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 /* 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
) {
126 if(*p2
|| (p1
< ctx
->end
&& is_identifier_char(*p1
)))
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')
145 if('a' <= c
&& c
<= 'f')
148 if('A' <= c
&& c
<= 'F')
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
;
161 r
= check_keyword(ctx
, keywords
[i
].word
, lval
);
163 return keywords
[i
].token
;
174 static void skip_spaces(parser_ctx_t
*ctx
)
176 while(ctx
->ptr
< ctx
->end
&& isspaceW(*ctx
->ptr
)) {
177 if(is_endline(*ctx
->ptr
++))
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))
191 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
++));
196 static BOOL
skip_comment(parser_ctx_t
*ctx
)
198 if(ctx
->ptr
+1 >= ctx
->end
|| *ctx
->ptr
!= '/')
201 switch(ctx
->ptr
[1]) {
204 while(ctx
->ptr
+1 < ctx
->end
&& (ctx
->ptr
[0] != '*' || ctx
->ptr
[1] != '/'))
207 if(ctx
->ptr
[0] == '*' && ctx
->ptr
[1] == '/') {
210 WARN("unexpected end of file (missing end of comment)\n");
216 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
))
226 static BOOL
unescape(WCHAR
*str
)
266 i
= hex_to_int(*++p
);
271 i
= hex_to_int(*++p
);
277 i
= hex_to_int(*++p
);
282 i
= hex_to_int(*++p
);
287 i
= hex_to_int(*++p
);
292 i
= hex_to_int(*++p
);
301 c
= c
*8 + (*p
++ - '0');
303 c
= c
*8 + (*p
++ - '0');
319 static int parse_identifier(parser_ctx_t
*ctx
, const WCHAR
**ret
)
321 const WCHAR
*ptr
= ctx
->ptr
++;
325 while(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
))
330 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
331 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
334 /* FIXME: unescape */
338 static int parse_string_literal(parser_ctx_t
*ctx
, const WCHAR
**ret
, WCHAR endch
)
340 const WCHAR
*ptr
= ++ctx
->ptr
;
344 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= endch
) {
345 if(*ctx
->ptr
++ == '\\')
349 if(ctx
->ptr
== ctx
->end
)
350 return lex_error(ctx
, IDS_UNTERMINATED_STR
);
354 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
355 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
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
));
378 static int parse_double_literal(parser_ctx_t
*ctx
, LONG int_part
, literal_t
**literal
)
383 if(ctx
->ptr
== ctx
->end
|| (!isdigitW(*ctx
->ptr
) &&
384 *ctx
->ptr
!='.' && *ctx
->ptr
!='e' && *ctx
->ptr
!='E')) {
385 ERR("Illegal character\n");
390 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
391 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
392 if(d
>LONGLONG_MAX
/10 || hlp
<0) {
399 while(ctx
->ptr
< ctx
->end
&& isdigitW(*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)
414 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
417 if(ctx
->ptr
< ctx
->end
&& (*ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')) {
421 if(ctx
->ptr
< ctx
->end
) {
422 if(*ctx
->ptr
== '+') {
424 }else if(*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)
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
;
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
)
460 l
= *ctx
->ptr
++ - '0';
461 if(ctx
->ptr
== ctx
->end
) {
462 *literal
= alloc_int_literal(ctx
, l
);
463 return tNumericLiteral
;
467 if(*ctx
->ptr
== 'x' || *ctx
->ptr
== 'X') {
468 if(++ctx
->ptr
== ctx
->end
) {
469 ERR("unexpexted end of file\n");
473 while(ctx
->ptr
< ctx
->end
&& (d
= hex_to_int(*ctx
->ptr
)) != -1) {
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
);
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
)
525 ctx
->nl
= ctx
->ptr
== ctx
->begin
;
529 if(ctx
->ptr
== ctx
->end
)
531 }while(skip_comment(ctx
) || skip_html_comment(ctx
));
533 if(isalphaW(*ctx
->ptr
)) {
534 ret
= check_keywords(ctx
, lval
);
538 return parse_identifier(ctx
, lval
);
541 if(isdigitW(*ctx
->ptr
))
542 return parse_numeric_literal(ctx
, lval
);
558 *(const WCHAR
**)lval
= ctx
->ptr
++;
562 if(++ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
563 return parse_double_literal(ctx
, 0, lval
);
567 if(++ctx
->ptr
== ctx
->end
) {
568 *(int*)lval
= EXPR_LESS
;
575 *(int*)lval
= EXPR_LESSEQ
;
578 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* <<= */
580 *(int*)lval
= EXPR_ASSIGNLSHIFT
;
583 *(int*)lval
= EXPR_LSHIFT
;
586 *(int*)lval
= EXPR_LESS
;
591 if(++ctx
->ptr
== ctx
->end
) { /* > */
592 *(int*)lval
= EXPR_GREATER
;
599 *(int*)lval
= EXPR_GREATEREQ
;
602 if(++ctx
->ptr
< ctx
->end
) {
603 if(*ctx
->ptr
== '=') { /* >>= */
605 *(int*)lval
= EXPR_ASSIGNRSHIFT
;
608 if(*ctx
->ptr
== '>') { /* >>> */
609 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* >>>= */
611 *(int*)lval
= EXPR_ASSIGNRRSHIFT
;
614 *(int*)lval
= EXPR_RRSHIFT
;
618 *(int*)lval
= EXPR_RSHIFT
;
621 *(int*)lval
= EXPR_GREATER
;
627 if(ctx
->ptr
< ctx
->end
) {
634 *(int*)lval
= EXPR_ASSIGNADD
;
642 if(ctx
->ptr
< ctx
->end
) {
644 case '-': /* -- or --> */
646 if(ctx
->is_html
&& ctx
->nl
&& ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '>') {
653 *(int*)lval
= EXPR_ASSIGNSUB
;
660 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* *= */
662 *(int*)lval
= EXPR_ASSIGNMUL
;
668 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* %= */
670 *(int*)lval
= EXPR_ASSIGNMOD
;
676 if(++ctx
->ptr
< ctx
->end
) {
680 *(int*)lval
= EXPR_ASSIGNAND
;
690 if(++ctx
->ptr
< ctx
->end
) {
694 *(int*)lval
= EXPR_ASSIGNOR
;
704 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* ^= */
706 *(int*)lval
= EXPR_ASSIGNXOR
;
712 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* != */
713 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* !== */
715 *(int*)lval
= EXPR_NOTEQEQ
;
718 *(int*)lval
= EXPR_NOTEQ
;
724 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* == */
725 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* === */
727 *(int*)lval
= EXPR_EQEQ
;
730 *(int*)lval
= EXPR_EQ
;
736 if(++ctx
->ptr
< ctx
->end
) {
737 if(*ctx
->ptr
== '=') { /* /= */
739 *(int*)lval
= EXPR_ASSIGNDIV
;
747 return parse_string_literal(ctx
, lval
, *ctx
->ptr
);
751 return parse_identifier(ctx
, lval
);
754 WARN("unexpected char '%c' %d\n", *ctx
->ptr
, *ctx
->ptr
);
758 literal_t
*parse_regexp(parser_ctx_t
*ctx
)
760 const WCHAR
*re
, *flags_ptr
;
767 while(*ctx
->ptr
!= '/')
771 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= '/') {
772 if(*ctx
->ptr
++ == '\\' && ctx
->ptr
< ctx
->end
)
776 if(ctx
->ptr
== ctx
->end
) {
777 WARN("unexpected end of file\n");
781 re_len
= ctx
->ptr
-re
;
783 flags_ptr
= ++ctx
->ptr
;
784 while(ctx
->ptr
< ctx
->end
&& isalnumW(*ctx
->ptr
))
787 hres
= parse_regexp_flags(flags_ptr
, ctx
->ptr
-flags_ptr
, &flags
);
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
;