1%{
2/*-
3 * Copyright (c) 2011 James Gritton
4 * All rights reserved.
5 *
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
8 * are met:
9 * 1. Redistributions of source code must retain the above copyright
10 *    notice, this list of conditions and the following disclaimer.
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 AND CONTRIBUTORS ``AS IS'' AND
16 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
17 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
18 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
19 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
20 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
21 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
22 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
23 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
24 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
25 * SUCH DAMAGE.
26 */
27
28#include <sys/cdefs.h>
29__FBSDID("$FreeBSD$");
30
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
39extern int yynerrs;
40
41static ssize_t text2lval(size_t triml, size_t trimr, int tovar);
42
43static int instr;
44static int lineno = 1;
45
46#define YY_DECL int yylex(void)
47%}
48
49%option noinput
50%option nounput
51
52%start _ DQ
53
54%%
55
56			/* Whitespace or equivalent */
57<_>[ \t]+		instr = 0;
58<_>#.*			;
59<_>\/\/.*		;
60<_>\/\*([^*]|(\*+([^*\/])))*\*+\/ {
61				const char *s;
62
63				for (s = yytext; s < yytext + yyleng; s++)
64					if (*s == '\n')
65						lineno++;
66				instr = 0;
67			}
68<_>\n			{
69				lineno++;
70				instr = 0;
71			}
72
73			/* Reserved tokens */
74<_>\+=			{
75				instr = 0;
76				return PLEQ;
77			}
78<_>[,;={}]		{
79				instr = 0;
80				return yytext[0];
81			}
82
83			/* Atomic (unquoted) strings */
84<_,DQ>[A-Za-z0-9_!%&()\-.:<>?@\[\]^`|~]+ |
85<_,DQ>\\(.|\n|[0-7]{1,3}|x[0-9A-Fa-f]{1,2}) |
86<_,DQ>[$*+/\\]		{
87				(void)text2lval(0, 0, 0);
88				return instr ? STR1 : (instr = 1, STR);
89			}
90
91			/* Single and double quoted strings */
92<_>'([^\'\\]|\\(.|\n))*' {
93				(void)text2lval(1, 1, 0);
94				return instr ? STR1 : (instr = 1, STR);
95			}
96<_>\"([^"\\]|\\(.|\n))*\" |
97<DQ>[^\"$\\]([^"\\]|\\(.|\n))*\" {
98				size_t skip;
99				ssize_t atvar;
100
101				skip = yytext[0] == '"' ? 1 : 0;
102				atvar = text2lval(skip, 1, 1);
103				if (atvar < 0)
104					BEGIN _;
105				else {
106					/*
107					 * The string has a variable inside it.
108					 * Go into DQ mode to get the variable
109					 * and then the rest of the string.
110					 */
111					BEGIN DQ;
112					yyless(atvar);
113				}
114				return instr ? STR1 : (instr = 1, STR);
115			}
116<DQ>\"			BEGIN _;
117
118			/* Variables, single-word or bracketed */
119<_,DQ>$[A-Za-z_][A-Za-z_0-9]* {
120				(void)text2lval(1, 0, 0);
121				return instr ? VAR1 : (instr = 1, VAR);
122			}
123<_>$\{([^\n{}]|\\(.|\n))*\} |
124<DQ>$\{([^\n\"{}]|\\(.|\n))*\} {
125				(void)text2lval(2, 1, 0);
126				return instr ? VAR1 : (instr = 1, VAR);
127			}
128
129			/* Partially formed bits worth complaining about */
130<_>\/\*([^*]|(\*+([^*\/])))*\** {
131				warnx("%s line %d: unterminated comment",
132				    cfname, lineno);
133				yynerrs++;
134			}
135<_>'([^\n'\\]|\\.)*	|
136<_>\"([^\n\"\\]|\\.)*	{
137				warnx("%s line %d: unterminated string",
138				    cfname, lineno);
139				yynerrs++;
140			}
141<_>$\{([^\n{}]|\\.)*	|
142<DQ>$\{([^\n\"{}]|\\.)*	{
143				warnx("%s line %d: unterminated variable",
144				    cfname, lineno);
145				yynerrs++;
146			}
147
148			/* A hack because "<0>" rules aren't allowed */
149<_>.			return yytext[0];
150.|\n			{
151				BEGIN _;
152				yyless(0);
153			}
154
155%%
156
157void
158yyerror(const char *s)
159{
160	if (!yytext)
161		warnx("%s line %d: %s", cfname, lineno, s);
162	else if (!yytext[0])
163		warnx("%s: unexpected EOF", cfname);
164	else
165		warnx("%s line %d: %s: %s", cfname, lineno, yytext, s);
166}
167
168/*
169 * Copy string from yytext to yylval, handling backslash escapes,
170 * and optionally stopping at the beginning of a variable.
171 */
172static ssize_t
173text2lval(size_t triml, size_t trimr, int tovar)
174{
175	char *d;
176	const char *s, *se;
177
178	yylval.cs = d = emalloc(yyleng - trimr - triml + 1);
179	se = yytext + (yyleng - trimr);
180	for (s = yytext + triml; s < se; s++, d++) {
181		if (*s != '\\') {
182			if (tovar && *s == '$') {
183				*d = '\0';
184				return s - yytext;
185			}
186			if (*s == '\n')
187				lineno++;
188			*d = *s;
189			continue;
190		}
191		s++;
192		if (*s >= '0' && *s <= '7') {
193			*d = *s - '0';
194			if (s + 1 < se && s[1] >= '0' && s[1] <= '7') {
195				*d = 010 * *d + (*++s - '0');
196				if (s + 1 < se && s[1] >= '0' && s[1] <= '7')
197					*d = 010 * *d + (*++s - '0');
198			}
199			continue;
200		}
201		switch (*s) {
202		case 'a':	*d = '\a';	break;
203		case 'b':	*d = '\b';	break;
204		case 'f':	*d = '\f';	break;
205		case 'n':	*d = '\n';	break;
206		case 'r':	*d = '\r';	break;
207		case 't':	*d = '\t';	break;
208		case 'v':	*d = '\v';	break;
209		case '\n':	d--; lineno++;	break;
210		default:	*d = *s;	break;
211		case 'x':
212			*d = 0;
213			if (s + 1 >= se)
214				break;
215			if (s[1] >= '0' && s[1] <= '9')
216				*d = *++s - '0';
217			else if (s[1] >= 'A' && s[1] <= 'F')
218				*d = *++s + (0xA - 'A');
219			else if (s[1] >= 'a' && s[1] <= 'a')
220				*d = *++s + (0xa - 'a');
221			else
222				break;
223			if (s + 1 >= se)
224				break;
225			if (s[1] >= '0' && s[1] <= '9')
226				*d = *d * 0x10 + (*++s - '0');
227			else if (s[1] >= 'A' && s[1] <= 'F')
228				*d = *d * 0x10 + (*++s + (0xA - 'A'));
229			else if (s[1] >= 'a' && s[1] <= 'a')
230				*d = *d * 0x10 + (*++s + (0xa - 'a'));
231		}
232	}
233	*d = '\0';
234	return -1;
235}
236