xref: /dragonfly/sys/ddb/db_lex.c (revision f9a239ec)
1 /*
2  * Mach Operating System
3  * Copyright (c) 1991,1990 Carnegie Mellon University
4  * All Rights Reserved.
5  *
6  * Permission to use, copy, modify and distribute this software and its
7  * documentation is hereby granted, provided that both the copyright
8  * notice and this permission notice appear in all copies of the
9  * software, derivative works or modified versions, and any portions
10  * thereof, and that both notices appear in supporting documentation.
11  *
12  * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS
13  * CONDITION.  CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
14  * ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
15  *
16  * Carnegie Mellon requests users of this software to return to
17  *
18  *  Software Distribution Coordinator  or  Software.Distribution@CS.CMU.EDU
19  *  School of Computer Science
20  *  Carnegie Mellon University
21  *  Pittsburgh PA 15213-3890
22  *
23  * any improvements or extensions that they make and grant Carnegie the
24  * rights to redistribute these changes.
25  *
26  * $FreeBSD: src/sys/ddb/db_lex.c,v 1.18 1999/08/28 00:41:08 peter Exp $
27  * $DragonFly: src/sys/ddb/db_lex.c,v 1.4 2005/02/01 17:53:01 eirikn Exp $
28  */
29 
30 /*
31  *	Author: David B. Golub, Carnegie Mellon University
32  *	Date:	7/90
33  */
34 /*
35  * Lexical analyzer.
36  */
37 #include <sys/param.h>
38 
39 #include <ddb/ddb.h>
40 #include <ddb/db_lex.h>
41 
42 static char	db_line[120];
43 static char *	db_lp, *db_endlp;
44 
45 static int	db_lex (void);
46 static void 	db_flush_line (void);
47 static int 	db_read_char (void);
48 static void 	db_unread_char (int);
49 
50 int
51 db_read_line(void)
52 {
53 	int	i;
54 
55 	i = db_readline(db_line, sizeof(db_line));
56 	if (i == 0)
57 	    return (0);	/* EOI */
58 	db_lp = db_line;
59 	db_endlp = db_lp + i;
60 	return (i);
61 }
62 
63 static void
64 db_flush_line(void)
65 {
66 	db_lp = db_line;
67 	db_endlp = db_line;
68 }
69 
70 static int	db_look_char = 0;
71 
72 static int
73 db_read_char(void)
74 {
75 	int	c;
76 
77 	if (db_look_char != 0) {
78 	    c = db_look_char;
79 	    db_look_char = 0;
80 	}
81 	else if (db_lp >= db_endlp)
82 	    c = -1;
83 	else
84 	    c = *db_lp++;
85 	return (c);
86 }
87 
88 static void
89 db_unread_char(c)
90 	int c;
91 {
92 	db_look_char = c;
93 }
94 
95 static int	db_look_token = 0;
96 
97 void
98 db_unread_token(int t)
99 {
100 	db_look_token = t;
101 }
102 
103 int
104 db_read_token(void)
105 {
106 	int	t;
107 
108 	if (db_look_token) {
109 	    t = db_look_token;
110 	    db_look_token = 0;
111 	}
112 	else
113 	    t = db_lex();
114 	return (t);
115 }
116 
117 db_expr_t	db_tok_number;
118 char	db_tok_string[TOK_STRING_SIZE];
119 
120 db_expr_t	db_radix = 16;
121 
122 void
123 db_flush_lex(void)
124 {
125 	db_flush_line();
126 	db_look_char = 0;
127 	db_look_token = 0;
128 }
129 
130 static int
131 db_lex(void)
132 {
133 	int	c;
134 
135 	c = db_read_char();
136 	while (c <= ' ' || c > '~') {
137 	    if (c == '\n' || c == -1)
138 		return (tEOL);
139 	    c = db_read_char();
140 	}
141 
142 	if (c >= '0' && c <= '9') {
143 	    /* number */
144 	    int	r, digit = 0;
145 
146 	    if (c > '0')
147 		r = db_radix;
148 	    else {
149 		c = db_read_char();
150 		if (c == 'O' || c == 'o')
151 		    r = 8;
152 		else if (c == 'T' || c == 't')
153 		    r = 10;
154 		else if (c == 'X' || c == 'x')
155 		    r = 16;
156 		else {
157 		    r = db_radix;
158 		    db_unread_char(c);
159 		}
160 		c = db_read_char();
161 	    }
162 	    db_tok_number = 0;
163 	    for (;;) {
164 		if (c >= '0' && c <= ((r == 8) ? '7' : '9'))
165 		    digit = c - '0';
166 		else if (r == 16 && ((c >= 'A' && c <= 'F') ||
167 				     (c >= 'a' && c <= 'f'))) {
168 		    if (c >= 'a')
169 			digit = c - 'a' + 10;
170 		    else if (c >= 'A')
171 			digit = c - 'A' + 10;
172 		}
173 		else
174 		    break;
175 		db_tok_number = db_tok_number * r + digit;
176 		c = db_read_char();
177 	    }
178 	    if ((c >= '0' && c <= '9') ||
179 		(c >= 'A' && c <= 'Z') ||
180 		(c >= 'a' && c <= 'z') ||
181 		(c == '_'))
182 	    {
183 		db_error("Bad character in number\n");
184 		db_flush_lex();
185 		return (tEOF);
186 	    }
187 	    db_unread_char(c);
188 	    return (tNUMBER);
189 	}
190 	if ((c >= 'A' && c <= 'Z') ||
191 	    (c >= 'a' && c <= 'z') ||
192 	    c == '_' || c == '\\')
193 	{
194 	    /* string */
195 	    char *cp;
196 
197 	    cp = db_tok_string;
198 	    if (c == '\\') {
199 		c = db_read_char();
200 		if (c == '\n' || c == -1)
201 		    db_error("Bad escape\n");
202 	    }
203 	    *cp++ = c;
204 	    while (1) {
205 		c = db_read_char();
206 		if ((c >= 'A' && c <= 'Z') ||
207 		    (c >= 'a' && c <= 'z') ||
208 		    (c >= '0' && c <= '9') ||
209 		    c == '_' || c == '\\' || c == ':')
210 		{
211 		    if (c == '\\') {
212 			c = db_read_char();
213 			if (c == '\n' || c == -1)
214 			    db_error("Bad escape\n");
215 		    }
216 		    *cp++ = c;
217 		    if (cp == db_tok_string+sizeof(db_tok_string)) {
218 			db_error("String too long\n");
219 			db_flush_lex();
220 			return (tEOF);
221 		    }
222 		    continue;
223 		}
224 		else {
225 		    *cp = '\0';
226 		    break;
227 		}
228 	    }
229 	    db_unread_char(c);
230 	    return (tIDENT);
231 	}
232 
233 	switch (c) {
234 	    case '+':
235 		return (tPLUS);
236 	    case '-':
237 		return (tMINUS);
238 	    case '.':
239 		c = db_read_char();
240 		if (c == '.')
241 		    return (tDOTDOT);
242 		db_unread_char(c);
243 		return (tDOT);
244 	    case '*':
245 		return (tSTAR);
246 	    case '/':
247 		return (tSLASH);
248 	    case '=':
249 		return (tEQ);
250 	    case '%':
251 		return (tPCT);
252 	    case '#':
253 		return (tHASH);
254 	    case '(':
255 		return (tLPAREN);
256 	    case ')':
257 		return (tRPAREN);
258 	    case ',':
259 		return (tCOMMA);
260 	    case '"':
261 		return (tDITTO);
262 	    case '$':
263 		return (tDOLLAR);
264 	    case '!':
265 		return (tEXCL);
266 	    case '<':
267 		c = db_read_char();
268 		if (c == '<')
269 		    return (tSHIFT_L);
270 		db_unread_char(c);
271 		break;
272 	    case '>':
273 		c = db_read_char();
274 		if (c == '>')
275 		    return (tSHIFT_R);
276 		db_unread_char(c);
277 		break;
278 	    case -1:
279 		return (tEOF);
280 	}
281 	db_printf("Bad character\n");
282 	db_flush_lex();
283 	return (tEOF);
284 }
285