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
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};
73 {continueW, kCONTINUE},
81 {functionW, kFUNCTION},
84 {instanceofW, kINSTANCEOF},
94 {undefinedW, kUNDEFINED},
101 static int lex_error(parser_ctx_t *ctx, HRESULT hres)
103 ctx->hres = JSCRIPT_ERROR|hres;
104 ctx->lexer_error = TRUE;
108 static int check_keyword(parser_ctx_t *ctx, const WCHAR *word, const WCHAR **lval)
110 const WCHAR *p1 = ctx->ptr;
111 const WCHAR *p2 = word;
113 while(p1 < ctx->end && *p2) {
120 if(*p2 || (p1 < ctx->end && isalnumW(*p1)))
128 /* ECMA-262 3rd Edition 7.3 */
129 static BOOL is_endline(WCHAR c)
131 return c == '\n' || c == '\r' || c == 0x2028 || c == 0x2029;
134 static BOOL is_identifier_char(WCHAR c)
136 return isalnumW(c) || c == '$' || c == '_' || c == '\\';
139 static int hex_to_int(WCHAR c)
141 if('0' <= c && c <= '9')
144 if('a' <= c && c <= 'f')
147 if('A' <= c && c <= 'F')
153 static int check_keywords(parser_ctx_t *ctx, const WCHAR **lval)
155 int min = 0, max = sizeof(keywords)/sizeof(keywords[0])-1, r, i;
160 r = check_keyword(ctx, keywords[i].word, lval);
162 return keywords[i].token;
173 static void skip_spaces(parser_ctx_t *ctx)
175 while(ctx->ptr < ctx->end && isspaceW(*ctx->ptr)) {
176 if(is_endline(*ctx->ptr++))
181 static BOOL skip_html_comment(parser_ctx_t *ctx)
183 const WCHAR html_commentW[] = {'<','!','-','-',0};
185 if(!ctx->is_html || ctx->ptr+3 >= ctx->end ||
186 memcmp(ctx->ptr, html_commentW, sizeof(WCHAR)*4))
190 while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr++));
195 static BOOL skip_comment(parser_ctx_t *ctx)
197 if(ctx->ptr+1 >= ctx->end || *ctx->ptr != '/')
200 switch(ctx->ptr[1]) {
203 while(ctx->ptr+1 < ctx->end && (ctx->ptr[0] != '*' || ctx->ptr[1] != '/'))
206 if(ctx->ptr[0] == '*' && ctx->ptr[1] == '/') {
209 WARN("unexpected end of file (missing end of comment)\n");
215 while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr))
225 static BOOL unescape(WCHAR *str)
265 i = hex_to_int(*++p);
270 i = hex_to_int(*++p);
276 i = hex_to_int(*++p);
281 i = hex_to_int(*++p);
286 i = hex_to_int(*++p);
291 i = hex_to_int(*++p);
300 c = c*10 + (*p++ - '0');
316 static int parse_identifier(parser_ctx_t *ctx, const WCHAR **ret)
318 const WCHAR *ptr = ctx->ptr++;
322 while(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr))
327 *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
328 memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
331 /* FIXME: unescape */
335 static int parse_string_literal(parser_ctx_t *ctx, const WCHAR **ret, WCHAR endch)
337 const WCHAR *ptr = ++ctx->ptr;
341 while(ctx->ptr < ctx->end && *ctx->ptr != endch) {
342 if(*ctx->ptr++ == '\\')
346 if(ctx->ptr == ctx->end)
347 return lex_error(ctx, IDS_UNTERMINATED_STR);
351 *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
352 memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
357 if(!unescape(wstr)) {
358 WARN("unescape failed\n");
359 return lex_error(ctx, E_FAIL);
362 return tStringLiteral;
365 static literal_t *alloc_int_literal(parser_ctx_t *ctx, LONG l)
367 literal_t *ret = parser_alloc(ctx, sizeof(literal_t));
375 static int parse_double_literal(parser_ctx_t *ctx, LONG int_part, literal_t **literal)
380 if(ctx->ptr == ctx->end || (!isdigitW(*ctx->ptr) &&
381 *ctx->ptr!='.' && *ctx->ptr!='e' && *ctx->ptr!='E')) {
382 ERR("Illegal character\n");
387 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
388 hlp = d*10 + *(ctx->ptr++) - '0';
389 if(d>LONGLONG_MAX/10 || hlp<0) {
396 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
401 if(*ctx->ptr == '.') ctx->ptr++;
403 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
404 hlp = d*10 + *(ctx->ptr++) - '0';
405 if(d>LONGLONG_MAX/10 || hlp<0)
411 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
414 if(ctx->ptr < ctx->end && (*ctx->ptr == 'e' || *ctx->ptr == 'E')) {
418 if(ctx->ptr < ctx->end) {
419 if(*ctx->ptr == '+') {
421 }else if(*ctx->ptr == '-') {
424 }else if(!isdigitW(*ctx->ptr)) {
425 WARN("Expected exponent part\n");
426 return lex_error(ctx, E_FAIL);
430 if(ctx->ptr == ctx->end) {
431 WARN("unexpected end of file\n");
432 return lex_error(ctx, E_FAIL);
435 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr)) {
436 if(e > INT_MAX/10 || (e = e*10 + *ctx->ptr++ - '0')<0)
441 if(exp<0 && e<0 && e+exp>0) exp = INT_MIN;
442 else if(exp>0 && e>0 && e+exp<0) exp = INT_MAX;
446 *literal = parser_alloc(ctx, sizeof(literal_t));
447 (*literal)->vt = VT_R8;
448 (*literal)->u.dval = (double)d*pow(10, exp);
450 return tNumericLiteral;
453 static int parse_numeric_literal(parser_ctx_t *ctx, literal_t **literal)
457 l = *ctx->ptr++ - '0';
458 if(ctx->ptr == ctx->end) {
459 *literal = alloc_int_literal(ctx, l);
460 return tNumericLiteral;
464 if(*ctx->ptr == 'x' || *ctx->ptr == 'X') {
465 if(++ctx->ptr == ctx->end) {
466 ERR("unexpexted end of file\n");
470 while(ctx->ptr < ctx->end && (d = hex_to_int(*ctx->ptr)) != -1) {
475 if(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr)) {
476 WARN("unexpected identifier char\n");
477 return lex_error(ctx, E_FAIL);
480 *literal = alloc_int_literal(ctx, l);
481 return tNumericLiteral;
484 if(isdigitW(*ctx->ptr) || is_identifier_char(*ctx->ptr)) {
485 WARN("wrong char after zero\n");
486 return lex_error(ctx, E_FAIL);
489 *literal = alloc_int_literal(ctx, 0);
492 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
494 d = l*10 + *(ctx->ptr)-'0';
496 /* Check for integer overflow */
497 if (l > INT_MAX/10 || d < 0)
498 return parse_double_literal(ctx, l, literal);
504 if(ctx->ptr < ctx->end) {
505 if(*ctx->ptr == '.' || *ctx->ptr == 'e' || *ctx->ptr == 'E')
506 return parse_double_literal(ctx, l, literal);
508 if(is_identifier_char(*ctx->ptr)) {
509 WARN("unexpected identifier char\n");
510 return lex_error(ctx, E_FAIL);
514 *literal = alloc_int_literal(ctx, l);
515 return tNumericLiteral;
518 int parser_lex(void *lval, parser_ctx_t *ctx)
522 ctx->nl = ctx->ptr == ctx->begin;
526 if(ctx->ptr == ctx->end)
528 }while(skip_comment(ctx) || skip_html_comment(ctx));
530 if(isalphaW(*ctx->ptr)) {
531 ret = check_keywords(ctx, lval);
535 return parse_identifier(ctx, lval);
538 if(isdigitW(*ctx->ptr))
539 return parse_numeric_literal(ctx, lval);
555 *(const WCHAR**)lval = ctx->ptr++;
559 if(++ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
560 return parse_double_literal(ctx, 0, lval);
564 if(++ctx->ptr == ctx->end) {
565 *(int*)lval = EXPR_LESS;
572 *(int*)lval = EXPR_LESSEQ;
575 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* <<= */
577 *(int*)lval = EXPR_ASSIGNLSHIFT;
580 *(int*)lval = EXPR_LSHIFT;
583 *(int*)lval = EXPR_LESS;
588 if(++ctx->ptr == ctx->end) { /* > */
589 *(int*)lval = EXPR_GREATER;
596 *(int*)lval = EXPR_GREATEREQ;
599 if(++ctx->ptr < ctx->end) {
600 if(*ctx->ptr == '=') { /* >>= */
602 *(int*)lval = EXPR_ASSIGNRSHIFT;
605 if(*ctx->ptr == '>') { /* >>> */
606 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* >>>= */
608 *(int*)lval = EXPR_ASSIGNRRSHIFT;
611 *(int*)lval = EXPR_RRSHIFT;
615 *(int*)lval = EXPR_RSHIFT;
618 *(int*)lval = EXPR_GREATER;
624 if(ctx->ptr < ctx->end) {
631 *(int*)lval = EXPR_ASSIGNADD;
639 if(ctx->ptr < ctx->end) {
641 case '-': /* -- or --> */
643 if(ctx->is_html && ctx->nl && ctx->ptr < ctx->end && *ctx->ptr == '>') {
650 *(int*)lval = EXPR_ASSIGNSUB;
657 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* *= */
659 *(int*)lval = EXPR_ASSIGNMUL;
665 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* %= */
667 *(int*)lval = EXPR_ASSIGNMOD;
673 if(++ctx->ptr < ctx->end) {
677 *(int*)lval = EXPR_ASSIGNAND;
687 if(++ctx->ptr < ctx->end) {
691 *(int*)lval = EXPR_ASSIGNOR;
701 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* ^= */
703 *(int*)lval = EXPR_ASSIGNXOR;
709 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* != */
710 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* !== */
712 *(int*)lval = EXPR_NOTEQEQ;
715 *(int*)lval = EXPR_NOTEQ;
721 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* == */
722 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* === */
724 *(int*)lval = EXPR_EQEQ;
727 *(int*)lval = EXPR_EQ;
733 if(++ctx->ptr < ctx->end) {
734 if(*ctx->ptr == '=') { /* /= */
736 *(int*)lval = EXPR_ASSIGNDIV;
744 return parse_string_literal(ctx, lval, *ctx->ptr);
748 return parse_identifier(ctx, lval);
751 WARN("unexpected char '%c' %d\n", *ctx->ptr, *ctx->ptr);
755 static void add_object_literal(parser_ctx_t *ctx, DispatchEx *obj)
757 obj_literal_t *literal = parser_alloc(ctx, sizeof(obj_literal_t));
760 literal->next = ctx->obj_literals;
761 ctx->obj_literals = literal;
764 literal_t *parse_regexp(parser_ctx_t *ctx)
766 const WCHAR *re, *flags;
775 while(ctx->ptr < ctx->end && *ctx->ptr != '/') {
776 if(*ctx->ptr++ == '\\' && ctx->ptr < ctx->end)
780 if(ctx->ptr == ctx->end) {
781 WARN("unexpected end of file\n");
785 re_len = ctx->ptr-re;
788 while(ctx->ptr < ctx->end && isalnumW(*ctx->ptr))
791 hres = create_regexp_str(ctx->script, re, re_len, flags, ctx->ptr-flags, ®exp);
795 add_object_literal(ctx, regexp);
797 ret = parser_alloc(ctx, sizeof(literal_t));
798 ret->vt = VT_DISPATCH;
799 ret->u.disp = (IDispatch*)_IDispatchEx_(regexp);