Search in sources :

Example 1 with GrammarAST

use of org.antlr.v4.tool.ast.GrammarAST in project antlr4 by antlr.

the class Grammar method getStateToGrammarRegionMap.

public static Map<Integer, Interval> getStateToGrammarRegionMap(GrammarRootAST ast, IntervalSet grammarTokenTypes) {
    Map<Integer, Interval> stateToGrammarRegionMap = new HashMap<Integer, Interval>();
    if (ast == null)
        return stateToGrammarRegionMap;
    List<GrammarAST> nodes = ast.getNodesWithType(grammarTokenTypes);
    for (GrammarAST n : nodes) {
        if (n.atnState != null) {
            Interval tokenRegion = Interval.of(n.getTokenStartIndex(), n.getTokenStopIndex());
            org.antlr.runtime.tree.Tree ruleNode = null;
            // RULEs, BLOCKs of transformed recursive rules point to original token interval
            switch(n.getType()) {
                case ANTLRParser.RULE:
                    ruleNode = n;
                    break;
                case ANTLRParser.BLOCK:
                case ANTLRParser.CLOSURE:
                    ruleNode = n.getAncestor(ANTLRParser.RULE);
                    break;
            }
            if (ruleNode instanceof RuleAST) {
                String ruleName = ((RuleAST) ruleNode).getRuleName();
                Rule r = ast.g.getRule(ruleName);
                if (r instanceof LeftRecursiveRule) {
                    RuleAST originalAST = ((LeftRecursiveRule) r).getOriginalAST();
                    tokenRegion = Interval.of(originalAST.getTokenStartIndex(), originalAST.getTokenStopIndex());
                }
            }
            stateToGrammarRegionMap.put(n.atnState.stateNumber, tokenRegion);
        }
    }
    return stateToGrammarRegionMap;
}
Also used : RuleAST(org.antlr.v4.tool.ast.RuleAST) OrderedHashMap(org.antlr.v4.misc.OrderedHashMap) HashMap(java.util.HashMap) LinkedHashMap(java.util.LinkedHashMap) GrammarAST(org.antlr.v4.tool.ast.GrammarAST) Interval(org.antlr.v4.runtime.misc.Interval)

Example 2 with GrammarAST

use of org.antlr.v4.tool.ast.GrammarAST in project antlr4 by antlr.

the class Grammar method loadImportedGrammars.

public void loadImportedGrammars() {
    if (ast == null)
        return;
    GrammarAST i = (GrammarAST) ast.getFirstChildWithType(ANTLRParser.IMPORT);
    if (i == null)
        return;
    Set<String> visited = new HashSet<>();
    visited.add(this.name);
    importedGrammars = new ArrayList<Grammar>();
    for (Object c : i.getChildren()) {
        GrammarAST t = (GrammarAST) c;
        String importedGrammarName = null;
        if (t.getType() == ANTLRParser.ASSIGN) {
            t = (GrammarAST) t.getChild(1);
            importedGrammarName = t.getText();
        } else if (t.getType() == ANTLRParser.ID) {
            importedGrammarName = t.getText();
        }
        if (visited.contains(importedGrammarName)) {
            // ignore circular refs
            continue;
        }
        Grammar g;
        try {
            g = tool.loadImportedGrammar(this, t);
        } catch (IOException ioe) {
            tool.errMgr.grammarError(ErrorType.ERROR_READING_IMPORTED_GRAMMAR, importedGrammarName, t.getToken(), importedGrammarName, name);
            continue;
        }
        // did it come back as error node or missing?
        if (g == null)
            continue;
        g.parent = this;
        importedGrammars.add(g);
        // recursively pursue any imports in this import
        g.loadImportedGrammars();
    }
}
Also used : GrammarAST(org.antlr.v4.tool.ast.GrammarAST) IOException(java.io.IOException) HashSet(java.util.HashSet) LinkedHashSet(java.util.LinkedHashSet)

Example 3 with GrammarAST

use of org.antlr.v4.tool.ast.GrammarAST in project antlr4 by antlr.

the class GrammarTransformPipeline method extractImplicitLexer.

/** Build lexer grammar from combined grammar that looks like:
	 *
	 *  (COMBINED_GRAMMAR A
	 *      (tokens { X (= Y 'y'))
	 *      (OPTIONS (= x 'y'))
	 *      (@ members {foo})
	 *      (@ lexer header {package jj;})
	 *      (RULES (RULE .+)))
	 *
	 *  Move rules and actions to new tree, don't dup. Split AST apart.
	 *  We'll have this Grammar share token symbols later; don't generate
	 *  tokenVocab or tokens{} section.  Copy over named actions.
	 *
	 *  Side-effects: it removes children from GRAMMAR &amp; RULES nodes
	 *                in combined AST.  Anything cut out is dup'd before
	 *                adding to lexer to avoid "who's ur daddy" issues
	 */
public GrammarRootAST extractImplicitLexer(Grammar combinedGrammar) {
    GrammarRootAST combinedAST = combinedGrammar.ast;
    //tool.log("grammar", "before="+combinedAST.toStringTree());
    GrammarASTAdaptor adaptor = new GrammarASTAdaptor(combinedAST.token.getInputStream());
    GrammarAST[] elements = combinedAST.getChildren().toArray(new GrammarAST[0]);
    // MAKE A GRAMMAR ROOT and ID
    String lexerName = combinedAST.getChild(0).getText() + "Lexer";
    GrammarRootAST lexerAST = new GrammarRootAST(new CommonToken(ANTLRParser.GRAMMAR, "LEXER_GRAMMAR"), combinedGrammar.ast.tokenStream);
    lexerAST.grammarType = ANTLRParser.LEXER;
    lexerAST.token.setInputStream(combinedAST.token.getInputStream());
    lexerAST.addChild((GrammarAST) adaptor.create(ANTLRParser.ID, lexerName));
    // COPY OPTIONS
    GrammarAST optionsRoot = (GrammarAST) combinedAST.getFirstChildWithType(ANTLRParser.OPTIONS);
    if (optionsRoot != null && optionsRoot.getChildCount() != 0) {
        GrammarAST lexerOptionsRoot = (GrammarAST) adaptor.dupNode(optionsRoot);
        lexerAST.addChild(lexerOptionsRoot);
        GrammarAST[] options = optionsRoot.getChildren().toArray(new GrammarAST[0]);
        for (GrammarAST o : options) {
            String optionName = o.getChild(0).getText();
            if (Grammar.lexerOptions.contains(optionName) && !Grammar.doNotCopyOptionsToLexer.contains(optionName)) {
                GrammarAST optionTree = (GrammarAST) adaptor.dupTree(o);
                lexerOptionsRoot.addChild(optionTree);
                lexerAST.setOption(optionName, (GrammarAST) optionTree.getChild(1));
            }
        }
    }
    // COPY all named actions, but only move those with lexer:: scope
    List<GrammarAST> actionsWeMoved = new ArrayList<GrammarAST>();
    for (GrammarAST e : elements) {
        if (e.getType() == ANTLRParser.AT) {
            lexerAST.addChild((Tree) adaptor.dupTree(e));
            if (e.getChild(0).getText().equals("lexer")) {
                actionsWeMoved.add(e);
            }
        }
    }
    for (GrammarAST r : actionsWeMoved) {
        combinedAST.deleteChild(r);
    }
    GrammarAST combinedRulesRoot = (GrammarAST) combinedAST.getFirstChildWithType(ANTLRParser.RULES);
    if (combinedRulesRoot == null)
        return lexerAST;
    // MOVE lexer rules
    GrammarAST lexerRulesRoot = (GrammarAST) adaptor.create(ANTLRParser.RULES, "RULES");
    lexerAST.addChild(lexerRulesRoot);
    List<GrammarAST> rulesWeMoved = new ArrayList<GrammarAST>();
    GrammarASTWithOptions[] rules;
    if (combinedRulesRoot.getChildCount() > 0) {
        rules = combinedRulesRoot.getChildren().toArray(new GrammarASTWithOptions[0]);
    } else {
        rules = new GrammarASTWithOptions[0];
    }
    for (GrammarASTWithOptions r : rules) {
        String ruleName = r.getChild(0).getText();
        if (Grammar.isTokenName(ruleName)) {
            lexerRulesRoot.addChild((Tree) adaptor.dupTree(r));
            rulesWeMoved.add(r);
        }
    }
    for (GrammarAST r : rulesWeMoved) {
        combinedRulesRoot.deleteChild(r);
    }
    // Will track 'if' from IF : 'if' ; rules to avoid defining new token for 'if'
    List<Pair<GrammarAST, GrammarAST>> litAliases = Grammar.getStringLiteralAliasesFromLexerRules(lexerAST);
    Set<String> stringLiterals = combinedGrammar.getStringLiterals();
    // add strings from combined grammar (and imported grammars) into lexer
    // put them first as they are keywords; must resolve ambigs to these rules
    //		tool.log("grammar", "strings from parser: "+stringLiterals);
    int insertIndex = 0;
    nextLit: for (String lit : stringLiterals) {
        // if lexer already has a rule for literal, continue
        if (litAliases != null) {
            for (Pair<GrammarAST, GrammarAST> pair : litAliases) {
                GrammarAST litAST = pair.b;
                if (lit.equals(litAST.getText()))
                    continue nextLit;
            }
        }
        // create for each literal: (RULE <uniquename> (BLOCK (ALT <lit>))
        String rname = combinedGrammar.getStringLiteralLexerRuleName(lit);
        // can't use wizard; need special node types
        GrammarAST litRule = new RuleAST(ANTLRParser.RULE);
        BlockAST blk = new BlockAST(ANTLRParser.BLOCK);
        AltAST alt = new AltAST(ANTLRParser.ALT);
        TerminalAST slit = new TerminalAST(new CommonToken(ANTLRParser.STRING_LITERAL, lit));
        alt.addChild(slit);
        blk.addChild(alt);
        CommonToken idToken = new CommonToken(ANTLRParser.TOKEN_REF, rname);
        litRule.addChild(new TerminalAST(idToken));
        litRule.addChild(blk);
        lexerRulesRoot.insertChild(insertIndex, litRule);
        //			lexerRulesRoot.getChildren().add(0, litRule);
        // reset indexes and set litRule parent
        lexerRulesRoot.freshenParentAndChildIndexes();
        // next literal will be added after the one just added
        insertIndex++;
    }
    // TODO: take out after stable if slow
    lexerAST.sanityCheckParentAndChildIndexes();
    combinedAST.sanityCheckParentAndChildIndexes();
    //		tool.log("grammar", combinedAST.toTokenString());
    combinedGrammar.tool.log("grammar", "after extract implicit lexer =" + combinedAST.toStringTree());
    combinedGrammar.tool.log("grammar", "lexer =" + lexerAST.toStringTree());
    if (lexerRulesRoot.getChildCount() == 0)
        return null;
    return lexerAST;
}
Also used : RuleAST(org.antlr.v4.tool.ast.RuleAST) GrammarRootAST(org.antlr.v4.tool.ast.GrammarRootAST) GrammarAST(org.antlr.v4.tool.ast.GrammarAST) ArrayList(java.util.ArrayList) BlockAST(org.antlr.v4.tool.ast.BlockAST) AltAST(org.antlr.v4.tool.ast.AltAST) TerminalAST(org.antlr.v4.tool.ast.TerminalAST) GrammarASTAdaptor(org.antlr.v4.parse.GrammarASTAdaptor) CommonToken(org.antlr.runtime.CommonToken) GrammarASTWithOptions(org.antlr.v4.tool.ast.GrammarASTWithOptions) Pair(org.antlr.v4.runtime.misc.Pair)

Example 4 with GrammarAST

use of org.antlr.v4.tool.ast.GrammarAST in project antlr4 by antlr.

the class GrammarTransformPipeline method setGrammarPtr.

/** Utility visitor that sets grammar ptr in each node */
public static void setGrammarPtr(final Grammar g, GrammarAST tree) {
    if (tree == null)
        return;
    // ensure each node has pointer to surrounding grammar
    TreeVisitor v = new TreeVisitor(new GrammarASTAdaptor());
    v.visit(tree, new TreeVisitorAction() {

        @Override
        public Object pre(Object t) {
            ((GrammarAST) t).g = g;
            return t;
        }

        @Override
        public Object post(Object t) {
            return t;
        }
    });
}
Also used : TreeVisitor(org.antlr.runtime.tree.TreeVisitor) GrammarAST(org.antlr.v4.tool.ast.GrammarAST) GrammarASTAdaptor(org.antlr.v4.parse.GrammarASTAdaptor) TreeVisitorAction(org.antlr.runtime.tree.TreeVisitorAction)

Example 5 with GrammarAST

use of org.antlr.v4.tool.ast.GrammarAST in project antlr4 by antlr.

the class GrammarTransformPipeline method expandParameterizedLoops.

/** Find and replace
     *      ID*[','] with ID (',' ID)*
     *      ID+[','] with ID (',' ID)+
     *      (x {action} y)+[','] with x {action} y (',' x {action} y)+
     *
     *  Parameter must be a token.
     *  todo: do we want?
     */
public void expandParameterizedLoops(GrammarAST root) {
    TreeVisitor v = new TreeVisitor(new GrammarASTAdaptor());
    v.visit(root, new TreeVisitorAction() {

        @Override
        public Object pre(Object t) {
            if (((GrammarAST) t).getType() == 3) {
                return expandParameterizedLoop((GrammarAST) t);
            }
            return t;
        }

        @Override
        public Object post(Object t) {
            return t;
        }
    });
}
Also used : TreeVisitor(org.antlr.runtime.tree.TreeVisitor) GrammarAST(org.antlr.v4.tool.ast.GrammarAST) GrammarASTAdaptor(org.antlr.v4.parse.GrammarASTAdaptor) TreeVisitorAction(org.antlr.runtime.tree.TreeVisitorAction)

Aggregations

GrammarAST (org.antlr.v4.tool.ast.GrammarAST)55 Rule (org.antlr.v4.tool.Rule)20 ATNState (org.antlr.v4.runtime.atn.ATNState)15 ArrayList (java.util.ArrayList)12 GrammarASTAdaptor (org.antlr.v4.parse.GrammarASTAdaptor)12 IntervalSet (org.antlr.v4.runtime.misc.IntervalSet)12 LeftRecursiveRule (org.antlr.v4.tool.LeftRecursiveRule)12 Grammar (org.antlr.v4.tool.Grammar)8 Decl (org.antlr.v4.codegen.model.decl.Decl)7 ActionAST (org.antlr.v4.tool.ast.ActionAST)7 AltAST (org.antlr.v4.tool.ast.AltAST)7 TerminalAST (org.antlr.v4.tool.ast.TerminalAST)7 LinkedHashMap (java.util.LinkedHashMap)6 Token (org.antlr.runtime.Token)6 RuleAST (org.antlr.v4.tool.ast.RuleAST)6 HashMap (java.util.HashMap)5 AddToLabelList (org.antlr.v4.codegen.model.AddToLabelList)5 Pair (org.antlr.v4.runtime.misc.Pair)5 LexerGrammar (org.antlr.v4.tool.LexerGrammar)5 GrammarASTWithOptions (org.antlr.v4.tool.ast.GrammarASTWithOptions)5