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
)
276 i
= hex_to_int(*++p
);
281 i
= hex_to_int(*++p
);
287 i
= hex_to_int(*++p
);
292 i
= hex_to_int(*++p
);
297 i
= hex_to_int(*++p
);
302 i
= hex_to_int(*++p
);
311 c
= c
*8 + (*p
++ - '0');
313 c
= c
*8 + (*p
++ - '0');
329 static int parse_identifier(parser_ctx_t
*ctx
, const WCHAR
**ret
)
331 const WCHAR
*ptr
= ctx
->ptr
++;
335 while(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
))
340 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
341 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
344 /* FIXME: unescape */
348 static int parse_string_literal(parser_ctx_t
*ctx
, const WCHAR
**ret
, WCHAR endch
)
350 const WCHAR
*ptr
= ++ctx
->ptr
;
354 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= endch
) {
355 if(*ctx
->ptr
++ == '\\')
359 if(ctx
->ptr
== ctx
->end
)
360 return lex_error(ctx
, JS_E_UNTERMINATED_STRING
);
364 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
365 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
370 if(!unescape(wstr
)) {
371 WARN("unescape failed\n");
372 return lex_error(ctx
, E_FAIL
);
375 return tStringLiteral
;
378 static literal_t
*new_int_literal(parser_ctx_t
*ctx
, LONG l
)
380 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
388 static literal_t
*new_double_literal(parser_ctx_t
*ctx
, DOUBLE d
)
390 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
392 ret
->type
= LT_DOUBLE
;
397 literal_t
*new_boolean_literal(parser_ctx_t
*ctx
, VARIANT_BOOL bval
)
399 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
407 static int parse_double_literal(parser_ctx_t
*ctx
, LONG int_part
, literal_t
**literal
)
412 if(ctx
->ptr
== ctx
->end
|| (!isdigitW(*ctx
->ptr
) &&
413 *ctx
->ptr
!='.' && *ctx
->ptr
!='e' && *ctx
->ptr
!='E')) {
414 ERR("Illegal character\n");
419 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
420 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
421 if(d
>LONGLONG_MAX
/10 || hlp
<0) {
428 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
433 if(*ctx
->ptr
== '.') ctx
->ptr
++;
435 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
436 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
437 if(d
>LONGLONG_MAX
/10 || hlp
<0)
443 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
446 if(ctx
->ptr
< ctx
->end
&& (*ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')) {
450 if(ctx
->ptr
< ctx
->end
) {
451 if(*ctx
->ptr
== '+') {
453 }else if(*ctx
->ptr
== '-') {
456 }else if(!isdigitW(*ctx
->ptr
)) {
457 WARN("Expected exponent part\n");
458 return lex_error(ctx
, E_FAIL
);
462 if(ctx
->ptr
== ctx
->end
) {
463 WARN("unexpected end of file\n");
464 return lex_error(ctx
, E_FAIL
);
467 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
468 if(e
> INT_MAX
/10 || (e
= e
*10 + *ctx
->ptr
++ - '0')<0)
473 if(exp
<0 && e
<0 && e
+exp
>0) exp
= INT_MIN
;
474 else if(exp
>0 && e
>0 && e
+exp
<0) exp
= INT_MAX
;
478 *literal
= new_double_literal(ctx
, (DOUBLE
)d
*pow(10, exp
));
479 return tNumericLiteral
;
482 static int parse_numeric_literal(parser_ctx_t
*ctx
, literal_t
**literal
)
486 l
= *ctx
->ptr
++ - '0';
487 if(ctx
->ptr
== ctx
->end
) {
488 *literal
= new_int_literal(ctx
, l
);
489 return tNumericLiteral
;
493 if(*ctx
->ptr
== 'x' || *ctx
->ptr
== 'X') {
494 if(++ctx
->ptr
== ctx
->end
) {
495 ERR("unexpexted end of file\n");
499 while(ctx
->ptr
< ctx
->end
&& (d
= hex_to_int(*ctx
->ptr
)) != -1) {
504 if(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
)) {
505 WARN("unexpected identifier char\n");
506 return lex_error(ctx
, E_FAIL
);
509 *literal
= new_int_literal(ctx
, l
);
510 return tNumericLiteral
;
513 if(isdigitW(*ctx
->ptr
) || is_identifier_char(*ctx
->ptr
)) {
514 WARN("wrong char after zero\n");
515 return lex_error(ctx
, E_FAIL
);
518 *literal
= new_int_literal(ctx
, 0);
521 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
523 d
= l
*10 + *(ctx
->ptr
)-'0';
525 /* Check for integer overflow */
526 if (l
> INT_MAX
/10 || d
< 0)
527 return parse_double_literal(ctx
, l
, literal
);
533 if(ctx
->ptr
< ctx
->end
) {
534 if(*ctx
->ptr
== '.' || *ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')
535 return parse_double_literal(ctx
, l
, literal
);
537 if(is_identifier_char(*ctx
->ptr
)) {
538 WARN("unexpected identifier char\n");
539 return lex_error(ctx
, E_FAIL
);
543 *literal
= new_int_literal(ctx
, l
);
544 return tNumericLiteral
;
547 static int next_token(parser_ctx_t
*ctx
, void *lval
)
551 if(ctx
->ptr
== ctx
->end
)
553 }while(skip_comment(ctx
) || skip_html_comment(ctx
));
555 if(isalphaW(*ctx
->ptr
)) {
556 int ret
= check_keywords(ctx
, lval
);
560 return parse_identifier(ctx
, lval
);
563 if(isdigitW(*ctx
->ptr
))
564 return parse_numeric_literal(ctx
, lval
);
580 *(const WCHAR
**)lval
= ctx
->ptr
++;
584 if(++ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
585 return parse_double_literal(ctx
, 0, lval
);
589 if(++ctx
->ptr
== ctx
->end
) {
590 *(int*)lval
= EXPR_LESS
;
597 *(int*)lval
= EXPR_LESSEQ
;
600 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* <<= */
602 *(int*)lval
= EXPR_ASSIGNLSHIFT
;
605 *(int*)lval
= EXPR_LSHIFT
;
608 *(int*)lval
= EXPR_LESS
;
613 if(++ctx
->ptr
== ctx
->end
) { /* > */
614 *(int*)lval
= EXPR_GREATER
;
621 *(int*)lval
= EXPR_GREATEREQ
;
624 if(++ctx
->ptr
< ctx
->end
) {
625 if(*ctx
->ptr
== '=') { /* >>= */
627 *(int*)lval
= EXPR_ASSIGNRSHIFT
;
630 if(*ctx
->ptr
== '>') { /* >>> */
631 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* >>>= */
633 *(int*)lval
= EXPR_ASSIGNRRSHIFT
;
636 *(int*)lval
= EXPR_RRSHIFT
;
640 *(int*)lval
= EXPR_RSHIFT
;
643 *(int*)lval
= EXPR_GREATER
;
649 if(ctx
->ptr
< ctx
->end
) {
656 *(int*)lval
= EXPR_ASSIGNADD
;
664 if(ctx
->ptr
< ctx
->end
) {
666 case '-': /* -- or --> */
668 if(ctx
->is_html
&& ctx
->nl
&& ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '>') {
675 *(int*)lval
= EXPR_ASSIGNSUB
;
682 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* *= */
684 *(int*)lval
= EXPR_ASSIGNMUL
;
690 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* %= */
692 *(int*)lval
= EXPR_ASSIGNMOD
;
698 if(++ctx
->ptr
< ctx
->end
) {
702 *(int*)lval
= EXPR_ASSIGNAND
;
712 if(++ctx
->ptr
< ctx
->end
) {
716 *(int*)lval
= EXPR_ASSIGNOR
;
726 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* ^= */
728 *(int*)lval
= EXPR_ASSIGNXOR
;
734 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* != */
735 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* !== */
737 *(int*)lval
= EXPR_NOTEQEQ
;
740 *(int*)lval
= EXPR_NOTEQ
;
746 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* == */
747 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* === */
749 *(int*)lval
= EXPR_EQEQ
;
752 *(int*)lval
= EXPR_EQ
;
758 if(++ctx
->ptr
< ctx
->end
) {
759 if(*ctx
->ptr
== '=') { /* /= */
761 *(int*)lval
= EXPR_ASSIGNDIV
;
769 return parse_string_literal(ctx
, lval
, *ctx
->ptr
);
773 return parse_identifier(ctx
, lval
);
779 WARN("unexpected char '%c' %d\n", *ctx
->ptr
, *ctx
->ptr
);
789 struct _cc_var_t
*next
;
794 void release_cc(cc_ctx_t
*cc
)
796 cc_var_t
*iter
, *next
;
798 for(iter
= cc
->vars
; iter
; iter
= next
) {
806 static BOOL
add_cc_var(cc_ctx_t
*cc
, const WCHAR
*name
, cc_var_t
*v
)
813 new_v
= heap_alloc(sizeof(cc_var_t
) + (len
+1)*sizeof(WCHAR
));
817 memcpy(new_v
, v
, sizeof(*v
));
818 memcpy(new_v
->name
, name
, (len
+1)*sizeof(WCHAR
));
819 new_v
->name_len
= len
;
820 new_v
->next
= cc
->vars
;
825 static cc_var_t
*find_cc_var(cc_ctx_t
*cc
, const WCHAR
*name
, unsigned name_len
)
829 for(iter
= cc
->vars
; iter
; iter
= iter
->next
) {
830 if(iter
->name_len
== name_len
&& !memcmp(iter
->name
, name
, name_len
*sizeof(WCHAR
)))
837 static int init_cc(parser_ctx_t
*ctx
)
842 static const WCHAR _win32W
[] = {'_','w','i','n','3','2',0};
843 static const WCHAR _win64W
[] = {'_','w','i','n','6','4',0};
844 static const WCHAR _x86W
[] = {'_','x','8','6',0};
845 static const WCHAR _amd64W
[] = {'_','a','m','d','6','4',0};
846 static const WCHAR _jscriptW
[] = {'_','j','s','c','r','i','p','t',0};
847 static const WCHAR _jscript_buildW
[] = {'_','j','s','c','r','i','p','t','_','b','u','i','l','d',0};
848 static const WCHAR _jscript_versionW
[] = {'_','j','s','c','r','i','p','t','_','v','e','r','s','i','o','n',0};
853 cc
= heap_alloc(sizeof(cc_ctx_t
));
855 return lex_error(ctx
, E_OUTOFMEMORY
);
859 v
.u
.b
= VARIANT_TRUE
;
860 if(!add_cc_var(cc
, _jscriptW
, &v
)
861 || !add_cc_var(cc
, sizeof(void*) == 8 ? _win64W
: _win32W
, &v
)
862 || !add_cc_var(cc
, sizeof(void*) == 8 ? _amd64W
: _x86W
, &v
)) {
864 return lex_error(ctx
, E_OUTOFMEMORY
);
868 v
.u
.n
= JSCRIPT_BUILD_VERSION
;
869 if(!add_cc_var(cc
, _jscript_buildW
, &v
)) {
871 return lex_error(ctx
, E_OUTOFMEMORY
);
874 v
.u
.n
= JSCRIPT_MAJOR_VERSION
+ (DOUBLE
)JSCRIPT_MINOR_VERSION
/10.0;
875 if(!add_cc_var(cc
, _jscript_versionW
, &v
)) {
877 return lex_error(ctx
, E_OUTOFMEMORY
);
880 ctx
->script
->cc
= cc
;
884 static int cc_token(parser_ctx_t
*ctx
, void *lval
)
889 static const WCHAR cc_onW
[] = {'c','c','_','o','n',0};
890 static const WCHAR setW
[] = {'s','e','t',0};
891 static const WCHAR elifW
[] = {'e','l','i','f',0};
892 static const WCHAR endW
[] = {'e','n','d',0};
896 if(!check_keyword(ctx
, cc_onW
, NULL
))
899 if(!check_keyword(ctx
, setW
, NULL
)) {
900 FIXME("@set not implemented\n");
901 return lex_error(ctx
, E_NOTIMPL
);
904 if(!check_keyword(ctx
, ifW
, NULL
)) {
905 FIXME("@if not implemented\n");
906 return lex_error(ctx
, E_NOTIMPL
);
909 if(!check_keyword(ctx
, elifW
, NULL
)) {
910 FIXME("@elif not implemented\n");
911 return lex_error(ctx
, E_NOTIMPL
);
914 if(!check_keyword(ctx
, elseW
, NULL
)) {
915 FIXME("@else not implemented\n");
916 return lex_error(ctx
, E_NOTIMPL
);
919 if(!check_keyword(ctx
, endW
, NULL
)) {
920 FIXME("@end not implemented\n");
921 return lex_error(ctx
, E_NOTIMPL
);
925 return lex_error(ctx
, JS_E_DISABLED_CC
);
927 while(ctx
->ptr
+id_len
< ctx
->end
&& is_identifier_char(ctx
->ptr
[id_len
]))
932 TRACE("var %s\n", debugstr_wn(ctx
->ptr
, id_len
));
934 var
= find_cc_var(ctx
->script
->cc
, ctx
->ptr
, id_len
);
936 if(!var
|| var
->is_num
) {
937 *(literal_t
**)lval
= new_double_literal(ctx
, var
? var
->u
.n
: ret_nan());
938 return tNumericLiteral
;
941 *(literal_t
**)lval
= new_boolean_literal(ctx
, var
->u
.b
);
942 return tBooleanLiteral
;
945 int parser_lex(void *lval
, parser_ctx_t
*ctx
)
949 ctx
->nl
= ctx
->ptr
== ctx
->begin
;
952 ret
= next_token(ctx
, lval
);
953 } while(ret
== '@' && !(ret
= cc_token(ctx
, lval
)));
958 literal_t
*parse_regexp(parser_ctx_t
*ctx
)
960 const WCHAR
*re
, *flags_ptr
;
967 while(*ctx
->ptr
!= '/')
971 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= '/') {
972 if(*ctx
->ptr
++ == '\\' && ctx
->ptr
< ctx
->end
)
976 if(ctx
->ptr
== ctx
->end
) {
977 WARN("unexpected end of file\n");
981 re_len
= ctx
->ptr
-re
;
983 flags_ptr
= ++ctx
->ptr
;
984 while(ctx
->ptr
< ctx
->end
&& isalnumW(*ctx
->ptr
))
987 hres
= parse_regexp_flags(flags_ptr
, ctx
->ptr
-flags_ptr
, &flags
);
991 ret
= parser_alloc(ctx
, sizeof(literal_t
));
992 ret
->type
= LT_REGEXP
;
993 ret
->u
.regexp
.str
= re
;
994 ret
->u
.regexp
.str_len
= re_len
;
995 ret
->u
.regexp
.flags
= flags
;