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
26 #include "parser.tab.h"
28 #include "wine/debug.h"
29 #include "wine/unicode.h"
31 WINE_DEFAULT_DEBUG_CHANNEL(jscript
);
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};
70 {continueW
, kCONTINUE
},
78 {functionW
, kFUNCTION
},
81 {instanceofW
, kINSTANCEOF
},
91 {undefinedW
, kUNDEFINED
},
98 static int lex_error(parser_ctx_t
*ctx
, HRESULT hres
)
104 static int check_keyword(parser_ctx_t
*ctx
, const WCHAR
*word
, const WCHAR
**lval
)
106 const WCHAR
*p1
= ctx
->ptr
;
107 const WCHAR
*p2
= word
;
109 while(p1
< ctx
->end
&& *p2
) {
116 if(*p2
|| (p1
< ctx
->end
&& isalnumW(*p1
)))
124 /* ECMA-262 3rd Edition 7.3 */
125 static BOOL
is_endline(WCHAR c
)
127 return c
== '\n' || c
== '\r' || c
== 0x2028 || c
== 0x2029;
130 static BOOL
is_identifier_char(WCHAR c
)
132 return isalnumW(c
) || c
== '$' || c
== '_' || c
== '\\';
135 static int hex_to_int(WCHAR c
)
137 if('0' <= c
&& c
<= '9')
140 if('a' <= c
&& c
<= 'f')
143 if('A' <= c
&& c
<= 'F')
149 static int check_keywords(parser_ctx_t
*ctx
, const WCHAR
**lval
)
151 int min
= 0, max
= sizeof(keywords
)/sizeof(keywords
[0])-1, r
, i
;
156 r
= check_keyword(ctx
, keywords
[i
].word
, lval
);
158 return keywords
[i
].token
;
169 static void skip_spaces(parser_ctx_t
*ctx
)
171 while(ctx
->ptr
< ctx
->end
&& isspaceW(*ctx
->ptr
)) {
172 if(is_endline(*ctx
->ptr
++))
177 static BOOL
skip_html_comment(parser_ctx_t
*ctx
)
179 const WCHAR html_commentW
[] = {'<','!','-','-',0};
181 if(!ctx
->is_html
|| ctx
->ptr
+3 >= ctx
->end
||
182 memcmp(ctx
->ptr
, html_commentW
, sizeof(WCHAR
)*4))
186 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
++));
191 static BOOL
skip_comment(parser_ctx_t
*ctx
)
193 if(ctx
->ptr
+1 >= ctx
->end
|| *ctx
->ptr
!= '/')
196 switch(ctx
->ptr
[1]) {
199 while(ctx
->ptr
+1 < ctx
->end
&& (ctx
->ptr
[0] != '*' || ctx
->ptr
[1] != '/'))
202 if(ctx
->ptr
[0] == '*' && ctx
->ptr
[1] == '/') {
205 WARN("unexpected end of file (missing end of comment)\n");
211 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
))
221 static BOOL
unescape(WCHAR
*str
)
261 i
= hex_to_int(*++p
);
266 i
= hex_to_int(*++p
);
272 i
= hex_to_int(*++p
);
277 i
= hex_to_int(*++p
);
282 i
= hex_to_int(*++p
);
287 i
= hex_to_int(*++p
);
296 c
= c
*10 + (*p
++ - '0');
312 static int parse_identifier(parser_ctx_t
*ctx
, const WCHAR
**ret
)
314 const WCHAR
*ptr
= ctx
->ptr
++;
318 while(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
))
323 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
324 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
327 /* FIXME: unescape */
331 static int parse_string_literal(parser_ctx_t
*ctx
, const WCHAR
**ret
, WCHAR endch
)
333 const WCHAR
*ptr
= ++ctx
->ptr
;
337 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= endch
) {
338 if(*ctx
->ptr
++ == '\\')
342 if(ctx
->ptr
== ctx
->end
) {
343 WARN("unexpected end of file\n");
344 return lex_error(ctx
, E_FAIL
);
349 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
350 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
355 if(!unescape(wstr
)) {
356 WARN("unescape failed\n");
357 return lex_error(ctx
, E_FAIL
);
360 return tStringLiteral
;
363 static literal_t
*alloc_int_literal(parser_ctx_t
*ctx
, LONG l
)
365 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
373 static int parse_double_literal(parser_ctx_t
*ctx
, LONG int_part
, literal_t
**literal
)
377 if(ctx
->ptr
== ctx
->end
|| !isdigitW(*ctx
->ptr
)) {
378 ERR("No digit after point\n");
383 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
384 d
+= (tmp
/= 10.0)*(*ctx
->ptr
++ - '0');
386 if(ctx
->ptr
< ctx
->end
&& (*ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')) {
390 if(ctx
->ptr
< ctx
->end
) {
391 if(*ctx
->ptr
== '+') {
393 }else if(*ctx
->ptr
== '-') {
396 }else if(!isdigitW(*ctx
->ptr
)) {
397 WARN("Expected exponent part\n");
398 return lex_error(ctx
, E_FAIL
);
402 if(ctx
->ptr
== ctx
->end
) {
403 WARN("unexpected end of file\n");
404 return lex_error(ctx
, E_FAIL
);
407 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
408 e
= e
*10 + *ctx
->ptr
++ - '0';
414 *literal
= parser_alloc(ctx
, sizeof(literal_t
));
415 (*literal
)->vt
= VT_R8
;
416 (*literal
)->u
.dval
= d
;
418 return tNumericLiteral
;
421 static int parse_numeric_literal(parser_ctx_t
*ctx
, literal_t
**literal
)
425 l
= *ctx
->ptr
++ - '0';
426 if(ctx
->ptr
== ctx
->end
) {
427 *literal
= alloc_int_literal(ctx
, l
);
428 return tNumericLiteral
;
432 if(*ctx
->ptr
== 'x' || *ctx
->ptr
== 'X') {
433 if(++ctx
->ptr
== ctx
->end
) {
434 ERR("unexpexted end of file\n");
438 while(ctx
->ptr
< ctx
->end
&& (d
= hex_to_int(*ctx
->ptr
)) != -1) {
443 if(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
)) {
444 WARN("unexpected identifier char\n");
445 return lex_error(ctx
, E_FAIL
);
448 *literal
= alloc_int_literal(ctx
, l
);
449 return tNumericLiteral
;
452 if(isdigitW(*ctx
->ptr
) || is_identifier_char(*ctx
->ptr
)) {
453 WARN("wrong char after zero\n");
454 return lex_error(ctx
, E_FAIL
);
457 *literal
= alloc_int_literal(ctx
, 0);
460 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
461 l
= l
*10 + *(ctx
->ptr
++)-'0';
463 if(ctx
->ptr
< ctx
->end
) {
464 if(*ctx
->ptr
== '.') {
466 return parse_double_literal(ctx
, l
, literal
);
469 if(is_identifier_char(*ctx
->ptr
)) {
470 WARN("unexpected identifier char\n");
471 return lex_error(ctx
, E_FAIL
);
475 *literal
= alloc_int_literal(ctx
, l
);
476 return tNumericLiteral
;
479 int parser_lex(void *lval
, parser_ctx_t
*ctx
)
483 ctx
->nl
= ctx
->ptr
== ctx
->begin
;
487 if(ctx
->ptr
== ctx
->end
)
489 }while(skip_comment(ctx
) || skip_html_comment(ctx
));
491 if(isalphaW(*ctx
->ptr
)) {
492 ret
= check_keywords(ctx
, lval
);
496 return parse_identifier(ctx
, lval
);
499 if(isdigitW(*ctx
->ptr
))
500 return parse_numeric_literal(ctx
, lval
);
516 *(const WCHAR
**)lval
= ctx
->ptr
++;
520 if(++ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
521 return parse_double_literal(ctx
, 0, lval
);
525 if(++ctx
->ptr
== ctx
->end
) {
526 *(int*)lval
= EXPR_LESS
;
533 *(int*)lval
= EXPR_LESSEQ
;
536 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* <<= */
538 *(int*)lval
= EXPR_ASSIGNLSHIFT
;
541 *(int*)lval
= EXPR_LSHIFT
;
544 *(int*)lval
= EXPR_LESS
;
549 if(++ctx
->ptr
== ctx
->end
) { /* > */
550 *(int*)lval
= EXPR_GREATER
;
557 *(int*)lval
= EXPR_GREATEREQ
;
560 if(++ctx
->ptr
< ctx
->end
) {
561 if(*ctx
->ptr
== '=') { /* >>= */
563 *(int*)lval
= EXPR_ASSIGNRSHIFT
;
566 if(*ctx
->ptr
== '>') { /* >>> */
567 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* >>>= */
569 *(int*)lval
= EXPR_ASSIGNRRSHIFT
;
572 *(int*)lval
= EXPR_RRSHIFT
;
576 *(int*)lval
= EXPR_RSHIFT
;
579 *(int*)lval
= EXPR_GREATER
;
585 if(ctx
->ptr
< ctx
->end
) {
592 *(int*)lval
= EXPR_ASSIGNADD
;
600 if(ctx
->ptr
< ctx
->end
) {
602 case '-': /* -- or --> */
604 if(ctx
->is_html
&& ctx
->nl
&& ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '>') {
611 *(int*)lval
= EXPR_ASSIGNSUB
;
618 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* *= */
620 *(int*)lval
= EXPR_ASSIGNMUL
;
626 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* %= */
628 *(int*)lval
= EXPR_ASSIGNMOD
;
634 if(++ctx
->ptr
< ctx
->end
) {
638 *(int*)lval
= EXPR_ASSIGNAND
;
648 if(++ctx
->ptr
< ctx
->end
) {
652 *(int*)lval
= EXPR_ASSIGNOR
;
662 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* ^= */
664 *(int*)lval
= EXPR_ASSIGNXOR
;
670 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* != */
671 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* !== */
673 *(int*)lval
= EXPR_NOTEQEQ
;
676 *(int*)lval
= EXPR_NOTEQ
;
682 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* == */
683 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* === */
685 *(int*)lval
= EXPR_EQEQ
;
688 *(int*)lval
= EXPR_EQ
;
694 if(++ctx
->ptr
< ctx
->end
) {
695 if(*ctx
->ptr
== '=') { /* /= */
697 *(int*)lval
= EXPR_ASSIGNDIV
;
705 return parse_string_literal(ctx
, lval
, *ctx
->ptr
);
709 return parse_identifier(ctx
, lval
);
712 WARN("unexpected char '%c' %d\n", *ctx
->ptr
, *ctx
->ptr
);
716 static void add_object_literal(parser_ctx_t
*ctx
, DispatchEx
*obj
)
718 obj_literal_t
*literal
= parser_alloc(ctx
, sizeof(obj_literal_t
));
721 literal
->next
= ctx
->obj_literals
;
722 ctx
->obj_literals
= literal
;
725 literal_t
*parse_regexp(parser_ctx_t
*ctx
)
727 const WCHAR
*re
, *flags
;
736 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= '/') {
737 if(*ctx
->ptr
++ == '\\' && ctx
->ptr
< ctx
->end
)
741 if(ctx
->ptr
== ctx
->end
) {
742 WARN("unexpected end of file\n");
746 re_len
= ctx
->ptr
-re
;
749 while(ctx
->ptr
< ctx
->end
&& isalnumW(*ctx
->ptr
))
752 hres
= create_regexp_str(ctx
->script
, re
, re_len
, flags
, ctx
->ptr
-flags
, ®exp
);
756 add_object_literal(ctx
, regexp
);
758 ret
= parser_alloc(ctx
, sizeof(literal_t
));
759 ret
->vt
= VT_DISPATCH
;
760 ret
->u
.disp
= (IDispatch
*)_IDispatchEx_(regexp
);