X-Git-Url: https://git.saurik.com/cycript.git/blobdiff_plain/e53322782610fa076081d974ac22712107a2fe0e..cac61857bcc00ddc1e354756e11b24e93b8edbf3:/Cycript.l diff --git a/Cycript.l b/Cycript.l index 10be43f..063e900 100644 --- a/Cycript.l +++ b/Cycript.l @@ -1,5 +1,54 @@ %{ -#include "Cycript.tab.h" +#define YYLTYPE cy::location +#include "Cycript.tab.hh" +typedef cy::parser::token tk; + +#define YY_EXTRA_TYPE CYDriver * + +#define T yylval->newline_ = yyextra->state_ == CYNewLine; +#define C T yyextra->state_ = CYClear; +#define R T yyextra->state_ = CYRestricted; + +#define N \ + if (yyextra->state_ != CYNewLine) { \ + bool restricted(yyextra->state_ == CYRestricted); \ + if (restricted) { \ + yyextra->state_ = CYClear; \ + return tk::NewLine; \ + } else \ + yyextra->state_ = CYNewLine; \ + } + +#define L { \ + yylloc->step(); \ + yylloc->columns(yyleng); \ +} + +int H(char c) { + if (c >= '0' && c <= '9') + return c - '0'; + if (c >= 'a' && c <= 'f') + return c - 'a' + 10; + if (c >= 'A' && c <= 'F') + return c - 'A' + 10; + return -1; +} + +#define YY_INPUT(data, value, size) { \ + if (yyextra->file_ != NULL) { \ + size_t copy(fread(data, 1, size, yyextra->file_)); \ + value = copy == 0 ? YY_NULL : copy; \ + } else if (yyextra->size_ == 0) \ + value = YY_NULL; \ + else { \ + size_t copy(std::min(size, yyextra->size_)); \ + memcpy(data, yyextra->data_, copy); \ + yyextra->data_ += copy; \ + yyextra->size_ -= copy; \ + value = copy; \ + } \ +} + %} %option prefix="cy" @@ -9,79 +58,201 @@ %option yylineno %option nounput %option interactive +%option reentrant -delim [ \t] -whitesp {delim}+ -digit [0-9] -number [-]?{digit}*[.]?{digit}+ +Exponent [eE][+-]?[0-9]+ +Escape \\[\\'"bfnrtv]|\\0|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\\n %% -"&" { return CYTokenAmpersand; } -"&&" { return CYTokenAmpersandAmpersand; } -"&=" { return CYTokenAmpersandEqual; } -"^" { return CYTokenCarrot; } -"^=" { return CYTokenCarrotEqual; } -"=" { return CYTokenEqual; } -"==" { return CYTokenEqualEqual; } -"===" { return CYTokenEqualEqualEqual; } -"!" { return CYTokenExclamation; } -"!=" { return CYTokenExclamationEqual; } -"!==" { return CYTokenExclamationEqualEqual; } -"-" { return CYTokenHyphen; } -"-=" { return CYTokenHyphenEqual; } -"--" { return CYTokenHyphenHyphen; } -"->" { return CYTokenHyphenRight; } -"<" { return CYTokenLeft; } -"<=" { return CYTokenLeftEqual; } -"<<" { return CYTokenLeftLeft; } -"<<=" { return CYTokenLeftLeftEqual; } -"%" { return CYTokenPercent; } -"%=" { return CYTokenPercentEqual; } -"." { return CYTokenPeriod; } -"|" { return CYTokenPipe; } -"|=" { return CYTokenPipeEqual; } -"||" { return CYTokenPipePipe; } -"+" { return CYTokenPlus; } -"+=" { return CYTokenPlusEqual; } -"++" { return CYTokenPlusPlus; } -">" { return CYTokenRight; } -">=" { return CYTokenRightEqual; } -">>" { return CYTokenRightRight; } -">>=" { return CYTokenRightRightEqual; } -">>>" { return CYTokenRightRightRight; } -">>>=" { return CYTokenRightRightRightEqual; } -"/" { return CYTokenSlash; } -"/=" { return CYTokenSlashEqual; } -"*" { return CYTokenStar; } -"*=" { return CYTokenStarEqual; } -"~" { return CYTokenTilde; } - -"break" { return CYTokenBreak; } -"case" { return CYTokenCase; } -"catch" { return CYTokenCatch; } -"continue" { return CYTokenContinue; } -"default" { return CYTokenDefault; } -"delete" { return CYTokenDelete; } -"do" { return CYTokenDo; } -"else" { return CYTokenElse; } -"false" { return CYTokenFalse; } -"finally" { return CYTokenFinally; } -"for" { return CYTokenFor; } -"function" { return CYTokenFunction; } -"if" { return CYTokenIf; } -"in" { return CYTokenIn; } -"instanceof" { return CYTokenInstanceOf; } -"new" { return CYTokenNew; } -"null" { return CYTokenNull; } -"return" { return CYTokenReturn; } -"switch" { return CYTokenSwitch; } -"this" { return CYTokenThis; } -"throw" { return CYTokenThrow; } -"true" { return CYTokenTrue; } -"try" { return CYTokenTry; } -"typeof" { return CYTokenTypeOf; } -"var" { return CYTokenVar; } -"void" { return CYTokenVoid; } -"while" { return CYTokenWhile; } -"with" { return CYTokenWith; } +\/\/[^\n]* ; +\/\*(\n|[^\*]|\*[^/])\*\/ if (memchr(yytext, '\n', yyleng) != NULL) N // XXX: supposedly I will be screwed on very very long multi-line comments and need to replace this with a manual lexer. http://websrv.cs.fsu.edu/~engelen/courses/COP5621/Pr2.pdf ; XXX: this rule doesn't work anyway, fucking A :( + +"&" L C return tk::Ampersand; +"&&" L C return tk::AmpersandAmpersand; +"&=" L C return tk::AmpersandEqual; +"^" L C return tk::Carrot; +"^=" L C return tk::CarrotEqual; +"=" L C return tk::Equal; +"==" L C return tk::EqualEqual; +"===" L C return tk::EqualEqualEqual; +"!" L C return tk::Exclamation; +"!=" L C return tk::ExclamationEqual; +"!==" L C return tk::ExclamationEqualEqual; +"-" L C return tk::Hyphen; +"-=" L C return tk::HyphenEqual; +"--" L C return yylval->newline_ ? tk::HyphenHyphen_ : tk::HyphenHyphen; +"->" L C return tk::HyphenRight; +"<" L C return tk::Left; +"<=" L C return tk::LeftEqual; +"<<" L C return tk::LeftLeft; +"<<=" L C return tk::LeftLeftEqual; +"%" L C return tk::Percent; +"%=" L C return tk::PercentEqual; +"." L C return tk::Period; +"|" L C return tk::Pipe; +"|=" L C return tk::PipeEqual; +"||" L C return tk::PipePipe; +"+" L C return tk::Plus; +"+=" L C return tk::PlusEqual; +"++" L C return yylval->newline_ ? tk::PlusPlus_ : tk::PlusPlus; +">" L C return tk::Right; +">=" L C return tk::RightEqual; +">>" L C return tk::RightRight; +">>=" L C return tk::RightRightEqual; +">>>" L C return tk::RightRightRight; +">>>=" L C return tk::RightRightRightEqual; +"/" L C return tk::Slash; +"/=" L C return tk::SlashEqual; +"*" L C return tk::Star; +"*=" L C return tk::StarEqual; +"~" L C return tk::Tilde; + +":" L C return tk::Colon; +"," L C return tk::Comma; +"?" L C return tk::Question; +";" L C return tk::SemiColon; + +"(" L C return tk::OpenParen; +")" L C return tk::CloseParen; + +"{" L C return tk::OpenBrace; +"}" L C return tk::CloseBrace; + +"[" L C return tk::OpenBracket; +"]" L C return tk::CloseBracket; + +"@class" L C return tk::AtClass; +"@end" L C return tk::AtEnd; +"@selector" L C return tk::AtSelector; + +"break" L R yylval->word_ = new CYWord("break"); return tk::Break; +"case" L C yylval->word_ = new CYWord("case"); return tk::Case; +"catch" L C yylval->word_ = new CYWord("catch"); return tk::Catch; +"continue" L R yylval->word_ = new CYWord("continue"); return tk::Continue; +"default" L C yylval->word_ = new CYWord("default"); return tk::Default; +"delete" L C yylval->word_ = new CYWord("delete"); return tk::Delete; +"do" L C yylval->word_ = new CYWord("do"); return tk::Do; +"else" L C yylval->word_ = new CYWord("else"); return tk::Else; +"false" L C yylval->false_ = new CYFalse(); return tk::False; +"finally" L C yylval->word_ = new CYWord("finally"); return tk::Finally; +"for" L C yylval->word_ = new CYWord("for"); return tk::For; +"function" L C yylval->word_ = new CYWord("function"); return tk::Function; +"if" L C yylval->word_ = new CYWord("if"); return tk::If; +"in" L C yylval->word_ = new CYWord("in"); return tk::In; +"instanceof" L C yylval->word_ = new CYWord("instanceof"); return tk::InstanceOf; +"new" L C yylval->word_ = new CYWord("new"); return tk::New; +"null" L C yylval->null_ = new CYNull(); return tk::Null; +"return" L R yylval->word_ = new CYWord("return"); return tk::Return; +"switch" L C yylval->word_ = new CYWord("switch"); return tk::Switch; +"this" L C yylval->this_ = new CYThis(); return tk::This; +"throw" L R yylval->word_ = new CYWord("throw"); return tk::Throw; +"true" L C yylval->true_ = new CYTrue(); return tk::True; +"try" L C yylval->word_ = new CYWord("try"); return tk::Try; +"typeof" L C yylval->word_ = new CYWord("typeof"); return tk::TypeOf; +"var" L C yylval->word_ = new CYWord("var"); return tk::Var; +"void" L C yylval->word_ = new CYWord("void"); return tk::Void; +"while" L C yylval->word_ = new CYWord("while"); return tk::While; +"with" L C yylval->word_ = new CYWord("with"); return tk::With; + +"abstract" L C yylval->word_ = new CYWord("abstract"); return tk::Abstract; +"boolean" L C yylval->word_ = new CYWord("boolean"); return tk::Boolean; +"byte" L C yylval->word_ = new CYWord("byte"); return tk::Byte; +"char" L C yylval->word_ = new CYWord("char"); return tk::Char; +"class" L C yylval->word_ = new CYWord("class"); return tk::Class; +"const" L C yylval->word_ = new CYWord("const"); return tk::Const; +"debugger" L C yylval->word_ = new CYWord("debugger"); return tk::Debugger; +"double" L C yylval->word_ = new CYWord("double"); return tk::Double; +"enum" L C yylval->word_ = new CYWord("enum"); return tk::Enum; +"export" L C yylval->word_ = new CYWord("export"); return tk::Export; +"extends" L C yylval->word_ = new CYWord("extends"); return tk::Extends; +"final" L C yylval->word_ = new CYWord("final"); return tk::Final; +"float" L C yylval->word_ = new CYWord("float"); return tk::Float; +"goto" L C yylval->word_ = new CYWord("goto"); return tk::Goto; +"implements" L C yylval->word_ = new CYWord("implements"); return tk::Implements; +"import" L C yylval->word_ = new CYWord("import"); return tk::Import; +"int" L C yylval->word_ = new CYWord("int"); return tk::Int; +"interface" L C yylval->word_ = new CYWord("interface"); return tk::Interface; +"long" L C yylval->word_ = new CYWord("long"); return tk::Long; +"native" L C yylval->word_ = new CYWord("native"); return tk::Native; +"package" L C yylval->word_ = new CYWord("package"); return tk::Package; +"private" L C yylval->word_ = new CYWord("private"); return tk::Private; +"protected" L C yylval->word_ = new CYWord("protected"); return tk::Protected; +"public" L C yylval->word_ = new CYWord("public"); return tk::Public; +"short" L C yylval->word_ = new CYWord("short"); return tk::Short; +"static" L C yylval->word_ = new CYWord("static"); return tk::Static; +"super" L C yylval->word_ = new CYWord("super"); return tk::Super; +"synchronized" L C yylval->word_ = new CYWord("synchronized"); return tk::Synchronized; +"throws" L C yylval->word_ = new CYWord("throws"); return tk::Throws; +"transient" L C yylval->word_ = new CYWord("transient"); return tk::Transient; +"volatile" L C yylval->word_ = new CYWord("volatile"); return tk::Volatile; + +"each" L C yylval->word_ = new CYIdentifier("each"); return tk::Each; +"let" L C yylval->word_ = new CYIdentifier("let"); return tk::Let; + +[a-zA-Z$_][a-zA-Z$_0-9]* yylval->identifier_ = new CYIdentifier(apr_pstrmemdup(yyextra->pool_, yytext, yyleng)); L C return tk::Identifier_; + +(\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?){Exponent}? yylval->number_ = new CYNumber(strtod(yytext, NULL)); L C return tk::NumericLiteral; + +0[xX][0-9a-fA-F]+ L C yylval->number_ = new CYNumber(strtoull(yytext + 2, NULL, 16)); return tk::NumericLiteral; + +0[bB][0-1]+ L C yylval->number_ = new CYNumber(strtoull(yytext + 2, NULL, 2)); return tk::NumericLiteral; + +\"([^"\\\n]|{Escape})*\"|'([^'\\\n]|{Escape})*' L C { + char *value(reinterpret_cast(apr_palloc(yyextra->pool_, yyleng))); + char *local(value); + + for (int i(1); i != yyleng - 1; ++i) { + char next(yytext[i]); + + if (yytext[i] == '\\') + switch (next = yytext[++i]) { + case '\n': continue; + case '\\': next = '\\'; break; + case '\'': next = '\''; break; + case '"': next = '"'; break; + case 'b': next = '\b'; break; + case 'f': next = '\f'; break; + case 'n': next = '\n'; break; + case 'r': next = '\r'; break; + case 't': next = '\t'; break; + case 'v': next = '\v'; break; + case '0': next = '\0'; break; + + case 'x': + next = H(yytext[i + 1]) << 4 | H(yytext[i + 2]); + i += 2; + break; + } + + *local++ = next; + } + + *local = '\0'; + yylval->string_ = new CYString(value, local - value); + return tk::StringLiteral; +} + +\n yylloc->end.lines(); yylloc->step(); N + +[ \t] L +<> L yyterminate(); + +. L { + CYDriver::Error error; + error.location_ = *yylloc; + error.message_ = "syntax error, unknown token"; + yyextra->errors_.push_back(error); + yyterminate(); +} + +%% + +void CYDriver::ScannerInit() { + cylex_init(&scanner_); + cyset_extra(this, scanner_); +} + +void CYDriver::ScannerDestroy() { + cylex_destroy(scanner_); +}