stdscan.c

   1 /* ----------------------------------------------------------------------- *
   2  *
   3  *   Copyright 1996-2016 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 #include <inttypes.h>
  41
  42 #include "nasm.h"
  43 #include "nasmlib.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_error(ERR_NONFATAL,
 117                     "`%s' is not a valid decorator with braces", tv->t_charptr);
 118         tv->t_type = TOKEN_INVALID;
 119     } else if (tv->t_flag & TFLAG_BRC_OPT) {
 120         if (is_reg_class(OPMASKREG, tv->t_integer)) {
 121             /* within braces, opmask register is now used as a mask */
 122             tv->t_type = TOKEN_OPMASK;
 123         }
 124     }
 125
 126     return tv->t_type;
 127 }
 128
 129 int stdscan(void *private_data, struct tokenval *tv)
 130 {
 131     char ourcopy[MAX_KEYWORD + 1], *r, *s;
 132
 133     (void)private_data;         /* Don't warn that this parameter is unused */
 134
 135     stdscan_bufptr = nasm_skip_spaces(stdscan_bufptr);
 136     if (!*stdscan_bufptr)
 137         return tv->t_type = TOKEN_EOS;
 138
 139     /* we have a token; either an id, a number or a char */
 140     if (isidstart(*stdscan_bufptr) ||
 141         (*stdscan_bufptr == '$' && isidstart(stdscan_bufptr[1]))) {
 142         /* now we've got an identifier */
 143         bool is_sym = false;
 144         int token_type;
 145
 146         if (*stdscan_bufptr == '$') {
 147             is_sym = true;
 148             stdscan_bufptr++;
 149         }
 150
 151         r = stdscan_bufptr++;
 152         /* read the entire buffer to advance the buffer pointer but... */
 153         while (isidchar(*stdscan_bufptr))
 154             stdscan_bufptr++;
 155
 156         /* ... copy only up to IDLEN_MAX-1 characters */
 157         tv->t_charptr = stdscan_copy(r, stdscan_bufptr - r < IDLEN_MAX ?
 158                                      stdscan_bufptr - r : IDLEN_MAX - 1);
 159
 160         if (is_sym || stdscan_bufptr - r > MAX_KEYWORD)
 161             return tv->t_type = TOKEN_ID;       /* bypass all other checks */
 162
 163         for (s = tv->t_charptr, r = ourcopy; *s; s++)
 164             *r++ = nasm_tolower(*s);
 165         *r = '\0';
 166         /* right, so we have an identifier sitting in temp storage. now,
 167          * is it actually a register or instruction name, or what? */
 168         token_type = nasm_token_hash(ourcopy, tv);
 169
 170         if (unlikely(tv->t_flag & TFLAG_WARN)) {
 171             nasm_error(ERR_WARNING|ERR_PASS1|ERR_WARN_PTR,
 172                        "`%s' is not a NASM keyword", tv->t_charptr);
 173         }
 174
 175         if (likely(!(tv->t_flag & TFLAG_BRC))) {
 176             /* most of the tokens fall into this case */
 177             return token_type;
 178         } else {
 179             return tv->t_type = TOKEN_ID;
 180         }
 181     } else if (*stdscan_bufptr == '$' && !isnumchar(stdscan_bufptr[1])) {
 182         /*
 183          * It's a $ sign with no following hex number; this must
 184          * mean it's a Here token ($), evaluating to the current
 185          * assembly location, or a Base token ($$), evaluating to
 186          * the base of the current segment.
 187          */
 188         stdscan_bufptr++;
 189         if (*stdscan_bufptr == '$') {
 190             stdscan_bufptr++;
 191             return tv->t_type = TOKEN_BASE;
 192         }
 193         return tv->t_type = TOKEN_HERE;
 194     } else if (isnumstart(*stdscan_bufptr)) {   /* now we've got a number */
 195         bool rn_error;
 196         bool is_hex = false;
 197         bool is_float = false;
 198         bool has_e = false;
 199         char c;
 200
 201         r = stdscan_bufptr;
 202
 203         if (*stdscan_bufptr == '$') {
 204             stdscan_bufptr++;
 205             is_hex = true;
 206         }
 207
 208         for (;;) {
 209             c = *stdscan_bufptr++;
 210
 211             if (!is_hex && (c == 'e' || c == 'E')) {
 212                 has_e = true;
 213                 if (*stdscan_bufptr == '+' || *stdscan_bufptr == '-') {
 214                     /*
 215                      * e can only be followed by +/- if it is either a
 216                      * prefixed hex number or a floating-point number
 217                      */
 218                     is_float = true;
 219                     stdscan_bufptr++;
 220                 }
 221             } else if (c == 'H' || c == 'h' || c == 'X' || c == 'x') {
 222                 is_hex = true;
 223             } else if (c == 'P' || c == 'p') {
 224                 is_float = true;
 225                 if (*stdscan_bufptr == '+' || *stdscan_bufptr == '-')
 226                     stdscan_bufptr++;
 227             } else if (isnumchar(c) || c == '_')
 228                 ; /* just advance */
 229             else if (c == '.')
 230                 is_float = true;
 231             else
 232                 break;
 233         }
 234         stdscan_bufptr--;       /* Point to first character beyond number */
 235
 236         if (has_e && !is_hex) {
 237             /* 1e13 is floating-point, but 1e13h is not */
 238             is_float = true;
 239         }
 240
 241         if (is_float) {
 242             tv->t_charptr = stdscan_copy(r, stdscan_bufptr - r);
 243             return tv->t_type = TOKEN_FLOAT;
 244         } else {
 245             r = stdscan_copy(r, stdscan_bufptr - r);
 246             tv->t_integer = readnum(r, &rn_error);
 247             stdscan_pop();
 248             if (rn_error) {
 249                 /* some malformation occurred */
 250                 return tv->t_type = TOKEN_ERRNUM;
 251             }
 252             tv->t_charptr = NULL;
 253             return tv->t_type = TOKEN_NUM;
 254         }
 255     } else if (*stdscan_bufptr == '\'' || *stdscan_bufptr == '"' ||
 256                *stdscan_bufptr == '`') {
 257         /* a quoted string */
 258         char start_quote = *stdscan_bufptr;
 259         tv->t_charptr = stdscan_bufptr;
 260         tv->t_inttwo = nasm_unquote(tv->t_charptr, &stdscan_bufptr);
 261         if (*stdscan_bufptr != start_quote)
 262             return tv->t_type = TOKEN_ERRSTR;
 263         stdscan_bufptr++;       /* Skip final quote */
 264         return tv->t_type = TOKEN_STR;
 265     } else if (*stdscan_bufptr == '{') {
 266         /* now we've got a decorator */
 267         int token_len;
 268
 269         stdscan_bufptr = nasm_skip_spaces(stdscan_bufptr);
 270
 271         r = ++stdscan_bufptr;
 272         /*
 273          * read the entire buffer to advance the buffer pointer
 274          * {rn-sae}, {rd-sae}, {ru-sae}, {rz-sae} contain '-' in tokens.
 275          */
 276         while (isbrcchar(*stdscan_bufptr))
 277             stdscan_bufptr++;
 278
 279         token_len = stdscan_bufptr - r;
 280
 281         /* ... copy only up to DECOLEN_MAX-1 characters */
 282         tv->t_charptr = stdscan_copy(r, token_len < DECOLEN_MAX ?
 283                                         token_len : DECOLEN_MAX - 1);
 284
 285         stdscan_bufptr = nasm_skip_spaces(stdscan_bufptr);
 286         /* if brace is not closed properly or token is too long  */
 287         if ((*stdscan_bufptr != '}') || (token_len > MAX_KEYWORD)) {
 288             nasm_error(ERR_NONFATAL,
 289                        "invalid decorator token inside braces");
 290             return tv->t_type = TOKEN_INVALID;
 291         }
 292
 293         stdscan_bufptr++;       /* skip closing brace */
 294
 295         for (s = tv->t_charptr, r = ourcopy; *s; s++)
 296             *r++ = nasm_tolower(*s);
 297         *r = '\0';
 298
 299         /* right, so we have a decorator sitting in temp storage. */
 300         nasm_token_hash(ourcopy, tv);
 301
 302         /* handle tokens inside braces */
 303         return stdscan_handle_brace(tv);
 304     } else if (*stdscan_bufptr == ';') {
 305         /* a comment has happened - stay */
 306         return tv->t_type = TOKEN_EOS;
 307     } else if (stdscan_bufptr[0] == '>' && stdscan_bufptr[1] == '>') {
 308         stdscan_bufptr += 2;
 309         return tv->t_type = TOKEN_SHR;
 310     } else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '<') {
 311         stdscan_bufptr += 2;
 312         return tv->t_type = TOKEN_SHL;
 313     } else if (stdscan_bufptr[0] == '/' && stdscan_bufptr[1] == '/') {
 314         stdscan_bufptr += 2;
 315         return tv->t_type = TOKEN_SDIV;
 316     } else if (stdscan_bufptr[0] == '%' && stdscan_bufptr[1] == '%') {
 317         stdscan_bufptr += 2;
 318         return tv->t_type = TOKEN_SMOD;
 319     } else if (stdscan_bufptr[0] == '=' && stdscan_bufptr[1] == '=') {
 320         stdscan_bufptr += 2;
 321         return tv->t_type = TOKEN_EQ;
 322     } else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '>') {
 323         stdscan_bufptr += 2;
 324         return tv->t_type = TOKEN_NE;
 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_LE;
 331     } else if (stdscan_bufptr[0] == '>' && stdscan_bufptr[1] == '=') {
 332         stdscan_bufptr += 2;
 333         return tv->t_type = TOKEN_GE;
 334     } else if (stdscan_bufptr[0] == '&' && stdscan_bufptr[1] == '&') {
 335         stdscan_bufptr += 2;
 336         return tv->t_type = TOKEN_DBL_AND;
 337     } else if (stdscan_bufptr[0] == '^' && stdscan_bufptr[1] == '^') {
 338         stdscan_bufptr += 2;
 339         return tv->t_type = TOKEN_DBL_XOR;
 340     } else if (stdscan_bufptr[0] == '|' && stdscan_bufptr[1] == '|') {
 341         stdscan_bufptr += 2;
 342         return tv->t_type = TOKEN_DBL_OR;
 343     } else                      /* just an ordinary char */
 344         return tv->t_type = (uint8_t)(*stdscan_bufptr++);
 345 }