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 static const WCHAR breakW[] = {'b','r','e','a','k',0};
35 static const WCHAR caseW[] = {'c','a','s','e',0};
36 static const WCHAR catchW[] = {'c','a','t','c','h',0};
37 static const WCHAR continueW[] = {'c','o','n','t','i','n','u','e',0};
38 static const WCHAR defaultW[] = {'d','e','f','a','u','l','t',0};
39 static const WCHAR deleteW[] = {'d','e','l','e','t','e',0};
40 static const WCHAR doW[] = {'d','o',0};
41 static const WCHAR elseW[] = {'e','l','s','e',0};
42 static const WCHAR falseW[] = {'f','a','l','s','e',0};
43 static const WCHAR finallyW[] = {'f','i','n','a','l','l','y',0};
44 static const WCHAR forW[] = {'f','o','r',0};
45 static const WCHAR functionW[] = {'f','u','n','c','t','i','o','n',0};
46 static const WCHAR ifW[] = {'i','f',0};
47 static const WCHAR inW[] = {'i','n',0};
48 static const WCHAR instanceofW[] = {'i','n','s','t','a','n','c','e','o','f',0};
49 static const WCHAR newW[] = {'n','e','w',0};
50 static const WCHAR nullW[] = {'n','u','l','l',0};
51 static const WCHAR returnW[] = {'r','e','t','u','r','n',0};
52 static const WCHAR switchW[] = {'s','w','i','t','c','h',0};
53 static const WCHAR thisW[] = {'t','h','i','s',0};
54 static const WCHAR throwW[] = {'t','h','r','o','w',0};
55 static const WCHAR trueW[] = {'t','r','u','e',0};
56 static const WCHAR tryW[] = {'t','r','y',0};
57 static const WCHAR typeofW[] = {'t','y','p','e','o','f',0};
58 static const WCHAR undefinedW[] = {'u','n','d','e','f','i','n','e','d',0};
59 static const WCHAR varW[] = {'v','a','r',0};
60 static const WCHAR voidW[] = {'v','o','i','d',0};
61 static const WCHAR whileW[] = {'w','h','i','l','e',0};
62 static const WCHAR withW[] = {'w','i','t','h',0};
71 {continueW, kCONTINUE},
79 {functionW, kFUNCTION},
82 {instanceofW, kINSTANCEOF},
92 {undefinedW, kUNDEFINED},
99 static int lex_error(parser_ctx_t *ctx, HRESULT hres)
105 static int check_keyword(parser_ctx_t *ctx, const WCHAR *word, const WCHAR **lval)
107 const WCHAR *p1 = ctx->ptr;
108 const WCHAR *p2 = word;
110 while(p1 < ctx->end && *p2) {
117 if(*p2 || (p1 < ctx->end && isalnumW(*p1)))
125 /* ECMA-262 3rd Edition 7.3 */
126 static BOOL is_endline(WCHAR c)
128 return c == '\n' || c == '\r' || c == 0x2028 || c == 0x2029;
131 static BOOL is_identifier_char(WCHAR c)
133 return isalnumW(c) || c == '$' || c == '_' || c == '\\';
136 static int hex_to_int(WCHAR c)
138 if('0' <= c && c <= '9')
141 if('a' <= c && c <= 'f')
144 if('A' <= c && c <= 'F')
150 static int check_keywords(parser_ctx_t *ctx, const WCHAR **lval)
152 int min = 0, max = sizeof(keywords)/sizeof(keywords[0])-1, r, i;
157 r = check_keyword(ctx, keywords[i].word, lval);
159 return keywords[i].token;
170 static void skip_spaces(parser_ctx_t *ctx)
172 while(ctx->ptr < ctx->end && isspaceW(*ctx->ptr)) {
173 if(is_endline(*ctx->ptr++))
178 static BOOL skip_html_comment(parser_ctx_t *ctx)
180 const WCHAR html_commentW[] = {'<','!','-','-',0};
182 if(!ctx->is_html || ctx->ptr+3 >= ctx->end ||
183 memcmp(ctx->ptr, html_commentW, sizeof(WCHAR)*4))
187 while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr++));
192 static BOOL skip_comment(parser_ctx_t *ctx)
194 if(ctx->ptr+1 >= ctx->end || *ctx->ptr != '/')
197 switch(ctx->ptr[1]) {
200 while(ctx->ptr+1 < ctx->end && (ctx->ptr[0] != '*' || ctx->ptr[1] != '/'))
203 if(ctx->ptr[0] == '*' && ctx->ptr[1] == '/') {
206 WARN("unexpected end of file (missing end of comment)\n");
212 while(ctx->ptr < ctx->end && !is_endline(*ctx->ptr))
222 static BOOL unescape(WCHAR *str)
262 i = hex_to_int(*++p);
267 i = hex_to_int(*++p);
273 i = hex_to_int(*++p);
278 i = hex_to_int(*++p);
283 i = hex_to_int(*++p);
288 i = hex_to_int(*++p);
297 c = c*10 + (*p++ - '0');
313 static int parse_identifier(parser_ctx_t *ctx, const WCHAR **ret)
315 const WCHAR *ptr = ctx->ptr++;
319 while(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr))
324 *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
325 memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
328 /* FIXME: unescape */
332 static int parse_string_literal(parser_ctx_t *ctx, const WCHAR **ret, WCHAR endch)
334 const WCHAR *ptr = ++ctx->ptr;
338 while(ctx->ptr < ctx->end && *ctx->ptr != endch) {
339 if(*ctx->ptr++ == '\\')
343 if(ctx->ptr == ctx->end) {
344 WARN("unexpected end of file\n");
345 return lex_error(ctx, E_FAIL);
350 *ret = wstr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
351 memcpy(wstr, ptr, (len+1)*sizeof(WCHAR));
356 if(!unescape(wstr)) {
357 WARN("unescape failed\n");
358 return lex_error(ctx, E_FAIL);
361 return tStringLiteral;
364 static literal_t *alloc_int_literal(parser_ctx_t *ctx, LONG l)
366 literal_t *ret = parser_alloc(ctx, sizeof(literal_t));
374 static int parse_double_literal(parser_ctx_t *ctx, LONG int_part, literal_t **literal)
378 if(ctx->ptr == ctx->end || (!isdigitW(*ctx->ptr) &&
379 *ctx->ptr!='.' && *ctx->ptr!='e' && *ctx->ptr!='E')) {
380 ERR("Illegal character\n");
385 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
386 d = d*10 + *(ctx->ptr++) - '0';
388 if(*ctx->ptr == '.') ctx->ptr++;
390 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
391 d += (tmp /= 10.0)*(*ctx->ptr++ - '0');
393 if(ctx->ptr < ctx->end && (*ctx->ptr == 'e' || *ctx->ptr == 'E')) {
397 if(ctx->ptr < ctx->end) {
398 if(*ctx->ptr == '+') {
400 }else if(*ctx->ptr == '-') {
403 }else if(!isdigitW(*ctx->ptr)) {
404 WARN("Expected exponent part\n");
405 return lex_error(ctx, E_FAIL);
409 if(ctx->ptr == ctx->end) {
410 WARN("unexpected end of file\n");
411 return lex_error(ctx, E_FAIL);
414 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
415 e = e*10 + *ctx->ptr++ - '0';
421 *literal = parser_alloc(ctx, sizeof(literal_t));
422 (*literal)->vt = VT_R8;
423 (*literal)->u.dval = d;
425 return tNumericLiteral;
428 static int parse_numeric_literal(parser_ctx_t *ctx, literal_t **literal)
432 l = *ctx->ptr++ - '0';
433 if(ctx->ptr == ctx->end) {
434 *literal = alloc_int_literal(ctx, l);
435 return tNumericLiteral;
439 if(*ctx->ptr == 'x' || *ctx->ptr == 'X') {
440 if(++ctx->ptr == ctx->end) {
441 ERR("unexpexted end of file\n");
445 while(ctx->ptr < ctx->end && (d = hex_to_int(*ctx->ptr)) != -1) {
450 if(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr)) {
451 WARN("unexpected identifier char\n");
452 return lex_error(ctx, E_FAIL);
455 *literal = alloc_int_literal(ctx, l);
456 return tNumericLiteral;
459 if(isdigitW(*ctx->ptr) || is_identifier_char(*ctx->ptr)) {
460 WARN("wrong char after zero\n");
461 return lex_error(ctx, E_FAIL);
464 *literal = alloc_int_literal(ctx, 0);
467 while(ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
469 d = l*10 + *(ctx->ptr)-'0';
471 /* Check for integer overflow */
472 if (l > INT_MAX/10 || d < 0)
473 return parse_double_literal(ctx, l, literal);
479 if(ctx->ptr < ctx->end) {
480 if(*ctx->ptr == '.' || *ctx->ptr == 'e' || *ctx->ptr == 'E')
481 return parse_double_literal(ctx, l, literal);
483 if(is_identifier_char(*ctx->ptr)) {
484 WARN("unexpected identifier char\n");
485 return lex_error(ctx, E_FAIL);
489 *literal = alloc_int_literal(ctx, l);
490 return tNumericLiteral;
493 int parser_lex(void *lval, parser_ctx_t *ctx)
497 ctx->nl = ctx->ptr == ctx->begin;
501 if(ctx->ptr == ctx->end)
503 }while(skip_comment(ctx) || skip_html_comment(ctx));
505 if(isalphaW(*ctx->ptr)) {
506 ret = check_keywords(ctx, lval);
510 return parse_identifier(ctx, lval);
513 if(isdigitW(*ctx->ptr))
514 return parse_numeric_literal(ctx, lval);
530 *(const WCHAR**)lval = ctx->ptr++;
534 if(++ctx->ptr < ctx->end && isdigitW(*ctx->ptr))
535 return parse_double_literal(ctx, 0, lval);
539 if(++ctx->ptr == ctx->end) {
540 *(int*)lval = EXPR_LESS;
547 *(int*)lval = EXPR_LESSEQ;
550 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* <<= */
552 *(int*)lval = EXPR_ASSIGNLSHIFT;
555 *(int*)lval = EXPR_LSHIFT;
558 *(int*)lval = EXPR_LESS;
563 if(++ctx->ptr == ctx->end) { /* > */
564 *(int*)lval = EXPR_GREATER;
571 *(int*)lval = EXPR_GREATEREQ;
574 if(++ctx->ptr < ctx->end) {
575 if(*ctx->ptr == '=') { /* >>= */
577 *(int*)lval = EXPR_ASSIGNRSHIFT;
580 if(*ctx->ptr == '>') { /* >>> */
581 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* >>>= */
583 *(int*)lval = EXPR_ASSIGNRRSHIFT;
586 *(int*)lval = EXPR_RRSHIFT;
590 *(int*)lval = EXPR_RSHIFT;
593 *(int*)lval = EXPR_GREATER;
599 if(ctx->ptr < ctx->end) {
606 *(int*)lval = EXPR_ASSIGNADD;
614 if(ctx->ptr < ctx->end) {
616 case '-': /* -- or --> */
618 if(ctx->is_html && ctx->nl && ctx->ptr < ctx->end && *ctx->ptr == '>') {
625 *(int*)lval = EXPR_ASSIGNSUB;
632 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* *= */
634 *(int*)lval = EXPR_ASSIGNMUL;
640 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* %= */
642 *(int*)lval = EXPR_ASSIGNMOD;
648 if(++ctx->ptr < ctx->end) {
652 *(int*)lval = EXPR_ASSIGNAND;
662 if(++ctx->ptr < ctx->end) {
666 *(int*)lval = EXPR_ASSIGNOR;
676 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* ^= */
678 *(int*)lval = EXPR_ASSIGNXOR;
684 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* != */
685 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* !== */
687 *(int*)lval = EXPR_NOTEQEQ;
690 *(int*)lval = EXPR_NOTEQ;
696 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* == */
697 if(++ctx->ptr < ctx->end && *ctx->ptr == '=') { /* === */
699 *(int*)lval = EXPR_EQEQ;
702 *(int*)lval = EXPR_EQ;
708 if(++ctx->ptr < ctx->end) {
709 if(*ctx->ptr == '=') { /* /= */
711 *(int*)lval = EXPR_ASSIGNDIV;
719 return parse_string_literal(ctx, lval, *ctx->ptr);
723 return parse_identifier(ctx, lval);
726 WARN("unexpected char '%c' %d\n", *ctx->ptr, *ctx->ptr);
730 static void add_object_literal(parser_ctx_t *ctx, DispatchEx *obj)
732 obj_literal_t *literal = parser_alloc(ctx, sizeof(obj_literal_t));
735 literal->next = ctx->obj_literals;
736 ctx->obj_literals = literal;
739 literal_t *parse_regexp(parser_ctx_t *ctx)
741 const WCHAR *re, *flags;
750 while(ctx->ptr < ctx->end && *ctx->ptr != '/') {
751 if(*ctx->ptr++ == '\\' && ctx->ptr < ctx->end)
755 if(ctx->ptr == ctx->end) {
756 WARN("unexpected end of file\n");
760 re_len = ctx->ptr-re;
763 while(ctx->ptr < ctx->end && isalnumW(*ctx->ptr))
766 hres = create_regexp_str(ctx->script, re, re_len, flags, ctx->ptr-flags, ®exp);
770 add_object_literal(ctx, regexp);
772 ret = parser_alloc(ctx, sizeof(literal_t));
773 ret->vt = VT_DISPATCH;
774 ret->u.disp = (IDispatch*)_IDispatchEx_(regexp);