3 Routines for compiling Flash2 AVM2 ABC Actionscript
5 Extension module for the rfxswf library.
6 Part of the swftools package.
8 Copyright (c) 2008 Matthias Kramm <kramm@quiss.org>
10 This program is free software; you can redistribute it and/or modify
11 it under the terms of the GNU General Public License as published by
12 the Free Software Foundation; either version 2 of the License, or
13 (at your option) any later version.
15 This program is distributed in the hope that it will be useful,
16 but WITHOUT ANY WARRANTY; without even the implied warranty of
17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 GNU General Public License for more details.
20 You should have received a copy of the GNU General Public License
21 along with this program; if not, write to the Free Software
22 Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
31 #include "tokenizer.h"
35 int as3_verbosity = 1;
36 unsigned int as3_tokencount = 0;
38 void as3_error(const char*format, ...)
45 va_start(arglist, format);
46 vsprintf(buf, format, arglist);
48 fprintf(stderr, "%s:%d:%d: error: %s\n", current_filename_short, current_line, current_column, buf);
52 void as3_warning(const char*format, ...)
59 va_start(arglist, format);
60 vsprintf(buf, format, arglist);
62 fprintf(stderr, "%s:%d:%d: warning: %s\n", current_filename_short, current_line, current_column, buf);
65 void as3_softwarning(const char*format, ...)
72 va_start(arglist, format);
73 vsprintf(buf, format, arglist);
75 fprintf(stderr, "%s:%d:%d: warning: %s\n", current_filename_short, current_line, current_column, buf);
78 static void dbg(const char*format, ...)
85 va_start(arglist, format);
86 vsprintf(buf, format, arglist);
89 while(l && buf[l-1]=='\n') {
93 printf("(tokenizer) ");
98 #ifndef YY_CURRENT_BUFFER
99 #define YY_CURRENT_BUFFER yy_current_buffer
102 static void*as3_buffer=0;
103 static int as3_buffer_pos=0;
104 static int as3_buffer_len=0;
105 void as3_file_input(FILE*fi)
110 void as3_buffer_input(void*buffer, int len)
113 syntaxerror("trying to parse zero bytearray");
115 as3_buffer_len = len;
120 #define YY_INPUT(buf,result,max_size) { \
123 while((result = fread(buf, 1, max_size, as3_in))==0 && ferror(as3_in)) \
124 { if(errno != EINTR) {YY_FATAL_ERROR("input in flex scanner failed"); break;} \
125 errno=0; clearerr(as3_in); \
128 int to_read = max_size; \
129 if(to_read + as3_buffer_pos > as3_buffer_len) \
130 to_read = as3_buffer_len - as3_buffer_pos; \
131 memcpy(buf, as3_buffer+as3_buffer_pos, to_read); \
132 as3_buffer_pos += to_read; \
137 void handleInclude(char*text, int len, char quotes)
141 char*p1 = strchr(text, '"');
142 char*p2 = strrchr(text, '"');
143 if(!p1 || !p2 || p1==p2) {
144 syntaxerror("Invalid include in line %d\n", current_line);
147 filename = strdup(p1+1);
151 while(!strchr(" \n\r\t", text[i1])) i1++;
153 while(strchr(" \n\r\t", text[i1])) i1++;
154 while(strchr(" \n\r\t", text[i2-1])) i2--;
155 if(i2!=len) text[i2]=0;
156 filename = strdup(&text[i1]);
159 char*fullfilename = find_file(filename, 1);
160 enter_file2(filename, fullfilename, YY_CURRENT_BUFFER);
161 yyin = fopen(fullfilename, "rb");
163 syntaxerror("Couldn't open include file \"%s\"\n", fullfilename);
166 yy_switch_to_buffer(yy_create_buffer( yyin, YY_BUF_SIZE ) );
167 //BEGIN(INITIAL); keep context
170 static int do_unescape(const char*s, const char*end, char*n)
176 if(o) o[len] = *s;len++;
181 if(s==end) syntaxerror("invalid \\ at end of string");
183 /* handle the various line endings (mac, dos, unix) */
196 case '\\': if(o) o[len] = '\\';s++;len++; break;
197 case '"': if(o) o[len] = '"';s++;len++; break;
198 case '\'': if(o) o[len] = '\'';s++;len++; break;
199 case 'b': if(o) o[len] = '\b';s++;len++; break;
200 case 'f': if(o) o[len] = '\f';s++;len++; break;
201 case 'n': if(o) o[len] = '\n';s++;len++; break;
202 case 'r': if(o) o[len] = '\r';s++;len++; break;
203 case 't': if(o) o[len] = '\t';s++;len++; break;
204 case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': {
207 while(strchr("01234567", *s) && nr<3 && s<end) {
214 syntaxerror("octal number out of range (0-255): %d", num);
215 if(o) o[len] = num;len++;
218 case 'x': case 'u': {
227 if(s==end) syntaxerror("invalid \\u or \\x at end of string");
230 if(s==end) syntaxerror("invalid \\u{ at end of string");
235 while(strchr("0123456789abcdefABCDEF", *s) && (bracket || nr < max) && s<end) {
237 if(*s>='0' && *s<='9') num |= *s - '0';
238 if(*s>='a' && *s<='f') num |= *s - 'a' + 10;
239 if(*s>='A' && *s<='F') num |= *s - 'A' + 10;
244 if(*s=='}' && s<end) {
247 syntaxerror("missing terminating '}'");
251 char*utf8 = getUTF8(num);
253 if(o) o[len] = *utf8;utf8++;len++;
257 syntaxerror("byte out of range (0-255): %d", num);
258 if(o) o[len] = num;len++;
277 static string_t string_unescape(const char*in, int l)
280 const char*end = &in[l];
282 int len = do_unescape(s, end, 0);
283 char*n = (char*)malloc(len+1);
284 do_unescape(s, end, n);
285 string_t out = string_new(n, len);
289 static void handleCData(char*s, int len)
291 a3_lval.str.str = s+9; // <![CDATA[
292 a3_lval.str.len = len-9-3;// ]]>
293 a3_lval.str.str = strdup_n(a3_lval.str.str, a3_lval.str.len);
296 static void handleString(char*s, int len)
299 // don't bother decoding strings in pass 1
300 memset(&a3_lval, 0, sizeof(a3_lval));
305 if(s[len-1]!='"') syntaxerror("String doesn't end with '\"'");
308 else if(s[0]=='\'') {
309 if(s[len-1]!='\'') syntaxerror("String doesn't end with '\"'");
312 else syntaxerror("String incorrectly terminated");
315 a3_lval.str = string_unescape(s, len);
319 char start_of_expression;
321 static inline int m(int type)
323 a3_lval.token = type;
328 static char numberbuf[64];
331 if(yyleng>sizeof(numberbuf)-1)
332 syntaxerror("decimal number overflow");
334 memcpy(s, yytext, yyleng);
339 static inline int setint(int v)
341 a3_lval.number_int = v;
349 static inline int setuint(unsigned int v)
351 a3_lval.number_uint = v;
359 static inline int setfloat(double v)
361 a3_lval.number_float = v;
365 static inline int handlefloat()
368 a3_lval.number_float = atof(s);
372 static inline int handleint()
375 char l = (yytext[0]=='-');
377 char*max = l?"1073741824":"2147483647";
379 as3_softwarning("integer overflow: %s (converted to Number)", s);
380 return handlefloat();
384 for(t=0;t<yyleng-l;t++) {
385 if(yytext[l+t]>max[t]) {
386 as3_softwarning("integer overflow: %s (converted to Number)", s);
387 return handlefloat();
389 else if(yytext[l+t]<max[t])
399 for(t=0;t<yyleng;t++) {
407 static inline int handlehexfloat()
409 char l = (yytext[0]=='-')+2;
414 for(t=l;t<yyleng;t++) {
427 else if((c>='a' && c<='f') || (c>='A' && c<='F'))
428 d+=((c&0x0f)+9)*base;
432 static inline int handlehex()
434 char l = (yytext[0]=='-')+2;
439 syntaxerror("integer overflow %s", s);
449 else if((c>='a' && c<='f') || (c>='A' && c<='F'))
452 if(l && v>1073741824) {
454 as3_softwarning("signed integer overflow: %s (converted to Number)", s);
457 if(!l && v>2147483647) {
459 as3_softwarning("unsigned integer overflow: %s (converted to Number)", s);
464 return setint(-(int)v);
470 void handleLabel(char*text, int len)
473 for(t=len-1;t>=0;--t) {
478 char*s = malloc(t+1);
479 memcpy(s, yytext, t);
484 static int handleregexp()
486 char*s = malloc(yyleng);
488 memcpy(s, yytext+1, len);
491 for(t=len;t>=0;--t) {
497 a3_lval.regexp.pattern = s;
499 a3_lval.regexp.options = 0;
501 a3_lval.regexp.options = s+t+1;
506 void initialize_scanner();
507 #define YY_USER_INIT initialize_scanner();
509 /* count the number of lines+columns consumed by this token */
510 static inline void l() {
512 for(t=0;t<yyleng;t++) {
513 if(yytext[t]=='\n') {
521 /* count the number of columns consumed by this token */
522 static inline void c() {
523 current_column+=yyleng;
526 static trie_t*namespaces = 0;
527 void tokenizer_register_namespace(const char*id)
529 trie_put(&namespaces, id);
531 static inline tokenizer_is_namespace(const char*id)
533 return trie_lookup(namespaces, id);
536 static inline int handleIdentifier()
538 char*s = malloc(yyleng+1);
539 memcpy(s, yytext, yyleng);
542 if(tokenizer_is_namespace(s))
549 //Boolean {c();return m(KW_BOOLEAN);}
550 //int {c();return m(KW_INT);}
551 //uint {c();return m(KW_UINT);}
552 //Number {c();return m(KW_NUMBER);}
559 NAME [a-zA-Z_][a-zA-Z0-9_\\]*
562 HEXINT 0x[a-zA-Z0-9]+
563 HEXFLOAT 0x[a-zA-Z0-9]*\.[a-zA-Z0-9]*
565 FLOAT ([0-9]+(\.[0-9]*)?|\.[0-9]+)(e[0-9]+)?
567 HEXWITHSIGN [+-]?({HEXINT})
568 HEXFLOATWITHSIGN [+-]?({HEXFLOAT})
569 INTWITHSIGN [+-]?({INT})
570 FLOATWITHSIGN [+-]?({FLOAT})
572 CDATA <!\[CDATA\[([^]]|\][^]]|\]\][^>])*\]*\]\]\>
573 STRING ["](\\[\x00-\xff]|[^\\"\n])*["]|['](\\[\x00-\xff]|[^\\'\n])*[']
575 MULTILINE_COMMENT [/][*]+([*][^/]|[^/*]|[^*][/]|[\x00-\x1f])*[*]+[/]
576 SINGLELINE_COMMENT \/\/[^\n\r]*[\n\r]
577 REGEXP [/]([^/\n]|\\[/])*[/][a-zA-Z]*
581 {SINGLELINE_COMMENT} {l(); /* single line comment */}
582 {MULTILINE_COMMENT} {l(); /* multi line comment */}
583 [/][*] {syntaxerror("syntax error: unterminated comment", yytext);}
585 ^include{S}+{STRING}{S}*/\n {l();handleInclude(yytext, yyleng, 1);}
586 ^include{S}+[^" \t\r\n][\x20-\xff]*{S}*/\n {l();handleInclude(yytext, yyleng, 0);}
587 {STRING} {l(); BEGIN(INITIAL);handleString(yytext, yyleng);return T_STRING;}
588 {CDATA} {l(); BEGIN(INITIAL);handleCData(yytext, yyleng);return T_STRING;}
590 <BEGINNING,REGEXPOK>{
591 {REGEXP} {c(); BEGIN(INITIAL);return handleregexp();}
592 {HEXWITHSIGN}/{_} {c(); BEGIN(INITIAL);return handlehex();}
593 {HEXFLOATWITHSIGN}/{_} {c(); BEGIN(INITIAL);return handlehexfloat();}
594 {INTWITHSIGN}/{_} {c(); BEGIN(INITIAL);return handleint();}
595 {FLOATWITHSIGN}/{_} {c(); BEGIN(INITIAL);return handlefloat();}
598 <REGEXPOK>[\{] {c(); BEGIN(REGEXPOK);return m(T_DICTSTART);}
599 [\{] {c(); BEGIN(INITIAL); return m('{');}
601 \xef\xbb\xbf {/* utf 8 bom */}
604 {HEXINT}/{_} {c(); BEGIN(INITIAL);return handlehex();}
605 {HEXFLOAT}/{_} {c(); BEGIN(INITIAL);return handlehexfloat();}
606 {INT}/{_} {c(); BEGIN(INITIAL);return handleint();}
607 {FLOAT}/{_} {c(); BEGIN(INITIAL);return handlefloat();}
609 3rr0r {/* for debugging: generates a tokenizer-level error */
610 syntaxerror("3rr0r");}
612 {NAME}{S}*:{S}*for/{_} {l();BEGIN(INITIAL);handleLabel(yytext, yyleng-3);return T_FOR;}
613 {NAME}{S}*:{S}*do/{_} {l();BEGIN(INITIAL);handleLabel(yytext, yyleng-2);return T_DO;}
614 {NAME}{S}*:{S}*while/{_} {l();BEGIN(INITIAL);handleLabel(yytext, yyleng-5);return T_WHILE;}
615 {NAME}{S}*:{S}*switch/{_} {l();BEGIN(INITIAL);handleLabel(yytext, yyleng-6);return T_SWITCH;}
616 for {c();BEGIN(INITIAL);a3_lval.id="";return T_FOR;}
617 do {c();BEGIN(INITIAL);a3_lval.id="";return T_DO;}
618 while {c();BEGIN(INITIAL);a3_lval.id="";return T_WHILE;}
619 switch {c();BEGIN(INITIAL);a3_lval.id="";return T_SWITCH;}
621 [&][&] {c();BEGIN(REGEXPOK);return m(T_ANDAND);}
622 [|][|] {c();BEGIN(REGEXPOK);return m(T_OROR);}
623 [!][=] {c();BEGIN(REGEXPOK);return m(T_NE);}
624 [!][=][=] {c();BEGIN(REGEXPOK);return m(T_NEE);}
625 [=][=][=] {c();BEGIN(REGEXPOK);return m(T_EQEQEQ);}
626 [=][=] {c();BEGIN(REGEXPOK);return m(T_EQEQ);}
627 [>][=] {c();BEGIN(REGEXPOK);return m(T_GE);}
628 [<][=] {c();BEGIN(REGEXPOK);return m(T_LE);}
629 [-][-] {c();BEGIN(INITIAL);return m(T_MINUSMINUS);}
630 [+][+] {c();BEGIN(INITIAL);return m(T_PLUSPLUS);}
631 [+][=] {c();BEGIN(REGEXPOK);return m(T_PLUSBY);}
632 [\^][=] {c();BEGIN(REGEXPOK);return m(T_XORBY);}
633 [-][=] {c();BEGIN(REGEXPOK);return m(T_MINUSBY);}
634 [/][=] {c();BEGIN(REGEXPOK);return m(T_DIVBY);}
635 [%][=] {c();BEGIN(REGEXPOK);return m(T_MODBY);}
636 [*][=] {c();BEGIN(REGEXPOK);return m(T_MULBY);}
637 [|][=] {c();BEGIN(REGEXPOK);return m(T_ORBY);}
638 [>][>][=] {c();BEGIN(REGEXPOK);return m(T_SHRBY);}
639 [<][<][=] {c();BEGIN(REGEXPOK);return m(T_SHLBY);}
640 [>][>][>][=] {c();BEGIN(REGEXPOK);return m(T_USHRBY);}
641 [<][<] {c();BEGIN(REGEXPOK);return m(T_SHL);}
642 [>][>][>] {c();BEGIN(REGEXPOK);return m(T_USHR);}
643 [>][>] {c();BEGIN(REGEXPOK);return m(T_SHR);}
644 \.\.\. {c();BEGIN(REGEXPOK);return m(T_DOTDOTDOT);}
645 \.\. {c();BEGIN(REGEXPOK);return m(T_DOTDOT);}
646 \. {c();BEGIN(REGEXPOK);return m('.');}
647 :: {c();BEGIN(REGEXPOK);return m(T_COLONCOLON);}
648 : {c();BEGIN(REGEXPOK);return m(':');}
649 instanceof {c();BEGIN(REGEXPOK);return m(KW_INSTANCEOF);}
650 implements {c();BEGIN(REGEXPOK);return m(KW_IMPLEMENTS);}
651 interface {c();BEGIN(INITIAL);return m(KW_INTERFACE);}
652 namespace {c();BEGIN(INITIAL);return m(KW_NAMESPACE);}
653 protected {c();BEGIN(INITIAL);return m(KW_PROTECTED);}
654 undefined {c();BEGIN(INITIAL);return m(KW_UNDEFINED);}
655 continue {c();BEGIN(INITIAL);return m(KW_CONTINUE);}
656 override {c();BEGIN(INITIAL);return m(KW_OVERRIDE);}
657 internal {c();BEGIN(INITIAL);return m(KW_INTERNAL);}
658 function {c();BEGIN(INITIAL);return m(KW_FUNCTION);}
659 finally {c();BEGIN(INITIAL);return m(KW_FINALLY);}
660 default {c();BEGIN(INITIAL);return m(KW_DEFAULT);}
661 package {c();BEGIN(INITIAL);return m(KW_PACKAGE);}
662 private {c();BEGIN(INITIAL);return m(KW_PRIVATE);}
663 dynamic {c();BEGIN(INITIAL);return m(KW_DYNAMIC);}
664 extends {c();BEGIN(INITIAL);return m(KW_EXTENDS);}
665 delete {c();BEGIN(REGEXPOK);return m(KW_DELETE);}
666 return {c();BEGIN(REGEXPOK);return m(KW_RETURN);}
667 public {c();BEGIN(INITIAL);return m(KW_PUBLIC);}
668 native {c();BEGIN(INITIAL);return m(KW_NATIVE);}
669 static {c();BEGIN(INITIAL);return m(KW_STATIC);}
670 import {c();BEGIN(REGEXPOK);return m(KW_IMPORT);}
671 typeof {c();BEGIN(REGEXPOK);return m(KW_TYPEOF);}
672 throw {c();BEGIN(REGEXPOK);return m(KW_THROW);}
673 class {c();BEGIN(INITIAL);return m(KW_CLASS);}
674 const {c();BEGIN(INITIAL);return m(KW_CONST);}
675 catch {c();BEGIN(INITIAL);return m(KW_CATCH);}
676 final {c();BEGIN(INITIAL);return m(KW_FINAL);}
677 false {c();BEGIN(INITIAL);return m(KW_FALSE);}
678 break {c();BEGIN(INITIAL);return m(KW_BREAK);}
679 super {c();BEGIN(INITIAL);return m(KW_SUPER);}
680 each {c();BEGIN(INITIAL);return m(KW_EACH);}
681 void {c();BEGIN(INITIAL);return m(KW_VOID);}
682 true {c();BEGIN(INITIAL);return m(KW_TRUE);}
683 null {c();BEGIN(INITIAL);return m(KW_NULL);}
684 else {c();BEGIN(INITIAL);return m(KW_ELSE);}
685 case {c();BEGIN(REGEXPOK);return m(KW_CASE);}
686 with {c();BEGIN(REGEXPOK);return m(KW_WITH);}
687 use {c();BEGIN(REGEXPOK);return m(KW_USE);}
688 new {c();BEGIN(REGEXPOK);return m(KW_NEW);}
689 get {c();BEGIN(INITIAL);return m(KW_GET);}
690 set {c();BEGIN(INITIAL);return m(KW_SET);}
691 var {c();BEGIN(INITIAL);return m(KW_VAR);}
692 try {c();BEGIN(INITIAL);return m(KW_TRY);}
693 is {c();BEGIN(REGEXPOK);return m(KW_IS) ;}
694 in {c();BEGIN(REGEXPOK);return m(KW_IN) ;}
695 if {c();BEGIN(INITIAL);return m(KW_IF) ;}
696 as {c();BEGIN(REGEXPOK);return m(KW_AS);}
697 $?{NAME} {c();BEGIN(INITIAL);return handleIdentifier();}
699 [\]\}*] {c();BEGIN(INITIAL);return m(yytext[0]);}
700 [+-\/^~@$!%&\(=\[|?:;,<>] {c();BEGIN(REGEXPOK);return m(yytext[0]);}
701 [\)\]] {c();BEGIN(INITIAL);return m(yytext[0]);}
709 char c = buf[t]=input();
710 if(c=='\n' || c==EOF) {
715 if(c1>='0' && c1<='9')
716 syntaxerror("syntax error: %s (identifiers must not start with a digit)");
718 syntaxerror("syntax error: %s", buf);
724 void*b = leave_file();
727 yy_delete_buffer(YY_CURRENT_BUFFER);
730 yy_delete_buffer(YY_CURRENT_BUFFER);
731 yy_switch_to_buffer(b);
742 static char mbuf[256];
743 char*token2string(enum yytokentype nr, YYSTYPE v)
746 char*s = malloc(v.str.len+10);
747 strcpy(s, "<string>");
748 memcpy(s+8, v.str.str, v.str.len);
749 sprintf(s+8+v.str.len, " (%d bytes)", v.str.len);
752 else if(nr==T_REGEXP) {
753 char*s = malloc(strlen(v.regexp.pattern)+10);
754 sprintf(s, "<regexp>%s", v.regexp.pattern);
757 else if(nr==T_IDENTIFIER) {
758 char*s = malloc(strlen(v.id)+10);
759 sprintf(s, "<ID>%s", v.id);
762 else if(nr==T_INT) return "<int>";
763 else if(nr==T_UINT) return "<uint>";
764 else if(nr==T_BYTE) return "<byte>";
765 else if(nr==T_FLOAT) return "<float>";
766 else if(nr==T_EOF) return "***END***";
767 else if(nr==T_GE) return ">=";
768 else if(nr==T_LE) return "<=";
769 else if(nr==T_MINUSMINUS) return "--";
770 else if(nr==T_PLUSPLUS) return "++";
771 else if(nr==KW_IMPLEMENTS) return "implements";
772 else if(nr==KW_INTERFACE) return "interface";
773 else if(nr==KW_NAMESPACE) return "namespace";
774 else if(nr==KW_PROTECTED) return "protected";
775 else if(nr==KW_OVERRIDE) return "override";
776 else if(nr==KW_INTERNAL) return "internal";
777 else if(nr==KW_FUNCTION) return "function";
778 else if(nr==KW_PACKAGE) return "package";
779 else if(nr==KW_PRIVATE) return "private";
780 else if(nr==KW_BOOLEAN) return "Boolean";
781 else if(nr==KW_DYNAMIC) return "dynamic";
782 else if(nr==KW_EXTENDS) return "extends";
783 else if(nr==KW_PUBLIC) return "public";
784 else if(nr==KW_NATIVE) return "native";
785 else if(nr==KW_STATIC) return "static";
786 else if(nr==KW_IMPORT) return "import";
787 else if(nr==KW_NUMBER) return "number";
788 else if(nr==KW_CLASS) return "class";
789 else if(nr==KW_CONST) return "const";
790 else if(nr==KW_FINAL) return "final";
791 else if(nr==KW_FALSE) return "False";
792 else if(nr==KW_TRUE) return "True";
793 else if(nr==KW_UINT) return "uint";
794 else if(nr==KW_NULL) return "null";
795 else if(nr==KW_ELSE) return "else";
796 else if(nr==KW_USE) return "use";
797 else if(nr==KW_INT) return "int";
798 else if(nr==KW_NEW) return "new";
799 else if(nr==KW_GET) return "get";
800 else if(nr==KW_SET) return "set";
801 else if(nr==KW_VAR) return "var";
802 else if(nr==KW_IS) return "is";
803 else if(nr==KW_AS) return "as";
805 sprintf(mbuf, "%d", nr);
810 void initialize_scanner()