dlls/jscript/lex.c

   1 /*
   2  * Copyright 2008 Jacek Caban for CodeWeavers
   3  *
   4  * This library is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU Lesser General Public
   6  * License as published by the Free Software Foundation; either
   7  * version 2.1 of the License, or (at your option) any later version.
   8  *
   9  * This library is distributed in the hope that it will be useful,
  10  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12  * Lesser General Public License for more details.
  13  *
  14  * You should have received a copy of the GNU Lesser General Public
  15  * License along with this library; if not, write to the Free Software
  16  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
  17  */
  18
  19 #include <math.h>
  20
  21 #include "jscript.h"
  22 #include "activscp.h"
  23 #include "objsafe.h"
  24 #include "engine.h"
  25
  26 #include "parser.tab.h"
  27
  28 #include "wine/debug.h"
  29 #include "wine/unicode.h"
  30
  31 WINE_DEFAULT_DEBUG_CHANNEL(jscript);
  32
  33 static const WCHAR breakW[] = {'b','r','e','a','k',0};
  34 static const WCHAR caseW[] = {'c','a','s','e',0};
  35 static const WCHAR catchW[] = {'c','a','t','c','h',0};
  36 static const WCHAR continueW[] = {'c','o','n','t','i','n','u','e',0};
  37 static const WCHAR defaultW[] = {'d','e','f','a','u','l','t',0};
  38 static const WCHAR deleteW[] = {'d','e','l','e','t','e',0};
  39 static const WCHAR doW[] = {'d','o',0};
  40 static const WCHAR elseW[] = {'e','l','s','e',0};
  41 static const WCHAR falseW[] = {'f','a','l','s','e',0};
  42 static const WCHAR finallyW[] = {'f','i','n','a','l','l','y',0};
  43 static const WCHAR forW[] = {'f','o','r',0};
  44 static const WCHAR functionW[] = {'f','u','n','c','t','i','o','n',0};
  45 static const WCHAR ifW[] = {'i','f',0};
  46 static const WCHAR inW[] = {'i','n',0};
  47 static const WCHAR instanceofW[] = {'i','n','s','t','a','n','c','e','o','f',0};
  48 static const WCHAR newW[] = {'n','e','w',0};
  49 static const WCHAR nullW[] = {'n','u','l','l',0};
  50 static const WCHAR returnW[] = {'r','e','t','u','r','n',0};
  51 static const WCHAR switchW[] = {'s','w','i','t','c','h',0};
  52 static const WCHAR thisW[] = {'t','h','i','s',0};
  53 static const WCHAR throwW[] = {'t','h','r','o','w',0};
  54 static const WCHAR trueW[] = {'t','r','u','e',0};
  55 static const WCHAR tryW[] = {'t','r','y',0};
  56 static const WCHAR typeofW[] = {'t','y','p','e','o','f',0};
  57 static const WCHAR undefinedW[] = {'u','n','d','e','f','i','n','e','d',0};
  58 static const WCHAR varW[] = {'v','a','r',0};
  59 static const WCHAR voidW[] = {'v','o','i','d',0};
  60 static const WCHAR whileW[] = {'w','h','i','l','e',0};
  61 static const WCHAR withW[] = {'w','i','t','h',0};
  62
  63 static const struct {
  64     const WCHAR *word;
  65     int token;
  66 } keywords[] = {
  67     {breakW,       kBREAK},
  68     {caseW,        kCASE},
  69     {catchW,       kCATCH},
  70     {continueW,    kCONTINUE},
  71     {defaultW,     kDEFAULT},
  72     {deleteW,      kDELETE},
  73     {doW,          kDO},
  74     {elseW,        kELSE},
  75     {falseW,       kFALSE},
  76     {finallyW,     kFINALLY},
  77     {forW,         kFOR},
  78     {functionW,    kFUNCTION},
  79     {ifW,          kIF},
  80     {inW,          kIN},
  81     {instanceofW,  kINSTANCEOF},
  82     {newW,         kNEW},
  83     {nullW,        kNULL},
  84     {returnW,      kRETURN},
  85     {switchW,      kSWITCH},
  86     {thisW,        kTHIS},
  87     {throwW,       kTHROW},
  88     {trueW,        kTRUE},
  89     {tryW,         kTRY},
  90     {typeofW,      kTYPEOF},
  91     {undefinedW,   kUNDEFINED},
  92     {varW,         kVAR},
  93     {voidW,        kVOID},
  94     {whileW,       kWHILE},
  95     {withW,        kWITH}
  96 };
  97
  98 static int lex_error(parser_ctx_t *ctx, HRESULT hres)
  99 {
 100     ctx->hres = hres;
 101     return -1;
 102 }
 103
 104 static int check_keyword(parser_ctx_t *ctx, const WCHAR *word)
 105 {
 106     const WCHAR *p1 = ctx->ptr;
 107     const WCHAR *p2 = word;
 108
 109     while(p1 < ctx->end && *p2) {
 110         if(*p1 != *p2)
 111             return *p1 - *p2;
 112         p1++;
 113         p2++;
 114     }
 115
 116     if(*p2 || (p1 < ctx->end && isalnumW(*p1)))
 117         return 1;
 118
 119     ctx->ptr = p1;
 120     return 0;
 121 }
 122
 123 /* ECMA-262 3rd Edition    7.3 */
 124 static BOOL is_endline(WCHAR c)
 125 {
 126     return c == '\n' || c == '\r' || c == 0x2028 || c == 0x2029;
 127 }
 128
 129 static BOOL is_identifier_char(WCHAR c)
 130 {
 131     return isalnumW(c) || c == '$' || c == '_' || c == '\\';
 132 }
 133
 134 static int hex_to_int(WCHAR c)
 135 {
 136     if('0' <= c && c <= '9')
 137         return c-'0';
 138
 139     if('a' <= c && c <= 'f')
 140         return c-'a'+10;
 141
 142     if('A' <= c && c <= 'F')
 143         return c-'A'+10;
 144
 145     return -1;
 146 }
 147
 148 static int check_keywords(parser_ctx_t *ctx)
 149 {
 150     int min = 0, max = sizeof(keywords)/sizeof(keywords[0])-1, r, i;
 151
 152     while(min <= max) {
 153         i = (min+max)/2;
 154
 155         r = check_keyword(ctx, keywords[i].word);
 156         if(!r)
 157             return keywords[i].token;
 158
 159         if(r > 0)
 160             min = i+1;
 161         else
 162             max = i-1;
 163     }
 164
 165     return 0;
 166 }
 167
 168 static void skip_spaces(parser_ctx_t *ctx)
 169 {
 170     while(ctx->ptr < ctx->end && isspaceW(*ctx->ptr)) {
 171         if(is_endline(*ctx->ptr++))
 172             ctx->nl = TRUE;
 173     }
 174 }
 175
 176 static BOOL skip_comment(parser_ctx_t *ctx)
 177 {
 178     if(ctx->ptr+1 >= ctx->end || *ctx->ptr != '/')
 179         return FALSE;
 180
 181     switch(ctx->ptr[1]) {
 182     case '*':
 183         ctx->ptr += 2;
 184         while(ctx->ptr+1 < ctx->end && (ctx->ptr[0] != '*' || ctx->ptr[1] != '/'))
 185             ctx->ptr++;
 186
 187         if(ctx->ptr[0] == '*' && ctx->ptr[1] == '/') {
 188             ctx->ptr += 2;
 189         }else {
 190             WARN("unexpected end of file (missing end of comment)\n");
 191             ctx->ptr = ctx->end;
 192         }
 193         break;
 194     case '/':
 195         ctx->ptr += 2;
 196         while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr))
 197             ctx->ptr++;
 198         break;
 199     default:
 200         return FALSE;
 201     }
 202
 203     return TRUE;
 204 }
 205
 206 static BOOL unescape(WCHAR *str)
 207 {
 208     WCHAR *pd, *p, c;
 209     int i;
 210
 211     pd = p = str;
 212     while(*p) {
 213         if(*p != '\\') {
 214             *pd++ = *p++;
 215             continue;
 216         }
 217
 218         p++;
 219         c = 0;
 220
 221         switch(*p) {
 222         case '\'':
 223         case '\"':
 224         case '\\':
 225             c = *p;
 226             break;
 227         case 'b':
 228             c = '\b';
 229             break;
 230         case 't':
 231             c = '\t';
 232             break;
 233         case 'n':
 234             c = '\n';
 235             break;
 236         case 'v':
 237             c = '\v';
 238             break;
 239         case 'f':
 240             c = '\f';
 241             break;
 242         case 'r':
 243             c = '\r';
 244             break;
 245         case '0':
 246             break;
 247         case 'x':
 248             i = hex_to_int(*++p);
 249             if(i == -1)
 250                 return FALSE;
 251             c = i << 16;
 252
 253             i = hex_to_int(*++p);
 254             if(i == -1)
 255                 return FALSE;
 256             c += i;
 257             break;
 258         case 'u':
 259             i = hex_to_int(*++p);
 260             if(i == -1)
 261                 return FALSE;
 262             c = i << 24;
 263
 264             i = hex_to_int(*++p);
 265             if(i == -1)
 266                 return FALSE;
 267             c += i << 16;
 268
 269             i = hex_to_int(*++p);
 270             if(i == -1)
 271                 return FALSE;
 272             c += 1 << 8;
 273
 274             i = hex_to_int(*++p);
 275             if(i == -1)
 276                 return FALSE;
 277             c += i;
 278             break;
 279         default:
 280             c = *p;
 281         }
 282
 283         *pd++ = c;
 284         p++;
 285     }
 286
 287     *pd = 0;
 288     return TRUE;
 289 }
 290
 291 static int parse_identifier(parser_ctx_t *ctx, const WCHAR **ret)
 292 {
 293     const WCHAR *ptr = ctx->ptr++;
 294     WCHAR *wstr;
 295     int len;
 296
 297     while(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr))
 298         ctx->ptr++;
 299
 300     len = ctx->ptr-ptr;
 301
 302     *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
 303     memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
 304     wstr[len] = 0;
 305
 306     /* FIXME: unescape */
 307     return tIdentifier;
 308 }
 309
 310 static int parse_string_literal(parser_ctx_t *ctx, const WCHAR **ret, WCHAR endch)
 311 {
 312     const WCHAR *ptr = ++ctx->ptr;
 313     WCHAR *wstr;
 314     int len;
 315
 316     while(ctx->ptr < ctx->end && *ctx->ptr != endch) {
 317         if(*ctx->ptr++ == '\\')
 318             ctx->ptr++;
 319     }
 320
 321     if(ctx->ptr == ctx->end) {
 322         WARN("unexpected end of file\n");
 323         return lex_error(ctx, E_FAIL);
 324     }
 325
 326     len = ctx->ptr-ptr;
 327
 328     *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
 329     memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
 330     wstr[len] = 0;
 331
 332     ctx->ptr++;
 333
 334     if(!unescape(wstr)) {
 335         WARN("unescape failed\n");
 336         return lex_error(ctx, E_FAIL);
 337     }
 338
 339     return tStringLiteral;
 340 }
 341
 342 static literal_t *alloc_int_literal(parser_ctx_t *ctx, LONG l)
 343 {
 344     literal_t *ret = parser_alloc(ctx, sizeof(literal_t));
 345
 346     ret->vt = VT_I4;
 347     ret->u.lval = l;
 348
 349     return ret;
 350 }
 351
 352 static int parse_double_literal(parser_ctx_t *ctx, LONG int_part, literal_t **literal)
 353 {
 354     double d, tmp = 1.0;
 355
 356     if(ctx->ptr == ctx->end || !isdigitW(*ctx->ptr)) {
 357         ERR("No digit after point\n");
 358         return 0;
 359     }
 360
 361     d = int_part;
 362     while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
 363         d += (tmp /= 10.0)*(*ctx->ptr++ - '0');
 364
 365     if(ctx->ptr < ctx->end && (*ctx->ptr == 'e' || *ctx->ptr == 'E')) {
 366         int sign = 1, e = 0;
 367
 368         ctx->ptr++;
 369         if(ctx->ptr < ctx->end) {
 370             if(*ctx->ptr == '+') {
 371                 ctx->ptr++;
 372             }else if(*ctx->ptr == '-') {
 373                 sign = -1;
 374                 ctx->ptr++;
 375             }else if(!isdigitW(*ctx->ptr)) {
 376                 WARN("Expected exponent part\n");
 377                 return lex_error(ctx, E_FAIL);
 378             }
 379         }
 380
 381         if(ctx->ptr == ctx->end) {
 382             WARN("unexpected end of file\n");
 383             return lex_error(ctx, E_FAIL);
 384         }
 385
 386         while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
 387             e = e*10 + *ctx->ptr++ - '0';
 388         e *= sign;
 389
 390         d = pow(d, e);
 391     }
 392
 393     *literal = parser_alloc(ctx, sizeof(literal_t));
 394     (*literal)->vt = VT_R8;
 395     (*literal)->u.dval = d;
 396
 397     return tNumericLiteral;
 398 }
 399
 400 static int parse_numeric_literal(parser_ctx_t *ctx, literal_t **literal)
 401 {
 402     LONG l, d;
 403
 404     l = *ctx->ptr++ - '0';
 405     if(ctx->ptr == ctx->end) {
 406         *literal = alloc_int_literal(ctx, l);
 407         return tNumericLiteral;
 408     }
 409
 410     if(!l) {
 411         if(*ctx->ptr == 'x' || *ctx->ptr == 'X') {
 412             if(++ctx->ptr == ctx->end) {
 413                 ERR("unexpexted end of file\n");
 414                 return 0;
 415             }
 416
 417             while(ctx->ptr < ctx->end && (d = hex_to_int(*ctx->ptr)) != -1) {
 418                 l = l*16 + d;
 419                 ctx->ptr++;
 420             }
 421
 422             if(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr)) {
 423                 WARN("unexpected identifier char\n");
 424                 return lex_error(ctx, E_FAIL);
 425             }
 426
 427             *literal = alloc_int_literal(ctx, l);
 428             return tNumericLiteral;
 429         }
 430
 431         if(isdigitW(*ctx->ptr) || is_identifier_char(*ctx->ptr)) {
 432             WARN("wrong char after zero\n");
 433             return lex_error(ctx, E_FAIL);
 434         }
 435
 436         *literal = alloc_int_literal(ctx, 0);
 437     }
 438
 439     while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
 440         l = l*10 + *(ctx->ptr++)-'0';
 441
 442     if(ctx->ptr < ctx->end) {
 443         if(*ctx->ptr == '.') {
 444             ctx->ptr++;
 445             return parse_double_literal(ctx, l, literal);
 446         }
 447
 448         if(is_identifier_char(*ctx->ptr)) {
 449             WARN("unexpected identifier char\n");
 450             return lex_error(ctx, E_FAIL);
 451         }
 452     }
 453
 454     *literal = alloc_int_literal(ctx, l);
 455     return tNumericLiteral;
 456 }
 457
 458 int parser_lex(void *lval, parser_ctx_t *ctx)
 459 {
 460     int ret;
 461
 462     ctx->nl = FALSE;
 463
 464     do {
 465         skip_spaces(ctx);
 466         if(ctx->ptr == ctx->end)
 467             return 0;
 468     }while(skip_comment(ctx));
 469
 470     if(isalphaW(*ctx->ptr)) {
 471         ret = check_keywords(ctx);
 472         if(ret)
 473             return ret;
 474
 475         return parse_identifier(ctx, (const WCHAR**)lval);
 476     }
 477
 478     if(isdigitW(*ctx->ptr))
 479         return parse_numeric_literal(ctx, lval);
 480
 481     switch(*ctx->ptr) {
 482     case '{':
 483     case '}':
 484     case '(':
 485     case ')':
 486     case '[':
 487     case ']':
 488     case ';':
 489     case ',':
 490     case '~':
 491     case '?':
 492     case ':':
 493         return *ctx->ptr++;
 494
 495     case '.':
 496         if(++ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
 497             return parse_double_literal(ctx, 0, lval);
 498         return '.';
 499
 500     case '<':
 501         if(++ctx->ptr == ctx->end) {
 502             *(int*)lval = EXPR_LESS;
 503             return tRelOper;
 504         }
 505
 506         switch(*ctx->ptr) {
 507         case '=':  /* <= */
 508             ctx->ptr++;
 509             *(int*)lval = EXPR_LESSEQ;
 510             return tRelOper;
 511         case '<':  /* << */
 512             if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* <<= */
 513                 ctx->ptr++;
 514                 *(int*)lval = EXPR_ASSIGNLSHIFT;
 515                 return tAssignOper;
 516             }
 517             *(int*)lval = EXPR_LSHIFT;
 518             return tShiftOper;
 519         default: /* < */
 520             *(int*)lval = EXPR_LESS;
 521             return tRelOper;
 522         }
 523
 524     case '>':
 525         if(++ctx->ptr == ctx->end) { /* > */
 526             *(int*)lval = EXPR_GREATER;
 527             return tRelOper;
 528         }
 529
 530         switch(*ctx->ptr) {
 531         case '=':  /* >= */
 532             ctx->ptr++;
 533             *(int*)lval = EXPR_GREATEREQ;
 534             return tRelOper;
 535         case '>':  /* >> */
 536             if(++ctx->ptr < ctx->end) {
 537                 if(*ctx->ptr == '=') {  /* >>= */
 538                     ctx->ptr++;
 539                     *(int*)lval = EXPR_ASSIGNRSHIFT;
 540                     return tAssignOper;
 541                 }
 542                 if(*ctx->ptr == '>') {  /* >>> */
 543                     if(++ctx->ptr < ctx->end && *ctx->ptr == '=') {  /* >>>= */
 544                         ctx->ptr++;
 545                         *(int*)lval = EXPR_ASSIGNRRSHIFT;
 546                         return tAssignOper;
 547                     }
 548                     *(int*)lval = EXPR_RRSHIFT;
 549                     return tRelOper;
 550                 }
 551             }
 552             *(int*)lval = EXPR_RSHIFT;
 553             return tShiftOper;
 554         default:
 555             *(int*)lval = EXPR_GREATER;
 556             return tRelOper;
 557         }
 558
 559     case '+':
 560         ctx->ptr++;
 561         if(ctx->ptr < ctx->end) {
 562             switch(*ctx->ptr) {
 563             case '+':  /* ++ */
 564                 ctx->ptr++;
 565                 return tINC;
 566             case '=':  /* += */
 567                 ctx->ptr++;
 568                 *(int*)lval = EXPR_ASSIGNADD;
 569                 return tAssignOper;
 570             }
 571         }
 572         return '+';
 573
 574     case '-':
 575         ctx->ptr++;
 576         if(ctx->ptr < ctx->end) {
 577             switch(*ctx->ptr) {
 578             case '-':  /* -- */
 579                 ctx->ptr++;
 580                 return tDEC;
 581             case '=':  /* -= */
 582                 ctx->ptr++;
 583                 *(int*)lval = EXPR_ASSIGNSUB;
 584                 return tAssignOper;
 585             }
 586         }
 587         return '-';
 588
 589     case '*':
 590         if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* *= */
 591             ctx->ptr++;
 592             *(int*)lval = EXPR_ASSIGNMUL;
 593             return tAssignOper;
 594         }
 595         return '*';
 596
 597     case '%':
 598         if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* %= */
 599             ctx->ptr++;
 600             *(int*)lval = EXPR_ASSIGNMOD;
 601             return tAssignOper;
 602         }
 603         return '%';
 604
 605     case '&':
 606         if(++ctx->ptr < ctx->end) {
 607             switch(*ctx->ptr) {
 608             case '=':  /* &= */
 609                 ctx->ptr++;
 610                 *(int*)lval = EXPR_ASSIGNAND;
 611                 return tAssignOper;
 612             case '&':  /* && */
 613                 ctx->ptr++;
 614                 return tANDAND;
 615             }
 616         }
 617         return '&';
 618
 619     case '|':
 620         if(++ctx->ptr < ctx->end) {
 621             switch(*ctx->ptr) {
 622             case '=':  /* |= */
 623                 ctx->ptr++;
 624                 *(int*)lval = EXPR_ASSIGNOR;
 625                 return tAssignOper;
 626             case '|':  /* || */
 627                 ctx->ptr++;
 628                 return tOROR;
 629             }
 630         }
 631         return '|';
 632
 633     case '^':
 634         if(++ctx->ptr < ctx->end && *ctx->ptr == '=') {  /* ^= */
 635             ctx->ptr++;
 636             *(int*)lval = EXPR_ASSIGNXOR;
 637             return tAssignOper;
 638         }
 639         return '^';
 640
 641     case '!':
 642         if(++ctx->ptr < ctx->end && *ctx->ptr == '=') {  /* != */
 643             if(++ctx->ptr < ctx->end && *ctx->ptr == '=') {  /* !== */
 644                 ctx->ptr++;
 645                 *(int*)lval = EXPR_NOTEQEQ;
 646                 return tEqOper;
 647             }
 648             *(int*)lval = EXPR_NOTEQ;
 649             return tEqOper;
 650         }
 651         return '!';
 652
 653     case '=':
 654         if(++ctx->ptr < ctx->end && *ctx->ptr == '=') {  /* == */
 655             if(++ctx->ptr < ctx->end && *ctx->ptr == '=') {  /* === */
 656                 ctx->ptr++;
 657                 *(int*)lval = EXPR_EQEQ;
 658                 return tEqOper;
 659             }
 660             *(int*)lval = EXPR_EQ;
 661             return tEqOper;
 662         }
 663         return '=';
 664
 665     case '/':
 666         if(++ctx->ptr < ctx->end) {
 667             if(*ctx->ptr == '=') {  /* /= */
 668                 ctx->ptr++;
 669                 *(int*)lval = EXPR_ASSIGNMUL;
 670                 return tAssignOper;
 671             }
 672         }
 673         return '/';
 674
 675     case '\"':
 676     case '\'':
 677         return parse_string_literal(ctx, (const WCHAR**)lval, *ctx->ptr);
 678
 679     case '_':
 680     case '$':
 681         return parse_identifier(ctx, lval);
 682     }
 683
 684     WARN("unexpected char '%c' %d\n", *ctx->ptr, *ctx->ptr);
 685     return 0;
 686 }