triml; s < se; s++, d++) {
166 		if (*s != '\\') {
167 			if (tovar && *s == '$') {
168 				*d = '\0';
169 				return s - yytext;
170 			}
171 			*d = *s;
172 			continue;
173 		}
174 		s++;
175 		if (*s >= '0' && *s <= '7') {
176 			*d = *s - '0';
177 			if (s + 1 < se && s[1] >= '0' && s[1] <= '7') {
178 				*d = 010 * *d + (*++s - '0');
179 				if (s + 1 < se && s[1] >= '0' && s[1] <= '7')
180 					*d = 010 * *d + (*++s - '0');
181 			}
182 			continue;
183 		}
184 		switch (*s) {
185 		case 'a':	*d = '\a';	break;
186 		case 'b':	*d = '\b';	break;
187 		case 'f':	*d = '\f';	break;
188 		case 'n':	*d = '\n';	break;
189 		case 'r':	*d = '\r';	break;
190 		case 't':	*d = '\t';	break;
191 		case 'v':	*d = '\v';	break;
192 		default:	*d = *s;	break;
193 		case '\n':	d--;		break;
194 		case 'x':
195 			*d = 0;
196 			if (s + 1 >= se)
197 				break;
198 			if (s[1] >= '0' && s[1] <= '9')
199 				*d = *++s - '0';
200 			else if (s[1] >= 'A' && s[1] <= 'F')
201 				*d = *++s + (0xA - 'A');
202 			else if (s[1] >= 'a' && s[1] <= 'f')
203 				*d = *++s + (0xa - 'a');
204 			else
205 				break;
206 			if (s + 1 >= se)
207 				break;
208 			if (s[1] >= '0' && s[1] <= '9')
209 				*d = *d * 0x10 + (*++s - '0');
210 			else if (s[1] >= 'A' && s[1] <= 'F')
211 				*d = *d * 0x10 + (*++s + (0xA - 'A'));
212 			else if (s[1] >= 'a' && s[1] <= 'f')
213 				*d = *d * 0x10 + (*++s + (0xa - 'a'));
214 		}
215 	}
216 	*d = '\0';
217 	return -1;
218 }
219