diff options
author | Kenneth Russel <[email protected]> | 2009-06-15 22:42:48 +0000 |
---|---|---|
committer | Kenneth Russel <[email protected]> | 2009-06-15 22:42:48 +0000 |
commit | c91f003551542c2aab62dd8ef89a7894c7e50689 (patch) | |
tree | e49c45b21c3ebeb8d238e8eb96415c745f9427da /src/java/com/sun/gluegen/cgram/GnuCParser.g | |
parent | 90bcb596e88898f807b39c9e7c85485ab8c006b6 (diff) |
Copied JOGL_2_SANDBOX r145 on to trunk; JOGL_2_SANDBOX branch is now closed
git-svn-id: file:///usr/local/projects/SUN/JOGL/git-svn/../svn-server-sync/gluegen/trunk@147 a78bb65f-1512-4460-ba86-f6dc96a7bf27
Diffstat (limited to 'src/java/com/sun/gluegen/cgram/GnuCParser.g')
-rw-r--r-- | src/java/com/sun/gluegen/cgram/GnuCParser.g | 865 |
1 files changed, 865 insertions, 0 deletions
diff --git a/src/java/com/sun/gluegen/cgram/GnuCParser.g b/src/java/com/sun/gluegen/cgram/GnuCParser.g new file mode 100644 index 0000000..7de057c --- /dev/null +++ b/src/java/com/sun/gluegen/cgram/GnuCParser.g @@ -0,0 +1,865 @@ +/*%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% + + Copyright (c) Non, Inc. 1998 -- All Rights Reserved + +PROJECT: C Compiler +MODULE: GnuCParser +FILE: GnuCParser.g + +AUTHOR: Monty Zukowski ([email protected]) April 28, 1998 + +DESCRIPTION: + This is a grammar for the GNU C compiler. It is a + grammar subclass of StdCParser, overriding only those + rules which are different from Standard C. + +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%*/ + + +header { + package com.sun.gluegen.cgram; + + import java.io.*; + + import antlr.CommonAST; + import antlr.DumpASTVisitor; +} + + +class GnuCParser extends StdCParser; + +options + { + k = 2; + exportVocab = GNUC; + buildAST = true; + ASTLabelType = "TNode"; + + // Copied following options from java grammar. + codeGenMakeSwitchThreshold = 2; + codeGenBitsetTestThreshold = 3; + } + + +{ + // Suppport C++-style single-line comments? + public static boolean CPPComments = true; + + // access to symbol table + public CSymbolTable symbolTable = new CSymbolTable(); + + // source for names to unnamed scopes + protected int unnamedScopeCounter = 0; + + public boolean isTypedefName(String name) { + boolean returnValue = false; + TNode node = symbolTable.lookupNameInCurrentScope(name); + for (; node != null; node = (TNode) node.getNextSibling() ) { + if(node.getType() == LITERAL_typedef) { + returnValue = true; + break; + } + } + return returnValue; + } + + + public String getAScopeName() { + return "" + (unnamedScopeCounter++); + } + + public void pushScope(String scopeName) { + symbolTable.pushScope(scopeName); + } + + public void popScope() { + symbolTable.popScope(); + } + + int traceDepth = 0; + public void reportError(RecognitionException ex) { + try { + System.err.println("ANTLR Parsing Error: "+ex + " token name:" + tokenNames[LA(1)]); + ex.printStackTrace(System.err); + } + catch (TokenStreamException e) { + System.err.println("ANTLR Parsing Error: "+ex); + ex.printStackTrace(System.err); + } + } + public void reportError(String s) { + System.err.println("ANTLR Parsing Error from String: " + s); + } + public void reportWarning(String s) { + System.err.println("ANTLR Parsing Warning from String: " + s); + } + public void match(int t) throws MismatchedTokenException { + boolean debugging = false; + + if ( debugging ) { + for (int x=0; x<traceDepth; x++) System.out.print(" "); + try { + System.out.println("Match("+tokenNames[t]+") with LA(1)="+ + tokenNames[LA(1)] + ((inputState.guessing>0)?" [inputState.guessing "+ inputState.guessing + "]":"")); + } + catch (TokenStreamException e) { + System.out.println("Match("+tokenNames[t]+") " + ((inputState.guessing>0)?" [inputState.guessing "+ inputState.guessing + "]":"")); + + } + + } + try { + if ( LA(1)!=t ) { + if ( debugging ){ + for (int x=0; x<traceDepth; x++) System.out.print(" "); + System.out.println("token mismatch: "+tokenNames[LA(1)] + + "!="+tokenNames[t]); + } + throw new MismatchedTokenException(tokenNames, LT(1), t, false, getFilename()); + + } else { + // mark token as consumed -- fetch next token deferred until LA/LT + consume(); + } + } + catch (TokenStreamException e) { + } + + } + public void traceIn(String rname) { + traceDepth += 1; + for (int x=0; x<traceDepth; x++) System.out.print(" "); + try { + System.out.println("> "+rname+"; LA(1)==("+ tokenNames[LT(1).getType()] + + ") " + LT(1).getText() + " [inputState.guessing "+ inputState.guessing + "]"); + } + catch (TokenStreamException e) { + } + } + public void traceOut(String rname) { + for (int x=0; x<traceDepth; x++) System.out.print(" "); + try { + System.out.println("< "+rname+"; LA(1)==("+ tokenNames[LT(1).getType()] + + ") "+LT(1).getText() + " [inputState.guessing "+ inputState.guessing + "]"); + } + catch (TokenStreamException e) { + } + traceDepth -= 1; + } + +} + + +translationUnit + : ( externalList )? /* Empty source files are allowed. */ + ; +asm_expr + : "asm"^ + ("volatile")? LCURLY expr RCURLY ( SEMI )+ + ; + +idList + : ID ( options{warnWhenFollowAmbig=false;}: COMMA ID )* + ; + +externalDef + : ( "typedef" | declaration )=> declaration + | ( functionPrefix )=> functionDef + | typelessDeclaration + | asm_expr + | SEMI + ; + +/* these two are here because GCC allows "cat = 13;" as a valid program! */ +functionPrefix + { String declName; } + : ( (functionDeclSpecifiers)=> ds:functionDeclSpecifiers + | //epsilon + ) + declName = d:declarator[true] + ( declaration )* (VARARGS)? ( SEMI )* + LCURLY + ; + +typelessDeclaration + { AST typeMissing = #[NTypeMissing]; } + : initDeclList[typeMissing] SEMI { ## = #( #[NTypeMissing], ##); } + ; + +initializer + : ( ( ( (initializerElementLabel)=> initializerElementLabel )? + ( assignExpr | lcurlyInitializer ) { ## = #( #[NInitializer], ## ); } + ) + | lcurlyInitializer + ) + ; + +// GCC allows more specific initializers +initializerElementLabel + : ( ( LBRACKET ((constExpr VARARGS)=> rangeExpr | constExpr) RBRACKET (ASSIGN)? ) + | ID COLON + | DOT ID ASSIGN + ) + { ## = #( #[NInitializerElementLabel], ##) ; } + ; + +// GCC allows empty initializer lists +lcurlyInitializer + : + LCURLY^ (initializerList ( COMMA! )? )? RCURLY + { ##.setType( NLcurlyInitializer ); } + ; + +initializerList + : initializer ( options{warnWhenFollowAmbig=false;}:COMMA! initializer )* + ; + + +declarator[boolean isFunctionDefinition] returns [String declName] + { declName = ""; } + : + ( pointerGroup )? + + ( id:ID { declName = id.getText(); } + | LPAREN declName = declarator[false] RPAREN + ) + + ( declaratorParamaterList[isFunctionDefinition, declName] + | LBRACKET ( expr )? RBRACKET + )* + { ## = #( #[NDeclarator], ## ); } + ; + +declaratorParamaterList[boolean isFunctionDefinition, String declName] + : + LPAREN^ + { + if (isFunctionDefinition) { + pushScope(declName); + } + else { + pushScope("!"+declName); + } + } + ( + (declSpecifiers)=> parameterTypeList + | (idList)? + ) + { + popScope(); + } + ( COMMA! )? + RPAREN + { ##.setType(NParameterTypeList); } + ; + +parameterTypeList + : parameterDeclaration + ( options { + warnWhenFollowAmbig = false; + } : + ( COMMA | SEMI ) + parameterDeclaration + )* + ( ( COMMA | SEMI ) + VARARGS + )? + ; + + +declarationList + : ( options { // this loop properly aborts when + // it finds a non-typedefName ID MBZ + warnWhenFollowAmbig = false; + } : + + localLabelDeclaration + | ( declarationPredictor )=> declaration + )+ + ; +localLabelDeclaration + : ( //GNU note: any __label__ declarations must come before regular declarations. + "__label__"^ ID (options{warnWhenFollowAmbig=false;}: COMMA! ID)* ( COMMA! )? ( SEMI! )+ + ) + ; + + +declaration + { AST ds1 = null; } + : ds:declSpecifiers { ds1 = astFactory.dupList(#ds); } + ( + initDeclList[ds1] + )? + ( SEMI )+ + { ## = #( #[NDeclaration], ##); } + + ; + +functionStorageClassSpecifier + : "extern" + | "static" + | "inline" + ; + +typeSpecifier [int specCount] returns [int retSpecCount] + { retSpecCount = specCount + 1; } + : + ( "void" + | "char" + | "short" + | "int" + | "__int32" + | "__int64" + | "long" + | "float" + | "double" + | "signed" + | "unsigned" + | structOrUnionSpecifier ( options{warnWhenFollowAmbig=false;}: attributeDecl )* + | enumSpecifier + | { specCount==0 }? typedefName + | "typeof"^ LPAREN + ( ( typeName )=> typeName + | expr + ) + RPAREN + | "__complex" + ) + ; + + +structOrUnionSpecifier + { String scopeName; } + : sou:structOrUnion! + ( ( ID LCURLY )=> i:ID l:LCURLY + { + scopeName = #sou.getText() + " " + #i.getText(); + #l.setText(scopeName); + pushScope(scopeName); + } + ( structDeclarationList )? + { popScope();} + RCURLY + | l1:LCURLY + { + scopeName = getAScopeName(); + #l1.setText(scopeName); + pushScope(scopeName); + } + ( structDeclarationList )? + { popScope(); } + RCURLY + | ID + ) + { + ## = #( #sou, ## ); + } + ; + + +structDeclaration + : specifierQualifierList structDeclaratorList ( COMMA! )? ( SEMI! )+ + ; + +structDeclaratorList + : structDeclarator ( options{warnWhenFollowAmbig=false;}: COMMA! structDeclarator )* + ; + +structDeclarator + : ( declarator[false] )? + ( COLON constExpr )? + ( attributeDecl )* + { ## = #( #[NStructDeclarator], ##); } + ; + + + +enumSpecifier + : "enum"^ + ( ( ID LCURLY )=> i:ID LCURLY enumList[i.getText()] RCURLY + | LCURLY enumList["anonymous"] RCURLY + | ID + ) + ; +enumList[String enumName] + : enumerator[enumName] ( options{warnWhenFollowAmbig=false;}: COMMA! enumerator[enumName] )* ( COMMA! )? + ; + + +initDeclList[AST declarationSpecifiers] + : initDecl[declarationSpecifiers] + ( options{warnWhenFollowAmbig=false;}: COMMA! initDecl[declarationSpecifiers] )* + ( COMMA! )? + ; + +initDecl[AST declarationSpecifiers] + { String declName = ""; } + : declName = d:declarator[false] + { AST ds1, d1; + ds1 = astFactory.dupList(declarationSpecifiers); + d1 = astFactory.dupList(#d); + symbolTable.add(declName, #(null, ds1, d1) ); + } + ( attributeDecl )* + ( ASSIGN initializer + | COLON expr + )? + { ## = #( #[NInitDecl], ## ); } + ; + +attributeDecl + : "__attribute"^ LPAREN LPAREN attributeList RPAREN RPAREN + | "asm"^ LPAREN stringConst RPAREN { ##.setType( NAsmAttribute ); } + ; + +attributeList + : attribute ( options{warnWhenFollowAmbig=false;}: COMMA attribute)* ( COMMA )? + ; + +attribute + : ( ~(LPAREN | RPAREN | COMMA) + | LPAREN attributeList RPAREN + )* + ; +compoundStatement[String scopeName] + : LCURLY^ + + { + pushScope(scopeName); + } + ( //this ambiguity is ok, declarationList and nestedFunctionDef end properly + options { + warnWhenFollowAmbig = false; + } : + ( "typedef" | "__label__" | declaration )=> declarationList + | (nestedFunctionDef)=> nestedFunctionDef + )* + ( statementList )? + { popScope(); } + RCURLY + { ##.setType( NCompoundStatement ); ##.setAttribute( "scopeName", scopeName ); } + ; + +nestedFunctionDef + { String declName; } + : ( "auto" )? //only for nested functions + ( (functionDeclSpecifiers)=> ds:functionDeclSpecifiers + )? + declName = d:declarator[false] + { + AST d2, ds2; + d2 = astFactory.dupList(#d); + ds2 = astFactory.dupList(#ds); + symbolTable.add(declName, #(null, ds2, d2)); + pushScope(declName); + } + ( declaration )* + { popScope(); } + compoundStatement[declName] + { ## = #( #[NFunctionDef], ## );} + ; + +statement + : SEMI // Empty statements + + | compoundStatement[getAScopeName()] // Group of statements + + | expr SEMI! { ## = #( #[NStatementExpr], ## );} // Expressions + +// Iteration statements: + + | "while"^ LPAREN! expr RPAREN! statement + | "do"^ statement "while"! LPAREN! expr RPAREN! SEMI! + |! "for" + LPAREN ( e1:expr )? SEMI ( e2:expr )? SEMI ( e3:expr )? RPAREN + s:statement + { + if ( #e1 == null) { #e1 = (TNode) #[ NEmptyExpression ]; } + if ( #e2 == null) { #e2 = (TNode) #[ NEmptyExpression ]; } + if ( #e3 == null) { #e3 = (TNode) #[ NEmptyExpression ]; } + ## = #( #[LITERAL_for, "for"], #e1, #e2, #e3, #s ); + } + + +// Jump statements: + + | "goto"^ expr SEMI! + | "continue" SEMI! + | "break" SEMI! + | "return"^ ( expr )? SEMI! + + + | ID COLON! (options {warnWhenFollowAmbig=false;}: statement)? { ## = #( #[NLabel], ## ); } +// GNU allows range expressions in case statements + | "case"^ ((constExpr VARARGS)=> rangeExpr | constExpr) COLON! ( options{warnWhenFollowAmbig=false;}:statement )? + | "default"^ COLON! ( options{warnWhenFollowAmbig=false;}: statement )? + +// Selection statements: + + | "if"^ + LPAREN! expr RPAREN! statement + ( //standard if-else ambiguity + options { + warnWhenFollowAmbig = false; + } : + "else" statement )? + | "switch"^ LPAREN! expr RPAREN! statement + ; + + + +conditionalExpr + : logicalOrExpr + ( QUESTION^ (expr)? COLON conditionalExpr )? + ; + +rangeExpr //used in initializers only + : constExpr VARARGS constExpr + { ## = #(#[NRangeExpr], ##); } + ; + +castExpr + : ( LPAREN typeName RPAREN )=> + LPAREN^ typeName RPAREN ( castExpr | lcurlyInitializer ) + { ##.setType(NCast); } + + | unaryExpr + ; +nonemptyAbstractDeclarator + : ( + pointerGroup + ( (LPAREN + ( nonemptyAbstractDeclarator + | parameterTypeList + )? + ( COMMA! )? + RPAREN) + | (LBRACKET (expr)? RBRACKET) + )* + + | ( (LPAREN + ( nonemptyAbstractDeclarator + | parameterTypeList + )? + ( COMMA! )? + RPAREN) + | (LBRACKET (expr)? RBRACKET) + )+ + ) + { ## = #( #[NNonemptyAbstractDeclarator], ## ); } + + ; + + + +unaryExpr + : postfixExpr + | INC^ castExpr + | DEC^ castExpr + | u:unaryOperator castExpr { ## = #( #[NUnaryExpr], ## ); } + + | "sizeof"^ + ( ( LPAREN typeName )=> LPAREN typeName RPAREN + | unaryExpr + ) + | "__alignof"^ + ( ( LPAREN typeName )=> LPAREN typeName RPAREN + | unaryExpr + ) + | gnuAsmExpr + ; + +unaryOperator + : BAND + | STAR + | PLUS + | MINUS + | BNOT //also stands for complex conjugation + | LNOT + | LAND //for label dereference (&&label) + | "__real" + | "__imag" + ; + +gnuAsmExpr + : "asm"^ ("volatile")? + LPAREN stringConst + ( options { warnWhenFollowAmbig = false; }: + COLON (strOptExprPair ( COMMA strOptExprPair)* )? + ( options { warnWhenFollowAmbig = false; }: + COLON (strOptExprPair ( COMMA strOptExprPair)* )? + )? + )? + ( COLON stringConst ( COMMA stringConst)* )? + RPAREN + { ##.setType(NGnuAsmExpr); } + ; + +//GCC requires the PARENs +strOptExprPair + : stringConst ( LPAREN expr RPAREN )? + ; + + +primaryExpr + : ID + | Number + | charConst + | stringConst +// JTC: +// ID should catch the enumerator +// leaving it in gives ambiguous err +// | enumerator + | (LPAREN LCURLY) => LPAREN^ compoundStatement[getAScopeName()] RPAREN + | LPAREN^ expr RPAREN { ##.setType(NExpressionGroup); } + ; + + +{ + import java.io.*; + import java.util.*; + import antlr.*; +} + +class GnuCLexer extends StdCLexer; +options + { + k = 3; + importVocab = GNUC; + testLiterals = false; + } +tokens { + LITERAL___extension__ = "__extension__"; +} + +{ + public void initialize(String src) + { + setOriginalSource(src); + initialize(); + } + + public void initialize() + { + literals.put(new ANTLRHashString("__alignof__", this), new Integer(LITERAL___alignof)); + literals.put(new ANTLRHashString("__asm", this), new Integer(LITERAL_asm)); + literals.put(new ANTLRHashString("__asm__", this), new Integer(LITERAL_asm)); + literals.put(new ANTLRHashString("__attribute__", this), new Integer(LITERAL___attribute)); + literals.put(new ANTLRHashString("__complex__", this), new Integer(LITERAL___complex)); + literals.put(new ANTLRHashString("__const", this), new Integer(LITERAL_const)); + literals.put(new ANTLRHashString("__const__", this), new Integer(LITERAL_const)); + literals.put(new ANTLRHashString("__imag__", this), new Integer(LITERAL___imag)); + literals.put(new ANTLRHashString("__inline", this), new Integer(LITERAL_inline)); + literals.put(new ANTLRHashString("__inline__", this), new Integer(LITERAL_inline)); + literals.put(new ANTLRHashString("__real__", this), new Integer(LITERAL___real)); + literals.put(new ANTLRHashString("__signed", this), new Integer(LITERAL_signed)); + literals.put(new ANTLRHashString("__signed__", this), new Integer(LITERAL_signed)); + literals.put(new ANTLRHashString("__typeof", this), new Integer(LITERAL_typeof)); + literals.put(new ANTLRHashString("__typeof__", this), new Integer(LITERAL_typeof)); + literals.put(new ANTLRHashString("__volatile", this), new Integer(LITERAL_volatile)); + literals.put(new ANTLRHashString("__volatile__", this), new Integer(LITERAL_volatile)); + } + + + LineObject lineObject = new LineObject(); + String originalSource = ""; + PreprocessorInfoChannel preprocessorInfoChannel = new PreprocessorInfoChannel(); + int tokenNumber = 0; + boolean countingTokens = true; + int deferredLineCount = 0; + List defines = new ArrayList(); + + public void setCountingTokens(boolean ct) + { + countingTokens = ct; + if ( countingTokens ) { + tokenNumber = 0; + } + else { + tokenNumber = 1; + } + } + + public void setOriginalSource(String src) + { + originalSource = src; + lineObject.setSource(src); + } + public void setSource(String src) + { + lineObject.setSource(src); + } + + public PreprocessorInfoChannel getPreprocessorInfoChannel() + { + return preprocessorInfoChannel; + } + + public void setPreprocessingDirective(String pre) + { + preprocessorInfoChannel.addLineForTokenNumber( pre, new Integer(tokenNumber) ); + } + + public void addDefine(String name, String value) + { + defines.add(new Define(name, value)); + } + + /** Returns a list of Define objects corresponding to the + preprocessor definitions seen during parsing. */ + public List getDefines() { + return defines; + } + + protected Token makeToken(int t) + { + if ( t != Token.SKIP && countingTokens) { + tokenNumber++; + } + CToken tok = (CToken) super.makeToken(t); + tok.setLine(lineObject.line); + tok.setSource(lineObject.source); + tok.setTokenNumber(tokenNumber); + + lineObject.line += deferredLineCount; + deferredLineCount = 0; + return tok; + } + + public void deferredNewline() { + deferredLineCount++; + } + + public void newline() { + lineObject.newline(); + } + + + + + + +} +Whitespace + : ( ( ' ' | '\t' | '\014') + | "\r\n" { newline(); } + | ( '\n' | '\r' ) { newline(); } + ) { _ttype = Token.SKIP; } + ; + + +protected +Escape + : '\\' + ( options{warnWhenFollowAmbig=false;}: + ~('0'..'7' | 'x') + | ('0'..'3') ( options{warnWhenFollowAmbig=false;}: Digit )* + | ('4'..'7') ( options{warnWhenFollowAmbig=false;}: Digit )* + | 'x' ( options{warnWhenFollowAmbig=false;}: Digit | 'a'..'f' | 'A'..'F' )+ + ) + ; + +protected IntSuffix + : 'L' + | 'l' + | 'U' + | 'u' + | 'I' + | 'i' + | 'J' + | 'j' + ; +protected NumberSuffix + : + IntSuffix + | 'F' + | 'f' + ; + +Number + : ( ('-')? ( Digit )+ ( '.' | 'e' | 'E' ) )=> ('-')? ( Digit )+ + ( '.' ( Digit )* ( Exponent )? + | Exponent + ) + ( NumberSuffix + )* + + | ( "..." )=> "..." { _ttype = VARARGS; } + + | '.' { _ttype = DOT; } + ( ( Digit )+ ( Exponent )? + { _ttype = Number; } + ( NumberSuffix + )* + )? + + | '0' ( '0'..'7' )* + ( NumberSuffix + )* + + | ('-')? '1'..'9' ( Digit )* + ( NumberSuffix + )* + + | '0' ( 'x' | 'X' ) ( 'a'..'f' | 'A'..'F' | Digit )+ + ( IntSuffix + )* + ; + +IDMEAT + : + i:ID { + + if ( i.getType() == LITERAL___extension__ ) { + $setType(Token.SKIP); + } + else { + $setType(i.getType()); + } + + } + ; + +protected ID + options + { + testLiterals = true; + } + : ( 'a'..'z' | 'A'..'Z' | '_' | '$') + ( 'a'..'z' | 'A'..'Z' | '_' | '$' | '0'..'9' )* + ; + +WideCharLiteral + : + 'L' CharLiteral + { $setType(CharLiteral); } + ; + + + +WideStringLiteral + : + 'L' StringLiteral + { $setType(StringLiteral); } + ; + +StringLiteral + : + '"' + ( ('\\' ~('\n'))=> Escape + | ( '\r' { newline(); } + | '\n' { + newline(); + } + | '\\' '\n' { + newline(); + } + ) + | ~( '"' | '\r' | '\n' | '\\' ) + )* + '"' + ; + + + + |