Examples with Action org.antlr.v4.codegen.model.Action used on opensource projects

Example 1 with Action

use of org.antlr.v4.codegen.model.Action in project antlr4 by antlr.

the class ATNSerializer method serialize.

/** Serialize state descriptors, edge descriptors, and decision→state map
	 *  into list of ints:
	 *
	 * 		grammar-type, (ANTLRParser.LEXER, ...)
	 *  	max token type,
	 *  	num states,
	 *  	state-0-type ruleIndex, state-1-type ruleIndex, ... state-i-type ruleIndex optional-arg ...
	 *  	num rules,
	 *  	rule-1-start-state rule-1-args, rule-2-start-state  rule-2-args, ...
	 *  	(args are token type,actionIndex in lexer else 0,0)
	 *      num modes,
	 *      mode-0-start-state, mode-1-start-state, ... (parser has 0 modes)
	 *      num unicode-bmp-sets
	 *      bmp-set-0-interval-count intervals, bmp-set-1-interval-count intervals, ...
	 *      num unicode-smp-sets
	 *      smp-set-0-interval-count intervals, smp-set-1-interval-count intervals, ...
	 *	num total edges,
	 *      src, trg, edge-type, edge arg1, optional edge arg2 (present always), ...
	 *      num decisions,
	 *      decision-0-start-state, decision-1-start-state, ...
	 *
	 *  Convenient to pack into unsigned shorts to make as Java string.
	 */
public IntegerList serialize() {
    IntegerList data = new IntegerList();
    data.add(ATNDeserializer.SERIALIZED_VERSION);
    serializeUUID(data, ATNDeserializer.SERIALIZED_UUID);
    // convert grammar type to ATN const to avoid dependence on ANTLRParser
    data.add(atn.grammarType.ordinal());
    data.add(atn.maxTokenType);
    int nedges = 0;
    // Note that we use a LinkedHashMap as a set to
    // maintain insertion order while deduplicating
    // entries with the same key.
    Map<IntervalSet, Boolean> sets = new LinkedHashMap<>();
    // dump states, count edges and collect sets while doing so
    IntegerList nonGreedyStates = new IntegerList();
    IntegerList precedenceStates = new IntegerList();
    data.add(atn.states.size());
    for (ATNState s : atn.states) {
        if (s == null) {
            // might be optimized away
            data.add(ATNState.INVALID_TYPE);
            continue;
        }
        int stateType = s.getStateType();
        if (s instanceof DecisionState && ((DecisionState) s).nonGreedy) {
            nonGreedyStates.add(s.stateNumber);
        }
        if (s instanceof RuleStartState && ((RuleStartState) s).isLeftRecursiveRule) {
            precedenceStates.add(s.stateNumber);
        }
        data.add(stateType);
        if (s.ruleIndex == -1) {
            data.add(Character.MAX_VALUE);
        } else {
            data.add(s.ruleIndex);
        }
        if (s.getStateType() == ATNState.LOOP_END) {
            data.add(((LoopEndState) s).loopBackState.stateNumber);
        } else if (s instanceof BlockStartState) {
            data.add(((BlockStartState) s).endState.stateNumber);
        }
        if (s.getStateType() != ATNState.RULE_STOP) {
            // the deserializer can trivially derive these edges, so there's no need to serialize them
            nedges += s.getNumberOfTransitions();
        }
        for (int i = 0; i < s.getNumberOfTransitions(); i++) {
            Transition t = s.transition(i);
            int edgeType = Transition.serializationTypes.get(t.getClass());
            if (edgeType == Transition.SET || edgeType == Transition.NOT_SET) {
                SetTransition st = (SetTransition) t;
                sets.put(st.set, true);
            }
        }
    }
    // non-greedy states
    data.add(nonGreedyStates.size());
    for (int i = 0; i < nonGreedyStates.size(); i++) {
        data.add(nonGreedyStates.get(i));
    }
    // precedence states
    data.add(precedenceStates.size());
    for (int i = 0; i < precedenceStates.size(); i++) {
        data.add(precedenceStates.get(i));
    }
    int nrules = atn.ruleToStartState.length;
    data.add(nrules);
    for (int r = 0; r < nrules; r++) {
        ATNState ruleStartState = atn.ruleToStartState[r];
        data.add(ruleStartState.stateNumber);
        if (atn.grammarType == ATNType.LEXER) {
            if (atn.ruleToTokenType[r] == Token.EOF) {
                data.add(Character.MAX_VALUE);
            } else {
                data.add(atn.ruleToTokenType[r]);
            }
        }
    }
    int nmodes = atn.modeToStartState.size();
    data.add(nmodes);
    if (nmodes > 0) {
        for (ATNState modeStartState : atn.modeToStartState) {
            data.add(modeStartState.stateNumber);
        }
    }
    List<IntervalSet> bmpSets = new ArrayList<>();
    List<IntervalSet> smpSets = new ArrayList<>();
    for (IntervalSet set : sets.keySet()) {
        if (set.getMaxElement() <= Character.MAX_VALUE) {
            bmpSets.add(set);
        } else {
            smpSets.add(set);
        }
    }
    serializeSets(data, bmpSets, new CodePointSerializer() {

        @Override
        public void serializeCodePoint(IntegerList data, int cp) {
            data.add(cp);
        }
    });
    serializeSets(data, smpSets, new CodePointSerializer() {

        @Override
        public void serializeCodePoint(IntegerList data, int cp) {
            serializeInt(data, cp);
        }
    });
    Map<IntervalSet, Integer> setIndices = new HashMap<>();
    int setIndex = 0;
    for (IntervalSet bmpSet : bmpSets) {
        setIndices.put(bmpSet, setIndex++);
    }
    for (IntervalSet smpSet : smpSets) {
        setIndices.put(smpSet, setIndex++);
    }
    data.add(nedges);
    for (ATNState s : atn.states) {
        if (s == null) {
            // might be optimized away
            continue;
        }
        if (s.getStateType() == ATNState.RULE_STOP) {
            continue;
        }
        for (int i = 0; i < s.getNumberOfTransitions(); i++) {
            Transition t = s.transition(i);
            if (atn.states.get(t.target.stateNumber) == null) {
                throw new IllegalStateException("Cannot serialize a transition to a removed state.");
            }
            int src = s.stateNumber;
            int trg = t.target.stateNumber;
            int edgeType = Transition.serializationTypes.get(t.getClass());
            int arg1 = 0;
            int arg2 = 0;
            int arg3 = 0;
            switch(edgeType) {
                case Transition.RULE:
                    trg = ((RuleTransition) t).followState.stateNumber;
                    arg1 = ((RuleTransition) t).target.stateNumber;
                    arg2 = ((RuleTransition) t).ruleIndex;
                    arg3 = ((RuleTransition) t).precedence;
                    break;
                case Transition.PRECEDENCE:
                    PrecedencePredicateTransition ppt = (PrecedencePredicateTransition) t;
                    arg1 = ppt.precedence;
                    break;
                case Transition.PREDICATE:
                    PredicateTransition pt = (PredicateTransition) t;
                    arg1 = pt.ruleIndex;
                    arg2 = pt.predIndex;
                    arg3 = pt.isCtxDependent ? 1 : 0;
                    break;
                case Transition.RANGE:
                    arg1 = ((RangeTransition) t).from;
                    arg2 = ((RangeTransition) t).to;
                    if (arg1 == Token.EOF) {
                        arg1 = 0;
                        arg3 = 1;
                    }
                    break;
                case Transition.ATOM:
                    arg1 = ((AtomTransition) t).label;
                    if (arg1 == Token.EOF) {
                        arg1 = 0;
                        arg3 = 1;
                    }
                    break;
                case Transition.ACTION:
                    ActionTransition at = (ActionTransition) t;
                    arg1 = at.ruleIndex;
                    arg2 = at.actionIndex;
                    if (arg2 == -1) {
                        arg2 = 0xFFFF;
                    }
                    arg3 = at.isCtxDependent ? 1 : 0;
                    break;
                case Transition.SET:
                    arg1 = setIndices.get(((SetTransition) t).set);
                    break;
                case Transition.NOT_SET:
                    arg1 = setIndices.get(((SetTransition) t).set);
                    break;
                case Transition.WILDCARD:
                    break;
            }
            data.add(src);
            data.add(trg);
            data.add(edgeType);
            data.add(arg1);
            data.add(arg2);
            data.add(arg3);
        }
    }
    int ndecisions = atn.decisionToState.size();
    data.add(ndecisions);
    for (DecisionState decStartState : atn.decisionToState) {
        data.add(decStartState.stateNumber);
    }
    //
    if (atn.grammarType == ATNType.LEXER) {
        data.add(atn.lexerActions.length);
        for (LexerAction action : atn.lexerActions) {
            data.add(action.getActionType().ordinal());
            switch(action.getActionType()) {
                case CHANNEL:
                    int channel = ((LexerChannelAction) action).getChannel();
                    data.add(channel != -1 ? channel : 0xFFFF);
                    data.add(0);
                    break;
                case CUSTOM:
                    int ruleIndex = ((LexerCustomAction) action).getRuleIndex();
                    int actionIndex = ((LexerCustomAction) action).getActionIndex();
                    data.add(ruleIndex != -1 ? ruleIndex : 0xFFFF);
                    data.add(actionIndex != -1 ? actionIndex : 0xFFFF);
                    break;
                case MODE:
                    int mode = ((LexerModeAction) action).getMode();
                    data.add(mode != -1 ? mode : 0xFFFF);
                    data.add(0);
                    break;
                case MORE:
                    data.add(0);
                    data.add(0);
                    break;
                case POP_MODE:
                    data.add(0);
                    data.add(0);
                    break;
                case PUSH_MODE:
                    mode = ((LexerPushModeAction) action).getMode();
                    data.add(mode != -1 ? mode : 0xFFFF);
                    data.add(0);
                    break;
                case SKIP:
                    data.add(0);
                    data.add(0);
                    break;
                case TYPE:
                    int type = ((LexerTypeAction) action).getType();
                    data.add(type != -1 ? type : 0xFFFF);
                    data.add(0);
                    break;
                default:
                    String message = String.format(Locale.getDefault(), "The specified lexer action type %s is not valid.", action.getActionType());
                    throw new IllegalArgumentException(message);
            }
        }
    }
    // don't adjust the first value since that's the version number
    for (int i = 1; i < data.size(); i++) {
        if (data.get(i) < Character.MIN_VALUE || data.get(i) > Character.MAX_VALUE) {
            throw new UnsupportedOperationException("Serialized ATN data element " + data.get(i) + " element " + i + " out of range " + (int) Character.MIN_VALUE + ".." + (int) Character.MAX_VALUE);
        }
        int value = (data.get(i) + 2) & 0xFFFF;
        data.set(i, value);
    }
    return data;
}

Example 2 with Action

use of org.antlr.v4.codegen.model.Action in project antlr4 by antlr.

the class DefaultErrorStrategy method getMissingSymbol.

/** Conjure up a missing token during error recovery.
	 *
	 *  The recognizer attempts to recover from single missing
	 *  symbols. But, actions might refer to that missing symbol.
	 *  For example, x=ID {f($x);}. The action clearly assumes
	 *  that there has been an identifier matched previously and that
	 *  $x points at that token. If that token is missing, but
	 *  the next token in the stream is what we want we assume that
	 *  this token is missing and we keep going. Because we
	 *  have to return some token to replace the missing token,
	 *  we have to conjure one up. This method gives the user control
	 *  over the tokens returned for missing tokens. Mostly,
	 *  you will want to create something special for identifier
	 *  tokens. For literals such as '{' and ',', the default
	 *  action in the parser or tree parser works. It simply creates
	 *  a CommonToken of the appropriate type. The text will be the token.
	 *  If you change what tokens must be created by the lexer,
	 *  override this method to create the appropriate tokens.
	 */
protected Token getMissingSymbol(Parser recognizer) {
    Token currentSymbol = recognizer.getCurrentToken();
    IntervalSet expecting = getExpectedTokens(recognizer);
    int expectedTokenType = Token.INVALID_TYPE;
    if (!expecting.isNil()) {
        // get any element
        expectedTokenType = expecting.getMinElement();
    }
    String tokenText;
    if (expectedTokenType == Token.EOF)
        tokenText = "<missing EOF>";
    else
        tokenText = "<missing " + recognizer.getVocabulary().getDisplayName(expectedTokenType) + ">";
    Token current = currentSymbol;
    Token lookback = recognizer.getInputStream().LT(-1);
    if (current.getType() == Token.EOF && lookback != null) {
        current = lookback;
    }
    return recognizer.getTokenFactory().create(new Pair<TokenSource, CharStream>(current.getTokenSource(), current.getTokenSource().getInputStream()), expectedTokenType, tokenText, Token.DEFAULT_CHANNEL, -1, -1, current.getLine(), current.getCharPositionInLine());
}

Example 3 with Action

use of org.antlr.v4.codegen.model.Action in project antlr4 by antlr.

the class ParserInterpreter method visitState.

protected void visitState(ATNState p) {
    //		System.out.println("visitState "+p.stateNumber);
    int predictedAlt = 1;
    if (p instanceof DecisionState) {
        predictedAlt = visitDecisionState((DecisionState) p);
    }
    Transition transition = p.transition(predictedAlt - 1);
    switch(transition.getSerializationType()) {
        case Transition.EPSILON:
            if (p.getStateType() == ATNState.STAR_LOOP_ENTRY && ((StarLoopEntryState) p).isPrecedenceDecision && !(transition.target instanceof LoopEndState)) {
                // We are at the start of a left recursive rule's (...)* loop
                // and we're not taking the exit branch of loop.
                InterpreterRuleContext localctx = createInterpreterRuleContext(_parentContextStack.peek().a, _parentContextStack.peek().b, _ctx.getRuleIndex());
                pushNewRecursionContext(localctx, atn.ruleToStartState[p.ruleIndex].stateNumber, _ctx.getRuleIndex());
            }
            break;
        case Transition.ATOM:
            match(((AtomTransition) transition).label);
            break;
        case Transition.RANGE:
        case Transition.SET:
        case Transition.NOT_SET:
            if (!transition.matches(_input.LA(1), Token.MIN_USER_TOKEN_TYPE, 65535)) {
                recoverInline();
            }
            matchWildcard();
            break;
        case Transition.WILDCARD:
            matchWildcard();
            break;
        case Transition.RULE:
            RuleStartState ruleStartState = (RuleStartState) transition.target;
            int ruleIndex = ruleStartState.ruleIndex;
            InterpreterRuleContext newctx = createInterpreterRuleContext(_ctx, p.stateNumber, ruleIndex);
            if (ruleStartState.isLeftRecursiveRule) {
                enterRecursionRule(newctx, ruleStartState.stateNumber, ruleIndex, ((RuleTransition) transition).precedence);
            } else {
                enterRule(newctx, transition.target.stateNumber, ruleIndex);
            }
            break;
        case Transition.PREDICATE:
            PredicateTransition predicateTransition = (PredicateTransition) transition;
            if (!sempred(_ctx, predicateTransition.ruleIndex, predicateTransition.predIndex)) {
                throw new FailedPredicateException(this);
            }
            break;
        case Transition.ACTION:
            ActionTransition actionTransition = (ActionTransition) transition;
            action(_ctx, actionTransition.ruleIndex, actionTransition.actionIndex);
            break;
        case Transition.PRECEDENCE:
            if (!precpred(_ctx, ((PrecedencePredicateTransition) transition).precedence)) {
                throw new FailedPredicateException(this, String.format("precpred(_ctx, %d)", ((PrecedencePredicateTransition) transition).precedence));
            }
            break;
        default:
            throw new UnsupportedOperationException("Unrecognized ATN transition type.");
    }
    setState(transition.target.stateNumber);
}

Example 4 with Action

use of org.antlr.v4.codegen.model.Action 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;
        }
    });
}

Example 5 with Action

use of org.antlr.v4.codegen.model.Action in project antlr4 by antlr.

the class BaseGoTest method testActions.

public void testActions(String templates, String actionName, String action, String expected) throws org.antlr.runtime.RecognitionException {
    int lp = templates.indexOf('(');
    String name = templates.substring(0, lp);
    STGroup group = new STGroupString(templates);
    ST st = group.getInstanceOf(name);
    st.add(actionName, action);
    String grammar = st.render();
    ErrorQueue equeue = new ErrorQueue();
    Grammar g = new Grammar(grammar, equeue);
    if (g.ast != null && !g.ast.hasErrors) {
        SemanticPipeline sem = new SemanticPipeline(g);
        sem.process();
        ATNFactory factory = new ParserATNFactory(g);
        if (g.isLexer())
            factory = new LexerATNFactory((LexerGrammar) g);
        g.atn = factory.createATN();
        CodeGenerator gen = new CodeGenerator(g);
        ST outputFileST = gen.generateParser();
        String output = outputFileST.render();
        // System.out.println(output);
        String b = "#" + actionName + "#";
        int start = output.indexOf(b);
        String e = "#end-" + actionName + "#";
        int end = output.indexOf(e);
        String snippet = output.substring(start + b.length(), end);
        assertEquals(expected, snippet);
    }
    if (equeue.size() > 0) {
        System.err.println(equeue.toString());
    }
}