asm/stdscan.c

   1 /* ----------------------------------------------------------------------- *
   2  *
   3  *   Copyright 1996-2018 The NASM Authors - All Rights Reserved
   4  *   See the file AUTHORS included with the NASM distribution for
   5  *   the specific copyright holders.
   6  *
   7  *   Redistribution and use in source and binary forms, with or without
   8  *   modification, are permitted provided that the following
   9  *   conditions are met:
  10  *
  11  *   * Redistributions of source code must retain the above copyright
  12  *     notice, this list of conditions and the following disclaimer.
  13  *   * Redistributions in binary form must reproduce the above
  14  *     copyright notice, this list of conditions and the following
  15  *     disclaimer in the documentation and/or other materials provided
  16  *     with the distribution.
  17  *
  18  *     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
  19  *     CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
  20  *     INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
  21  *     MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
  22  *     DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
  23  *     CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  24  *     SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  25  *     NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
  26  *     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  27  *     HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
  28  *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
  29  *     OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
  30  *     EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  31  *
  32  * ----------------------------------------------------------------------- */
  33
  34 #include "compiler.h"
  35
  36 #include <stdio.h>
  37 #include <stdlib.h>
  38 #include <string.h>
  39 #include <ctype.h>
  40
  41 #include "nasm.h"
  42 #include "nasmlib.h"
  43 #include "error.h"
  44 #include "quote.h"
  45 #include "stdscan.h"
  46 #include "insns.h"
  47
  48 /*
  49  * Standard scanner routine used by parser.c and some output
  50  * formats. It keeps a succession of temporary-storage strings in
  51  * stdscan_tempstorage, which can be cleared using stdscan_reset.
  52  */
  53 static char *stdscan_bufptr = NULL;
  54 static char **stdscan_tempstorage = NULL;
  55 static int stdscan_tempsize = 0, stdscan_templen = 0;
  56 #define STDSCAN_TEMP_DELTA 256
  57
  58 void stdscan_set(char *str)
  59 {
  60         stdscan_bufptr = str;
  61 }
  62
  63 char *stdscan_get(void)
  64 {
  65         return stdscan_bufptr;
  66 }
  67
  68 static void stdscan_pop(void)
  69 {
  70     nasm_free(stdscan_tempstorage[--stdscan_templen]);
  71 }
  72
  73 void stdscan_reset(void)
  74 {
  75     while (stdscan_templen > 0)
  76         stdscan_pop();
  77 }
  78
  79 /*
  80  * Unimportant cleanup is done to avoid confusing people who are trying
  81  * to debug real memory leaks
  82  */
  83 void stdscan_cleanup(void)
  84 {
  85     stdscan_reset();
  86     nasm_free(stdscan_tempstorage);
  87 }
  88
  89 static char *stdscan_copy(char *p, int len)
  90 {
  91     char *text;
  92
  93     text = nasm_malloc(len + 1);
  94     memcpy(text, p, len);
  95     text[len] = '\0';
  96
  97     if (stdscan_templen >= stdscan_tempsize) {
  98         stdscan_tempsize += STDSCAN_TEMP_DELTA;
  99         stdscan_tempstorage = nasm_realloc(stdscan_tempstorage,
 100                                            stdscan_tempsize *
 101                                            sizeof(char *));
 102     }
 103     stdscan_tempstorage[stdscan_templen++] = text;
 104
 105     return text;
 106 }
 107
 108 /*
 109  * a token is enclosed with braces. proper token type will be assigned
 110  * accordingly with the token flag.
 111  */
 112 static int stdscan_handle_brace(struct tokenval *tv)
 113 {
 114     if (!(tv->t_flag & TFLAG_BRC_ANY)) {
 115         /* invalid token is put inside braces */
 116         nasm_nonfatal("`%s' is not a valid decorator with braces", tv->t_charptr);
 117         tv->t_type = TOKEN_INVALID;
 118     } else if (tv->t_flag & TFLAG_BRC_OPT) {
 119         if (is_reg_class(OPMASKREG, tv->t_integer)) {
 120             /* within braces, opmask register is now used as a mask */
 121             tv->t_type = TOKEN_OPMASK;
 122         }
 123     }
 124
 125     return tv->t_type;
 126 }
 127
 128 int stdscan(void *private_data, struct tokenval *tv)
 129 {
 130     char ourcopy[MAX_KEYWORD + 1], *r, *s;
 131
 132     (void)private_data;         /* Don't warn that this parameter is unused */
 133
 134     stdscan_bufptr = nasm_skip_spaces(stdscan_bufptr);
 135     if (!*stdscan_bufptr)
 136         return tv->t_type = TOKEN_EOS;
 137
 138     /* we have a token; either an id, a number or a char */
 139     if (nasm_isidstart(*stdscan_bufptr) ||
 140         (*stdscan_bufptr == '$' && nasm_isidstart(stdscan_bufptr[1]))) {
 141         /* now we've got an identifier */
 142         bool is_sym = false;
 143         int token_type;
 144
 145         if (*stdscan_bufptr == '$') {
 146             is_sym = true;
 147             stdscan_bufptr++;
 148         }
 149
 150         r = stdscan_bufptr++;
 151         /* read the entire buffer to advance the buffer pointer but... */
 152         while (nasm_isidchar(*stdscan_bufptr))
 153             stdscan_bufptr++;
 154
 155         /* ... copy only up to IDLEN_MAX-1 characters */
 156         tv->t_charptr = stdscan_copy(r, stdscan_bufptr - r < IDLEN_MAX ?
 157                                      stdscan_bufptr - r : IDLEN_MAX - 1);
 158
 159         if (is_sym || stdscan_bufptr - r > MAX_KEYWORD)
 160             return tv->t_type = TOKEN_ID;       /* bypass all other checks */
 161
 162         for (s = tv->t_charptr, r = ourcopy; *s; s++)
 163             *r++ = nasm_tolower(*s);
 164         *r = '\0';
 165         /* right, so we have an identifier sitting in temp storage. now,
 166          * is it actually a register or instruction name, or what? */
 167         token_type = nasm_token_hash(ourcopy, tv);
 168
 169         if (unlikely(tv->t_flag & TFLAG_WARN)) {
 170             nasm_warnf(ERR_PASS1|ERR_WARN_PTR, "`%s' is not a NASM keyword",
 171                        tv->t_charptr);
 172         }
 173
 174         if (likely(!(tv->t_flag & TFLAG_BRC))) {
 175             /* most of the tokens fall into this case */
 176             return token_type;
 177         } else {
 178             return tv->t_type = TOKEN_ID;
 179         }
 180     } else if (*stdscan_bufptr == '$' && !nasm_isnumchar(stdscan_bufptr[1])) {
 181         /*
 182          * It's a $ sign with no following hex number; this must
 183          * mean it's a Here token ($), evaluating to the current
 184          * assembly location, or a Base token ($$), evaluating to
 185          * the base of the current segment.
 186          */
 187         stdscan_bufptr++;
 188         if (*stdscan_bufptr == '$') {
 189             stdscan_bufptr++;
 190             return tv->t_type = TOKEN_BASE;
 191         }
 192         return tv->t_type = TOKEN_HERE;
 193     } else if (nasm_isnumstart(*stdscan_bufptr)) {   /* now we've got a number */
 194         bool rn_error;
 195         bool is_hex = false;
 196         bool is_float = false;
 197         bool has_e = false;
 198         char c;
 199
 200         r = stdscan_bufptr;
 201
 202         if (*stdscan_bufptr == '$') {
 203             stdscan_bufptr++;
 204             is_hex = true;
 205         }
 206
 207         for (;;) {
 208             c = *stdscan_bufptr++;
 209
 210             if (!is_hex && (c == 'e' || c == 'E')) {
 211                 has_e = true;
 212                 if (*stdscan_bufptr == '+' || *stdscan_bufptr == '-') {
 213                     /*
 214                      * e can only be followed by +/- if it is either a
 215                      * prefixed hex number or a floating-point number
 216                      */
 217                     is_float = true;
 218                     stdscan_bufptr++;
 219                 }
 220             } else if (c == 'H' || c == 'h' || c == 'X' || c == 'x') {
 221                 is_hex = true;
 222             } else if (c == 'P' || c == 'p') {
 223                 is_float = true;
 224                 if (*stdscan_bufptr == '+' || *stdscan_bufptr == '-')
 225                     stdscan_bufptr++;
 226             } else if (nasm_isnumchar(c))
 227                 ; /* just advance */
 228             else if (c == '.')
 229                 is_float = true;
 230             else
 231                 break;
 232         }
 233         stdscan_bufptr--;       /* Point to first character beyond number */
 234
 235         if (has_e && !is_hex) {
 236             /* 1e13 is floating-point, but 1e13h is not */
 237             is_float = true;
 238         }
 239
 240         if (is_float) {
 241             tv->t_charptr = stdscan_copy(r, stdscan_bufptr - r);
 242             return tv->t_type = TOKEN_FLOAT;
 243         } else {
 244             r = stdscan_copy(r, stdscan_bufptr - r);
 245             tv->t_integer = readnum(r, &rn_error);
 246             stdscan_pop();
 247             if (rn_error) {
 248                 /* some malformation occurred */
 249                 return tv->t_type = TOKEN_ERRNUM;
 250             }
 251             tv->t_charptr = NULL;
 252             return tv->t_type = TOKEN_NUM;
 253         }
 254     } else if (*stdscan_bufptr == '\'' || *stdscan_bufptr == '"' ||
 255                *stdscan_bufptr == '`') {
 256         /* a quoted string */
 257         char start_quote = *stdscan_bufptr;
 258         tv->t_charptr = stdscan_bufptr;
 259         tv->t_inttwo = nasm_unquote(tv->t_charptr, &stdscan_bufptr);
 260         if (*stdscan_bufptr != start_quote)
 261             return tv->t_type = TOKEN_ERRSTR;
 262         stdscan_bufptr++;       /* Skip final quote */
 263         return tv->t_type = TOKEN_STR;
 264     } else if (*stdscan_bufptr == '{') {
 265         /* now we've got a decorator */
 266         int token_len;
 267
 268         stdscan_bufptr = nasm_skip_spaces(stdscan_bufptr);
 269
 270         r = ++stdscan_bufptr;
 271         /*
 272          * read the entire buffer to advance the buffer pointer
 273          * {rn-sae}, {rd-sae}, {ru-sae}, {rz-sae} contain '-' in tokens.
 274          */
 275         while (nasm_isbrcchar(*stdscan_bufptr))
 276             stdscan_bufptr++;
 277
 278         token_len = stdscan_bufptr - r;
 279
 280         /* ... copy only up to DECOLEN_MAX-1 characters */
 281         tv->t_charptr = stdscan_copy(r, token_len < DECOLEN_MAX ?
 282                                         token_len : DECOLEN_MAX - 1);
 283
 284         stdscan_bufptr = nasm_skip_spaces(stdscan_bufptr);
 285         /* if brace is not closed properly or token is too long  */
 286         if ((*stdscan_bufptr != '}') || (token_len > MAX_KEYWORD)) {
 287             nasm_nonfatal("invalid decorator token inside braces");
 288             return tv->t_type = TOKEN_INVALID;
 289         }
 290
 291         stdscan_bufptr++;       /* skip closing brace */
 292
 293         for (s = tv->t_charptr, r = ourcopy; *s; s++)
 294             *r++ = nasm_tolower(*s);
 295         *r = '\0';
 296
 297         /* right, so we have a decorator sitting in temp storage. */
 298         nasm_token_hash(ourcopy, tv);
 299
 300         /* handle tokens inside braces */
 301         return stdscan_handle_brace(tv);
 302     } else if (*stdscan_bufptr == ';') {
 303         /* a comment has happened - stay */
 304         return tv->t_type = TOKEN_EOS;
 305     } else if (stdscan_bufptr[0] == '>' && stdscan_bufptr[1] == '>') {
 306         if (stdscan_bufptr[2] == '>') {
 307             stdscan_bufptr += 3;
 308             return tv->t_type = TOKEN_SAR;
 309         } else {
 310             stdscan_bufptr += 2;
 311             return tv->t_type = TOKEN_SHR;
 312         }
 313     } else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '<') {
 314         stdscan_bufptr += stdscan_bufptr[2] == '<' ? 3 : 2;
 315         return tv->t_type = TOKEN_SHL;
 316     } else if (stdscan_bufptr[0] == '/' && stdscan_bufptr[1] == '/') {
 317         stdscan_bufptr += 2;
 318         return tv->t_type = TOKEN_SDIV;
 319     } else if (stdscan_bufptr[0] == '%' && stdscan_bufptr[1] == '%') {
 320         stdscan_bufptr += 2;
 321         return tv->t_type = TOKEN_SMOD;
 322     } else if (stdscan_bufptr[0] == '=' && stdscan_bufptr[1] == '=') {
 323         stdscan_bufptr += 2;
 324         return tv->t_type = TOKEN_EQ;
 325     } else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '>') {
 326         stdscan_bufptr += 2;
 327         return tv->t_type = TOKEN_NE;
 328     } else if (stdscan_bufptr[0] == '!' && stdscan_bufptr[1] == '=') {
 329         stdscan_bufptr += 2;
 330         return tv->t_type = TOKEN_NE;
 331     } else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '=') {
 332         stdscan_bufptr += 2;
 333         return tv->t_type = TOKEN_LE;
 334     } else if (stdscan_bufptr[0] == '>' && stdscan_bufptr[1] == '=') {
 335         stdscan_bufptr += 2;
 336         return tv->t_type = TOKEN_GE;
 337     } else if (stdscan_bufptr[0] == '&' && stdscan_bufptr[1] == '&') {
 338         stdscan_bufptr += 2;
 339         return tv->t_type = TOKEN_DBL_AND;
 340     } else if (stdscan_bufptr[0] == '^' && stdscan_bufptr[1] == '^') {
 341         stdscan_bufptr += 2;
 342         return tv->t_type = TOKEN_DBL_XOR;
 343     } else if (stdscan_bufptr[0] == '|' && stdscan_bufptr[1] == '|') {
 344         stdscan_bufptr += 2;
 345         return tv->t_type = TOKEN_DBL_OR;
 346     } else                      /* just an ordinary char */
 347         return tv->t_type = (uint8_t)(*stdscan_bufptr++);
 348 }