xref: /netbsd-src/sys/ddb/db_lex.c (revision 9fbd88883c38d0c0fbfcbe66d76fe6b0fab3f9de)
1 /*	$NetBSD: db_lex.c,v 1.15 2001/11/12 22:54:05 lukem Exp $	*/
2 
3 /*
4  * Mach Operating System
5  * Copyright (c) 1991,1990 Carnegie Mellon University
6  * All Rights Reserved.
7  *
8  * Permission to use, copy, modify and distribute this software and its
9  * documentation is hereby granted, provided that both the copyright
10  * notice and this permission notice appear in all copies of the
11  * software, derivative works or modified versions, and any portions
12  * thereof, and that both notices appear in supporting documentation.
13  *
14  * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
15  * CONDITION.  CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
16  * ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
17  *
18  * Carnegie Mellon requests users of this software to return to
19  *
20  *  Software Distribution Coordinator  or  Software.Distribution@CS.CMU.EDU
21  *  School of Computer Science
22  *  Carnegie Mellon University
23  *  Pittsburgh PA 15213-3890
24  *
25  * any improvements or extensions that they make and grant Carnegie the
26  * rights to redistribute these changes.
27  *
28  *	Author: David B. Golub, Carnegie Mellon University
29  *	Date:	7/90
30  */
31 
32 /*
33  * Lexical analyzer.
34  */
35 
36 #include <sys/cdefs.h>
37 __KERNEL_RCSID(0, "$NetBSD: db_lex.c,v 1.15 2001/11/12 22:54:05 lukem Exp $");
38 
39 #include <sys/param.h>
40 #include <sys/systm.h>
41 
42 #include <machine/db_machdep.h>
43 
44 #include <ddb/db_lex.h>
45 #include <ddb/db_output.h>
46 #include <ddb/db_command.h>
47 #include <ddb/db_sym.h>
48 #include <ddb/db_extern.h>
49 
50 char	db_line[120];
51 char *	db_lp, *db_endlp;
52 
53 int
54 db_read_line()
55 {
56 	int	i;
57 
58 	i = db_readline(db_line, sizeof(db_line));
59 	if (i == 0)
60 	    return (0);	/* EOI */
61 	db_lp = db_line;
62 	db_endlp = db_lp + i;
63 	return (i);
64 }
65 
66 void
67 db_flush_line()
68 {
69 	db_lp = db_line;
70 	db_endlp = db_line;
71 }
72 
73 int	db_look_char = 0;
74 
75 int
76 db_read_char()
77 {
78 	int	c;
79 
80 	if (db_look_char != 0) {
81 	    c = db_look_char;
82 	    db_look_char = 0;
83 	}
84 	else if (db_lp >= db_endlp)
85 	    c = -1;
86 	else
87 	    c = *db_lp++;
88 	return (c);
89 }
90 
91 void
92 db_unread_char(c)
93 	int c;
94 {
95 	db_look_char = c;
96 }
97 
98 int	db_look_token = 0;
99 
100 void
101 db_unread_token(t)
102 	int	t;
103 {
104 	db_look_token = t;
105 }
106 
107 int
108 db_read_token()
109 {
110 	int	t;
111 
112 	if (db_look_token) {
113 	    t = db_look_token;
114 	    db_look_token = 0;
115 	}
116 	else
117 	    t = db_lex();
118 	return (t);
119 }
120 
121 int	db_radix = 16;
122 
123 /*
124  * Convert the number to a string in the current radix.
125  * This replaces the non-standard %n printf() format.
126  */
127 
128 char *
129 db_num_to_str(val)
130 db_expr_t val;
131 {
132 	/*
133 	 * 2 chars for "0x", 1 for a sign ("-")
134 	 * up to 21 chars for a 64-bit number:
135 	 *   % echo 2^64 | bc | wc -c
136 	 *   21
137 	 * and 1 char for a terminal NUL
138 	 * 2+1+21+1 => 25
139 	 */
140 	static char buf[25];
141 
142 	if (db_radix == 16)
143 		snprintf(buf, sizeof(buf), "%#10lx", val);
144 	else if (db_radix == 8)
145 		snprintf(buf, sizeof(buf), "%#10lo", val);
146 	else
147 		snprintf(buf, sizeof(buf), "%10lu", val);
148 	return buf;
149 }
150 
151 void
152 db_flush_lex()
153 {
154 	db_flush_line();
155 	db_look_char = 0;
156 	db_look_token = 0;
157 }
158 
159 int
160 db_lex()
161 {
162 	int	c;
163 
164 	c = db_read_char();
165 	while (c <= ' ' || c > '~') {
166 	    if (c == '\n' || c == -1)
167 		return (tEOL);
168 	    c = db_read_char();
169 	}
170 
171 	if (c >= '0' && c <= '9') {
172 	    /* number */
173 	    db_expr_t	r, digit = 0;
174 
175 	    if (c > '0')
176 		r = db_radix;
177 	    else {
178 		c = db_read_char();
179 		if (c == 'O' || c == 'o')
180 		    r = 8;
181 		else if (c == 'T' || c == 't')
182 		    r = 10;
183 		else if (c == 'X' || c == 'x')
184 		    r = 16;
185 		else {
186 		    r = db_radix;
187 		    db_unread_char(c);
188 		}
189 		c = db_read_char();
190 	    }
191 	    db_tok_number = 0;
192 	    for (;;) {
193 		if (c >= '0' && c <= ((r == 8) ? '7' : '9'))
194 		    digit = c - '0';
195 		else if (r == 16 && ((c >= 'A' && c <= 'F') ||
196 				     (c >= 'a' && c <= 'f'))) {
197 		    if (c >= 'a')
198 			digit = c - 'a' + 10;
199 		    else if (c >= 'A')
200 			digit = c - 'A' + 10;
201 		}
202 		else
203 		    break;
204 		db_tok_number = db_tok_number * r + digit;
205 		c = db_read_char();
206 	    }
207 	    if ((c >= '0' && c <= '9') ||
208 		(c >= 'A' && c <= 'Z') ||
209 		(c >= 'a' && c <= 'z') ||
210 		(c == '_'))
211 	    {
212 		db_error("Bad character in number\n");
213 		/*NOTREACHED*/
214 	    }
215 	    db_unread_char(c);
216 	    return (tNUMBER);
217 	}
218 	if ((c >= 'A' && c <= 'Z') ||
219 	    (c >= 'a' && c <= 'z') ||
220 	    c == '_' || c == '\\')
221 	{
222 	    /* string */
223 	    char *cp;
224 
225 	    cp = db_tok_string;
226 	    if (c == '\\') {
227 		c = db_read_char();
228 		if (c == '\n' || c == -1) {
229 		    db_error("Bad escape\n");
230 		    /*NOTREACHED*/
231 		}
232 	    }
233 	    *cp++ = c;
234 	    while (1) {
235 		c = db_read_char();
236 		if ((c >= 'A' && c <= 'Z') ||
237 		    (c >= 'a' && c <= 'z') ||
238 		    (c >= '0' && c <= '9') ||
239 		    c == '_' || c == '\\' || c == ':')
240 		{
241 		    if (c == '\\') {
242 			c = db_read_char();
243 			if (c == '\n' || c == -1) {
244 			    db_error("Bad escape\n");
245 			    /*NOTREACHED*/
246 			}
247 		    }
248 		    *cp++ = c;
249 		    if (cp == db_tok_string+sizeof(db_tok_string)) {
250 			db_error("String too long\n");
251 			/*NOTREACHED*/
252 		    }
253 		    continue;
254 		}
255 		else {
256 		    *cp = '\0';
257 		    break;
258 		}
259 	    }
260 	    db_unread_char(c);
261 	    return (tIDENT);
262 	}
263 
264 	switch (c) {
265 	    case '+':
266 		return (tPLUS);
267 	    case '-':
268 		return (tMINUS);
269 	    case '.':
270 		c = db_read_char();
271 		if (c == '.')
272 		    return (tDOTDOT);
273 		db_unread_char(c);
274 		return (tDOT);
275 	    case '*':
276 		return (tSTAR);
277 	    case '/':
278 		return (tSLASH);
279 	    case '=':
280 		return (tEQ);
281 	    case '%':
282 		return (tPCT);
283 	    case '#':
284 		return (tHASH);
285 	    case '(':
286 		return (tLPAREN);
287 	    case ')':
288 		return (tRPAREN);
289 	    case ',':
290 		return (tCOMMA);
291 	    case '"':
292 		return (tDITTO);
293 	    case '$':
294 		return (tDOLLAR);
295 	    case '!':
296 		return (tEXCL);
297 	    case '<':
298 		c = db_read_char();
299 		if (c == '<')
300 		    return (tSHIFT_L);
301 		db_unread_char(c);
302 		break;
303 	    case '>':
304 		c = db_read_char();
305 		if (c == '>')
306 		    return (tSHIFT_R);
307 		db_unread_char(c);
308 		break;
309 	    case -1:
310 		return (tEOF);
311 	}
312 	db_printf("Bad character\n");
313 	db_flush_lex();
314 	return (tEOF);
315 }
316