sys/ddb/db_lex.c

   1 /*      $OpenBSD: db_lex.c,v 1.9 2006/03/13 06:23:20 jsg Exp $  */
   2 /*      $NetBSD: db_lex.c,v 1.8 1996/02/05 01:57:05 christos Exp $      */
   3
   4 /*
   5  * Mach Operating System
   6  * Copyright (c) 1993,1992,1991,1990 Carnegie Mellon University
   7  * All Rights Reserved.
   8  *
   9  * Permission to use, copy, modify and distribute this software and its
  10  * documentation is hereby granted, provided that both the copyright
  11  * notice and this permission notice appear in all copies of the
  12  * software, derivative works or modified versions, and any portions
  13  * thereof, and that both notices appear in supporting documentation.
  14  *
  15  * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
  16  * CONDITION.  CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
  17  * ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
  18  *
  19  * Carnegie Mellon requests users of this software to return to
  20  *
  21  *  Software Distribution Coordinator  or  Software.Distribution@CS.CMU.EDU
  22  *  School of Computer Science
  23  *  Carnegie Mellon University
  24  *  Pittsburgh PA 15213-3890
  25  *
  26  * any improvements or extensions that they make and grant Carnegie Mellon
  27  * the rights to redistribute these changes.
  28  *
  29  *      Author: David B. Golub, Carnegie Mellon University
  30  *      Date:   7/90
  31  */
  32
  33 /*
  34  * Lexical analyzer.
  35  */
  36 #include <sys/param.h>
  37 #include <sys/proc.h>
  38
  39 #include <uvm/uvm_extern.h>
  40
  41 #include <machine/db_machdep.h>
  42
  43 #include <ddb/db_lex.h>
  44 #include <ddb/db_output.h>
  45 #include <ddb/db_command.h>
  46 #include <ddb/db_sym.h>
  47 #include <ddb/db_extern.h>
  48 #include <ddb/db_var.h>
  49
  50 char    db_line[120];
  51 char *  db_lp, *db_endlp;
  52
  53 db_expr_t db_tok_number;
  54 char    db_tok_string[TOK_STRING_SIZE];
  55
  56 int
  57 db_read_line(void)
  58 {
  59         int     i;
  60
  61         i = db_readline(db_line, sizeof(db_line));
  62         if (i == 0)
  63             return (0); /* EOI */
  64         db_lp = db_line;
  65         db_endlp = db_lp + i;
  66         return (i);
  67 }
  68
  69 void
  70 db_flush_line(void)
  71 {
  72         db_lp = db_line;
  73         db_endlp = db_line;
  74 }
  75
  76 int     db_look_char = 0;
  77
  78 int
  79 db_read_char(void)
  80 {
  81         int     c;
  82
  83         if (db_look_char != 0) {
  84             c = db_look_char;
  85             db_look_char = 0;
  86         }
  87         else if (db_lp >= db_endlp)
  88             c = -1;
  89         else
  90             c = *db_lp++;
  91         return (c);
  92 }
  93
  94 void
  95 db_unread_char(int c)
  96 {
  97         db_look_char = c;
  98 }
  99
 100 int     db_look_token = 0;
 101
 102 void
 103 db_unread_token(int t)
 104 {
 105         db_look_token = t;
 106 }
 107
 108 int
 109 db_read_token(void)
 110 {
 111         int     t;
 112
 113         if (db_look_token) {
 114             t = db_look_token;
 115             db_look_token = 0;
 116         }
 117         else
 118             t = db_lex();
 119         return (t);
 120 }
 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 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                 /*NOTREACHED*/
 185             }
 186             db_unread_char(c);
 187             return (tNUMBER);
 188         }
 189         if ((c >= 'A' && c <= 'Z') ||
 190             (c >= 'a' && c <= 'z') ||
 191             c == '_' || c == '\\')
 192         {
 193             /* string */
 194             char *cp;
 195
 196             cp = db_tok_string;
 197             if (c == '\\') {
 198                 c = db_read_char();
 199                 if (c == '\n' || c == -1) {
 200                     db_error("Bad escape\n");
 201                     /*NOTREACHED*/
 202                 }
 203             }
 204             *cp++ = c;
 205             while (1) {
 206                 c = db_read_char();
 207                 if ((c >= 'A' && c <= 'Z') ||
 208                     (c >= 'a' && c <= 'z') ||
 209                     (c >= '0' && c <= '9') ||
 210                     c == '_' || c == '\\' || c == ':')
 211                 {
 212                     if (c == '\\') {
 213                         c = db_read_char();
 214                         if (c == '\n' || c == -1) {
 215                             db_error("Bad escape\n");
 216                             /*NOTREACHED*/
 217                         }
 218                     }
 219                     *cp++ = c;
 220                     if (cp == db_tok_string+sizeof(db_tok_string)) {
 221                         db_error("String too long\n");
 222                         /*NOTREACHED*/
 223                     }
 224                     continue;
 225                 }
 226                 else {
 227                     *cp = '\0';
 228                     break;
 229                 }
 230             }
 231             db_unread_char(c);
 232             return (tIDENT);
 233         }
 234
 235         switch (c) {
 236             case '+':
 237                 return (tPLUS);
 238             case '-':
 239                 return (tMINUS);
 240             case '.':
 241                 c = db_read_char();
 242                 if (c == '.')
 243                     return (tDOTDOT);
 244                 db_unread_char(c);
 245                 return (tDOT);
 246             case '*':
 247                 return (tSTAR);
 248             case '/':
 249                 return (tSLASH);
 250             case '=':
 251                 return (tEQ);
 252             case '%':
 253                 return (tPCT);
 254             case '#':
 255                 return (tHASH);
 256             case '(':
 257                 return (tLPAREN);
 258             case ')':
 259                 return (tRPAREN);
 260             case ',':
 261                 return (tCOMMA);
 262             case '"':
 263                 return (tDITTO);
 264             case '$':
 265                 return (tDOLLAR);
 266             case '!':
 267                 return (tEXCL);
 268             case '<':
 269                 c = db_read_char();
 270                 if (c == '<')
 271                     return (tSHIFT_L);
 272                 db_unread_char(c);
 273                 break;
 274             case '>':
 275                 c = db_read_char();
 276                 if (c == '>')
 277                     return (tSHIFT_R);
 278                 db_unread_char(c);
 279                 break;
 280             case -1:
 281                 return (tEOF);
 282         }
 283         db_printf("Bad character\n");
 284         db_flush_lex();
 285         return (tEOF);
 286 }