SatherCodeGenerator.java
上传用户:afrynkmhm
上传日期:2007-01-06
资源大小:1262k
文件大小:100k
源码类别:
编译器/解释器
开发平台:
Others
- package antlr;
- /* ANTLR Translator Generator
- * Project led by Terence Parr at http://www.jGuru.com
- * Software rights: http://www.antlr.org/RIGHTS.html
- *
- * $Id: //depot/code/org.antlr/release/antlr-2.7.0/antlr/SatherCodeGenerator.java#1 $
- */
- import java.util.Enumeration;
- import java.util.Hashtable;
- import antlr.collections.impl.BitSet;
- import antlr.collections.impl.Vector;
- import java.io.PrintWriter; //SAS: changed for proper text file io
- import java.io.IOException;
- import java.io.FileWriter;
- /**Generate MY_PARSER.sa, MY_LEXER.sa and MY_PARSER_TOKENTYPES.sa */
- public class SatherCodeGenerator extends CodeGenerator
- {
- // non-zero if inside syntactic predicate generation
- protected int syntacticPredLevel = 0;
- // Are we generating ASTs (for parsers and tree parsers) right now?
- protected boolean genAST = false;
- // Are we saving the text consumed (for lexers) right now?
- protected boolean saveText = false;
- // Grammar parameters set up to handle different grammar classes.
- // These are used to get instanceof tests out of code generation
- String labeledElementType;
- String labeledElementASTType;
- String labeledElementInit;
- String commonExtraArgs;
- String commonExtraParams;
- String commonLocalVars;
- String lt1Value;
- String exceptionThrown;
- String throwNoViable;
- // Tracks the rule being generated. Used for mapTreeId
- RuleBlock currentRule;
- // Tracks the rule or labeled subrule being generated. Used for AST generation.
- String currentASTResult;
- // Mapping between the ids used in the current alt, and the
- // names of variables used to represent their AST values.
- Hashtable treeVariableMap = new Hashtable();
- // Count of unnamed generated variables
- int astVarNumber = 1;
- // Special value used to mark duplicate in treeVariableMap
- protected static final String NONUNIQUE = new String();
- private Vector semPreds;
- /** Create a Java code-generator using the given Grammar.
- * The caller must still call setTool, setBehavior, and setAnalyzer
- * before generating code.
- */
- public SatherCodeGenerator() {
- super();
- charFormatter = new SatherCharFormatter();
- }
- /** Adds a semantic predicate string to the sem pred vector
- These strings will be used to build an array of sem pred names
- when building a debugging parser. This method should only be
- called when the debug option is specified
- */
- protected int addSemPred(String predicate) {
- semPreds.appendElement(predicate);
- return semPreds.size()-1;
- }
- public void exitIfError() {
- if (tool.hasError) {
- System.out.println("Exiting due to errors.");
- System.exit(1);
- }
- }
- /**Generate the parser, lexer, treeparser, and token types in Java */
- public void gen() {
- // Do the code generation
- try {
- // Loop over all grammars
- Enumeration grammarIter = behavior.grammars.elements();
- while (grammarIter.hasMoreElements()) {
- Grammar g = (Grammar)grammarIter.nextElement();
- // Connect all the components to each other
- g.setGrammarAnalyzer(analyzer);
- g.setCodeGenerator(this);
- analyzer.setGrammar(g);
- // To get right overloading behavior across hetrogeneous grammars
- setupGrammarParameters(g);
- g.generate();
- exitIfError();
- }
- // Loop over all token managers (some of which are lexers)
- Enumeration tmIter = behavior.tokenManagers.elements();
- while (tmIter.hasMoreElements()) {
- TokenManager tm = (TokenManager)tmIter.nextElement();
- if (!tm.isReadOnly()) {
- // Write the token manager tokens as Java
- // this must appear before genTokenInterchange so that
- // labels are set on string literals
- genTokenTypes(tm);
- // Write the token manager tokens as plain text
- genTokenInterchange(tm);
- }
- exitIfError();
- }
- }
- catch (IOException e) {
- System.out.println(e.getMessage());
- }
- }
- /** Generate code for the given grammar element.
- * @param blk The {...} action to generate
- */
- public void gen(ActionElement action) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genAction("+action+")");
- if ( action.isSemPred ) {
- genSemPred(action.actionText, action.line);
- }
- else {
- if ( grammar.hasSyntacticPredicate ) {
- println("if ( input_state.guessing = 0 ) then");
- tabs++;
- }
- ActionTransInfo tInfo = new ActionTransInfo();
- String actionStr = processActionForTreeSpecifiers(action.actionText, action.getLine(), currentRule, tInfo);
- if ( tInfo.refRuleRoot!=null ) {
- // Somebody referenced "#rule", make sure translated var is valid
- // assignment to #rule is left as a ref also, meaning that assignments
- // with no other refs like "#rule = foo();" still forces this code to be
- // generated (unnecessarily).
- println(tInfo.refRuleRoot + " := current_ast.root;");
- }
- // dump the translated action
- printAction(actionStr);
- if ( tInfo.assignToRoot ) {
- // Somebody did a "#rule=", reset internal currentAST.root
- println("current_ast.root := "+ tInfo.refRuleRoot + ";");
- // reset the child pointer too to be last sibling in sibling list
- println("if ( ~void( " + tInfo.refRuleRoot + " ) and ~void( "
- + tInfo.refRuleRoot + ".first_child ) ) then" );
- tabs++;
- println("current_ast.child := " + tInfo.refRuleRoot + ".first_child");
- tabs--;
- println("else");
- tabs++;
- println("current_ast.child := " + tInfo.refRuleRoot + ";");
- tabs--;
- println("end; -- if");
- println("current_ast.advance_child_to_end;");
- }
- if ( grammar.hasSyntacticPredicate ) {
- tabs--;
- println("end; -- if");
- }
- }
- }
- /** Generate code for the given grammar element.
- * @param blk The "x|y|z|..." block to generate
- */
- public void gen(AlternativeBlock blk) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("gen("+blk+")");
- // println("{");
- genBlockPreamble(blk);
- // Tell AST generation to build subrule result
- String saveCurrentASTResult = currentASTResult;
- if (blk.getLabel() != null) {
- currentASTResult = blk.getLabel();
- }
- boolean ok = grammar.theLLkAnalyzer.deterministic(blk);
- JavaBlockFinishingInfo howToFinish = genCommonBlock(blk, true);
- genBlockFinish(howToFinish, throwNoViable);
- println("");
- // println("}");
- // Restore previous AST generation
- currentASTResult = saveCurrentASTResult;
- }
- /** Generate code for the given grammar element.
- * @param blk The block-end element to generate. Block-end
- * elements are synthesized by the grammar parser to represent
- * the end of a block.
- */
- public void gen(BlockEndElement end) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genRuleEnd("+end+")");
- }
- /** Generate code for the given grammar element.
- * @param blk The character literal reference to generate
- */
- public void gen(CharLiteralElement atom) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genChar("+atom+")");
- if ( atom.getLabel()!=null ) {
- println(atom.getLabel() + " := " + lt1Value + ";");
- }
- boolean oldsaveText = saveText;
- saveText = saveText && atom.getAutoGenType()==GrammarElement.AUTO_GEN_NONE;
- genMatch(atom);
- saveText = oldsaveText;
- }
- /** Generate code for the given grammar element.
- * @param blk The character-range reference to generate
- */
- public void gen(CharRangeElement r) {
- if ( r.getLabel()!=null && syntacticPredLevel == 0) {
- println(r.getLabel() + " := " + lt1Value + ";");
- }
- println("match_range( " + r.beginText+ ", " + r.endText+ " );");
- }
- /** Generate the lexer Java file */
- public void gen(LexerGrammar g) throws IOException {
- // If debugging, create a new sempred vector for this grammar
- if (g.debuggingOutput)
- semPreds = new Vector();
- setGrammar(g);
- if (!(grammar instanceof LexerGrammar)) {
- tool.panic("Internal error generating lexer");
- }
- // SAS: moved output creation to method so a subclass can change
- // how the output is generated (for VAJ interface)
- setupOutput(grammar.getClassName());
- genAST = false; // no way to gen trees.
- saveText = true; // save consumed characters.
- tabs=0;
- // Generate header common to all Java output files
- genHeader();
- // Do not use printAction because we assume tabs==0
- println((String)behavior.headerActions.get(""));
- // Generate user-defined lexer file preamble
- println(grammar.preambleAction);
- // Generate lexer class definition
- String sup=null;
- if ( grammar.superClass!=null ) {
- sup = grammar.superClass;
- }
- else {
- sup = "ANTLR_CHAR_SCANNER{TOKEN}";
- }
- // print javadoc comment if any
- if ( grammar.comment!=null ) {
- _println(grammar.comment);
- }
- println("class " + grammar.getClassName() +
- "{TOKEN} < $ANTLR_TOKEN_STREAM{TOKEN} , $ANTLR_FILE_CURSOR is " );
- tabs++;
- println("include " + sup + " create -> private char_scanner_create;");
- println("include " + grammar.tokenManager.getName() + "_TOKENTYPES;");
- /*
- Token tsuffix = (Token)grammar.options.get("classHeaderSuffix");
- if ( tsuffix != null ) {
- String suffix = Tool.stripFrontBack(tsuffix.getText(),""",""");
- if ( suffix != null ) {
- print(", "+suffix); // must be an interface name for Java
- }
- }
- */
- println("");
- // Generate user-defined lexer class members
- print(
- processActionForTreeSpecifiers(grammar.classMemberAction, 0, currentRule, null)
- );
- //
- // Generate the constructor from ISTREAM, which in turn
- // calls the ByteBuffer constructor
- //
- println("create ( istr : $ISTREAM ) : SAME is");
- tabs++;
- println("inp : ANTLR_BYTE_BUFFER := #ANTLR_BYTE_BUFFER( istr );");
- println("res : SAME := #SAME( inp );");
- println("res.EOF_CHAR := istr.eof_char;");
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- //
- // Generate the constructor from Reader, which in turn
- // calls the CharBuffer constructor
- //
- // println("public " + grammar.getClassName() + "(Reader in) {");
- // tabs++;
- // println("this(new CharBuffer(in));");
- // tabs--;
- // println("}");
- println("create ( bb : ANTLR_BYTE_BUFFER ) : SAME is");
- tabs++;
- // if debugging, wrap the input buffer in a debugger
- // if (grammar.debuggingOutput)
- // println("this(new LexerSharedInputState(new antlr.debug.DebuggingInputBuffer(ib)));");
- // else
- println("state : ANTLR_LEXER_SHARED_INPUT_STATE := #ANTLR_LEXER_SHARED_INPUT_STATE( bb );");
- println("res: SAME := #SAME( state );");
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- //
- // Generate the constructor from InputBuffer (char or byte)
- //
- println("create ( state : ANTLR_LEXER_SHARED_INPUT_STATE ) : SAME is ");
- tabs++;
- println("res : SAME := char_scanner_create( state );");
- // if debugging, set up array variables and call user-overridable
- // debugging setup method
- // if ( grammar.debuggingOutput ) {
- // println("rule_names := sa_rule_names;");
- // println("sem_pred_names := sa_sem_pred_names;");
- // println("setup_debugging;");
- // }
- // Generate the initialization of a hashtable
- // containing the string literals used in the lexer
- // The literals variable itself is in CharScanner
- println("res.literals := #MAP{STR,INT};");
- Enumeration keys = grammar.tokenManager.getTokenSymbolKeys();
- while ( keys.hasMoreElements() ) {
- String key = (String)keys.nextElement();
- if ( key.charAt(0) != '"' ) {
- continue;
- }
- TokenSymbol sym = grammar.tokenManager.getTokenSymbol(key);
- if ( sym instanceof StringLiteralSymbol ) {
- StringLiteralSymbol s = (StringLiteralSymbol)sym;
- println("res.literals[ " + s.getId() + " ] := " + s.getTokenType() + ";");
- }
- }
- Enumeration ids;
- // Generate the setting of various generated options.
- println("res.case_sensitive_literals := " + g.caseSensitiveLiterals + ";");
- println("res.case_sensitive := " + g.caseSensitive + ";");
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- // generate the rule name array for debugging
- if (grammar.debuggingOutput) {
- println("private const sa_rule_names : ARRAY{STR} := |");
- ids = grammar.rules.elements();
- int ruleNum=0;
- while ( ids.hasMoreElements() ) {
- GrammarSymbol sym = (GrammarSymbol) ids.nextElement();
- if ( sym instanceof RuleSymbol)
- println(" ""+((RuleSymbol)sym).getId()+"",");
- }
- println("|;");
- }
- // Generate nextToken() rule.
- // nextToken() is a synthetic lexer rule that is the implicit OR of all
- // user-defined lexer rules.
- genNextToken();
- // Generate code for each rule in the lexer
- ids = grammar.rules.elements();
- int ruleNum=0;
- while ( ids.hasMoreElements() ) {
- RuleSymbol sym = (RuleSymbol) ids.nextElement();
- // Don't generate the synthetic rules
- if (!sym.getId().equals("mnextToken")) {
- genRule(sym, false, ruleNum++);
- }
- exitIfError();
- }
- // Generate the semantic predicate map for debugging
- if (grammar.debuggingOutput)
- genSemPredMap();
- // Generate the bitsets used throughout the lexer
- genBitsets(bitsetsUsed, ((LexerGrammar)grammar).charVocabulary.size());
- println("");
- tabs--;
- println("end; -- class");
- // Close the lexer output stream
- currentOutput.close();
- currentOutput = null;
- }
- /** Generate code for the given grammar element.
- * @param blk The (...)+ block to generate
- */
- public void gen(OneOrMoreBlock blk) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("gen+("+blk+")");
- String label;
- String cnt;
- // println("{"); // cannot nest Sather scopes within a routine
- genBlockPreamble(blk);
- if ( blk.getLabel() != null ) {
- cnt = getNextSatherPrefix() + "_cnt_"+ blk.getLabel();
- }
- else {
- cnt = getNextSatherPrefix() + "_cnt" + blk.ID;
- }
- println( cnt + " : INT := 0;");
- // if ( blk.getLabel() != null ) {
- // label = blk.getLabel();
- // }
- // else {
- // label = "_loop" + blk.ID;
- // }
- // println(label+":");
- println("loop");
- tabs++;
- // Tell AST generation to build subrule result
- String saveCurrentASTResult = currentASTResult;
- if (blk.getLabel() != null) {
- currentASTResult = blk.getLabel();
- }
- boolean ok = grammar.theLLkAnalyzer.deterministic(blk);
- // generate exit test if greedy set to false
- // and an alt is ambiguous with exit branch
- // or when lookahead derived purely from end-of-file
- // Lookahead analysis stops when end-of-file is hit,
- // returning set {epsilon}. Since {epsilon} is not
- // ambig with any real tokens, no error is reported
- // by deterministic() routines and we have to check
- // for the case where the lookahead depth didn't get
- // set to NONDETERMINISTIC (this only happens when the
- // FOLLOW contains real atoms + epsilon).
- boolean generateNonGreedyExitPath = false;
- int nonGreedyExitDepth = grammar.maxk;
- if ( !blk.greedy &&
- blk.exitLookaheadDepth<=grammar.maxk &&
- blk.exitCache[blk.exitLookaheadDepth].containsEpsilon() )
- {
- generateNonGreedyExitPath = true;
- nonGreedyExitDepth = blk.exitLookaheadDepth;
- }
- else if ( !blk.greedy &&
- blk.exitLookaheadDepth==LLkGrammarAnalyzer.NONDETERMINISTIC )
- {
- generateNonGreedyExitPath = true;
- }
- // generate exit test if greedy set to false
- // and an alt is ambiguous with exit branch
- if ( generateNonGreedyExitPath ) {
- if ( DEBUG_CODE_GENERATOR ) {
- System.out.println("nongreedy (...)+ loop; exit depth is "+
- blk.exitLookaheadDepth);
- }
- String predictExit =
- getLookaheadTestExpression(blk.exitCache,
- nonGreedyExitDepth);
- println("-- nongreedy exit test");
- println("if ( " + cnt + " >= 1 and " + predictExit + " ) then break! end; -- if");
- }
- JavaBlockFinishingInfo howToFinish = genCommonBlock(blk, false);
- genBlockFinish(
- howToFinish,
- "if ( " + cnt + " >= 1 ) then break! else " + throwNoViable + " end; -- if"
- );
- println( cnt + " := " + cnt + " + 1;" );
- tabs--;
- println("end; -- loop");
- // println("}"); // cannot nest Sather scopes within a routine
- // Restore previous AST generation
- currentASTResult = saveCurrentASTResult;
- }
- /** Generate the parser Java file */
- public void gen(ParserGrammar g) throws IOException {
- // if debugging, set up a new vector to keep track of sempred
- // strings for this grammar
- if (g.debuggingOutput)
- semPreds = new Vector();
- setGrammar(g);
- if (!(grammar instanceof ParserGrammar)) {
- tool.panic("Internal error generating parser");
- }
- // Open the output stream for the parser and set the currentOutput
- // SAS: moved file setup so subclass could do it (for VAJ interface)
- setupOutput(grammar.getClassName());
- genAST = grammar.buildAST;
- tabs = 0;
- // Generate the header common to all output files.
- genHeader();
- // Do not use printAction because we assume tabs==0
- println((String)behavior.headerActions.get(""));
- // Output the user-defined parser preamble
- println(grammar.preambleAction);
- // Generate parser class definition
- String sup=null;
- if ( grammar.superClass != null )
- sup = grammar.superClass.toUpperCase();
- else
- sup = "ANTLR_" + grammar.getSuperClass().toUpperCase();
- // print javadoc comment if any
- if ( grammar.comment!=null ) {
- _println(grammar.comment);
- }
- println("class " + grammar.getClassName() + "{ TOKEN < $ANTLR_TOKEN, AST < $ANTLR_AST{AST} } is");
- tabs++;
- println("include " + sup + "{ TOKEN, " + labeledElementASTType + " } create -> super_create;" );
- println("include " + grammar.tokenManager.getName() + "_"
- + TokenTypesFileSuffix.toUpperCase() + ";" );
- println("");
- /*
- Token tsuffix = (Token)grammar.options.get("classHeaderSuffix");
- if ( tsuffix != null ) {
- String suffix = Tool.stripFrontBack(tsuffix.getText(),""",""");
- if ( suffix != null )
- print(", "+suffix); // must be an interface name for Java
- }
- */
- // set up an array of all the rule names so the debugger can
- // keep track of them only by number -- less to store in tree...
- if (grammar.debuggingOutput) {
- println("const sa_rule_names : ARRAY{STR} := |");
- Enumeration ids = grammar.rules.elements();
- int ruleNum=0;
- while ( ids.hasMoreElements() ) {
- GrammarSymbol sym = (GrammarSymbol) ids.nextElement();
- if ( sym instanceof RuleSymbol)
- println(" ""+((RuleSymbol)sym).getId()+"",");
- }
- println("|;");
- }
- // Generate user-defined parser class members
- print(
- processActionForTreeSpecifiers(grammar.classMemberAction, 0, currentRule, null)
- );
- // Generate parser class constructor from TokenBuffer
- println("");
- println( "create ( token_buf : ANTLR_TOKEN_BUFFER{TOKEN} , k : INT ) : SAME is");
- tabs++;
- println("res : SAME := super_create( token_buf, k );");
- println("res.token_names := sa_token_names;");
- // println("res.ast_factory := #ANTLR_COMMON_AST_FACTORY;");
- // if debugging, set up arrays and call the user-overridable
- // debugging setup method
- if ( grammar.debuggingOutput ) {
- println("res.rule_names := sa_rule_names;");
- println("res.sem_pred_names := sa_sem_pred_names;");
- println("res.setup_debugging( token_buf );");
- }
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- println( "create ( token_buf : ANTLR_TOKEN_BUFFER{TOKEN} ) : SAME is");
- tabs++;
- println("return #SAME( token_buf, " + grammar.maxk + ");");
- tabs--;
- println("end; -- create");
- println("");
- // Generate parser class constructor from TokenStream
- println( "create ( lexer : $ANTLR_TOKEN_STREAM{TOKEN} , k : INT ) : SAME is");
- tabs++;
- println("res : SAME := super_create( lexer, k );");
- println("res.token_names := sa_token_names;");
- // println("res.ast_factory := #ANTLR_COMMON_AST_FACTORY;");
- // if debugging, set up arrays and call the user-overridable
- // debugging setup method
- if ( grammar.debuggingOutput ) {
- println("res.rule_names := sa_rule_names;");
- println("res.sem_pred_names := sa_sem_pred_names;");
- println("res.setup_debugging( lexer );");
- }
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- println( "create( lexer : $ANTLR_TOKEN_STREAM{TOKEN} ) : SAME is");
- tabs++;
- println("res : SAME := #SAME( lexer, " + grammar.maxk + ");");
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- println( "create ( state : ANTLR_PARSER_SHARED_INPUT_STATE{TOKEN} ) : SAME is ");
- tabs++;
- println("res : SAME := super_create( state," + grammar.maxk + ");");
- println("res.token_names := sa_token_names;");
- // println("res.ast_factory := #ANTLR_COMMON_AST_FACTORY;");
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- // Generate code for each rule in the grammar
- Enumeration ids = grammar.rules.elements();
- int ruleNum=0;
- while ( ids.hasMoreElements() ) {
- GrammarSymbol sym = (GrammarSymbol) ids.nextElement();
- if ( sym instanceof RuleSymbol) {
- RuleSymbol rs = (RuleSymbol)sym;
- genRule(rs, rs.references.size()==0, ruleNum++);
- }
- exitIfError();
- }
- // Generate the token names
- genTokenStrings();
- // Generate the bitsets used throughout the grammar
- genBitsets(bitsetsUsed, grammar.tokenManager.maxTokenType());
- // Generate the semantic predicate map for debugging
- if (grammar.debuggingOutput)
- genSemPredMap();
- // Close class definition
- println("");
- tabs--;
- println("end; -- class");
- // Close the parser output stream
- currentOutput.close();
- currentOutput = null;
- }
- /** Generate code for the given grammar element.
- * @param blk The rule-reference to generate
- */
- public void gen(RuleRefElement rr) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genRR("+rr+")");
- RuleSymbol rs = (RuleSymbol)grammar.getSymbol(rr.targetRule);
- if (rs == null || !rs.isDefined())
- {
- // Is this redundant???
- tool.error("Rule '" + rr.targetRule + "' is not defined", grammar.getFilename(), rr.getLine());
- return;
- }
- if (!(rs instanceof RuleSymbol))
- {
- // Is this redundant???
- tool.error("'" + rr.targetRule + "' does not name a grammar rule", grammar.getFilename(), rr.getLine());
- return;
- }
- genErrorTryForElement(rr);
- // AST value for labeled rule refs in tree walker.
- // This is not AST construction; it is just the input tree node value.
- if ( grammar instanceof TreeWalkerGrammar &&
- rr.getLabel() != null &&
- syntacticPredLevel == 0 )
- {
- println("if ( SYS::is_eq( sa_t , " + labeledElementASTType + "::ASTNULL ) ) then");
- tabs++;
- println( rr.getLabel() + " := void;");
- tabs--;
- println("else");
- println(rr.getLabel() + " := " + lt1Value + ";" );
- println("end; -- if");
- }
- // if in lexer and ! on rule ref or alt or rule, save buffer index to kill later
- if ( grammar instanceof LexerGrammar && (!saveText||rr.getAutoGenType()==GrammarElement.AUTO_GEN_BANG) ) {
- println("sa_save_index := text.length;");
- }
- // Process return value assignment if any
- printTabs();
- if (rr.idAssign != null)
- {
- // Warn if the rule has no return type
- if (rs.block.returnAction == null)
- {
- tool.warning("Rule '" + rr.targetRule + "' has no return type", grammar.getFilename(), rr.getLine());
- }
- _print(rr.idAssign + ":=");
- } else {
- // Warn about return value if any, but not inside syntactic predicate
- if ( !(grammar instanceof LexerGrammar) && syntacticPredLevel == 0 && rs.block.returnAction != null)
- {
- tool.warning("Rule '" + rr.targetRule + "' returns a value", grammar.getFilename(), rr.getLine());
- }
- }
- // Call the rule
- GenRuleInvocation(rr);
- // if in lexer and ! on element or alt or rule, save buffer index to kill later
- if ( grammar instanceof LexerGrammar && (!saveText||rr.getAutoGenType()==GrammarElement.AUTO_GEN_BANG) ) {
- println("text := text.substring( 0, sa_save_index); -- truncate");
- }
- // if not in a syntactic predicate
- if (syntacticPredLevel == 0) {
- boolean doNoGuessTest = (
- grammar.hasSyntacticPredicate &&
- (
- grammar.buildAST && rr.getLabel() != null ||
- (genAST && rr.getAutoGenType() == GrammarElement.AUTO_GEN_NONE)
- )
- );
- if (doNoGuessTest) {
- println("if ( input_state.guessing = 0 ) then");
- tabs++;
- }
- if (grammar.buildAST && rr.getLabel() != null) {
- // always gen variable for rule return on labeled rules
- println( rr.getLabel() + "_ast := return_ast;");
- }
- if (genAST) {
- switch (rr.getAutoGenType()) {
- case GrammarElement.AUTO_GEN_NONE:
- println("current_ast.add_child( return_ast );");
- break;
- case GrammarElement.AUTO_GEN_CARET:
- tool.error("Internal: encountered ^ after rule reference");
- break;
- default:
- break;
- }
- }
- // if a lexer and labeled, Token label defined at rule level, just set it here
- if ( grammar instanceof LexerGrammar && rr.getLabel() != null ) {
- println(rr.getLabel()+" := sa_return_token;");
- }
- if (doNoGuessTest) {
- tabs--;
- println("end;");
- }
- }
- genErrorCatchForElement(rr);
- }
- /** Generate code for the given grammar element.
- * @param blk The string-literal reference to generate
- */
- public void gen(StringLiteralElement atom) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genString("+atom+")");
- // Variable declarations for labeled elements
- if (atom.getLabel()!=null && syntacticPredLevel == 0) {
- println(atom.getLabel() + " := " + lt1Value + ";");
- }
- // AST
- genElementAST(atom);
- // is there a bang on the literal?
- boolean oldsaveText = saveText;
- saveText = saveText && atom.getAutoGenType()==GrammarElement.AUTO_GEN_NONE;
- // matching
- genMatch(atom);
- saveText = oldsaveText;
- // tack on tree cursor motion if doing a tree walker
- if (grammar instanceof TreeWalkerGrammar) {
- println("sa_t := sa_t.next_sibling;");
- }
- }
- /** Generate code for the given grammar element.
- * @param blk The token-range reference to generate
- */
- public void gen(TokenRangeElement r) {
- genErrorTryForElement(r);
- if ( r.getLabel()!=null && syntacticPredLevel == 0) {
- println(r.getLabel() + " := " + lt1Value + ";");
- }
- // AST
- genElementAST(r);
- // match
- println("match_range( " + r.beginText + ", " + r.endText + " );");
- genErrorCatchForElement(r);
- }
- /** Generate code for the given grammar element.
- * @param blk The token-reference to generate
- */
- public void gen(TokenRefElement atom) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genTokenRef("+atom+")");
- if ( grammar instanceof LexerGrammar ) {
- tool.panic("Token reference found in lexer");
- }
- genErrorTryForElement(atom);
- // Assign Token value to token label variable
- if ( atom.getLabel()!=null && syntacticPredLevel == 0) {
- println(atom.getLabel() + " := " + lt1Value + ";");
- }
- // AST
- genElementAST(atom);
- // matching
- genMatch(atom);
- genErrorCatchForElement(atom);
- // tack on tree cursor motion if doing a tree walker
- if (grammar instanceof TreeWalkerGrammar) {
- println("sa_t := sa_t.next_sibling;");
- }
- }
- public void gen(TreeElement t) {
- // save AST cursor
- println("sa__t" + t.ID + " : " + labeledElementASTType + " := sa_t;");
- // If there is a label on the root, then assign that to the variable
- if (t.root.getLabel() != null) {
- println("if ( SYS::is_eq( sa_t , AST::ASTNULL ) ) then");
- tabs++;
- println(t.root.getLabel() + " := void;");
- println("else");
- println(t.root.getLabel() + " := sa_t;");
- println("end; -- if");
- }
- // Generate AST variables
- genElementAST(t.root);
- if (grammar.buildAST) {
- // Save the AST construction state
- println("sa__current_ast" + t.ID + " : ANTLR_AST_PAIR{AST} := current_ast.copy;");
- // Make the next item added a child of the TreeElement root
- println("current_ast.root := current_ast.child;");
- println("current_ast.child := void;");
- }
- // match root
- genMatch(t.root);
- // move to list of children
- println("sa_t := sa_t.first_child;");
- // walk list of children, generating code for each
- for (int i=0; i<t.getAlternatives().size(); i++) {
- Alternative a = t.getAlternativeAt(i);
- AlternativeElement e = a.head;
- while ( e != null ) {
- e.generate();
- e = e.next;
- }
- }
- if (grammar.buildAST) {
- // restore the AST construction state to that just after the
- // tree root was added
- println("current_ast := sa__current_ast" + t.ID + ";");
- }
- // restore AST cursor
- println("sa_t := sa__t" + t.ID + ";");
- // move cursor to sibling of tree just parsed
- println("sa_t := sa_t.next_sibling;");
- }
- /** Generate the tree-parser Java file */
- public void gen(TreeWalkerGrammar g) throws IOException {
- // SAS: debugging stuff removed for now...
- setGrammar(g);
- if (!(grammar instanceof TreeWalkerGrammar)) {
- tool.panic("Internal error generating tree-walker");
- }
- // Open the output stream for the parser and set the currentOutput
- // SAS: move file open to method so subclass can override it
- // (mainly for VAJ interface)
- setupOutput(grammar.getClassName());
- genAST = grammar.buildAST;
- tabs = 0;
- // Generate the header common to all output files.
- genHeader();
- // Do not use printAction because we assume tabs==0
- println((String)behavior.headerActions.get(""));
- // Output the user-defined parser premamble
- println(grammar.preambleAction);
- // Generate parser class definition
- String sup=null;
- if ( grammar.superClass!=null ) {
- sup = grammar.superClass.toUpperCase();
- }
- else {
- sup = "ANTLR_TREE_PARSER";
- }
- // print javadoc comment if any
- if ( grammar.comment!=null ) {
- _println(grammar.comment);
- }
- println("class " + grammar.getClassName() + "{AST < $ANTLR_AST{AST} } is" );
- println("");
- tabs++;
- println("include " + sup + "{" + labeledElementASTType + "} create -> tree_parser_create;" );
- println("include " + grammar.tokenManager.getName() + "_"
- + TokenTypesFileSuffix.toUpperCase() + ";" );
- println("");
- /*
- Token tsuffix = (Token)grammar.options.get("classHeaderSuffix");
- if ( tsuffix != null ) {
- String suffix = Tool.stripFrontBack(tsuffix.getText(),""",""");
- if ( suffix != null ) {
- print(", "+suffix); // must be an interface name for Java
- }
- }
- */
- // Generate user-defined parser class members
- print(
- processActionForTreeSpecifiers(grammar.classMemberAction, 0, currentRule, null)
- );
- println("attr token_names : ARRAY{STR};");
- println("");
- // Generate default parser class constructor
- println("create : SAME is" );
- tabs++;
- println("res : SAME := tree_parser_create;");
- println("res.token_names := sa_token_names;");
- println("return res;");
- tabs--;
- println("end; -- create");
- println("");
- // Generate code for each rule in the grammar
- Enumeration ids = grammar.rules.elements();
- int ruleNum=0;
- String ruleNameInits = "";
- while ( ids.hasMoreElements() ) {
- GrammarSymbol sym = (GrammarSymbol) ids.nextElement();
- if ( sym instanceof RuleSymbol) {
- RuleSymbol rs = (RuleSymbol)sym;
- genRule(rs, rs.references.size()==0, ruleNum++);
- }
- exitIfError();
- }
- // Generate the token names
- genTokenStrings();
- // Generate the bitsets used throughout the grammar
- genBitsets(bitsetsUsed, grammar.tokenManager.maxTokenType());
- // Close class definition
- tabs--;
- println("end; -- class");
- println("");
- // Close the parser output stream
- currentOutput.close();
- currentOutput = null;
- }
- /** Generate code for the given grammar element.
- * @param wc The wildcard element to generate
- */
- public void gen(WildcardElement wc) {
- // Variable assignment for labeled elements
- if (wc.getLabel()!=null && syntacticPredLevel == 0) {
- println(wc.getLabel() + " := " + lt1Value + ";");
- }
- // AST
- genElementAST(wc);
- // Match anything but EOF
- if (grammar instanceof TreeWalkerGrammar) {
- println("if ( void(sa_t) ) then");
- tabs++;
- println("raise #ANTLR_MISMATCHED_TOKEN_EXCEPTION;");
- tabs--;
- println("end;");
- }
- else if (grammar instanceof LexerGrammar) {
- if ( grammar instanceof LexerGrammar &&
- (!saveText||wc.getAutoGenType()==GrammarElement.AUTO_GEN_BANG) ) {
- println("sa_save_index := text.length;");
- }
- println("match_not(EOF_CHAR);");
- if ( grammar instanceof LexerGrammar &&
- (!saveText||wc.getAutoGenType()==GrammarElement.AUTO_GEN_BANG) ) {
- // kill text atom put in buffer
- println("text := text.substring( 0 , sa_save_index);");
- }
- }
- else {
- println("match_not(" + getValueString(Token.EOF_TYPE) + ");");
- }
- // tack on tree cursor motion if doing a tree walker
- if (grammar instanceof TreeWalkerGrammar) {
- println("sa_t := sa_t.next_sibling;");
- }
- }
- /** Generate code for the given grammar element.
- * @param blk The (...)* block to generate
- */
- public void gen(ZeroOrMoreBlock blk) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("gen*("+blk+")");
- // println("{");
- genBlockPreamble(blk);
- String label;
- if ( blk.getLabel() != null ) {
- label = blk.getLabel();
- }
- else {
- label = "_loop" + blk.ID;
- }
- // println(label+":");
- println("loop");
- tabs++;
- // Tell AST generation to build subrule result
- String saveCurrentASTResult = currentASTResult;
- if (blk.getLabel() != null) {
- currentASTResult = blk.getLabel();
- }
- boolean ok = grammar.theLLkAnalyzer.deterministic(blk);
- // generate exit test if greedy set to false
- // and an alt is ambiguous with exit branch
- // or when lookahead derived purely from end-of-file
- // Lookahead analysis stops when end-of-file is hit,
- // returning set {epsilon}. Since {epsilon} is not
- // ambig with any real tokens, no error is reported
- // by deterministic() routines and we have to check
- // for the case where the lookahead depth didn't get
- // set to NONDETERMINISTIC (this only happens when the
- // FOLLOW contains real atoms + epsilon).
- boolean generateNonGreedyExitPath = false;
- int nonGreedyExitDepth = grammar.maxk;
- if ( !blk.greedy &&
- blk.exitLookaheadDepth<=grammar.maxk &&
- blk.exitCache[blk.exitLookaheadDepth].containsEpsilon() )
- {
- generateNonGreedyExitPath = true;
- nonGreedyExitDepth = blk.exitLookaheadDepth;
- }
- else if ( !blk.greedy &&
- blk.exitLookaheadDepth==LLkGrammarAnalyzer.NONDETERMINISTIC )
- {
- generateNonGreedyExitPath = true;
- }
- if ( generateNonGreedyExitPath ) {
- if ( DEBUG_CODE_GENERATOR ) {
- System.out.println("nongreedy (...)* loop; exit depth is "+
- blk.exitLookaheadDepth);
- }
- String predictExit =
- getLookaheadTestExpression(blk.exitCache,
- nonGreedyExitDepth);
- println("-- nongreedy exit test");
- println("if ( " + predictExit + " ) then break! end; -- if");
- }
- JavaBlockFinishingInfo howToFinish = genCommonBlock(blk, false);
- genBlockFinish(howToFinish, "break!;" );
- tabs--;
- println("end; -- loop ");
- // Restore previous AST generation
- currentASTResult = saveCurrentASTResult;
- }
- /** Generate an alternative.
- * @param alt The alternative to generate
- * @param blk The block to which the alternative belongs
- */
- protected void genAlt(Alternative alt, AlternativeBlock blk) {
- // Save the AST generation state, and set it to that of the alt
- boolean savegenAST = genAST;
- genAST = genAST && alt.getAutoGen();
- boolean oldsaveTest = saveText;
- saveText = saveText && alt.getAutoGen();
- // Reset the variable name map for the alternative
- Hashtable saveMap = treeVariableMap;
- treeVariableMap = new Hashtable();
- // Generate try block around the alt for error handling
- if (alt.exceptionSpec != null) {
- println("protect -- for error handling");
- tabs++;
- }
- AlternativeElement elem = alt.head;
- while ( !(elem instanceof BlockEndElement) ) {
- elem.generate(); // alt can begin with anything. Ask target to gen.
- elem = elem.next;
- }
- if ( genAST) {
- if (blk instanceof RuleBlock) {
- // Set the AST return value for the rule
- RuleBlock rblk = (RuleBlock)blk;
- println(rblk.getRuleName() + "_ast := current_ast.root;");
- }
- else if (blk.getLabel() != null) {
- // ### future: also set AST value for labeled subrules.
- // println(blk.getLabel() + "_ast = ("+labeledElementASTType+")currentAST.root;");
- }
- }
- if (alt.exceptionSpec != null) {
- // close try block
- tabs--;
- genErrorHandler(alt.exceptionSpec);
- }
- genAST = savegenAST;
- saveText = oldsaveTest;
- treeVariableMap = saveMap;
- }
- /** Generate all the bitsets to be used in the parser or lexer
- * Generate the raw bitset data like "long _tokenSet1_data[] = {...};"
- * and the BitSet object declarations like "BitSet _tokenSet1 = new BitSet(_tokenSet1_data);"
- * Note that most languages do not support object initialization inside a
- * class definition, so other code-generators may have to separate the
- * bitset declarations from the initializations (e.g., put the initializations
- * in the generated constructor instead).
- * @param bitsetList The list of bitsets to generate.
- * @param maxVocabulary Ensure that each generated bitset can contain at least this value.
- */
- protected void genBitsets(
- Vector bitsetList,
- int maxVocabulary
- ) {
- SatherCharFormatter satherCharFormatter = new SatherCharFormatter();
- println("");
- // here, I differ from the Java code generator. Lexer's bitsets are implemented as
- // Sather sets of CHAR's, Parser bitesets as Sather sets INT's
- if ( grammar instanceof LexerGrammar ) {
- for ( int i = 0 ; i < bitsetList.size() ; i++) {
- BitSet p = (BitSet)bitsetList.elementAt(i);
- // Ensure that generated BitSet is large enough for vocabulary
- p.growToInclude(maxVocabulary);
- String boolList = satherCharFormatter.BitSet2BoolList( p, ", " );
- String bitsetName = "sa" + getBitsetName(i);
- String bitsetData = bitsetName + "_data_";
- // initialization data
- println(
- "const " + bitsetData +
- " : ARRAY{BOOL} := " +
- "| " +
- boolList +
- " |;"
- );
- println( "const " + bitsetName + " : CHAR_SET := bitset( " +
- bitsetData + " );" );
- }
- }
- else {
- for ( int i = 0 ; i < bitsetList.size() ; i++) {
- BitSet p = (BitSet)bitsetList.elementAt(i);
- // Ensure that generated BitSet is large enough for vocabulary
- p.growToInclude(maxVocabulary);
- String charList = satherCharFormatter.BitSet2IntList( p, ", " );
- String bitsetName = "sa" + getBitsetName(i);
- String bitsetData = bitsetName + "_data_";
- // initialization data
- println(
- "const " + bitsetData +
- " : ARRAY{INT} := " +
- "| " +
- charList +
- " |;"
- );
- println( "const " + bitsetName + " : INT_SET := int_set( " +
- bitsetData + " );" );
- }
- }
- }
- /** Generate the finish of a block, using a combination of the info
- * returned from genCommonBlock() and the action to perform when
- * no alts were taken
- * @param howToFinish The return of genCommonBlock()
- * @param noViableAction What to generate when no alt is taken
- */
- private void genBlockFinish(JavaBlockFinishingInfo howToFinish, String noViableAction)
- {
- if (howToFinish.needAnErrorClause &&
- (howToFinish.generatedAnIf || howToFinish.generatedSwitch)) {
- if ( howToFinish.generatedAnIf ) {
- tabs++;
- println("else");
- }
- else {
- // println("{");
- }
- tabs++;
- println(noViableAction);
- tabs--;
- if ( howToFinish.generatedAnIf ) {
- println("end; -- if");
- tabs--;
- }
- if ( howToFinish.generatedSwitch )
- println("end; -- case");
- }
- if ( !howToFinish.needAnErrorClause && howToFinish.generatedSwitch )
- println("end; -- case");
- if ( howToFinish.postscript!=null ) {
- println(howToFinish.postscript);
- }
- }
- /** Generate the header for a block, which may be a RuleBlock or a
- * plain AlternativeBLock. This generates any variable declarations,
- * init-actions, and syntactic-predicate-testing variables.
- * @blk The block for which the preamble is to be generated.
- */
- protected void genBlockPreamble(AlternativeBlock blk) {
- // define labels for rule blocks.
- if ( blk instanceof RuleBlock ) {
- RuleBlock rblk = (RuleBlock)blk;
- if ( rblk.labeledElements!=null ) {
- for (int i=0; i<rblk.labeledElements.size(); i++) {
- AlternativeElement a = (AlternativeElement)rblk.labeledElements.elementAt(i);
- //System.out.println("looking at labeled element: "+a);
- // Variables for labeled rule refs and subrules are different than
- // variables for grammar atoms. This test is a little tricky because
- // we want to get all rule refs and ebnf, but not rule blocks or
- // syntactic predicates
- if (
- a instanceof RuleRefElement ||
- a instanceof AlternativeBlock &&
- !(a instanceof RuleBlock) &&
- !(a instanceof SynPredBlock)
- ) {
- if (
- !(a instanceof RuleRefElement) &&
- ((AlternativeBlock)a).not &&
- analyzer.subruleCanBeInverted(((AlternativeBlock)a), grammar instanceof LexerGrammar)
- ) {
- // Special case for inverted subrules that will be inlined.
- // Treat these like token or char literal references
- println( a.getLabel() + " : " + labeledElementType + " := " + labeledElementInit + ";");
- if (grammar.buildAST) {
- println( a.getLabel() + "_ast : " + labeledElementASTType+ ";" );
- }
- }
- else {
- if (grammar.buildAST) {
- // Always gen AST variables for labeled elements, even if the
- // element itself is marked with !
- println( a.getLabel() + "_ast : " + labeledElementASTType+ ";" );
- }
- if ( grammar instanceof LexerGrammar ) {
- println( a.getLabel() + " : $ANTLR_TOKEN; " );
- }
- if (grammar instanceof TreeWalkerGrammar) {
- // always generate rule-ref variables for tree walker
- println( a.getLabel() + " : " + labeledElementType + " := " + labeledElementInit + ";");
- }
- }
- }
- else {
- // It is a token or literal reference. Generate the
- // correct variable type for this grammar
- println( a.getLabel() + " : " + labeledElementType + " := " + labeledElementInit + ";");
- // In addition, generate *_AST variables if building ASTs
- if (grammar.buildAST) {
- println( a.getLabel() + "_ast : " + labeledElementASTType + ";");
- }
- }
- }
- }
- }
- // dump out init action
- if ( blk.initAction!=null ) {
- printAction(
- processActionForTreeSpecifiers(blk.initAction, 0, currentRule, null)
- );
- }
- }
- /** Generate a series of case statements that implement a BitSet test.
- * @param p The Bitset for which cases are to be generated
- */
- protected void genCases(BitSet p) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genCases("+p+")");
- int[] elems;
- elems = p.toArray();
- // Wrap cases four-per-line for lexer, one-per-line for parser
- int wrap = (grammar instanceof LexerGrammar) ? 4 : 1;
- int j=1;
- boolean startOfLine = true;
- for (int i = 0; i < elems.length; i++) {
- if (j==1) {
- print("");
- } else {
- _print(" ");
- }
- if ( i == 0 )
- _print("when " + getValueString(elems[i]) );
- else
- _print(", " + getValueString(elems[i]) );
- if (j==wrap) {
- _println("");
- startOfLine = true;
- j=1;
- }
- else {
- j++;
- startOfLine = false;
- }
- }
- if (!startOfLine) {
- _println("");
- }
- }
- /**Generate common code for a block of alternatives; return a postscript
- * that needs to be generated at the end of the block. Other routines
- * may append else-clauses and such for error checking before the postfix
- * is generated.
- * If the grammar is a lexer, then generate alternatives in an order where
- * alternatives requiring deeper lookahead are generated first, and
- * EOF in the lookahead set reduces the depth of the lookahead.
- * @param blk The block to generate
- * @param noTestForSingle If true, then it does not generate a test for a single alternative.
- */
- public JavaBlockFinishingInfo genCommonBlock(
- AlternativeBlock blk,
- boolean noTestForSingle)
- {
- int nIF=0;
- boolean createdLL1Switch = false;
- int closingBracesOfIFSequence = 0;
- JavaBlockFinishingInfo finishingInfo = new JavaBlockFinishingInfo();
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genCommonBlk("+blk+")");
- // Save the AST generation state, and set it to that of the block
- boolean savegenAST = genAST;
- genAST = genAST && blk.getAutoGen();
- boolean oldsaveTest = saveText;
- saveText = saveText && blk.getAutoGen();
- // Is this block inverted? If so, generate special-case code
- if (
- blk.not &&
- analyzer.subruleCanBeInverted(blk, grammar instanceof LexerGrammar)
- ) {
- Lookahead p = analyzer.look(1, blk);
- // Variable assignment for labeled elements
- if (blk.getLabel() != null && syntacticPredLevel == 0) {
- println(blk.getLabel() + " := " + lt1Value + ";");
- }
- // AST
- genElementAST(blk);
- String astArgs="";
- if (grammar instanceof TreeWalkerGrammar) {
- astArgs="sa_t,";
- }
- // match the bitset for the alternative
- println("match( sa" + astArgs + getBitsetName(markBitsetForGen(p.fset)) + ");");
- // tack on tree cursor motion if doing a tree walker
- if (grammar instanceof TreeWalkerGrammar) {
- println("sa_t := sa_t.next_sibling;");
- }
- return finishingInfo;
- }
- // Special handling for single alt
- if (blk.getAlternatives().size() == 1) {
- Alternative alt = blk.getAlternativeAt(0);
- // Generate a warning if there is a synPred for single alt.
- if (alt.synPred != null)
- {
- tool.warning(
- "Syntactic predicate superfluous for single alternative",
- grammar.getFilename(),
- blk.getAlternativeAt(0).synPred.getLine()
- );
- }
- if (noTestForSingle) {
- if (alt.semPred != null) {
- // Generate validating predicate
- genSemPred(alt.semPred, blk.line);
- }
- genAlt(alt, blk);
- return finishingInfo;
- }
- }
- // count number of simple LL(1) cases; only do switch for
- // many LL(1) cases (no preds, no end of token refs)
- // We don't care about exit paths for (...)*, (...)+
- // because we don't explicitly have a test for them
- // as an alt in the loop.
- int nLL1 = 0;
- for (int i=0; i<blk.getAlternatives().size(); i++) {
- Alternative a = blk.getAlternativeAt(i);
- if ( a.lookaheadDepth == 1 && a.semPred == null &&
- !a.cache[1].containsEpsilon()) {
- nLL1++;
- }
- }
- // do LL(1) cases
- if ( nLL1 >= makeSwitchThreshold) {
- // Determine the name of the item to be compared
- String testExpr = lookaheadString(1);
- createdLL1Switch = true;
- // when parsing trees, convert null to valid tree node with NULL lookahead
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("if ( void(sa_t) ) then");
- tabs++;
- println("sa_t := AST::ASTNULL;");
- tabs--;
- println("end; -- if");
- }
- println("case ( " + testExpr + " )" );
- for (int i=0; i<blk.alternatives.size(); i++) {
- Alternative alt = blk.getAlternativeAt(i);
- // ignore any non-LL(1) alts, predicated alts or end-of-token alts
- if ( alt.lookaheadDepth!=1 || alt.semPred != null ||
- alt.cache[1].containsEpsilon() ) {
- continue;
- }
- Lookahead p = alt.cache[1];
- if (p.fset.degree() == 0 && !p.containsEpsilon()) {
- tool.warning("Alternate omitted due to empty prediction set",
- grammar.getFilename(),
- alt.head.getLine());
- }
- else {
- genCases(p.fset);
- println("then");
- tabs++;
- genAlt(alt, blk);
- // println("break;");
- tabs--;
- // println("}");
- }
- }
- println("else -- default");
- tabs++;
- }
- // do non-LL(1) and nondeterministic cases
- // This is tricky in the lexer, because of cases like:
- // STAR : '*' ;
- // ASSIGN_STAR : "*=";
- // Since nextToken is generated without a loop, then the STAR will
- // have end-of-token as it's lookahead set for LA(2). So, we must generate the
- // alternatives containing trailing end-of-token in their lookahead sets *after*
- // the alternatives without end-of-token. This implements the usual
- // lexer convention that longer matches come before shorter ones, e.g.
- // "*=" matches ASSIGN_STAR not STAR
- //
- // For non-lexer grammars, this does not sort the alternates by depth
- // Note that alts whose lookahead is purely end-of-token at k=1 end up
- // as default or else clauses.
- int startDepth = (grammar instanceof LexerGrammar) ? grammar.maxk : 0;
- for (int altDepth = startDepth; altDepth >= 0; altDepth--) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("checking depth "+altDepth);
- for (int i=0; i<blk.alternatives.size(); i++) {
- Alternative alt = blk.getAlternativeAt(i);
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genAlt: "+i);
- // if we made a switch above, ignore what we already took care
- // of. Specifically, LL(1) alts with no preds
- // that do not have end-of-token in their prediction set
- if ( createdLL1Switch &&
- (alt.lookaheadDepth==1 && alt.semPred == null &&
- !alt.cache[1].containsEpsilon()) ) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("ignoring alt because it was in the switch");
- continue;
- }
- String e;
- boolean unpredicted = false;
- if (grammar instanceof LexerGrammar) {
- // Calculate the "effective depth" of the alt, which is the max
- // depth at which cache[depth]!=end-of-token
- int effectiveDepth = alt.lookaheadDepth;
- if (effectiveDepth == GrammarAnalyzer.NONDETERMINISTIC) {
- // use maximum lookahead
- effectiveDepth = grammar.maxk;
- }
- while (
- effectiveDepth >= 1 &&
- alt.cache[effectiveDepth].containsEpsilon()
- )
- {
- effectiveDepth--;
- }
- // Ignore alts whose effective depth is other than the ones we
- // are generating for this iteration.
- if (effectiveDepth != altDepth) {
- if ( DEBUG_CODE_GENERATOR )
- System.out.println("ignoring alt because effectiveDepth!=altDepth;"+effectiveDepth+"/="+altDepth);
- continue;
- }
- unpredicted = lookaheadIsEmpty(alt, effectiveDepth);
- e = getLookaheadTestExpression(alt, effectiveDepth);
- } else {
- unpredicted = lookaheadIsEmpty(alt, grammar.maxk);
- e = getLookaheadTestExpression(alt, grammar.maxk);
- }
- boolean defaultBlock = true;
- if (unpredicted && alt.semPred==null && alt.synPred==null) {
- // The alt has empty prediction set and no predicate to help out.
- // if we have not generated
- // a previous if, just put {...} around the end-of-token clause
- if ( nIF==0 ) {
- // println("{");
- }
- else {
- println("else ");
- defaultBlock = false; // else is for an if statement, not a case statement.
- }
- finishingInfo.needAnErrorClause = false;
- // continue;
- }
- else { // check for sem and syn preds
- // Add any semantic predicate expression to the lookahead test
- if ( alt.semPred != null ) {
- // if debugging, wrap the evaluation of the predicate in a method
- // translate $ and # references
- ActionTransInfo tInfo = new ActionTransInfo();
- String actionStr = processActionForTreeSpecifiers(alt.semPred,
- blk.line,
- currentRule,
- tInfo);
- // ignore translation info...we don't need to do anything with it.
- // call that will inform SemanticPredicateListeners of the result
- if (((grammar instanceof ParserGrammar) || (grammar instanceof LexerGrammar)) && grammar.debuggingOutput)
- e = "("+e+" and fireSemanticPredicateEvaluated(antlr.debug.SemanticPredicateEvent.PREDICTING,"+
- addSemPred(charFormatter.escapeString(actionStr))+","+actionStr+"))";
- else
- e = "("+e+" and ("+actionStr +"))";
- }
- // Generate any syntactic predicates
- if ( nIF>0 ) {
- if ( alt.synPred != null ) {
- println("else");
- tabs++;
- genSynPred( alt.synPred, e );
- closingBracesOfIFSequence++;
- }
- else {
- println("elsif " + e + " then" );
- }
- }
- else {
- if ( alt.synPred != null ) {
- genSynPred( alt.synPred, e );
- }
- else {
- // when parsing trees, convert null to valid tree node
- // with NULL lookahead.
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("if ( void(sa_t) ) then");
- tabs++;
- println("sa_t := AST::ASTNULL;");
- tabs--;
- println("end; -- if");
- }
- println("if " + e + " then");
- }
- }
- }
- nIF++;
- tabs++;
- genAlt(alt, blk);
- tabs--;
- if ( !defaultBlock )
- println("end; -- if");
- }
- }
- String ps = "";
- for (int i=1; i<=closingBracesOfIFSequence; i++) {
- ps+="end;";
- }
- // Restore the AST generation state
- genAST = savegenAST;
- // restore save text state
- saveText=oldsaveTest;
- // Return the finishing info.
- if ( createdLL1Switch ) {
- tabs--;
- finishingInfo.postscript = ps; // + "end; -- case";
- finishingInfo.generatedSwitch = true;
- finishingInfo.generatedAnIf = nIF>0;
- //return new JavaBlockFinishingInfo(ps+"}",true,nIF>0); // close up switch statement
- }
- else {
- finishingInfo.postscript = ps;
- finishingInfo.generatedSwitch = false;
- finishingInfo.generatedAnIf = nIF>0;
- // return new JavaBlockFinishingInfo(ps, false,nIF>0);
- }
- return finishingInfo;
- }
- /** Generate code to link an element reference into the AST */
- private void genElementAST(AlternativeElement el) {
- // handle case where you're not building trees, but are in tree walker.
- // Just need to get labels set up.
- if ( grammar instanceof TreeWalkerGrammar && !grammar.buildAST ) {
- String elementRef;
- String astName;
- // Generate names and declarations of the AST variable(s)
- if (el.getLabel() == null) {
- elementRef = lt1Value;
- // Generate AST variables for unlabeled stuff
- astName = "tmp" + astVarNumber + "_ast";
- astVarNumber++;
- // Map the generated AST variable in the alternate
- mapTreeVariable(el, astName);
- // Generate an "input" AST variable also
- println( astName + "_in : " + labeledElementASTType + " := " + elementRef + ";");
- }
- return;
- }
- if (grammar.buildAST && syntacticPredLevel == 0) {
- boolean doNoGuessTest = (
- grammar.hasSyntacticPredicate &&
- (
- el.getLabel() != null ||
- el.getAutoGenType() != GrammarElement.AUTO_GEN_BANG
- )
- );
- String elementRef;
- String astName;
- // Generate names and declarations of the AST variable(s)
- if (el.getLabel() != null) {
- elementRef = el.getLabel();
- astName = el.getLabel() + "_ast";
- } else {
- elementRef = lt1Value;
- // Generate AST variables for unlabeled stuff
- astName = "tmp" + astVarNumber + "_ast";
- astVarNumber++;
- // Generate the declaration (can only build ASTs for atoms)
- GrammarAtom ga = (GrammarAtom)el;
- if ( ga.getASTNodeType()!=null ) {
- println( astName + " : " + ga.getASTNodeType() + " := void;");
- }
- else {
- println( astName + " : " + labeledElementASTType + ";" );
- }
- // Map the generated AST variable in the alternate
- mapTreeVariable(el, astName);
- if (grammar instanceof TreeWalkerGrammar) {
- // Generate an "input" AST variable also
- println(astName + "_in : " + labeledElementASTType + ";" );
- }
- }
- // Enclose actions with !guessing
- if (doNoGuessTest) {
- println("if ( input_state.guessing = 0 ) then");
- tabs++;
- }
- // we need to find out the type of argument of
- // toke_factory::create. Sather cannot
- // overload function unless their signatures
- // can clearly be disambiguated. Therefore the
- // AST::create( $ANTLR_AST ) : $ANTLR_AST
- // and
- // AST::create( $ANTLR_TOKEN ) : $ANTLR_AST
- // cannot coexist. Therefore we rename them
- // create_ast_from_ast and create_ast_from_token, respectively.
- String astType = labeledElementASTType;
- GrammarAtom atom = (GrammarAtom)el;
- if ( atom != null && atom.getASTNodeType() != null ) {
- astType = atom.getASTNodeType();
- // make this the current AST type, used
- // for supsequent AST create's, even though
- // the may be temporary AST's
- labeledElementASTType = astType;
- }
- String astCreateString;
- if ( grammar instanceof TreeWalkerGrammar )
- astCreateString = astType + "::create_from_ast( " + elementRef + " )";
- else
- astCreateString = astType + "::create_from_token( " + elementRef + " )";
- if (el.getLabel() != null) {
- println(astName + " := " + astCreateString + ";" );
- } else {
- elementRef = lt1Value;
- println(astName + " := " + astCreateString + ";");
- // Map the generated AST variable in the alternate
- if (grammar instanceof TreeWalkerGrammar) {
- // set "input" AST variable also
- println(astName + "_in := " + elementRef + ";");
- }
- }
- if (genAST) {
- switch (el.getAutoGenType()) {
- case GrammarElement.AUTO_GEN_NONE:
- println("current_ast.add_child( " + astName + " );");
- break;
- case GrammarElement.AUTO_GEN_CARET:
- println("current_ast.make_root( " + astName + " );");
- break;
- default:
- break;
- }
- }
- if (doNoGuessTest) {
- tabs--;
- println("end; -- if?");
- }
- }
- }
- /** Close the try block and generate catch phrases
- * if the element has a labeled handler in the rule
- */
- private void genErrorCatchForElement(AlternativeElement el) {
- if (el.getLabel() == null) return;
- String r = el.enclosingRuleName;
- if ( grammar instanceof LexerGrammar ) {
- r = CodeGenerator.lexerRuleName(el.enclosingRuleName);
- }
- RuleSymbol rs = (RuleSymbol)grammar.getSymbol(r);
- if (rs == null) {
- tool.panic("Enclosing rule not found!");
- }
- ExceptionSpec ex = rs.block.findExceptionSpec(el.getLabel());
- if (ex != null) {
- tabs--;
- println("}*28");
- genErrorHandler(ex);
- }
- }
- /** Generate the catch phrases for a user-specified error handler */
- private void genErrorHandler(ExceptionSpec ex) {
- // Each ExceptionHandler in the ExceptionSpec is a separate catch
- for (int i = 0; i < ex.handlers.size(); i++)
- {
- ExceptionHandler handler = (ExceptionHandler)ex.handlers.elementAt(i);
- // Generate catch phrase
- println("when " + extractTypeOfAction(handler.exceptionTypeAndName) + " then");
- tabs++;
- if (grammar.hasSyntacticPredicate) {
- println("if ( input_state.guessing = 0 ) then");
- tabs++;
- }
- // When not guessing, execute user handler action
- printAction(
- processActionForTreeSpecifiers(handler.action, 0, currentRule, null)
- );
- if (grammar.hasSyntacticPredicate) {
- tabs--;
- println("else");
- tabs++;
- // When guessing, rethrow exception
- println("raise exception");
- // extractIdOfAction(handler.exceptionTypeAndName) + ";"
- tabs--;
- println("end; -- if");
- }
- tabs--;
- }
- // Close catch phrase
- println("end; -- protect");
- }
- /** Generate a try { opening if the element has a labeled handler in the rule */
- private void genErrorTryForElement(AlternativeElement el) {
- if (el.getLabel() == null) return;
- String r = el.enclosingRuleName;
- if ( grammar instanceof LexerGrammar ) {
- r = CodeGenerator.lexerRuleName(el.enclosingRuleName);
- }
- RuleSymbol rs = (RuleSymbol)grammar.getSymbol(r);
- if (rs == null) {
- tool.panic("Enclosing rule not found!");
- }
- ExceptionSpec ex = rs.block.findExceptionSpec(el.getLabel());
- if (ex != null) {
- println("try { // for error handling");
- tabs++;
- }
- }
- /** Generate a header that is common to all Sather files */
- protected void genHeader() {
- println("-- $ANTLR "+Tool.version+": "+
- """+Tool.fileMinusPath(tool.grammarFile)+"""+
- " -> "+
- """+grammar.getClassName()+".sa"$");
- }
- private void genLiteralsTest() {
- println("sa_ttype := test_literals_table(sa_ttype);");
- }
- private void genLiteralsTestForPartialToken() {
- println("sa_ttype := test_literals_table( text.substring( sa_begin, text.length - sa_begin ), sa_ttype );");
- }
- protected void genMatch(BitSet b) {
- }
- protected void genMatch(GrammarAtom atom) {
- if ( atom instanceof StringLiteralElement ) {
- if ( grammar instanceof LexerGrammar ) {
- genMatchUsingAtomText(atom);
- }
- else {
- genMatchUsingAtomTokenType(atom);
- }
- }
- else if ( atom instanceof CharLiteralElement ) {
- if ( grammar instanceof LexerGrammar ) {
- genMatchUsingAtomText(atom);
- }
- else {
- tool.error("cannot ref character literals in grammar: "+atom);
- }
- }
- else if ( atom instanceof TokenRefElement ) {
- genMatchUsingAtomText(atom);
- }
- }
- protected void genMatchUsingAtomText(GrammarAtom atom) {
- // match() for trees needs the _t cursor
- String astArgs="";
- if (grammar instanceof TreeWalkerGrammar) {
- astArgs="sa_t,";
- }
- // if in lexer and ! on element, save buffer index to kill later
- if ( grammar instanceof LexerGrammar && (!saveText||atom.getAutoGenType()==GrammarElement.AUTO_GEN_BANG) ) {
- println("sa_save_index := text.length;");
- }
- print(atom.not ? "match_not(" : "match(");
- _print(astArgs);
- // print out what to match
- if (atom.atomText.equals("EOF")) {
- // horrible hack to handle EOF case
- _print("ANTLR_COMMON_TOKEN::EOF_TYPE");
- }
- else {
- _print(atom.atomText);
- }
- _println(");");
- if ( grammar instanceof LexerGrammar && (!saveText||atom.getAutoGenType()==GrammarElement.AUTO_GEN_BANG) ) {
- println("text := text.substring( 0 , sa_save_index);"); // kill text atom put in buffer
- }
- }
- protected void genMatchUsingAtomTokenType(GrammarAtom atom) {
- // match() for trees needs the _t cursor
- String astArgs="";
- if (grammar instanceof TreeWalkerGrammar) {
- astArgs="sa_t,";
- }
- // If the literal can be mangled, generate the symbolic constant instead
- String mangledName = null;
- String s = astArgs + getValueString(atom.getType());
- // matching
- println( (atom.not ? "match_not(" : "match(") + s + ");");
- }
- /** Generate the nextToken() rule.
- * nextToken() is a synthetic lexer rule that is the implicit OR of all
- * user-defined lexer rules.
- * @param RuleBlock
- */
- public void genNextToken() {
- // Are there any public rules? If not, then just generate a
- // fake nextToken().
- boolean hasPublicRules = false;
- for (int i = 0; i < grammar.rules.size(); i++) {
- RuleSymbol rs = (RuleSymbol)grammar.rules.elementAt(i);
- if ( rs.isDefined() && rs.access.equals("public") ) {
- hasPublicRules = true;
- break;
- }
- }
- if (!hasPublicRules) {
- println("");
- println("next_token : TOKEN is " );
- tabs++;
- println("protect");
- tabs++;
- println("upon_eof;");
- tabs--;
- println("when $ANTLR_CHAR_STREAM_EXCEPTION then");
- tabs++;
- println("raise #ANTLR_TOKEN_STREAM_EXCEPTION( exception.str );");
- tabs--;
- println("end; -- protect");
- println("return #ANTLR_COMMON_TOKEN( ANTLR_COMMON_TOKEN::EOF_TYPE, "");");
- tabs--;
- println("end;");
- println("");
- return;
- }
- // Create the synthesized nextToken() rule
- RuleBlock nextTokenBlk = MakeGrammar.createNextTokenRule(grammar, grammar.rules, "nextToken");
- // Define the nextToken rule symbol
- RuleSymbol nextTokenRs = new RuleSymbol("mnextToken");
- nextTokenRs.setDefined();
- nextTokenRs.setBlock(nextTokenBlk);
- nextTokenRs.access = "private";
- grammar.define(nextTokenRs);
- // Analyze the nextToken rule
- boolean ok = grammar.theLLkAnalyzer.deterministic(nextTokenBlk);
- // Generate the next token rule
- String filterRule=null;
- if ( ((LexerGrammar)grammar).filterMode ) {
- filterRule = ((LexerGrammar)grammar).filterRule;
- }
- println("");
- println("next_token : TOKEN is");
- tabs++;
- println("theRetToken : TOKEN;");
- println("continue : BOOL := true;");
- // _println("tryAgain:");
- println("loop");
- tabs++;
- // println("Token _token = null;");
- println("sa_ttype : INT := ANTLR_COMMON_TOKEN::INVALID_TYPE;");
- if ( ((LexerGrammar)grammar).filterMode ) {
- println("commit_to_path := false;");
- println("continue := true;");
- if ( filterRule!=null ) {
- // Here's a good place to ensure that the filter rule actually exists
- if ( !grammar.isDefined(CodeGenerator.lexerRuleName(filterRule)) ) {
- grammar.tool.error("Filter rule "+filterRule+" does not exist in this lexer");
- }
- else {
- RuleSymbol rs = (RuleSymbol)grammar.getSymbol(CodeGenerator.lexerRuleName(filterRule));
- if ( !rs.isDefined() ) {
- grammar.tool.error("Filter rule "+filterRule+" does not exist in this lexer");
- }
- else if ( rs.access.equals("public") ) {
- grammar.tool.error("Filter rule "+filterRule+" must be protected");
- }
- }
- println("sa_m : INT := mark;");
- }
- }
- println("reset_text;");
- println("protect -- for char stream error handling");
- tabs++;
- // Generate try around whole thing to trap scanner errors
- println("protect -- for lexical error handling");
- tabs++;
- // Test for public lexical rules with empty paths
- for (int i=0; i<nextTokenBlk.getAlternatives().size(); i++) {
- Alternative a = nextTokenBlk.getAlternativeAt(i);
- if ( a.cache[1].containsEpsilon() ) {
- tool.warning("found optional path in nextToken()");
- }
- }
- // Generate the block
- String newline = System.getProperty("line.separator");
- JavaBlockFinishingInfo howToFinish = genCommonBlock(nextTokenBlk, false);
- String errFinish = "if ( LA(1) = EOF_CHAR ) then upon_eof; sa_return_token := make_token( ANTLR_COMMON_TOKEN::EOF_TYPE);";
- errFinish += newline+"tttt";
- if ( ((LexerGrammar)grammar).filterMode ) {
- if ( filterRule==null ) {
- errFinish += "telse consume; continue := false; end; -- if";
- }
- else {
- errFinish += "telse"+newline+
- "ttttttcommit;"+newline+
- "ttttttprotect" +newline+
- "tttttttm"+filterRule+"(false);"+newline+
- "ttttttwhen $ANTLR_RECOGNITION_EXCEPTION then"+newline+
- "ttttttt-- catastrophic failure"+newline+
- "tttttttreport_error( exception );"+newline+
- "tttttttconsume;"+newline+
- "ttttttend; -- protect"+newline+
- "ttttttcontinue := false;"+newline+
- "tttttend; -- if";
- }
- }
- else {
- errFinish += "ttelse " + throwNoViable + " end; -- if";
- }
- genBlockFinish(howToFinish, errFinish);
- // at this point a valid token has been matched, undo "mark" that was done
- if ( ((LexerGrammar)grammar).filterMode && filterRule!=null ) {
- println("if continue then");
- tabs++;
- println("commit;");
- tabs--;
- println("end; -- if");
- }
- // Generate literals test if desired
- // make sure _ttype is set first; note _returnToken must be
- // non-null as the rule was required to create it.
- println("if ( ~void(sa_return_token) and continue ) then;");
- tabs++;
- println("sa_ttype := sa_return_token.ttype;");
- if ( ((LexerGrammar)grammar).getTestLiterals()) {
- genLiteralsTest();
- }
- // return token created by rule reference in switch
- println("sa_return_token.ttype := sa_ttype;");
- println("return sa_return_token;");
- tabs--;
- println("end; -- if");
- // Close try block
- tabs--;
- println("when $ANTLR_RECOGNITION_EXCEPTION then");
- tabs++;
- if ( ((LexerGrammar)grammar).filterMode ) {
- if ( filterRule==null ) {
- println("if ( ~commit_to_path ) then");
- tabs++;
- println("consume;");
- tabs--;
- println("end; -- if");
- }
- else {
- println("if ( ~commit_to_path ) then");
- tabs++;
- println("rewind( sa_m );");
- println("reset_text;");
- println("protect");
- tabs++;
- println("m" + filterRule + "(false);");
- tabs--;
- println("when $ANTLR_RECOGNITION_EXCEPTION then");
- tabs++;
- println("-- horrendous failure: error in filter rule");
- println("report_error( exception );");
- println("consume;");
- tabs--;
- println("end; -- protect");
- tabs--;
- println("end; -- if");
- }
- }
- else {
- if ( nextTokenBlk.getDefaultErrorHandler() ) {
- println("report_error( exception );");
- println("consume;");
- }
- else {
- // pass on to invoking routine
- println("raise #ANTLR_TOKEN_STREAM_RECOGNITION_EXCEPTION( exception.str );");
- }
- }
- tabs--;
- println("end; -- protect");
- // close CharStreamException try
- tabs--;
- println("when $ANTLR_CHAR_STREAM_EXCEPTION then");
- tabs++;
- println("raise #ANTLR_TOKEN_STREAM_EXCEPTION( exception.message );");
- tabs--;
- println("end; -- protect");
- // close for-loop
- tabs--;
- println("end; -- loop");
- // close method nextToken
- tabs--;
- // println("}");
- println("end; -- next_token");
- println("");
- }
- /** Gen a named rule block.
- * ASTs are generated for each element of an alternative unless
- * the rule or the alternative have a '!' modifier.
- *
- * If an alternative defeats the default tree construction, it
- * must set <rule>_AST to the root of the returned AST.
- *
- * Each alternative that does automatic tree construction, builds
- * up root and child list pointers in an ASTPair structure.
- *
- * A rule finishes by setting the returnAST variable from the
- * ASTPair.
- *
- * @param rule The name of the rule to generate
- * @param startSymbol true if the rule is a start symbol (i.e., not referenced elsewhere)
- */
- public void genRule(RuleSymbol s, boolean startSymbol, int ruleNum) {
- tabs=1;
- if ( DEBUG_CODE_GENERATOR ) System.out.println("genRule("+ s.getId() +")");
- if ( !s.isDefined() ) {
- tool.error("undefined rule: "+ s.getId());
- return;
- }
- // reset the AST type to the one specified in the
- // PARSER's or TREE_WALKER's class parameter list
- labeledElementASTType = "AST";
- // Generate rule return type, name, arguments
- RuleBlock rblk = s.getBlock();
- currentRule = rblk;
- currentASTResult = s.getId();
- // Save the AST generation state, and set it to that of the rule
- boolean savegenAST = genAST;
- genAST = genAST && rblk.getAutoGen();
- // boolean oldsaveTest = saveText;
- saveText = rblk.getAutoGen();
- // print javadoc comment if any
- if ( s.comment!=null ) {
- _println(s.comment);
- }
- // Gen method name
- print( s.getId() );
- // sather requires enclosing parens only when the arglist is non-empty
- boolean hasArgs = false;
- // Additional rule parameters common to all rules for this grammar
- if (commonExtraParams.length() != 0 ) {
- hasArgs = true;
- _print("( " + commonExtraParams);
- if ( rblk.argAction != null )
- _print(",");
- }
- if ( rblk.argAction != null || rblk.returnAction != null )
- if ( !hasArgs ) {
- hasArgs = true;
- _print("( ");
- }
- // Gen arguments
- if (rblk.argAction != null)
- {
- // Has specified arguments
- print(rblk.argAction);
- }
- if ( hasArgs )
- _print(" )");
- // Gen method return type (note lexer return action set at rule creation)
- if (rblk.returnAction != null)
- {
- // Has specified return value
- _print( " : " + extractSatherTypeOfAction(rblk.returnAction) );
- }
- _println(" is");
- tabs++;
- // Convert return action to variable declaration
- if (rblk.returnAction != null)
- println(rblk.returnAction + ";");
- // print out definitions needed by rules for various grammar types
- println(commonLocalVars);
- if (grammar.traceRules) {
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("trace_in(""+ s.getId() +"",sa_t);");
- }
- else {
- println("trace_in(""+ s.getId() +"");");
- }
- }
- if ( grammar instanceof LexerGrammar ) {
- // lexer rule default return value is the rule's token name
- // This is a horrible hack to support the built-in EOF lexer rule.
- if (s.getId().equals("mEOF"))
- println("sa_ttype := ANTLR_COMMON_TOKEN::EOF_TYPE;");
- else
- println("sa_ttype := "+ s.getId().substring(1)+";");
- println("sa_save_index : INT;"); // used for element! (so we can kill text matched for element)
- /*
- println("boolean old_saveConsumedInput=saveConsumedInput;");
- if ( !rblk.getAutoGen() ) { // turn off "save input" if ! on rule
- println("saveConsumedInput=false;");
- }
- */
- }
- // if debugging, write code to mark entry to the rule
- if ( grammar.debuggingOutput)
- if (grammar instanceof ParserGrammar)
- println("fire_enter_rule( " + ruleNum + ", 0 );");
- else if (grammar instanceof LexerGrammar)
- println("fire_enter_rule( " + ruleNum + ", sa_ttype );");
- // Generate trace code if desired
- if ( grammar.debuggingOutput || grammar.traceRules) {
- println("protect -- debugging output");
- tabs++;
- }
- // Initialize AST variables
- if (grammar instanceof TreeWalkerGrammar) {
- // "Input" value for rule
- println( s.getId() + "_ast_in : " + labeledElementASTType + " := sa_t;");
- }
- if (grammar.buildAST) {
- // Parser member used to pass AST returns from rule invocations
- println("return_ast := void;");
- // Tracks AST construction
- println("current_ast ::= #ANTLR_AST_PAIR{AST};");
- // User-settable return value for rule.
- println( s.getId() + "_ast : " + labeledElementASTType + ";" );
- }
- genBlockPreamble(rblk);
- println("");
- // Search for an unlabeled exception specification attached to the rule
- ExceptionSpec unlabeledUserSpec = rblk.findExceptionSpec("");
- // Generate try block around the entire rule for error handling
- if (unlabeledUserSpec != null || rblk.getDefaultErrorHandler() ) {
- println("protect -- for error handling");
- tabs++;
- }
- // Generate the alternatives
- if ( rblk.alternatives.size()==1 ) {
- // One alternative -- use simple form
- Alternative alt = rblk.getAlternativeAt(0);
- String pred = alt.semPred;
- if ( pred!=null )
- genSemPred(pred, currentRule.line);
- if (alt.synPred != null) {
- tool.warning(
- "Syntactic predicate ignored for single alternative",
- grammar.getFilename(), alt.synPred.getLine()
- );
- }
- genAlt(alt, rblk);
- }
- else {
- // Multiple alternatives -- generate complex form
- boolean ok = grammar.theLLkAnalyzer.deterministic(rblk);
- JavaBlockFinishingInfo howToFinish = genCommonBlock(rblk, false);
- genBlockFinish(howToFinish, throwNoViable);
- }
- // Generate catch phrase for error handling
- if (unlabeledUserSpec != null || rblk.getDefaultErrorHandler() ) {
- // Close the try block
- tabs--;
- }
- // Generate user-defined or default catch phrases
- if (unlabeledUserSpec != null) {
- genErrorHandler(unlabeledUserSpec);
- }
- else if (rblk.getDefaultErrorHandler()) {
- // Generate default catch phrase
- println("when " + exceptionThrown + " then");
- tabs++;
- // Generate code to handle error if not guessing
- if (grammar.hasSyntacticPredicate) {
- println("if ( input_state.guessing = 0 ) then");
- tabs++;
- }
- println("report_error( exception );");
- if ( !(grammar instanceof TreeWalkerGrammar) ) {
- // Generate code to consume until token in k==1 follow set
- Lookahead follow = grammar.theLLkAnalyzer.FOLLOW(1, rblk.endNode);
- String followSetName = "sa" + getBitsetName(markBitsetForGen(follow.fset));
- println("consume;");
- println("consume_until( " + followSetName + " );");
- } else {
- // Just consume one token
- println("if ( ~void(sa_t) ) then");
- tabs++;
- println("sa_t := sa_t.next_sibling;");
- tabs--;
- println("end; -- if");
- }
- if (grammar.hasSyntacticPredicate) {
- tabs--;
- // When guessing, rethrow exception
- println("else");
- tabs++;
- println("raise exception;");
- tabs--;
- println("end; -- if");
- }
- // Close catch phrase
- tabs--;
- println("end; -- protect");
- }
- // Squirrel away the AST "return" value
- if (grammar.buildAST) {
- println("return_ast := " + s.getId() + "_ast;");
- }
- // Set return tree value for tree walkers
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("sa_ret_tree := sa_t;");
- }
- // Generate literals test for lexer rules so marked
- if (rblk.getTestLiterals()) {
- if ( s.access.equals("protected") ) {
- genLiteralsTestForPartialToken();
- }
- else {
- genLiteralsTest();
- }
- }
- // if doing a lexer rule, dump code to create token if necessary
- if ( grammar instanceof LexerGrammar ) {
- println("if ( sa_create_token and void(sa_token) and sa_ttype /= ANTLR_COMMON_TOKEN::SKIP ) then");
- tabs++;
- println("sa_token := make_token( sa_ttype );");
- println("sa_token.text := text.substring( sa_begin, text.length - sa_begin );");
- tabs--;
- println("end; -- if");
- println("sa_return_token := sa_token;");
- }
- // Gen the return statement if there is one (lexer has hard-wired return action)
- if (rblk.returnAction != null) {
- println("return " + extractSatherIdOfAction( rblk.returnAction, rblk.getLine() ) + ";");
- }
- if ( grammar.debuggingOutput || grammar.traceRules) {
- // since Sather doesn't have anything similar to
- // Java's try..finally, we use a protect..when
- // instead. However, we mimic finally by printing
- // debugging statements regardless of whether an
- // exception is thrown
- tabs--;
- println("when $STR then -- assume this will catch everything");
- tabs++;
- // cache debugging statements since they will have
- // to printed twice
- String fire = null;
- String trace = null;
- // If debugging, generate calls to mark exit of rule
- if ( grammar.debuggingOutput)
- if (grammar instanceof ParserGrammar)
- fire = "fire_exit_rule(" + ruleNum + ",0);";
- else if (grammar instanceof LexerGrammar)
- fire = "fire_exit_rule(" + ruleNum + ", sa_ttype);";
- if (grammar.traceRules) {
- if ( grammar instanceof TreeWalkerGrammar ) {
- trace = "trace_out(""+ s.getId() +"", sa_t);";
- }
- else {
- trace = "trace_out(""+ s.getId() +"");";
- }
- }
- // these are printed _inside_ the when class of the
- // Sather protect statement.
- if ( fire != null )
- println(fire);
- if ( trace != null )
- println(trace);
- // rethrow exception
- println("raise exception;");
- tabs--;
- println("end; -- protect");
- // these are printed _outside_ the when class of the
- // Sather protect statement.
- if ( fire != null )
- println(fire);
- if ( trace != null )
- println(trace);
- }
- tabs--;
- println("end; -- rule");
- println("");
- // Restore the AST generation state
- genAST = savegenAST;
- // restore char save state
- // saveText = oldsaveTest;
- }
- private void GenRuleInvocation(RuleRefElement rr) {
- // dump rule name
- _print( rr.targetRule );
- boolean hasArgs = false; // flag to let us know if we need to close the arg list
- // lexers must tell rule if it should set _returnToken
- if ( grammar instanceof LexerGrammar ) {
- // if labeled, could access Token, so tell rule to create
- hasArgs = true;
- if ( rr.getLabel() != null ) {
- _print("( true");
- }
- else {
- _print("( false");
- }
- if (commonExtraArgs.length() != 0 || rr.args!=null ) {
- _print(",");
- }
- }
- else if ( commonExtraArgs.length() != 0 || rr.args != null ) {
- hasArgs = true;
- _print( "( " );
- }
- // Extra arguments common to all rules for this grammar
- _print(commonExtraArgs);
- if (commonExtraArgs.length() != 0 && rr.args!=null ) {
- _print(",");
- }
- // Process arguments to method, if any
- RuleSymbol rs = (RuleSymbol)grammar.getSymbol(rr.targetRule);
- if (rr.args != null)
- {
- // When not guessing, execute user arg action
- ActionTransInfo tInfo = new ActionTransInfo();
- String args = processActionForTreeSpecifiers(rr.args, 0, currentRule, tInfo);
- if ( tInfo.assignToRoot || tInfo.refRuleRoot!=null ) {
- tool.error("Arguments of rule reference '" + rr.targetRule + "' cannot set or ref #"+
- currentRule.getRuleName()+" on line "+rr.getLine());
- }
- _print(args);
- // Warn if the rule accepts no arguments
- if (rs.block.argAction == null) {
- tool.warning("Rule '" + rr.targetRule + "' accepts no arguments", grammar.getFilename(), rr.getLine());
- }
- } else {
- // For C++, no warning if rule has parameters, because there may be default
- // values for all of the parameters
- if (rs.block.argAction != null) {
- tool.warning("Missing parameters on reference to rule "+rr.targetRule, grammar.getFilename(), rr.getLine());
- }
- }
- if ( hasArgs )
- _print(" )");
- _println(";");
- // move down to the first child while parsing
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("sa_t := sa_ret_tree;");
- }
- }
- protected void genSemPred(String pred, int line) {
- // translate $ and # references
- ActionTransInfo tInfo = new ActionTransInfo();
- pred = processActionForTreeSpecifiers(pred, line, currentRule, tInfo);
- // ignore translation info...we don't need to do anything with it.
- String escapedPred = charFormatter.escapeString(pred);
- // if debugging, wrap the semantic predicate evaluation in a method
- // that can tell SemanticPredicateListeners the result
- if (grammar.debuggingOutput && ((grammar instanceof ParserGrammar) || (grammar instanceof LexerGrammar)))
- pred = "fireSemanticPredicateEvaluated(antlr.debug.SemanticPredicateEvent.VALIDATING,"
- + addSemPred(escapedPred) + "," + pred + ")";
- println("if (!(" + pred + "))");
- println(" throw new SemanticException("" + escapedPred + "");");
- }
- /** Write an array of Strings which are the semantic predicate
- * expressions. The debugger will reference them by number only
- */
- protected void genSemPredMap() {
- Enumeration e = semPreds.elements();
- println("private String _semPredNames[] = {");
- while(e.hasMoreElements())
- println("""+e.nextElement()+"",");
- println("};");
- }
- protected void genSynPred(SynPredBlock blk, String lookaheadExpr) {
- if ( DEBUG_CODE_GENERATOR ) System.out.println("gen=>("+blk+")");
- // Dump synpred result variable
- println("syn_pred_matched" + blk.ID + " : BOOL := false;");
- // Gen normal lookahead test
- println("if (" + lookaheadExpr + ") then");
- tabs++;
- // Save input state
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("sa__t" + blk.ID + " " + labeledElementASTType + " := sa_t;");
- }
- else {
- println("sa_m" + blk.ID + " : INT := mark;");
- }
- // Once inside the try, assume synpred works unless exception caught
- println("syn_pred_matched" + blk.ID + " := true;");
- println("input_state.guessing := input_state.guessing + 1;");
- // if debugging, tell listeners that a synpred has started
- if (grammar.debuggingOutput && ((grammar instanceof ParserGrammar) ||
- (grammar instanceof LexerGrammar))) {
- println("fireSyntacticPredicateStarted();");
- }
- syntacticPredLevel++;
- println("protect");
- tabs++;
- gen((AlternativeBlock)blk); // gen code to test predicate
- tabs--;
- //println("System.out.println("pred "+blk+" succeeded");");
- println( "when " + exceptionThrown + " then" );
- tabs++;
- println( "syn_pred_matched" + blk.ID +" := false;");
- //println("System.out.println("pred "+blk+" failed");");
- tabs--;
- println("end; -- protect");
- // Restore input state
- if ( grammar instanceof TreeWalkerGrammar ) {
- println("sa_t := sa__t" + blk.ID + ";");
- }
- else {
- println("rewind( sa_m" + blk.ID + " );");
- }
- println("input_state.guessing := input_state.guessing - 1;");
- // if debugging, tell listeners how the synpred turned out
- if (grammar.debuggingOutput && ((grammar instanceof ParserGrammar) ||
- (grammar instanceof LexerGrammar))) {
- println("if ( syn_pred_matched" + blk.ID + " ) then" );
- tabs++;
- println("fireSyntacticPredicateSucceeded();" );
- tabs--;
- println("else");
- tabs++;
- println(" fireSyntacticPredicateFailed();");
- tabs--;
- println("end; -- if");
- }
- syntacticPredLevel--;
- tabs--;
- // Close lookahead test
- println("end; -- if");
- // Test synred result
- println("if ( syn_pred_matched" + blk.ID + " ) then");
- }
- /** Generate a static array containing the names of the tokens,
- * indexed by the token type values. This static array is used
- * to format error messages so that the token identifers or literal
- * strings are displayed instead of the token numbers.
- *
- * If a lexical rule has a paraphrase, use it rather than the
- * token label.
- */
- public void genTokenStrings() {
- // Generate a string for each token. This creates a static
- // array of Strings indexed by token type.
- println("");
- println("const sa_token_names : ARRAY{STR} := |");
- tabs++;
- // Walk the token vocabulary and generate a Vector of strings
- // from the tokens.
- Vector v = grammar.tokenManager.getVocabulary();
- for (int i = 0; i < v.size(); i++)
- {
- String s = (String)v.elementAt(i);
- if (s == null)
- {
- s = "<"+String.valueOf(i)+">";
- }
- if ( !s.startsWith(""") && !s.startsWith("<") ) {
- TokenSymbol ts = (TokenSymbol)grammar.tokenManager.getTokenSymbol(s);
- if ( ts!=null && ts.getParaphrase()!=null ) {
- s = antlr.Tool.stripFrontBack(ts.getParaphrase(), """, """);
- }
- }
- print(charFormatter.literalString(s));
- if (i != v.size()-1) {
- _print(",");
- }
- _println("");
- }
- // Close the string array initailizer
- tabs--;
- println("|;");
- }
- /** Generate the token types Java file */
- protected void genTokenTypes(TokenManager tm) throws IOException {
- // Open the token output Java file and set the currentOutput stream
- // SAS: file open was moved to a method so a subclass can override
- // This was mainly for the VAJ interface
- setupOutput(tm.getName() + "_" + TokenTypesFileSuffix.toUpperCase() );
- tabs = 0;
- // Generate the header common to all Java files
- genHeader();
- // Do not use printAction because we assume tabs==0
- println((String)behavior.headerActions.get(""));
- // Encapsulate the definitions in an interface. This can be done
- // because they are all constants.
- println("class " + tm.getName() + "_" + TokenTypesFileSuffix.toUpperCase() +" is");
- tabs++;
- // Generate a definition for each token type
- Vector v = tm.getVocabulary();
- // Do special tokens manually
- println("const EOF : INT := " + Token.EOF_TYPE + ";");
- println("const NULL_TREE_LOOKAHEAD : INT := " + Token.NULL_TREE_LOOKAHEAD + ";");
- for (int i = Token.MIN_USER_TYPE; i < v.size(); i++) {
- String s = (String)v.elementAt(i);
- if (s != null) {
- if ( s.startsWith(""") ) {
- // a string literal
- StringLiteralSymbol sl = (StringLiteralSymbol)tm.getTokenSymbol(s);
- if ( sl==null ) {
- antlr.Tool.panic("String literal "+s+" not in symbol table");
- }
- else if ( sl.label != null ) {
- println( "const " + sl.label + " : INT" + " := " + i + ";");
- }
- else {
- String mangledName = mangleLiteral(s);
- if (mangledName != null) {
- // We were able to create a meaningful mangled token name
- println( "const " + mangledName + " : INT" + " := " + i + ";");
- // if no label specified, make the label equal to the mangled name
- sl.label = mangledName;
- }
- else {
- println("-- " + s + " := " + i);
- }
- }
- }
- else if ( !s.startsWith("<") ) {
- println( "const " + s + " : INT" + " := " + i + ";");
- }
- }
- }
- // convert the CHAR array into CHAR_SET instance
- println("");
- println("bitset ( bool_array : ARRAY{BOOL} ) : CHAR_SET is");
- tabs++;
- println( "return #CHAR_SET( bool_array );" );
- tabs--;
- println("end;");
- // convert the INT array into INT_SET instance
- println("");
- println("int_set ( int_array : ARRAY{INT} ) : INT_SET is");
- tabs++;
- println( "return #INT_SET( int_array );" );
- tabs--;
- println("end;");
- // Close the interface
- tabs--;
- println("end; -- class");
- // Close the tokens output file
- currentOutput.close();
- currentOutput = null;
- exitIfError();
- }
- /** Get a string for an expression to generate creation of an AST subtree.
- * @param v A Vector of String, where each element is an expression in the target language yielding an AST node.
- */
- public String getASTCreateString(Vector v) {
- if (v.size() == 0) {
- return "";
- }
- StringBuffer buf = new StringBuffer();
- buf.append("ANTLR_AST_UTIL{AST}::make( ( #ANTLR_AST_ARRAY{AST}(" + v.size() + "))");
- for (int i = 0; i < v.size(); i++) {
- buf.append(".add(" + v.elementAt(i) + ")");
- }
- buf.append(")");
- return buf.toString();
- }
- /** Get a string for an expression to generate creating of an AST node
- * @param str The arguments to the AST constructor
- */
- public String getASTCreateString(GrammarAtom atom, String str) {
- if ( atom!=null && atom.getASTNodeType() != null ) {
- return atom.getASTNodeType() + "::create("+str + ")";
- }
- else {
- return labeledElementASTType + "::create( " + str + " )";
- }
- }
- protected String getLookaheadTestExpression(Lookahead[] look, int k)
- {
- StringBuffer e = new StringBuffer(100);
- boolean first = true;
- e.append("(");
- for (int i = 1; i <= k; i++) {
- BitSet p = look[i].fset;
- if (!first) {
- e.append(") and (");
- }
- first = false;
- // Syn preds can yield <end-of-syn-pred> (epsilon) lookahead.
- // There is no way to predict what that token would be. Just
- // allow anything instead.
- if (look[i].containsEpsilon()) {
- e.append("true");
- } else {
- e.append(getLookaheadTestTerm(i, p));
- }
- }
- e.append(")");
- return e.toString();
- }
- /**Generate a lookahead test expression for an alternate. This
- * will be a series of tests joined by '&&' and enclosed by '()',
- * the number of such tests being determined by the depth of the lookahead.
- */
- protected String getLookaheadTestExpression(Alternative alt, int maxDepth) {
- int depth = alt.lookaheadDepth;
- if ( depth == GrammarAnalyzer.NONDETERMINISTIC ) {
- // if the decision is nondeterministic, do the best we can: LL(k)
- // any predicates that are around will be generated later.
- depth = grammar.maxk;
- }
- if ( maxDepth==0 ) {
- // empty lookahead can result from alt with sem pred
- // that can see end of token. E.g., A : {pred}? ('a')? ;
- return "true";
- }
- /*
- boolean first = true;
- for (int i=1; i<=depth && i<=maxDepth; i++) {
- BitSet p = alt.cache[i].fset;
- if (!first) {
- e.append(") and (");
- }
- first = false;
- // Syn preds can yield <end-of-syn-pred> (epsilon) lookahead.
- // There is no way to predict what that token would be. Just
- // allow anything instead.
- if ( alt.cache[i].containsEpsilon() ) {
- e.append("true");
- }
- else {
- e.append(getLookaheadTestTerm(i, p));
- }
- }
- e.append(")");
- */
- return "(" + getLookaheadTestExpression(alt.cache,depth) + ")";
- }
- /**Generate a depth==1 lookahead test expression given the BitSet.
- * This may be one of:
- * 1) a series of 'x==X||' tests
- * 2) a range test using >= && <= where possible,
- * 3) a bitset membership test for complex comparisons
- * @param k The lookahead level
- * @param p The lookahead set for level k
- */
- protected String getLookaheadTestTerm(int k, BitSet p) {
- // Determine the name of the item to be compared
- String ts = lookaheadString(k);
- // Generate a range expression if possible
- int[] elems = p.toArray();
- if (elementsAreRange(elems)) {
- return getRangeExpression(k, elems);
- }
- // Generate a bitset membership test if possible
- StringBuffer e;
- int degree = p.degree();
- if ( degree == 0 ) {
- return "true";
- }
- if ( degree >= bitsetTestThreshold) {
- int bitsetIdx = markBitsetForGen(p);
- return "sa" + getBitsetName(bitsetIdx) + ".member(" + ts + ")";
- }
- // Otherwise, generate the long-winded series of "x==X||" tests
- e = new StringBuffer();
- for (int i = 0; i < elems.length; i++) {
- // Get the compared-to item (token or character value)
- String cs = getValueString(elems[i]);
- // Generate the element comparison
- if ( i>0 ) e.append(" or ");
- e.append(ts);
- e.append("=");
- e.append(cs);
- }
- return e.toString();
- }
- /** Return an expression for testing a contiguous renage of elements
- * @param k The lookahead level
- * @param elems The elements representing the set, usually from BitSet.toArray().
- * @return String containing test expression.
- */
- public String getRangeExpression(int k, int[] elems) {
- if (!elementsAreRange(elems)) {
- tool.panic("getRangeExpression called with non-range");
- }
- int begin = elems[0];
- int end = elems[elems.length-1];
- return
- "(" + lookaheadString(k) + " >= " + getValueString(begin) + " and " +
- lookaheadString(k) + " <= " + getValueString(end) + ")";
- }
- /** getValueString: get a string representation of a token or char value
- * @param value The token or char value
- */
- private String getValueString(int value) {
- String cs;
- if ( grammar instanceof LexerGrammar ) {
- cs = charFormatter.literalChar(value);
- }
- else {
- TokenSymbol ts = grammar.tokenManager.getTokenSymbolAt(value);
- if ( ts == null ) {
- return ""+value; // return token type as string
- // tool.panic("vocabulary for token type " + value + " is null");
- }
- String tId = ts.getId();
- if ( ts instanceof StringLiteralSymbol ) {
- // if string literal, use predefined label if any
- // if no predefined, try to mangle into LITERAL_xxx.
- // if can't mangle, use int value as last resort
- StringLiteralSymbol sl = (StringLiteralSymbol)ts;
- String label = sl.getLabel();
- if ( label!=null ) {
- cs = label;
- }
- else {
- cs = mangleLiteral(tId);
- if (cs == null) {
- cs = String.valueOf(value);
- }
- }
- }
- else {
- cs = tId;
- }
- }
- return cs;
- }
- /**Is the lookahead for this alt empty? */
- protected boolean lookaheadIsEmpty(Alternative alt, int maxDepth) {
- int depth = alt.lookaheadDepth;
- if ( depth == GrammarAnalyzer.NONDETERMINISTIC ) {
- depth = grammar.maxk;
- }
- for (int i=1; i<=depth && i<=maxDepth; i++) {
- BitSet p = alt.cache[i].fset;
- if (p.degree() != 0) {
- return false;
- }
- }
- return true;
- }
- private String lookaheadString(int k) {
- if (grammar instanceof TreeWalkerGrammar) {
- return "sa_t.ttype";
- }
- return "LA(" + k + ")";
- }
- /** Mangle a string literal into a meaningful token name. This is
- * only possible for literals that are all characters. The resulting
- * mangled literal name is literalsPrefix with the text of the literal
- * appended.
- * @return A string representing the mangled literal, or null if not possible.
- */
- private String mangleLiteral(String s) {
- String mangled = antlr.Tool.literalsPrefix;
- for (int i = 1; i < s.length()-1; i++) {
- if (!Character.isLetter(s.charAt(i)) &&
- s.charAt(i) != '_') {
- return null;
- }
- mangled += s.charAt(i);
- }
- if ( antlr.Tool.upperCaseMangledLiterals ) {
- mangled = mangled.toUpperCase();
- }
- return mangled;
- }
- /** Map an identifier to it's corresponding tree-node variable.
- * This is context-sensitive, depending on the rule and alternative
- * being generated
- * @param idParam The identifier name to map
- * @return The mapped id (which may be the same as the input), or null if the mapping is invalid due to duplicates
- */
- public String mapTreeId(String idParam, ActionTransInfo transInfo) {
- // if not in an action of a rule, nothing to map.
- if ( currentRule==null ) return idParam;
- boolean in_var = false;
- String id = idParam;
- if (grammar instanceof TreeWalkerGrammar) {
- if ( !grammar.buildAST ) {
- in_var = true;
- }
- // If the id ends with "_in", then map it to the input variable
- else if (id.length() > 3 && id.lastIndexOf("_in") == id.length()-3) {
- // Strip off the "_in"
- id = id.substring(0, id.length()-3);
- in_var = true;
- }
- }
- // Check the rule labels. If id is a label, then the output
- // variable is label_AST, and the input variable is plain label.
- for (int i = 0; i < currentRule.labeledElements.size(); i++) {
- AlternativeElement elt = (AlternativeElement)currentRule.labeledElements.elementAt(i);
- if (elt.getLabel().equals(id)) {
- return in_var ? id : id + "_ast";
- }
- }
- // Failing that, check the id-to-variable map for the alternative.
- // If the id is in the map, then output variable is the name in the
- // map, and input variable is name_in
- String s = (String)treeVariableMap.get(id);
- if (s != null) {
- if (s == NONUNIQUE) {
- // There is more than one element with this id
- return null;
- } else if (s.equals(currentRule.getRuleName())) {
- // a recursive call to the enclosing rule is
- // ambiguous with the rule itself.
- return null;
- } else {
- return in_var ? s + "_in" : s;
- }
- }
- // Failing that, check the rule name itself. Output variable
- // is rule_AST; input variable is rule_AST_in (treeparsers).
- if (id.equals(currentRule.getRuleName())) {
- String r = in_var ? id + "_ast_in" : id + "_ast";
- if ( transInfo!=null ) {
- if ( !in_var ) {
- transInfo.refRuleRoot = r;
- }
- }
- return r;
- } else {
- // id does not map to anything -- return itself.
- return id;
- }
- }
- /** Given an element and the name of an associated AST variable,
- * create a mapping between the element "name" and the variable name.
- */
- private void mapTreeVariable(AlternativeElement e, String name)
- {
- // For tree elements, defer to the root
- if (e instanceof TreeElement) {
- mapTreeVariable( ((TreeElement)e).root, name);
- return;
- }
- // Determine the name of the element, if any, for mapping purposes
- String elName = null;
- // Don't map labeled items
- if (e.getLabel() == null) {
- if (e instanceof TokenRefElement) {
- // use the token id
- elName = ((TokenRefElement)e).atomText;
- }
- else if (e instanceof RuleRefElement) {
- // use the rule name
- elName = ((RuleRefElement)e).targetRule;
- }
- }
- // Add the element to the tree variable map if it has a name
- if (elName != null) {
- if (treeVariableMap.get(elName) != null) {
- // Name is already in the map -- mark it as duplicate
- treeVariableMap.remove(elName);
- treeVariableMap.put(elName, NONUNIQUE);
- }
- else {
- treeVariableMap.put(elName, name);
- }
- }
- }
- private void setupGrammarParameters(Grammar g) {
- if (g instanceof ParserGrammar) {
- labeledElementASTType = "AST";
- /*
- if ( g.hasOption("ASTLabelType") ) {
- Token tsuffix = g.getOption("ASTLabelType");
- if ( tsuffix != null ) {
- String suffix = Tool.stripFrontBack(tsuffix.getText(),""",""");
- if ( suffix != null ) {
- labeledElementASTType = suffix;
- }
- }
- }
- */
- labeledElementType = "$ANTLR_TOKEN";
- labeledElementInit = "void";
- commonExtraArgs = "";
- commonExtraParams = "";
- commonLocalVars = "";
- lt1Value = "LT(1)";
- exceptionThrown = "$ANTLR_RECOGNITION_EXCEPTION";
- throwNoViable = "raise ANTLR_NO_VIABLE_ALT_EXCEPTION{AST}::create_from_token(LT(1), file_name );";
- }
- else if (g instanceof LexerGrammar) {
- labeledElementType = "CHAR ";
- labeledElementInit = "'\0'";
- commonExtraArgs = "";
- commonExtraParams = "sa_create_token : BOOL";
- commonLocalVars = "sa_ttype : INT; sa_token : TOKEN; sa_begin : INT := text.length;";
- lt1Value = "LA(1)";
- exceptionThrown = "$ANTLR_RECOGNITION_EXCEPTION";
- throwNoViable = "raise #ANTLR_NO_VIABLE_ALT_FOR_CHAR_EXCEPTION( LA(1), file_name, line );";
- }
- else if (g instanceof TreeWalkerGrammar) {
- labeledElementASTType = "AST";
- labeledElementType = labeledElementASTType;
- /*
- if ( g.hasOption("ASTLabelType") ) {
- Token tsuffix = g.getOption("ASTLabelType");
- if ( tsuffix != null ) {
- String suffix = Tool.stripFrontBack(tsuffix.getText(),""",""");
- if ( suffix != null ) {
- labeledElementASTType = suffix;
- labeledElementType = suffix;
- }
- }
- }
- */
- if ( !g.hasOption("ASTLabelType") ) {
- g.setOption("ASTLabelType", new Token(ANTLRTokenTypes.STRING_LITERAL,"AST"));
- }
- labeledElementInit = "void";
- commonExtraArgs = "sa_t";
- commonExtraParams = "sa_t : " + labeledElementASTType;
- commonLocalVars = "";
- lt1Value = "sa_t";
- exceptionThrown = "$ANTLR_RECOGNITION_EXCEPTION";
- throwNoViable = "raise #ANTLR_NO_VIABLE_ALT_EXCEPTION{AST}(sa_t);";
- }
- else {
- tool.panic("Unknown grammar type");
- }
- }
- /** This method exists so a subclass, namely VAJCodeGenerator,
- * can open the file in its own evil way. JavaCodeGenerator
- * simply opens a text file...
- */
- public void setupOutput(String className) throws IOException {
- currentOutput = antlr.Tool.openOutputFile(className + ".sa");
- }
- private static int satherBlockId = 0;
- private static synchronized String getNextSatherPrefix() {
- // This method is required to overcome Sather's lack of
- // multiple/nested scopes within a method. It is used as a
- // prefix to newly declared variables. The prefix disambiguates
- // varaible names that in the Java/Cpp code generators would have
- // resided in their own scope.
- String label = "sa" + satherBlockId;
- satherBlockId++;
- return label;
- }
- protected String extractSatherTypeOfAction(String s) {
- int s_length = s.length();
- for (int i = s_length - 1 ; i >=0 ; i-- ) {
- if ( s.charAt(i) == ':' ) {
- return s.substring( i+1 , s_length - 1 );
- }
- }
- tool.warning("Unable to determine Sather type" );
- return "";
- }
- protected String extractSatherIdOfAction(String s, int line) {
- int s_length = s.length();
- for (int i = s_length - 1 ; i >=0 ; i-- ) {
- if ( s.charAt(i) == ':' ) {
- return s.substring( 0 , i - 1 );
- }
- }
- tool.warning("Unable to determine Sather return identifier");
- return "";
- }
- /** Lexically process tree-specifiers in the action.
- * This will replace @id and @(...) with the appropriate
- * function calls and/or variables.
- *
- * Override the default implementation inherited from CodeGenerator
- * in order to instantiate the Sather's ActionLexer rather than Java's
- */
- protected String processActionForTreeSpecifiers( String actionStr,
- int line,
- RuleBlock currentRule,
- ActionTransInfo tInfo )
- {
- if ( actionStr==null ) return null;
- // The action trans info tells us (at the moment) whether an
- // assignment was done to the rule's tree root.
- if (grammar==null) return actionStr;
- if ( (grammar.buildAST && actionStr.indexOf('@') != -1) ||
- grammar instanceof TreeWalkerGrammar ||
- (grammar instanceof LexerGrammar && actionStr.indexOf('%') != -1) ) {
- // Create a lexer to read an action and return the translated version
- antlr.actions.sather.ActionLexer lexer =
- new antlr.actions.sather.ActionLexer(actionStr, currentRule, this, tInfo);
- lexer.setLineOffset(line);
- lexer.setTool(tool);
- try {
- lexer.mACTION(true);
- actionStr = lexer.getTokenObject().getText();
- // System.out.println("action translated: "+actionStr);
- // System.out.println("trans info is "+tInfo);
- }
- catch (RecognitionException ex) {
- lexer.reportError(ex);
- return actionStr;
- }
- catch (TokenStreamException tex) {
- antlr.Tool.panic("Error reading action:"+actionStr);
- return actionStr;
- }
- catch (CharStreamException io) {
- antlr.Tool.panic("Error reading action:"+actionStr);
- return actionStr;
- }
- }
- return actionStr;
- }
- }