/* Cycript - Optimizing JavaScript Compiler/Runtime * Copyright (C) 2009-2015 Jay Freeman (saurik) */ /* GNU Affero General Public License, Version 3 {{{ */ /* * This program is free software: you can redistribute it and/or modify * it under the terms of the GNU Affero General Public License as published by * the Free Software Foundation, either version 3 of the License, or * (at your option) any later version. * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU Affero General Public License for more details. * You should have received a copy of the GNU Affero General Public License * along with this program. If not, see . **/ /* }}} */ /* XXX: supposedly I will be screwed on very very long multi-line comments and need to replace these with a manual lexer. http://websrv.cs.fsu.edu/~engelen/courses/COP5621/Pr2.pdf */ %{ #if defined(__clang__) #pragma clang diagnostic push #pragma clang diagnostic ignored "-Wunknown-pragmas" #pragma clang diagnostic ignored "-Wdeprecated-register" #endif #define YYLTYPE CYLocation #include "Cycript.tab.hh" typedef cy::parser::token tk; #include "Highlight.hpp" #define YY_EXTRA_TYPE CYDriver * // do /not/ fold token to the return: this is a macro and the ordering is dependent #define F(value, highlight) do { \ int token(value); \ @begin ObjectiveC yyextra->no_.AtImplementation = false; \ @end yyextra->no_.Function = false; \ yyextra->no_.OpenBrace = false; \ yylval->highlight_ = highlight; \ return token; \ } while (false) #define A new($pool) #define Y $pool.strmemdup(yytext, yyleng) #define I(type, Type, value, highlight) do { \ yylval->type ## _ = A CY ## Type; \ F(value, highlight); \ } while (false) #define T yylval->newline_ = yyextra->state_ == CYNewLine; BEGIN(Div); #define C T yyextra->state_ = CYClear; #define R T yyextra->state_ = CYRestricted; #define N \ if (yyextra->state_ != CYNewLine) { \ if (yyextra->state_ != CYRestricted) \ yyextra->state_ = CYNewLine; \ else { \ yyextra->state_ = CYClear; \ F(tk::NewLine, hi::Nothing); \ } \ } #define V(more) { \ if (const char *nl = reinterpret_cast(memchr(yytext, '\n', yyleng))) { \ unsigned lines(0); \ size_t left; \ do { \ ++lines; \ left = yyleng - (nl - yytext) - 1; \ nl = reinterpret_cast(memchr(nl + 1, '\n', left)); \ } while (nl != NULL); \ yylloc->step(); \ yylloc->end.lines(lines); \ yylloc->end.columns(left); \ more \ } else L \ } #define L { \ yylloc->step(); \ yylloc->end.columns(yyleng); \ } #define M { \ if (yyextra->commented_) { \ I(comment, Comment(Y), tk::Comment, hi::Comment); \ } \ } #define E(message) { \ CYDriver::Error error; \ error.location_ = *yylloc; \ error.message_ = "syntax error, " message; \ yyextra->errors_.push_back(error); \ yyterminate(); \ } 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; } static void U(char *&local, unsigned point) { if (false) { } else if (point < 0x000080) { *local++ = point; } else if (point < 0x000800) { *local++ = 0xc0 | point >> 0x06 & 0x1f; goto one; } else if (point < 0x010000) { *local++ = 0xe0 | point >> 0x0c & 0x0f; goto two; } else if (point < 0x110000) { *local++ = 0xf0 | point >> 0x12 & 0x07; *local++ = 0x80 | point >> 0x0c & 0x3f; two: *local++ = 0x80 | point >> 0x06 & 0x3f; one: *local++ = 0x80 | point >> 0x00 & 0x3f; } else _assert(false); } static void U(char *&local, const char *text, yy_size_t &i) { unsigned point; char next(text[++i]); if (next != '{') { point = H(text[i + 0]) << 12 | H(text[i + 1]) << 8 | H(text[i + 2]) << 4 | H(text[i + 3]); i += 3; } else { point = 0; for (;;) { next = text[++i]; if (next == '}') break; point = (point << 4) | H(next); } } U(local, point); } #define YY_INPUT(data, value, size) { \ if (yyextra->data_.eof()) \ value = YY_NULL; \ else { \ yyextra->data_.read(data, size); \ size_t copy(yyextra->data_.gcount()); \ value = copy == 0 ? YY_NULL : copy; \ } \ } %} %option prefix="cy" %option bison-bridge %option bison-locations %option nodefault %option noyywrap %option noyylineno %option nounput %option nounistd %option 8bit %option backup %option batch %option never-interactive %option pointer %option reentrant %option stack %option full %option ecs U1 [\x00-\x7f] U0 [\x80-\xbf] U2 [\xc2-\xdf] U3 [\xe0-\xef] U4 [\xf0-\xf4] HexDigit [0-9a-fA-F] LineTerminatorSequence \r?\n|\r|\xe2\x80[\xa8\xa9] WhiteSpace [\x09\x0b\x0c\x20]|\xc2\xa0|\xef\xbb\xbf UnicodeEscape \\u({HexDigit}{4}|\{{HexDigit}+\}) OctalEscape \\[1-7]|\\[4-7][0-7]|\\[0-3][0-7][0-7]? StringEscape \\['"\\bfnrtv]|\\0|{OctalEscape}|\\x{HexDigit}{2}|{UnicodeEscape} StringExtra {StringEscape}|\\{LineTerminatorSequence} SingleString ([^'\\\n]|{StringExtra})* DoubleString ([^"\\\n]|{StringExtra})* StringPrefix '{SingleString}|\"{DoubleString} @include UnicodeIDStart.l @include UnicodeIDContinue.l IdentifierMore [$_] UnicodeStart {IdentifierMore}|{UnicodeIDStart} UnicodePart {IdentifierMore}|\xe2\x80[\x8c\x8d]|{UnicodeIDContinue} UnicodeFail {U2}|{U3}|{U3}{U0}|{U4}|{U4}{U0}|{U4}{U0}{U0} UnicodeScrap {UnicodePart}*{UnicodeFail}? IdentifierStart {UnicodeStart}|{UnicodeEscape} IdentifierPart {UnicodePart}|{UnicodeEscape} IdentifierFail {UnicodeFail}|\\(u({HexDigit}{0,3}|\{{HexDigit}*))? IdentifierScrap {IdentifierPart}*{IdentifierFail}? NonTerminator [^\n] BackslashSequence \\{NonTerminator} RegularExpressionFirstChar [^\n*\\/]|{BackslashSequence} RegularExpressionChar [^\n\\/]|{BackslashSequence} RegularExpressionFlags {UnicodePart}* RegularExpressionChars {RegularExpressionChar}* RegularExpressionBody {RegularExpressionFirstChar}{RegularExpressionChars} @begin E4X XMLNameStart [a-zA-Z_:] XMLNamePart [a-zA-Z0-9.-_:] XMLName {XMLNameStart}{XMLNamePart}* @end %s Div %s RegExp @begin E4X %x XMLContent %x XMLTag @end %% \/{RegularExpressionBody}\/{RegularExpressionFlags} L C I(literal, RegEx(Y), tk::RegularExpressionLiteral, hi::Constant); \/{RegularExpressionBody}\/{RegularExpressionFlags}{UnicodeFail} L E("invalid flags") \/{RegularExpressionBody}?\\? L E("unterminated regex") #![^\n]* L M \/\/[^\n]* L M /* http://ostermiller.org/findcomment.html */ /* XXX: unify these two rules using !? */ \/\*!([^*]|[\r\n]|(\*+([^*/]|[\r\n])))*\*+\/ V() C I(comment, Comment(Y), tk::Comment, hi::Comment); \/\*([^*]|[\r\n]|(\*+([^*/]|[\r\n])))*\*+\/ V(N) M \/\*([^*]|[\r\n]|(\*+([^*/]|[\r\n])))*\** V() E("invalid comment") @begin E4X "<>" L F(tk::LeftRight, hi::Structure); "" L F(tk::LeftSlashRight, hi::Structure); \])*]]> V() F(tk::XMLCDATA, hi::Constant); \ V() F(tk::XMLComment, hi::Comment); \])*?> V() F(tk::XMLPI, hi::Meta); "=" L F(tk::Equal, hi::Structure); ">" L F(tk::Right, hi::Structure); "/>" L F(tk::SlashRight, hi::Structure); "{" L F(tk::OpenBrace, hi::Structure); \"(\n|[^"])*\"|'(\n|[^'])*' V() F(tk::XMLAttributeValue, hi::Constant); {XMLName} L F(tk::XMLName, hi::Identifier); [ \t\r\n] V() F(tk::XMLWhitespace, hi::Nothing); "{" L F(tk::OpenBrace, hi::Structure); "<" L F(tk::Left, hi::Structure); "" L C F(yylval->newline_ ? tk::EqualRight_ : tk::EqualRight, hi::Operator); "!" L C F(tk::Exclamation, hi::Operator); "!=" L C F(tk::ExclamationEqual, hi::Operator); "!==" L C F(tk::ExclamationEqualEqual, hi::Operator); "-" L C F(tk::Hyphen, hi::Operator); "-=" L C F(tk::HyphenEqual, hi::Operator); "--" L C F(yylval->newline_ ? tk::HyphenHyphen_ : tk::HyphenHyphen, hi::Operator); "->" L C F(tk::HyphenRight, hi::Operator); "<" L C F(tk::Left, hi::Operator); "<=" L C F(tk::LeftEqual, hi::Operator); "<<" L C F(tk::LeftLeft, hi::Operator); "<<=" L C F(tk::LeftLeftEqual, hi::Operator); "%" L C F(tk::Percent, hi::Operator); "%=" L C F(tk::PercentEqual, hi::Operator); "." L C F(tk::Period, hi::Operator); "|" L C F(tk::Pipe, hi::Operator); "|=" L C F(tk::PipeEqual, hi::Operator); "||" L C F(tk::PipePipe, hi::Operator); "+" L C F(tk::Plus, hi::Operator); "+=" L C F(tk::PlusEqual, hi::Operator); "++" L C F(yylval->newline_ ? tk::PlusPlus_ : tk::PlusPlus, hi::Operator); ">" L C F(tk::Right, hi::Operator); ">=" L C F(tk::RightEqual, hi::Operator); ">>" L C F(tk::RightRight, hi::Operator); ">>=" L C F(tk::RightRightEqual, hi::Operator); ">>>" L C F(tk::RightRightRight, hi::Operator); ">>>=" L C F(tk::RightRightRightEqual, hi::Operator); "*" L C F(tk::Star, hi::Operator); "*=" L C F(tk::StarEqual, hi::Operator); "~" L C F(tk::Tilde, hi::Operator);
"/" L C F(tk::Slash, hi::Operator);
"/=" L C F(tk::SlashEqual, hi::Operator); ":" L C F(tk::Colon, hi::Structure); "," L C F(tk::Comma, hi::Structure); "?" L C F(tk::Question, hi::Structure); ";" L C F(tk::SemiColon, hi::Structure); "(" L C F(tk::OpenParen, hi::Structure); ")" L C F(tk::CloseParen, hi::Structure); "{" L C F(yyextra->no_.OpenBrace ? tk::OpenBrace__ : yylval->newline_ ? tk::OpenBrace_ : tk::OpenBrace, hi::Structure); "}" L C F(tk::CloseBrace, hi::Structure); "[" L C F(tk::OpenBracket, hi::Structure); "]" L C F(tk::CloseBracket, hi::Structure); "@error" L C F(tk::AtError, hi::Error); @begin Java "@class" L C F(tk::AtClass, hi::Meta); @end @begin C "typedef" L C I(identifier, Identifier("typedef"), tk::Typedef, hi::Meta); "unsigned" L C I(identifier, Identifier("unsigned"), tk::Unsigned, hi::Type); "signed" L C I(identifier, Identifier("signed"), tk::Signed, hi::Type); "extern" L C I(identifier, Identifier("extern"), tk::Extern, hi::Type); @end @begin C "@encode" L C F(tk::AtEncode, hi::Meta); @end @begin ObjectiveC "@end" L C F(tk::AtEnd, hi::Meta); "@implementation" L C F(yyextra->no_.AtImplementation ? tk::AtImplementation_ : tk::AtImplementation, hi::Meta); "@import" L C F(tk::AtImport, hi::Special); "@selector" L C F(tk::AtSelector, hi::Meta); "@null" L C F(tk::AtNull, hi::Constant); "@YES" L C F(tk::AtYes, hi::Constant); "@NO" L C F(tk::AtNo, hi::Constant); "@true" L C F(tk::AtTrue, hi::Constant); "@false" L C F(tk::AtFalse, hi::Constant); "NULL" L C I(identifier, Identifier("NULL"), tk::Identifier_, hi::Constant); "nil" L C I(identifier, Identifier("nil"), tk::Identifier_, hi::Constant); "YES" L C I(identifier, Identifier("YES"), tk::Yes, hi::Constant); "NO" L C I(identifier, Identifier("NO"), tk::No, hi::Constant); "bool" L C I(identifier, Identifier("bool"), tk::Identifier_, hi::Type); "BOOL" L C I(identifier, Identifier("BOOL"), tk::Identifier_, hi::Type); "id" L C I(identifier, Identifier("id"), tk::Identifier_, hi::Type); "SEL" L C I(identifier, Identifier("SEL"), tk::Identifier_, hi::Type); @end "undefined" L C I(identifier, Identifier("undefined"), tk::Identifier_, hi::Operator); "false" L C I(false, False(), tk::False, hi::Constant); "null" L C I(null, Null(), tk::Null, hi::Constant); "true" L C I(true, True(), tk::True, hi::Constant); "auto" L C I(word, Word("auto"), tk::Auto, hi::Meta); "break" L R I(word, Word("break"), tk::Break, hi::Control); "case" L C I(word, Word("case"), tk::Case, hi::Control); "catch" L C I(word, Word("catch"), tk::Catch, hi::Control); "continue" L R I(word, Word("continue"), tk::Continue, hi::Control); "default" L C I(word, Word("default"), tk::Default, hi::Control); "delete" L C I(word, Word("delete"), tk::Delete, hi::Operator); "do" L C I(word, Word("do"), tk::Do, hi::Control); "else" L C I(word, Word("else"), tk::Else, hi::Control); "finally" L C I(word, Word("finally"), tk::Finally, hi::Control); "for" L C I(word, Word("for"), tk::For, hi::Control); "function" L C I(word, Word("function"), yyextra->no_.Function ? tk::Function_ : tk::Function, hi::Meta); "if" L C I(word, Word("if"), tk::If, hi::Control); "in" L C I(word, Word("in"), yyextra->in_.top() ? tk::In_ : tk::In, hi::Operator); "instanceof" L C I(word, Word("instanceof"), tk::InstanceOf, hi::Operator); "new" L C I(word, Word("new"), tk::New, hi::Operator); "return" L R I(word, Word("return"), tk::Return, hi::Control); "switch" L C I(word, Word("switch"), tk::Switch, hi::Control); "this" L C I(this, This(), tk::This, hi::Constant); "throw" L R I(word, Word("throw"), tk::Throw, hi::Control); "try" L C I(word, Word("try"), tk::Try, hi::Control); "typeof" L C I(word, Word("typeof"), tk::TypeOf, hi::Operator); "var" L C I(word, Word("var"), tk::Var, hi::Meta); "void" L C I(word, Word("void"), tk::Void, hi::Operator); "while" L C I(word, Word("while"), tk::While, hi::Control); "with" L C I(word, Word("with"), tk::With, hi::Control); "debugger" L C I(word, Word("debugger"), tk::Debugger, hi::Meta); "const" L C I(word, Word("const"), tk::Const, hi::Meta); "class" L C I(word, Word("class"), tk::Class, hi::Meta); "enum" L C I(word, Word("enum"), tk::Enum, hi::Meta); "export" L C I(word, Word("export"), tk::Export, hi::Meta); "extends" L C I(word, Word("extends"), tk::Extends, hi::Meta); "import" L C I(word, Word("import"), tk::Import, hi::Meta); "super" L C I(word, Word("super"), tk::Super, hi::Constant); "implements" L C I(identifier, Identifier("implements"), tk::Implements, hi::Meta); "interface" L C I(identifier, Identifier("interface"), tk::Interface, hi::Meta); "package" L C I(identifier, Identifier("package"), tk::Package, hi::Meta); "private" L C I(identifier, Identifier("private"), tk::Private, hi::Meta); "protected" L C I(identifier, Identifier("protected"), tk::Protected, hi::Meta); "public" L C I(identifier, Identifier("public"), tk::Public, hi::Meta); "static" L C I(identifier, Identifier("static"), tk::Static, hi::Meta); "abstract" L C I(identifier, Identifier("abstract"), tk::Abstract, hi::Meta); "boolean" L C I(identifier, Identifier("boolean"), tk::Boolean, hi::Type); "byte" L C I(identifier, Identifier("byte"), tk::Byte, hi::Type); "char" L C I(identifier, Identifier("char"), tk::Char, hi::Type); "double" L C I(identifier, Identifier("double"), tk::Double, hi::Type); "final" L C I(identifier, Identifier("final"), tk::Final, hi::Meta); "float" L C I(identifier, Identifier("float"), tk::Float, hi::Type); "goto" L C I(identifier, Identifier("goto"), tk::Goto, hi::Control); "int" L C I(identifier, Identifier("int"), tk::Int, hi::Type); "long" L C I(identifier, Identifier("long"), tk::Long, hi::Type); "native" L C I(identifier, Identifier("native"), tk::Native, hi::Meta); "short" L C I(identifier, Identifier("short"), tk::Short, hi::Type); "synchronized" L C I(identifier, Identifier("synchronized"), tk::Synchronized, hi::Meta); "throws" L C I(identifier, Identifier("throws"), tk::Throws, hi::Meta); "transient" L C I(identifier, Identifier("transient"), tk::Transient, hi::Meta); "volatile" L C I(identifier, Identifier("volatile"), tk::Volatile, hi::Meta); "let" L C I(identifier, Identifier("let"), tk::Let, hi::Meta); "yield" L R I(identifier, Identifier("yield"), tk::Yield, hi::Control); "each" L C I(identifier, Identifier("each"), tk::Each, hi::Control); "of" L C I(identifier, Identifier("of"), tk::Of, hi::Operator); @begin E4X "namespace" L C I(identifier, Identifier("namespace"), tk::Namespace, hi::Meta); "xml" L C I(identifier, Identifier("xml"), tk::XML, hi::Meta); @end {UnicodeStart}{UnicodePart}* L C I(identifier, Identifier(Y), tk::Identifier_, hi::Identifier); {IdentifierStart}{IdentifierPart}* L C { char *value(A char[yyleng + 1]); char *local(value); for (yy_size_t i(0), e(yyleng); i != e; ++i) { char next(yytext[i]); if (next != '\\') *local++ = next; else U(local, yytext, ++i); } *local = '\0'; I(identifier, Identifier(value), tk::Identifier_, hi::Identifier); } ({IdentifierStart}{IdentifierPart}*)?{IdentifierFail} L E("invalid identifier") 0[0-7]+ L C I(number, Number(strtoull(yytext + 1, NULL, 8)), tk::NumericLiteral, hi::Constant); 0[0-9]+ L C I(number, Number(strtoull(yytext + 1, NULL, 10)), tk::NumericLiteral, hi::Constant); 0[xX][0-9a-fA-F]+ L C I(number, Number(strtoull(yytext + 2, NULL, 16)), tk::NumericLiteral, hi::Constant); 0[oO][0-7]+ L C I(number, Number(strtoull(yytext + 2, NULL, 8)), tk::NumericLiteral, hi::Constant); 0[bB][0-1]+ L C I(number, Number(strtoull(yytext + 2, NULL, 2)), tk::NumericLiteral, hi::Constant); (\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?)([eE][+-]?[0-9]+)? L C I(number, Number(strtod(yytext, NULL)), tk::NumericLiteral, hi::Constant); (\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?)[eE][+-]?{IdentifierScrap} L E("invalid exponent") (\.?[0-9]|(0|[1-9][0-9]*)\.){IdentifierScrap} L E("invalid number") '{SingleString}'|\"{DoubleString}\" L C { char *value(A char[yyleng]); char *local(value); for (yy_size_t i(1), e(yyleng - 1); i != e; ++i) { char next(yytext[i]); if (yytext[i] == '\\') // XXX: support more line continuation characters if (false) line: { yylloc->end.lines(1); yylloc->end.columns(yyleng - i); } else switch (next = yytext[++i]) { case '\n': goto line; 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': case '1': case '2': case '3': if (yytext[i + 1] < '0' || yytext[i + 1] > '7') next = H(yytext[i]), i += 0; else if (yytext[i + 2] < '0' || yytext[i + 2] > '7') next = H(yytext[i]) << 3 | H(yytext[i + 1]), i += 1; else next = H(yytext[i]) << 6 | H(yytext[i + 1]) << 3 | H(yytext[i + 2]), i += 2; break; case '4': case '5': case '6': case '7': if (yytext[i + 1] < '0' || yytext[i + 1] > '7') next = H(yytext[i]), i += 0; else next = H(yytext[i]) << 3 | H(yytext[i + 1]), i += 1; break; case 'x': U(local, H(yytext[i + 1]) << 4 | H(yytext[i + 2])); i += 2; continue; case 'u': U(local, yytext, i); continue; } *local++ = next; } *local = '\0'; I(string, String(value, local - value), tk::StringLiteral, hi::Constant); } {StringPrefix}\\(x.{0,2}|u([^{].{0,3}|\{[^}]*)?|{UnicodeFail})? L E("invalid escape") {StringPrefix} L E("invalid string") {LineTerminatorSequence} yylloc->step(); yylloc->end.lines(); N {WhiteSpace} L <> if (yyextra->auto_) { yyextra->auto_ = false; F(tk::AutoComplete, hi::Nothing); } L yyterminate(); @({UnicodeStart}{UnicodeScrap}|{UnicodeFail}) L E("invalid keyword") . L E("invalid character") %% void CYDriver::ScannerInit() { cylex_init(&scanner_); cyset_extra(this, scanner_); } void CYDriver::ScannerDestroy() { cylex_destroy(scanner_); } CYDriver::Condition CYDriver::GetCondition() { switch (yy_top_state(scanner_)) { case RegExp: return RegExpCondition; @begin E4X case XMLContent: return XMLContentCondition; case XMLTag: return XMLTagCondition; @end default: _assert(false); } } void CYDriver::SetCondition(Condition condition) { struct yyguts_t *yyg(reinterpret_cast(scanner_)); switch (condition) { case RegExpCondition: BEGIN(RegExp); break; @begin E4X case XMLContentCondition: BEGIN(XMLContent); break; case XMLTagCondition: BEGIN(XMLTag); break; @end default: _assert(false); } } void CYDriver::PushCondition(Condition condition) { switch (condition) { case RegExpCondition: yy_push_state(RegExp, scanner_); break; @begin E4X case XMLContentCondition: yy_push_state(XMLContent, scanner_); break; case XMLTagCondition: yy_push_state(XMLTag, scanner_); break; @end default: _assert(false); } } void CYDriver::PopCondition() { yy_pop_state(scanner_); } #if defined(__clang__) #pragma clang diagnostic pop #endif