shlwapi/tests: Avoid size_t in traces.
[wine/multimedia.git] / dlls / jscript / lex.c
blob4e04dea041687fbfe8668bf2d63703b36441027f
1 /*
2 * Copyright 2008 Jacek Caban for CodeWeavers
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.
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.
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
19 #include <math.h>
20 #include <limits.h>
22 #include "jscript.h"
23 #include "activscp.h"
24 #include "objsafe.h"
25 #include "engine.h"
27 #include "parser.tab.h"
29 #include "wine/debug.h"
30 #include "wine/unicode.h"
32 WINE_DEFAULT_DEBUG_CHANNEL(jscript);
34 #define LONGLONG_MAX (((LONGLONG)0x7fffffff<<32)|0xffffffff)
36 static const WCHAR breakW[] = {'b','r','e','a','k',0};
37 static const WCHAR caseW[] = {'c','a','s','e',0};
38 static const WCHAR catchW[] = {'c','a','t','c','h',0};
39 static const WCHAR continueW[] = {'c','o','n','t','i','n','u','e',0};
40 static const WCHAR defaultW[] = {'d','e','f','a','u','l','t',0};
41 static const WCHAR deleteW[] = {'d','e','l','e','t','e',0};
42 static const WCHAR doW[] = {'d','o',0};
43 static const WCHAR elseW[] = {'e','l','s','e',0};
44 static const WCHAR falseW[] = {'f','a','l','s','e',0};
45 static const WCHAR finallyW[] = {'f','i','n','a','l','l','y',0};
46 static const WCHAR forW[] = {'f','o','r',0};
47 static const WCHAR functionW[] = {'f','u','n','c','t','i','o','n',0};
48 static const WCHAR ifW[] = {'i','f',0};
49 static const WCHAR inW[] = {'i','n',0};
50 static const WCHAR instanceofW[] = {'i','n','s','t','a','n','c','e','o','f',0};
51 static const WCHAR newW[] = {'n','e','w',0};
52 static const WCHAR nullW[] = {'n','u','l','l',0};
53 static const WCHAR returnW[] = {'r','e','t','u','r','n',0};
54 static const WCHAR switchW[] = {'s','w','i','t','c','h',0};
55 static const WCHAR thisW[] = {'t','h','i','s',0};
56 static const WCHAR throwW[] = {'t','h','r','o','w',0};
57 static const WCHAR trueW[] = {'t','r','u','e',0};
58 static const WCHAR tryW[] = {'t','r','y',0};
59 static const WCHAR typeofW[] = {'t','y','p','e','o','f',0};
60 static const WCHAR undefinedW[] = {'u','n','d','e','f','i','n','e','d',0};
61 static const WCHAR varW[] = {'v','a','r',0};
62 static const WCHAR voidW[] = {'v','o','i','d',0};
63 static const WCHAR whileW[] = {'w','h','i','l','e',0};
64 static const WCHAR withW[] = {'w','i','t','h',0};
66 static const struct {
67 const WCHAR *word;
68 int token;
69 } keywords[] = {
70 {breakW, kBREAK},
71 {caseW, kCASE},
72 {catchW, kCATCH},
73 {continueW, kCONTINUE},
74 {defaultW, kDEFAULT},
75 {deleteW, kDELETE},
76 {doW, kDO},
77 {elseW, kELSE},
78 {falseW, kFALSE},
79 {finallyW, kFINALLY},
80 {forW, kFOR},
81 {functionW, kFUNCTION},
82 {ifW, kIF},
83 {inW, kIN},
84 {instanceofW, kINSTANCEOF},
85 {newW, kNEW},
86 {nullW, kNULL},
87 {returnW, kRETURN},
88 {switchW, kSWITCH},
89 {thisW, kTHIS},
90 {throwW, kTHROW},
91 {trueW, kTRUE},
92 {tryW, kTRY},
93 {typeofW, kTYPEOF},
94 {undefinedW, kUNDEFINED},
95 {varW, kVAR},
96 {voidW, kVOID},
97 {whileW, kWHILE},
98 {withW, kWITH}
101 static int lex_error(parser_ctx_t *ctx, HRESULT hres)
103 ctx->hres = JSCRIPT_ERROR|hres;
104 ctx->lexer_error = TRUE;
105 return -1;
108 /* ECMA-262 3rd Edition 7.6 */
109 static BOOL is_identifier_char(WCHAR c)
111 return isalnumW(c) || c == '$' || c == '_' || c == '\\';
114 static int check_keyword(parser_ctx_t *ctx, const WCHAR *word, const WCHAR **lval)
116 const WCHAR *p1 = ctx->ptr;
117 const WCHAR *p2 = word;
119 while(p1 < ctx->end && *p2) {
120 if(*p1 != *p2)
121 return *p1 - *p2;
122 p1++;
123 p2++;
126 if(*p2 || (p1 < ctx->end && is_identifier_char(*p1)))
127 return 1;
129 *lval = ctx->ptr;
130 ctx->ptr = p1;
131 return 0;
134 /* ECMA-262 3rd Edition 7.3 */
135 static BOOL is_endline(WCHAR c)
137 return c == '\n' || c == '\r' || c == 0x2028 || c == 0x2029;
140 static int hex_to_int(WCHAR c)
142 if('0' <= c && c <= '9')
143 return c-'0';
145 if('a' <= c && c <= 'f')
146 return c-'a'+10;
148 if('A' <= c && c <= 'F')
149 return c-'A'+10;
151 return -1;
154 static int check_keywords(parser_ctx_t *ctx, const WCHAR **lval)
156 int min = 0, max = sizeof(keywords)/sizeof(keywords[0])-1, r, i;
158 while(min <= max) {
159 i = (min+max)/2;
161 r = check_keyword(ctx, keywords[i].word, lval);
162 if(!r)
163 return keywords[i].token;
165 if(r > 0)
166 min = i+1;
167 else
168 max = i-1;
171 return 0;
174 static void skip_spaces(parser_ctx_t *ctx)
176 while(ctx->ptr < ctx->end && isspaceW(*ctx->ptr)) {
177 if(is_endline(*ctx->ptr++))
178 ctx->nl = TRUE;
182 static BOOL skip_html_comment(parser_ctx_t *ctx)
184 const WCHAR html_commentW[] = {'<','!','-','-',0};
186 if(!ctx->is_html || ctx->ptr+3 >= ctx->end ||
187 memcmp(ctx->ptr, html_commentW, sizeof(WCHAR)*4))
188 return FALSE;
190 ctx->nl = TRUE;
191 while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr++));
193 return TRUE;
196 static BOOL skip_comment(parser_ctx_t *ctx)
198 if(ctx->ptr+1 >= ctx->end || *ctx->ptr != '/')
199 return FALSE;
201 switch(ctx->ptr[1]) {
202 case '*':
203 ctx->ptr += 2;
204 while(ctx->ptr+1 < ctx->end && (ctx->ptr[0] != '*' || ctx->ptr[1] != '/'))
205 ctx->ptr++;
207 if(ctx->ptr[0] == '*' && ctx->ptr[1] == '/') {
208 ctx->ptr += 2;
209 }else {
210 WARN("unexpected end of file (missing end of comment)\n");
211 ctx->ptr = ctx->end;
213 break;
214 case '/':
215 ctx->ptr += 2;
216 while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr))
217 ctx->ptr++;
218 break;
219 default:
220 return FALSE;
223 return TRUE;
226 static BOOL unescape(WCHAR *str)
228 WCHAR *pd, *p, c;
229 int i;
231 pd = p = str;
232 while(*p) {
233 if(*p != '\\') {
234 *pd++ = *p++;
235 continue;
238 p++;
239 c = 0;
241 switch(*p) {
242 case '\'':
243 case '\"':
244 case '\\':
245 c = *p;
246 break;
247 case 'b':
248 c = '\b';
249 break;
250 case 't':
251 c = '\t';
252 break;
253 case 'n':
254 c = '\n';
255 break;
256 case 'v':
257 c = '\v';
258 break;
259 case 'f':
260 c = '\f';
261 break;
262 case 'r':
263 c = '\r';
264 break;
265 case 'x':
266 i = hex_to_int(*++p);
267 if(i == -1)
268 return FALSE;
269 c = i << 4;
271 i = hex_to_int(*++p);
272 if(i == -1)
273 return FALSE;
274 c += i;
275 break;
276 case 'u':
277 i = hex_to_int(*++p);
278 if(i == -1)
279 return FALSE;
280 c = i << 12;
282 i = hex_to_int(*++p);
283 if(i == -1)
284 return FALSE;
285 c += i << 8;
287 i = hex_to_int(*++p);
288 if(i == -1)
289 return FALSE;
290 c += i << 4;
292 i = hex_to_int(*++p);
293 if(i == -1)
294 return FALSE;
295 c += i;
296 break;
297 default:
298 if(isdigitW(*p)) {
299 c = *p++ - '0';
300 if(isdigitW(*p)) {
301 c = c*8 + (*p++ - '0');
302 if(isdigitW(*p))
303 c = c*8 + (*p++ - '0');
305 p--;
307 else
308 c = *p;
311 *pd++ = c;
312 p++;
315 *pd = 0;
316 return TRUE;
319 static int parse_identifier(parser_ctx_t *ctx, const WCHAR **ret)
321 const WCHAR *ptr = ctx->ptr++;
322 WCHAR *wstr;
323 int len;
325 while(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr))
326 ctx->ptr++;
328 len = ctx->ptr-ptr;
330 *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
331 memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
332 wstr[len] = 0;
334 /* FIXME: unescape */
335 return tIdentifier;
338 static int parse_string_literal(parser_ctx_t *ctx, const WCHAR **ret, WCHAR endch)
340 const WCHAR *ptr = ++ctx->ptr;
341 WCHAR *wstr;
342 int len;
344 while(ctx->ptr < ctx->end && *ctx->ptr != endch) {
345 if(*ctx->ptr++ == '\\')
346 ctx->ptr++;
349 if(ctx->ptr == ctx->end)
350 return lex_error(ctx, IDS_UNTERMINATED_STR);
352 len = ctx->ptr-ptr;
354 *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
355 memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
356 wstr[len] = 0;
358 ctx->ptr++;
360 if(!unescape(wstr)) {
361 WARN("unescape failed\n");
362 return lex_error(ctx, E_FAIL);
365 return tStringLiteral;
368 static literal_t *alloc_int_literal(parser_ctx_t *ctx, LONG l)
370 literal_t *ret = parser_alloc(ctx, sizeof(literal_t));
372 ret->type = LT_INT;
373 ret->u.lval = l;
375 return ret;
378 static int parse_double_literal(parser_ctx_t *ctx, LONG int_part, literal_t **literal)
380 LONGLONG d, hlp;
381 int exp = 0;
383 if(ctx->ptr == ctx->end || (!isdigitW(*ctx->ptr) &&
384 *ctx->ptr!='.' && *ctx->ptr!='e' && *ctx->ptr!='E')) {
385 ERR("Illegal character\n");
386 return 0;
389 d = int_part;
390 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
391 hlp = d*10 + *(ctx->ptr++) - '0';
392 if(d>LONGLONG_MAX/10 || hlp<0) {
393 exp++;
394 break;
396 else
397 d = hlp;
399 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
400 exp++;
401 ctx->ptr++;
404 if(*ctx->ptr == '.') ctx->ptr++;
406 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
407 hlp = d*10 + *(ctx->ptr++) - '0';
408 if(d>LONGLONG_MAX/10 || hlp<0)
409 break;
411 d = hlp;
412 exp--;
414 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
415 ctx->ptr++;
417 if(ctx->ptr < ctx->end && (*ctx->ptr == 'e' || *ctx->ptr == 'E')) {
418 int sign = 1, e = 0;
420 ctx->ptr++;
421 if(ctx->ptr < ctx->end) {
422 if(*ctx->ptr == '+') {
423 ctx->ptr++;
424 }else if(*ctx->ptr == '-') {
425 sign = -1;
426 ctx->ptr++;
427 }else if(!isdigitW(*ctx->ptr)) {
428 WARN("Expected exponent part\n");
429 return lex_error(ctx, E_FAIL);
433 if(ctx->ptr == ctx->end) {
434 WARN("unexpected end of file\n");
435 return lex_error(ctx, E_FAIL);
438 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
439 if(e > INT_MAX/10 || (e = e*10 + *ctx->ptr++ - '0')<0)
440 e = INT_MAX;
442 e *= sign;
444 if(exp<0 && e<0 && e+exp>0) exp = INT_MIN;
445 else if(exp>0 && e>0 && e+exp<0) exp = INT_MAX;
446 else exp += e;
449 *literal = parser_alloc(ctx, sizeof(literal_t));
450 (*literal)->type = LT_DOUBLE;
451 (*literal)->u.dval = (double)d*pow(10, exp);
453 return tNumericLiteral;
456 static int parse_numeric_literal(parser_ctx_t *ctx, literal_t **literal)
458 LONG l, d;
460 l = *ctx->ptr++ - '0';
461 if(ctx->ptr == ctx->end) {
462 *literal = alloc_int_literal(ctx, l);
463 return tNumericLiteral;
466 if(!l) {
467 if(*ctx->ptr == 'x' || *ctx->ptr == 'X') {
468 if(++ctx->ptr == ctx->end) {
469 ERR("unexpexted end of file\n");
470 return 0;
473 while(ctx->ptr < ctx->end && (d = hex_to_int(*ctx->ptr)) != -1) {
474 l = l*16 + d;
475 ctx->ptr++;
478 if(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr)) {
479 WARN("unexpected identifier char\n");
480 return lex_error(ctx, E_FAIL);
483 *literal = alloc_int_literal(ctx, l);
484 return tNumericLiteral;
487 if(isdigitW(*ctx->ptr) || is_identifier_char(*ctx->ptr)) {
488 WARN("wrong char after zero\n");
489 return lex_error(ctx, E_FAIL);
492 *literal = alloc_int_literal(ctx, 0);
495 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
497 d = l*10 + *(ctx->ptr)-'0';
499 /* Check for integer overflow */
500 if (l > INT_MAX/10 || d < 0)
501 return parse_double_literal(ctx, l, literal);
503 l = d;
504 ctx->ptr++;
507 if(ctx->ptr < ctx->end) {
508 if(*ctx->ptr == '.' || *ctx->ptr == 'e' || *ctx->ptr == 'E')
509 return parse_double_literal(ctx, l, literal);
511 if(is_identifier_char(*ctx->ptr)) {
512 WARN("unexpected identifier char\n");
513 return lex_error(ctx, E_FAIL);
517 *literal = alloc_int_literal(ctx, l);
518 return tNumericLiteral;
521 int parser_lex(void *lval, parser_ctx_t *ctx)
523 int ret;
525 ctx->nl = ctx->ptr == ctx->begin;
527 do {
528 skip_spaces(ctx);
529 if(ctx->ptr == ctx->end)
530 return 0;
531 }while(skip_comment(ctx) || skip_html_comment(ctx));
533 if(isalphaW(*ctx->ptr)) {
534 ret = check_keywords(ctx, lval);
535 if(ret)
536 return ret;
538 return parse_identifier(ctx, lval);
541 if(isdigitW(*ctx->ptr))
542 return parse_numeric_literal(ctx, lval);
544 switch(*ctx->ptr) {
545 case '{':
546 case '(':
547 case ')':
548 case '[':
549 case ']':
550 case ';':
551 case ',':
552 case '~':
553 case '?':
554 case ':':
555 return *ctx->ptr++;
557 case '}':
558 *(const WCHAR**)lval = ctx->ptr++;
559 return '}';
561 case '.':
562 if(++ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
563 return parse_double_literal(ctx, 0, lval);
564 return '.';
566 case '<':
567 if(++ctx->ptr == ctx->end) {
568 *(int*)lval = EXPR_LESS;
569 return tRelOper;
572 switch(*ctx->ptr) {
573 case '=': /* <= */
574 ctx->ptr++;
575 *(int*)lval = EXPR_LESSEQ;
576 return tRelOper;
577 case '<': /* << */
578 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* <<= */
579 ctx->ptr++;
580 *(int*)lval = EXPR_ASSIGNLSHIFT;
581 return tAssignOper;
583 *(int*)lval = EXPR_LSHIFT;
584 return tShiftOper;
585 default: /* < */
586 *(int*)lval = EXPR_LESS;
587 return tRelOper;
590 case '>':
591 if(++ctx->ptr == ctx->end) { /* > */
592 *(int*)lval = EXPR_GREATER;
593 return tRelOper;
596 switch(*ctx->ptr) {
597 case '=': /* >= */
598 ctx->ptr++;
599 *(int*)lval = EXPR_GREATEREQ;
600 return tRelOper;
601 case '>': /* >> */
602 if(++ctx->ptr < ctx->end) {
603 if(*ctx->ptr == '=') { /* >>= */
604 ctx->ptr++;
605 *(int*)lval = EXPR_ASSIGNRSHIFT;
606 return tAssignOper;
608 if(*ctx->ptr == '>') { /* >>> */
609 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* >>>= */
610 ctx->ptr++;
611 *(int*)lval = EXPR_ASSIGNRRSHIFT;
612 return tAssignOper;
614 *(int*)lval = EXPR_RRSHIFT;
615 return tRelOper;
618 *(int*)lval = EXPR_RSHIFT;
619 return tShiftOper;
620 default:
621 *(int*)lval = EXPR_GREATER;
622 return tRelOper;
625 case '+':
626 ctx->ptr++;
627 if(ctx->ptr < ctx->end) {
628 switch(*ctx->ptr) {
629 case '+': /* ++ */
630 ctx->ptr++;
631 return tINC;
632 case '=': /* += */
633 ctx->ptr++;
634 *(int*)lval = EXPR_ASSIGNADD;
635 return tAssignOper;
638 return '+';
640 case '-':
641 ctx->ptr++;
642 if(ctx->ptr < ctx->end) {
643 switch(*ctx->ptr) {
644 case '-': /* -- or --> */
645 ctx->ptr++;
646 if(ctx->is_html && ctx->nl && ctx->ptr < ctx->end && *ctx->ptr == '>') {
647 ctx->ptr++;
648 return tHTMLCOMMENT;
650 return tDEC;
651 case '=': /* -= */
652 ctx->ptr++;
653 *(int*)lval = EXPR_ASSIGNSUB;
654 return tAssignOper;
657 return '-';
659 case '*':
660 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* *= */
661 ctx->ptr++;
662 *(int*)lval = EXPR_ASSIGNMUL;
663 return tAssignOper;
665 return '*';
667 case '%':
668 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* %= */
669 ctx->ptr++;
670 *(int*)lval = EXPR_ASSIGNMOD;
671 return tAssignOper;
673 return '%';
675 case '&':
676 if(++ctx->ptr < ctx->end) {
677 switch(*ctx->ptr) {
678 case '=': /* &= */
679 ctx->ptr++;
680 *(int*)lval = EXPR_ASSIGNAND;
681 return tAssignOper;
682 case '&': /* && */
683 ctx->ptr++;
684 return tANDAND;
687 return '&';
689 case '|':
690 if(++ctx->ptr < ctx->end) {
691 switch(*ctx->ptr) {
692 case '=': /* |= */
693 ctx->ptr++;
694 *(int*)lval = EXPR_ASSIGNOR;
695 return tAssignOper;
696 case '|': /* || */
697 ctx->ptr++;
698 return tOROR;
701 return '|';
703 case '^':
704 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* ^= */
705 ctx->ptr++;
706 *(int*)lval = EXPR_ASSIGNXOR;
707 return tAssignOper;
709 return '^';
711 case '!':
712 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* != */
713 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* !== */
714 ctx->ptr++;
715 *(int*)lval = EXPR_NOTEQEQ;
716 return tEqOper;
718 *(int*)lval = EXPR_NOTEQ;
719 return tEqOper;
721 return '!';
723 case '=':
724 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* == */
725 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* === */
726 ctx->ptr++;
727 *(int*)lval = EXPR_EQEQ;
728 return tEqOper;
730 *(int*)lval = EXPR_EQ;
731 return tEqOper;
733 return '=';
735 case '/':
736 if(++ctx->ptr < ctx->end) {
737 if(*ctx->ptr == '=') { /* /= */
738 ctx->ptr++;
739 *(int*)lval = EXPR_ASSIGNDIV;
740 return kDIVEQ;
743 return '/';
745 case '\"':
746 case '\'':
747 return parse_string_literal(ctx, lval, *ctx->ptr);
749 case '_':
750 case '$':
751 return parse_identifier(ctx, lval);
754 WARN("unexpected char '%c' %d\n", *ctx->ptr, *ctx->ptr);
755 return 0;
758 literal_t *parse_regexp(parser_ctx_t *ctx)
760 const WCHAR *re, *flags_ptr;
761 DWORD re_len, flags;
762 literal_t *ret;
763 HRESULT hres;
765 TRACE("\n");
767 while(*ctx->ptr != '/')
768 ctx->ptr--;
770 re = ++ctx->ptr;
771 while(ctx->ptr < ctx->end && *ctx->ptr != '/') {
772 if(*ctx->ptr++ == '\\' && ctx->ptr < ctx->end)
773 ctx->ptr++;
776 if(ctx->ptr == ctx->end) {
777 WARN("unexpected end of file\n");
778 return NULL;
781 re_len = ctx->ptr-re;
783 flags_ptr = ++ctx->ptr;
784 while(ctx->ptr < ctx->end && isalnumW(*ctx->ptr))
785 ctx->ptr++;
787 hres = parse_regexp_flags(flags_ptr, ctx->ptr-flags_ptr, &flags);
788 if(FAILED(hres))
789 return NULL;
791 ret = parser_alloc(ctx, sizeof(literal_t));
792 ret->type = LT_REGEXP;
793 ret->u.regexp.str = re;
794 ret->u.regexp.str_len = re_len;
795 ret->u.regexp.flags = flags;
796 return ret;