sys/ddb/db_lex.c

   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  */
  28
  29 /*
  30  *      Author: David B. Golub, Carnegie Mellon University
  31  *      Date:   7/90
  32  */
  33 /*
  34  * Lexical analyzer.
  35  */
  36 #include <sys/param.h>
  37 #include <sys/systm.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(int c)
  90 {
  91         db_look_char = c;
  92 }
  93
  94 static int      db_look_token = 0;
  95
  96 void
  97 db_unread_token(int t)
  98 {
  99         db_look_token = t;
 100 }
 101
 102 int
 103 db_read_token(void)
 104 {
 105         int     t;
 106
 107         if (db_look_token) {
 108             t = db_look_token;
 109             db_look_token = 0;
 110         }
 111         else
 112             t = db_lex();
 113         return (t);
 114 }
 115
 116 db_expr_t       db_tok_number;
 117 char    db_tok_string[TOK_STRING_SIZE];
 118
 119 db_expr_t       db_radix = 16;
 120
 121 /*
 122  * Convert the number to a string in the current radix.
 123  * This replaces the non-standard %r kprintf() format.
 124  */
 125
 126 char *
 127 db_num_to_str(db_expr_t val)
 128 {
 129
 130         /*
 131          * 2 chars for "0x", 1 for a sign ("-")
 132          * up to 21 chars for a 64-bit number:
 133          *   % echo 2^64 | bc | wc -c
 134          *   21
 135          * and 1 char for a terminal NUL
 136          * 2+1+21+1 => 25
 137          */
 138         static char buf[25];
 139
 140         if (db_radix == 16)
 141                 ksnprintf(buf, sizeof(buf), "%" DDB_EXPR_FMT "x", val);
 142         else if (db_radix == 8)
 143                 ksnprintf(buf, sizeof(buf), "%" DDB_EXPR_FMT "o", val);
 144         else
 145                 ksnprintf(buf, sizeof(buf), "%" DDB_EXPR_FMT "u", val);
 146
 147         return (buf);
 148 }
 149
 150 void
 151 db_flush_lex(void)
 152 {
 153         db_flush_line();
 154         db_look_char = 0;
 155         db_look_token = 0;
 156 }
 157
 158 static int
 159 db_lex(void)
 160 {
 161         int     c;
 162
 163         c = db_read_char();
 164         while (c <= ' ' || c > '~') {
 165             if (c == '\n' || c == -1)
 166                 return (tEOL);
 167             c = db_read_char();
 168         }
 169
 170         if (c >= '0' && c <= '9') {
 171             /* number */
 172             int r, digit = 0;
 173
 174             if (c > '0')
 175                 r = db_radix;
 176             else {
 177                 c = db_read_char();
 178                 if (c == 'O' || c == 'o')
 179                     r = 8;
 180                 else if (c == 'T' || c == 't')
 181                     r = 10;
 182                 else if (c == 'X' || c == 'x')
 183                     r = 16;
 184                 else {
 185                     r = db_radix;
 186                     db_unread_char(c);
 187                 }
 188                 c = db_read_char();
 189             }
 190             db_tok_number = 0;
 191             for (;;) {
 192                 if (c >= '0' && c <= ((r == 8) ? '7' : '9'))
 193                     digit = c - '0';
 194                 else if (r == 16 && ((c >= 'A' && c <= 'F') ||
 195                                      (c >= 'a' && c <= 'f'))) {
 196                     if (c >= 'a')
 197                         digit = c - 'a' + 10;
 198                     else if (c >= 'A')
 199                         digit = c - 'A' + 10;
 200                 }
 201                 else
 202                     break;
 203                 db_tok_number = db_tok_number * r + digit;
 204                 c = db_read_char();
 205             }
 206             if ((c >= '0' && c <= '9') ||
 207                 (c >= 'A' && c <= 'Z') ||
 208                 (c >= 'a' && c <= 'z') ||
 209                 (c == '_'))
 210             {
 211                 db_error("Bad character in number\n");
 212                 db_flush_lex();
 213                 return (tEOF);
 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             }
 231             *cp++ = c;
 232             while (1) {
 233                 c = db_read_char();
 234                 if ((c >= 'A' && c <= 'Z') ||
 235                     (c >= 'a' && c <= 'z') ||
 236                     (c >= '0' && c <= '9') ||
 237                     c == '_' || c == '\\' || c == ':')
 238                 {
 239                     if (c == '\\') {
 240                         c = db_read_char();
 241                         if (c == '\n' || c == -1)
 242                             db_error("Bad escape\n");
 243                     }
 244                     *cp++ = c;
 245                     if (cp == db_tok_string+sizeof(db_tok_string)) {
 246                         db_error("String too long\n");
 247                         db_flush_lex();
 248                         return (tEOF);
 249                     }
 250                     continue;
 251                 }
 252                 else {
 253                     *cp = '\0';
 254                     break;
 255                 }
 256             }
 257             db_unread_char(c);
 258             return (tIDENT);
 259         }
 260
 261         switch (c) {
 262             case '+':
 263                 return (tPLUS);
 264             case '-':
 265                 return (tMINUS);
 266             case '.':
 267                 c = db_read_char();
 268                 if (c == '.')
 269                     return (tDOTDOT);
 270                 db_unread_char(c);
 271                 return (tDOT);
 272             case '*':
 273                 return (tSTAR);
 274             case '/':
 275                 return (tSLASH);
 276             case '=':
 277                 return (tEQ);
 278             case '%':
 279                 return (tPCT);
 280             case '#':
 281                 return (tHASH);
 282             case '(':
 283                 return (tLPAREN);
 284             case ')':
 285                 return (tRPAREN);
 286             case ',':
 287                 return (tCOMMA);
 288             case '"':
 289                 return (tDITTO);
 290             case '$':
 291                 return (tDOLLAR);
 292             case '!':
 293                 return (tEXCL);
 294             case '<':
 295                 c = db_read_char();
 296                 if (c == '<')
 297                     return (tSHIFT_L);
 298                 db_unread_char(c);
 299                 break;
 300             case '>':
 301                 c = db_read_char();
 302                 if (c == '>')
 303                     return (tSHIFT_R);
 304                 db_unread_char(c);
 305                 break;
 306             case -1:
 307                 return (tEOF);
 308         }
 309         db_printf("Bad character\n");
 310         db_flush_lex();
 311         return (tEOF);
 312 }