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
},
100 static int lex_error(parser_ctx_t
*ctx
, HRESULT hres
)
103 ctx
->lexer_error
= TRUE
;
107 /* ECMA-262 3rd Edition 7.6 */
108 static BOOL
is_identifier_char(WCHAR c
)
110 return isalnumW(c
) || c
== '$' || c
== '_' || c
== '\\';
113 static int check_keyword(parser_ctx_t
*ctx
, const WCHAR
*word
, const WCHAR
**lval
)
115 const WCHAR
*p1
= ctx
->ptr
;
116 const WCHAR
*p2
= word
;
118 while(p1
< ctx
->end
&& *p2
) {
125 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
)
201 if(*ctx
->ptr
!= '/') {
202 if(*ctx
->ptr
== '@' && ctx
->ptr
+2 < ctx
->end
&& ctx
->ptr
[1] == '*' && ctx
->ptr
[2] == '/') {
210 switch(ctx
->ptr
[1]) {
213 if(ctx
->ptr
+2 < ctx
->end
&& *ctx
->ptr
== '@' && is_identifier_char(ctx
->ptr
[1]))
215 while(ctx
->ptr
+1 < ctx
->end
&& (ctx
->ptr
[0] != '*' || ctx
->ptr
[1] != '/'))
218 if(ctx
->ptr
[0] == '*' && ctx
->ptr
[1] == '/') {
221 WARN("unexpected end of file (missing end of comment)\n");
227 if(ctx
->ptr
+2 < ctx
->end
&& *ctx
->ptr
== '@' && is_identifier_char(ctx
->ptr
[1]))
229 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
))
239 static BOOL
unescape(WCHAR
*str
)
275 i
= hex_to_int(*++p
);
280 i
= hex_to_int(*++p
);
286 i
= hex_to_int(*++p
);
291 i
= hex_to_int(*++p
);
296 i
= hex_to_int(*++p
);
301 i
= hex_to_int(*++p
);
310 c
= c
*8 + (*p
++ - '0');
312 c
= c
*8 + (*p
++ - '0');
328 static int parse_identifier(parser_ctx_t
*ctx
, const WCHAR
**ret
)
330 const WCHAR
*ptr
= ctx
->ptr
++;
334 while(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
))
339 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
340 memcpy(wstr
, ptr
, len
*sizeof(WCHAR
));
343 /* FIXME: unescape */
347 static int parse_string_literal(parser_ctx_t
*ctx
, const WCHAR
**ret
, WCHAR endch
)
349 const WCHAR
*ptr
= ++ctx
->ptr
;
353 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= endch
) {
354 if(*ctx
->ptr
++ == '\\')
358 if(ctx
->ptr
== ctx
->end
)
359 return lex_error(ctx
, JS_E_UNTERMINATED_STRING
);
363 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
364 memcpy(wstr
, ptr
, len
*sizeof(WCHAR
));
369 if(!unescape(wstr
)) {
370 WARN("unescape failed\n");
371 return lex_error(ctx
, E_FAIL
);
374 return tStringLiteral
;
377 static literal_t
*new_int_literal(parser_ctx_t
*ctx
, LONG l
)
379 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
387 static literal_t
*new_double_literal(parser_ctx_t
*ctx
, DOUBLE d
)
389 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
391 ret
->type
= LT_DOUBLE
;
396 literal_t
*new_boolean_literal(parser_ctx_t
*ctx
, VARIANT_BOOL bval
)
398 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
406 static int parse_double_literal(parser_ctx_t
*ctx
, LONG int_part
, literal_t
**literal
)
411 if(ctx
->ptr
== ctx
->end
|| (!isdigitW(*ctx
->ptr
) &&
412 *ctx
->ptr
!='.' && *ctx
->ptr
!='e' && *ctx
->ptr
!='E')) {
413 ERR("Illegal character\n");
418 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
419 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
420 if(d
>LONGLONG_MAX
/10 || hlp
<0) {
427 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
432 if(*ctx
->ptr
== '.') ctx
->ptr
++;
434 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
435 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
436 if(d
>LONGLONG_MAX
/10 || hlp
<0)
442 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
445 if(ctx
->ptr
< ctx
->end
&& (*ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')) {
449 if(ctx
->ptr
< ctx
->end
) {
450 if(*ctx
->ptr
== '+') {
452 }else if(*ctx
->ptr
== '-') {
455 }else if(!isdigitW(*ctx
->ptr
)) {
456 WARN("Expected exponent part\n");
457 return lex_error(ctx
, E_FAIL
);
461 if(ctx
->ptr
== ctx
->end
) {
462 WARN("unexpected end of file\n");
463 return lex_error(ctx
, E_FAIL
);
466 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
467 if(e
> INT_MAX
/10 || (e
= e
*10 + *ctx
->ptr
++ - '0')<0)
472 if(exp
<0 && e
<0 && e
+exp
>0) exp
= INT_MIN
;
473 else if(exp
>0 && e
>0 && e
+exp
<0) exp
= INT_MAX
;
477 *literal
= new_double_literal(ctx
, exp
>=0 ? d
*pow(10, exp
) : d
/pow(10, -exp
));
478 return tNumericLiteral
;
481 static int parse_numeric_literal(parser_ctx_t
*ctx
, literal_t
**literal
)
485 l
= *ctx
->ptr
++ - '0';
486 if(ctx
->ptr
== ctx
->end
) {
487 *literal
= new_int_literal(ctx
, l
);
488 return tNumericLiteral
;
492 if(*ctx
->ptr
== 'x' || *ctx
->ptr
== 'X') {
493 if(++ctx
->ptr
== ctx
->end
) {
494 ERR("unexpected end of file\n");
498 while(ctx
->ptr
< ctx
->end
&& (d
= hex_to_int(*ctx
->ptr
)) != -1) {
503 if(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
)) {
504 WARN("unexpected identifier char\n");
505 return lex_error(ctx
, E_FAIL
);
508 *literal
= new_int_literal(ctx
, l
);
509 return tNumericLiteral
;
512 if(isdigitW(*ctx
->ptr
) || is_identifier_char(*ctx
->ptr
)) {
513 WARN("wrong char after zero\n");
514 return lex_error(ctx
, E_FAIL
);
517 *literal
= new_int_literal(ctx
, 0);
520 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
522 d
= l
*10 + *(ctx
->ptr
)-'0';
524 /* Check for integer overflow */
525 if (l
> INT_MAX
/10 || d
< 0)
526 return parse_double_literal(ctx
, l
, literal
);
532 if(ctx
->ptr
< ctx
->end
) {
533 if(*ctx
->ptr
== '.' || *ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')
534 return parse_double_literal(ctx
, l
, literal
);
536 if(is_identifier_char(*ctx
->ptr
)) {
537 WARN("unexpected identifier char\n");
538 return lex_error(ctx
, E_FAIL
);
542 *literal
= new_int_literal(ctx
, l
);
543 return tNumericLiteral
;
546 static int next_token(parser_ctx_t
*ctx
, void *lval
)
550 if(ctx
->ptr
== ctx
->end
)
552 }while(skip_comment(ctx
) || skip_html_comment(ctx
));
554 if(isalphaW(*ctx
->ptr
)) {
555 int ret
= check_keywords(ctx
, lval
);
559 return parse_identifier(ctx
, lval
);
562 if(isdigitW(*ctx
->ptr
))
563 return parse_numeric_literal(ctx
, lval
);
579 *(const WCHAR
**)lval
= ctx
->ptr
++;
583 if(++ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
584 return parse_double_literal(ctx
, 0, lval
);
588 if(++ctx
->ptr
== ctx
->end
) {
589 *(int*)lval
= EXPR_LESS
;
596 *(int*)lval
= EXPR_LESSEQ
;
599 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* <<= */
601 *(int*)lval
= EXPR_ASSIGNLSHIFT
;
604 *(int*)lval
= EXPR_LSHIFT
;
607 *(int*)lval
= EXPR_LESS
;
612 if(++ctx
->ptr
== ctx
->end
) { /* > */
613 *(int*)lval
= EXPR_GREATER
;
620 *(int*)lval
= EXPR_GREATEREQ
;
623 if(++ctx
->ptr
< ctx
->end
) {
624 if(*ctx
->ptr
== '=') { /* >>= */
626 *(int*)lval
= EXPR_ASSIGNRSHIFT
;
629 if(*ctx
->ptr
== '>') { /* >>> */
630 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* >>>= */
632 *(int*)lval
= EXPR_ASSIGNRRSHIFT
;
635 *(int*)lval
= EXPR_RRSHIFT
;
639 *(int*)lval
= EXPR_RSHIFT
;
642 *(int*)lval
= EXPR_GREATER
;
648 if(ctx
->ptr
< ctx
->end
) {
655 *(int*)lval
= EXPR_ASSIGNADD
;
663 if(ctx
->ptr
< ctx
->end
) {
665 case '-': /* -- or --> */
667 if(ctx
->is_html
&& ctx
->nl
&& ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '>') {
674 *(int*)lval
= EXPR_ASSIGNSUB
;
681 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* *= */
683 *(int*)lval
= EXPR_ASSIGNMUL
;
689 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* %= */
691 *(int*)lval
= EXPR_ASSIGNMOD
;
697 if(++ctx
->ptr
< ctx
->end
) {
701 *(int*)lval
= EXPR_ASSIGNAND
;
711 if(++ctx
->ptr
< ctx
->end
) {
715 *(int*)lval
= EXPR_ASSIGNOR
;
725 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* ^= */
727 *(int*)lval
= EXPR_ASSIGNXOR
;
733 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* != */
734 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* !== */
736 *(int*)lval
= EXPR_NOTEQEQ
;
739 *(int*)lval
= EXPR_NOTEQ
;
745 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* == */
746 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* === */
748 *(int*)lval
= EXPR_EQEQ
;
751 *(int*)lval
= EXPR_EQ
;
757 if(++ctx
->ptr
< ctx
->end
) {
758 if(*ctx
->ptr
== '=') { /* /= */
760 *(int*)lval
= EXPR_ASSIGNDIV
;
768 return parse_string_literal(ctx
, lval
, *ctx
->ptr
);
772 return parse_identifier(ctx
, lval
);
778 WARN("unexpected char '%c' %d\n", *ctx
->ptr
, *ctx
->ptr
);
788 struct _cc_var_t
*next
;
793 void release_cc(cc_ctx_t
*cc
)
795 cc_var_t
*iter
, *next
;
797 for(iter
= cc
->vars
; iter
; iter
= next
) {
805 static BOOL
add_cc_var(cc_ctx_t
*cc
, const WCHAR
*name
, cc_var_t
*v
)
812 new_v
= heap_alloc(sizeof(cc_var_t
) + (len
+1)*sizeof(WCHAR
));
816 memcpy(new_v
, v
, sizeof(*v
));
817 memcpy(new_v
->name
, name
, (len
+1)*sizeof(WCHAR
));
818 new_v
->name_len
= len
;
819 new_v
->next
= cc
->vars
;
824 static cc_var_t
*find_cc_var(cc_ctx_t
*cc
, const WCHAR
*name
, unsigned name_len
)
828 for(iter
= cc
->vars
; iter
; iter
= iter
->next
) {
829 if(iter
->name_len
== name_len
&& !memcmp(iter
->name
, name
, name_len
*sizeof(WCHAR
)))
836 static int init_cc(parser_ctx_t
*ctx
)
841 static const WCHAR _win32W
[] = {'_','w','i','n','3','2',0};
842 static const WCHAR _win64W
[] = {'_','w','i','n','6','4',0};
843 static const WCHAR _x86W
[] = {'_','x','8','6',0};
844 static const WCHAR _amd64W
[] = {'_','a','m','d','6','4',0};
845 static const WCHAR _jscriptW
[] = {'_','j','s','c','r','i','p','t',0};
846 static const WCHAR _jscript_buildW
[] = {'_','j','s','c','r','i','p','t','_','b','u','i','l','d',0};
847 static const WCHAR _jscript_versionW
[] = {'_','j','s','c','r','i','p','t','_','v','e','r','s','i','o','n',0};
852 cc
= heap_alloc(sizeof(cc_ctx_t
));
854 return lex_error(ctx
, E_OUTOFMEMORY
);
858 v
.u
.b
= VARIANT_TRUE
;
859 if(!add_cc_var(cc
, _jscriptW
, &v
)
860 || !add_cc_var(cc
, sizeof(void*) == 8 ? _win64W
: _win32W
, &v
)
861 || !add_cc_var(cc
, sizeof(void*) == 8 ? _amd64W
: _x86W
, &v
)) {
863 return lex_error(ctx
, E_OUTOFMEMORY
);
867 v
.u
.n
= JSCRIPT_BUILD_VERSION
;
868 if(!add_cc_var(cc
, _jscript_buildW
, &v
)) {
870 return lex_error(ctx
, E_OUTOFMEMORY
);
873 v
.u
.n
= JSCRIPT_MAJOR_VERSION
+ (DOUBLE
)JSCRIPT_MINOR_VERSION
/10.0;
874 if(!add_cc_var(cc
, _jscript_versionW
, &v
)) {
876 return lex_error(ctx
, E_OUTOFMEMORY
);
879 ctx
->script
->cc
= cc
;
883 static int cc_token(parser_ctx_t
*ctx
, void *lval
)
888 static const WCHAR cc_onW
[] = {'c','c','_','o','n',0};
889 static const WCHAR setW
[] = {'s','e','t',0};
890 static const WCHAR elifW
[] = {'e','l','i','f',0};
891 static const WCHAR endW
[] = {'e','n','d',0};
895 if(!check_keyword(ctx
, cc_onW
, NULL
))
898 if(!check_keyword(ctx
, setW
, NULL
)) {
899 FIXME("@set not implemented\n");
900 return lex_error(ctx
, E_NOTIMPL
);
903 if(!check_keyword(ctx
, ifW
, NULL
)) {
904 FIXME("@if not implemented\n");
905 return lex_error(ctx
, E_NOTIMPL
);
908 if(!check_keyword(ctx
, elifW
, NULL
)) {
909 FIXME("@elif not implemented\n");
910 return lex_error(ctx
, E_NOTIMPL
);
913 if(!check_keyword(ctx
, elseW
, NULL
)) {
914 FIXME("@else not implemented\n");
915 return lex_error(ctx
, E_NOTIMPL
);
918 if(!check_keyword(ctx
, endW
, NULL
)) {
919 FIXME("@end not implemented\n");
920 return lex_error(ctx
, E_NOTIMPL
);
924 return lex_error(ctx
, JS_E_DISABLED_CC
);
926 while(ctx
->ptr
+id_len
< ctx
->end
&& is_identifier_char(ctx
->ptr
[id_len
]))
931 TRACE("var %s\n", debugstr_wn(ctx
->ptr
, id_len
));
933 var
= find_cc_var(ctx
->script
->cc
, ctx
->ptr
, id_len
);
935 if(!var
|| var
->is_num
) {
936 *(literal_t
**)lval
= new_double_literal(ctx
, var
? var
->u
.n
: NAN
);
937 return tNumericLiteral
;
940 *(literal_t
**)lval
= new_boolean_literal(ctx
, var
->u
.b
);
941 return tBooleanLiteral
;
944 int parser_lex(void *lval
, parser_ctx_t
*ctx
)
948 ctx
->nl
= ctx
->ptr
== ctx
->begin
;
951 ret
= next_token(ctx
, lval
);
952 } while(ret
== '@' && !(ret
= cc_token(ctx
, lval
)));
957 literal_t
*parse_regexp(parser_ctx_t
*ctx
)
959 const WCHAR
*re
, *flags_ptr
;
960 BOOL in_class
= FALSE
;
967 while(*--ctx
->ptr
!= '/');
969 /* Simple regexp pre-parser; '/' if used in char class does not terminate regexp literal */
971 while(ctx
->ptr
< ctx
->end
) {
972 if(*ctx
->ptr
== '\\') {
973 if(++ctx
->ptr
== ctx
->end
)
976 if(*ctx
->ptr
== '\n')
990 if(ctx
->ptr
== ctx
->end
|| *ctx
->ptr
!= '/') {
991 WARN("pre-parsing failed\n");
995 re_len
= ctx
->ptr
-re
;
997 flags_ptr
= ++ctx
->ptr
;
998 while(ctx
->ptr
< ctx
->end
&& isalnumW(*ctx
->ptr
))
1001 hres
= parse_regexp_flags(flags_ptr
, ctx
->ptr
-flags_ptr
, &flags
);
1005 ret
= parser_alloc(ctx
, sizeof(literal_t
));
1006 ret
->type
= LT_REGEXP
;
1007 ret
->u
.regexp
.str
= re
;
1008 ret
->u
.regexp
.str_len
= re_len
;
1009 ret
->u
.regexp
.flags
= flags
;