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
)
107 static int check_keyword(parser_ctx_t
*ctx
, const WCHAR
*word
, const WCHAR
**lval
)
109 const WCHAR
*p1
= ctx
->ptr
;
110 const WCHAR
*p2
= word
;
112 while(p1
< ctx
->end
&& *p2
) {
119 if(*p2
|| (p1
< ctx
->end
&& isalnumW(*p1
)))
127 /* ECMA-262 3rd Edition 7.3 */
128 static BOOL
is_endline(WCHAR c
)
130 return c
== '\n' || c
== '\r' || c
== 0x2028 || c
== 0x2029;
133 static BOOL
is_identifier_char(WCHAR c
)
135 return isalnumW(c
) || c
== '$' || c
== '_' || c
== '\\';
138 static int hex_to_int(WCHAR c
)
140 if('0' <= c
&& c
<= '9')
143 if('a' <= c
&& c
<= 'f')
146 if('A' <= c
&& c
<= 'F')
152 static int check_keywords(parser_ctx_t
*ctx
, const WCHAR
**lval
)
154 int min
= 0, max
= sizeof(keywords
)/sizeof(keywords
[0])-1, r
, i
;
159 r
= check_keyword(ctx
, keywords
[i
].word
, lval
);
161 return keywords
[i
].token
;
172 static void skip_spaces(parser_ctx_t
*ctx
)
174 while(ctx
->ptr
< ctx
->end
&& isspaceW(*ctx
->ptr
)) {
175 if(is_endline(*ctx
->ptr
++))
180 static BOOL
skip_html_comment(parser_ctx_t
*ctx
)
182 const WCHAR html_commentW
[] = {'<','!','-','-',0};
184 if(!ctx
->is_html
|| ctx
->ptr
+3 >= ctx
->end
||
185 memcmp(ctx
->ptr
, html_commentW
, sizeof(WCHAR
)*4))
189 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
++));
194 static BOOL
skip_comment(parser_ctx_t
*ctx
)
196 if(ctx
->ptr
+1 >= ctx
->end
|| *ctx
->ptr
!= '/')
199 switch(ctx
->ptr
[1]) {
202 while(ctx
->ptr
+1 < ctx
->end
&& (ctx
->ptr
[0] != '*' || ctx
->ptr
[1] != '/'))
205 if(ctx
->ptr
[0] == '*' && ctx
->ptr
[1] == '/') {
208 WARN("unexpected end of file (missing end of comment)\n");
214 while(ctx
->ptr
< ctx
->end
&& !is_endline(*ctx
->ptr
))
224 static BOOL
unescape(WCHAR
*str
)
264 i
= hex_to_int(*++p
);
269 i
= hex_to_int(*++p
);
275 i
= hex_to_int(*++p
);
280 i
= hex_to_int(*++p
);
285 i
= hex_to_int(*++p
);
290 i
= hex_to_int(*++p
);
299 c
= c
*10 + (*p
++ - '0');
315 static int parse_identifier(parser_ctx_t
*ctx
, const WCHAR
**ret
)
317 const WCHAR
*ptr
= ctx
->ptr
++;
321 while(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
))
326 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
327 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
330 /* FIXME: unescape */
334 static int parse_string_literal(parser_ctx_t
*ctx
, const WCHAR
**ret
, WCHAR endch
)
336 const WCHAR
*ptr
= ++ctx
->ptr
;
340 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= endch
) {
341 if(*ctx
->ptr
++ == '\\')
345 if(ctx
->ptr
== ctx
->end
) {
346 WARN("unexpected end of file\n");
347 return lex_error(ctx
, E_FAIL
);
352 *ret
= wstr
= parser_alloc(ctx
, (len
+1)*sizeof(WCHAR
));
353 memcpy(wstr
, ptr
, (len
+1)*sizeof(WCHAR
));
358 if(!unescape(wstr
)) {
359 WARN("unescape failed\n");
360 return lex_error(ctx
, E_FAIL
);
363 return tStringLiteral
;
366 static literal_t
*alloc_int_literal(parser_ctx_t
*ctx
, LONG l
)
368 literal_t
*ret
= parser_alloc(ctx
, sizeof(literal_t
));
376 static int parse_double_literal(parser_ctx_t
*ctx
, LONG int_part
, literal_t
**literal
)
381 if(ctx
->ptr
== ctx
->end
|| (!isdigitW(*ctx
->ptr
) &&
382 *ctx
->ptr
!='.' && *ctx
->ptr
!='e' && *ctx
->ptr
!='E')) {
383 ERR("Illegal character\n");
388 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
389 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
390 if(d
>LONGLONG_MAX
/10 || hlp
<0) {
397 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
402 if(*ctx
->ptr
== '.') ctx
->ptr
++;
404 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
405 hlp
= d
*10 + *(ctx
->ptr
++) - '0';
406 if(d
>LONGLONG_MAX
/10 || hlp
<0)
412 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
415 if(ctx
->ptr
< ctx
->end
&& (*ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')) {
419 if(ctx
->ptr
< ctx
->end
) {
420 if(*ctx
->ptr
== '+') {
422 }else if(*ctx
->ptr
== '-') {
425 }else if(!isdigitW(*ctx
->ptr
)) {
426 WARN("Expected exponent part\n");
427 return lex_error(ctx
, E_FAIL
);
431 if(ctx
->ptr
== ctx
->end
) {
432 WARN("unexpected end of file\n");
433 return lex_error(ctx
, E_FAIL
);
436 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
)) {
437 if(e
> INT_MAX
/10 || (e
= e
*10 + *ctx
->ptr
++ - '0')<0)
442 if(exp
<0 && e
<0 && e
+exp
>0) exp
= INT_MIN
;
443 else if(exp
>0 && e
>0 && e
+exp
<0) exp
= INT_MAX
;
447 *literal
= parser_alloc(ctx
, sizeof(literal_t
));
448 (*literal
)->vt
= VT_R8
;
449 (*literal
)->u
.dval
= (double)d
*pow(10, exp
);
451 return tNumericLiteral
;
454 static int parse_numeric_literal(parser_ctx_t
*ctx
, literal_t
**literal
)
458 l
= *ctx
->ptr
++ - '0';
459 if(ctx
->ptr
== ctx
->end
) {
460 *literal
= alloc_int_literal(ctx
, l
);
461 return tNumericLiteral
;
465 if(*ctx
->ptr
== 'x' || *ctx
->ptr
== 'X') {
466 if(++ctx
->ptr
== ctx
->end
) {
467 ERR("unexpexted end of file\n");
471 while(ctx
->ptr
< ctx
->end
&& (d
= hex_to_int(*ctx
->ptr
)) != -1) {
476 if(ctx
->ptr
< ctx
->end
&& is_identifier_char(*ctx
->ptr
)) {
477 WARN("unexpected identifier char\n");
478 return lex_error(ctx
, E_FAIL
);
481 *literal
= alloc_int_literal(ctx
, l
);
482 return tNumericLiteral
;
485 if(isdigitW(*ctx
->ptr
) || is_identifier_char(*ctx
->ptr
)) {
486 WARN("wrong char after zero\n");
487 return lex_error(ctx
, E_FAIL
);
490 *literal
= alloc_int_literal(ctx
, 0);
493 while(ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
495 d
= l
*10 + *(ctx
->ptr
)-'0';
497 /* Check for integer overflow */
498 if (l
> INT_MAX
/10 || d
< 0)
499 return parse_double_literal(ctx
, l
, literal
);
505 if(ctx
->ptr
< ctx
->end
) {
506 if(*ctx
->ptr
== '.' || *ctx
->ptr
== 'e' || *ctx
->ptr
== 'E')
507 return parse_double_literal(ctx
, l
, literal
);
509 if(is_identifier_char(*ctx
->ptr
)) {
510 WARN("unexpected identifier char\n");
511 return lex_error(ctx
, E_FAIL
);
515 *literal
= alloc_int_literal(ctx
, l
);
516 return tNumericLiteral
;
519 int parser_lex(void *lval
, parser_ctx_t
*ctx
)
523 ctx
->nl
= ctx
->ptr
== ctx
->begin
;
527 if(ctx
->ptr
== ctx
->end
)
529 }while(skip_comment(ctx
) || skip_html_comment(ctx
));
531 if(isalphaW(*ctx
->ptr
)) {
532 ret
= check_keywords(ctx
, lval
);
536 return parse_identifier(ctx
, lval
);
539 if(isdigitW(*ctx
->ptr
))
540 return parse_numeric_literal(ctx
, lval
);
556 *(const WCHAR
**)lval
= ctx
->ptr
++;
560 if(++ctx
->ptr
< ctx
->end
&& isdigitW(*ctx
->ptr
))
561 return parse_double_literal(ctx
, 0, lval
);
565 if(++ctx
->ptr
== ctx
->end
) {
566 *(int*)lval
= EXPR_LESS
;
573 *(int*)lval
= EXPR_LESSEQ
;
576 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* <<= */
578 *(int*)lval
= EXPR_ASSIGNLSHIFT
;
581 *(int*)lval
= EXPR_LSHIFT
;
584 *(int*)lval
= EXPR_LESS
;
589 if(++ctx
->ptr
== ctx
->end
) { /* > */
590 *(int*)lval
= EXPR_GREATER
;
597 *(int*)lval
= EXPR_GREATEREQ
;
600 if(++ctx
->ptr
< ctx
->end
) {
601 if(*ctx
->ptr
== '=') { /* >>= */
603 *(int*)lval
= EXPR_ASSIGNRSHIFT
;
606 if(*ctx
->ptr
== '>') { /* >>> */
607 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* >>>= */
609 *(int*)lval
= EXPR_ASSIGNRRSHIFT
;
612 *(int*)lval
= EXPR_RRSHIFT
;
616 *(int*)lval
= EXPR_RSHIFT
;
619 *(int*)lval
= EXPR_GREATER
;
625 if(ctx
->ptr
< ctx
->end
) {
632 *(int*)lval
= EXPR_ASSIGNADD
;
640 if(ctx
->ptr
< ctx
->end
) {
642 case '-': /* -- or --> */
644 if(ctx
->is_html
&& ctx
->nl
&& ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '>') {
651 *(int*)lval
= EXPR_ASSIGNSUB
;
658 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* *= */
660 *(int*)lval
= EXPR_ASSIGNMUL
;
666 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* %= */
668 *(int*)lval
= EXPR_ASSIGNMOD
;
674 if(++ctx
->ptr
< ctx
->end
) {
678 *(int*)lval
= EXPR_ASSIGNAND
;
688 if(++ctx
->ptr
< ctx
->end
) {
692 *(int*)lval
= EXPR_ASSIGNOR
;
702 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* ^= */
704 *(int*)lval
= EXPR_ASSIGNXOR
;
710 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* != */
711 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* !== */
713 *(int*)lval
= EXPR_NOTEQEQ
;
716 *(int*)lval
= EXPR_NOTEQ
;
722 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* == */
723 if(++ctx
->ptr
< ctx
->end
&& *ctx
->ptr
== '=') { /* === */
725 *(int*)lval
= EXPR_EQEQ
;
728 *(int*)lval
= EXPR_EQ
;
734 if(++ctx
->ptr
< ctx
->end
) {
735 if(*ctx
->ptr
== '=') { /* /= */
737 *(int*)lval
= EXPR_ASSIGNDIV
;
745 return parse_string_literal(ctx
, lval
, *ctx
->ptr
);
749 return parse_identifier(ctx
, lval
);
752 WARN("unexpected char '%c' %d\n", *ctx
->ptr
, *ctx
->ptr
);
756 static void add_object_literal(parser_ctx_t
*ctx
, DispatchEx
*obj
)
758 obj_literal_t
*literal
= parser_alloc(ctx
, sizeof(obj_literal_t
));
761 literal
->next
= ctx
->obj_literals
;
762 ctx
->obj_literals
= literal
;
765 literal_t
*parse_regexp(parser_ctx_t
*ctx
)
767 const WCHAR
*re
, *flags
;
776 while(ctx
->ptr
< ctx
->end
&& *ctx
->ptr
!= '/') {
777 if(*ctx
->ptr
++ == '\\' && ctx
->ptr
< ctx
->end
)
781 if(ctx
->ptr
== ctx
->end
) {
782 WARN("unexpected end of file\n");
786 re_len
= ctx
->ptr
-re
;
789 while(ctx
->ptr
< ctx
->end
&& isalnumW(*ctx
->ptr
))
792 hres
= create_regexp_str(ctx
->script
, re
, re_len
, flags
, ctx
->ptr
-flags
, ®exp
);
796 add_object_literal(ctx
, regexp
);
798 ret
= parser_alloc(ctx
, sizeof(literal_t
));
799 ret
->vt
= VT_DISPATCH
;
800 ret
->u
.disp
= (IDispatch
*)_IDispatchEx_(regexp
);