Examples with ATNState org.antlr.v4.runtime.atn.ATNState used on opensource projects

Example 81 with ATNState

use of org.antlr.v4.runtime.atn.ATNState in project antlr4 by tunnelvisionlabs.

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<IntervalSet, Boolean>();
    // dump states, count edges and collect sets while doing so
    IntegerList nonGreedyStates = new IntegerList();
    IntegerList sllStates = 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 decisionState = (DecisionState) s;
            if (decisionState.nonGreedy) {
                nonGreedyStates.add(s.stateNumber);
            }
            if (decisionState.sll) {
                sllStates.add(s.stateNumber);
            }
        }
        if (s instanceof RuleStartState && ((RuleStartState) s).isPrecedenceRule) {
            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));
    }
    // SLL decisions
    data.add(sllStates.size());
    for (int i = 0; i < sllStates.size(); i++) {
        data.add(sllStates.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);
        boolean leftFactored = ruleNames.get(ruleStartState.ruleIndex).indexOf(ATNSimulator.RULE_VARIANT_DELIMITER) >= 0;
        data.add(leftFactored ? 1 : 0);
        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<IntervalSet>();
    List<IntervalSet> smpSets = new ArrayList<IntervalSet>();
    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<IntervalSet, Integer>();
    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 82 with ATNState

use of org.antlr.v4.runtime.atn.ATNState in project antlr4 by tunnelvisionlabs.

the class Parser method isExpectedToken.

/**
 * Checks whether or not {@code symbol} can follow the current state in the
 * ATN. The behavior of this method is equivalent to the following, but is
 * implemented such that the complete context-sensitive follow set does not
 * need to be explicitly constructed.
 *
 * <pre>
 * return getExpectedTokens().contains(symbol);
 * </pre>
 *
 * @param symbol the symbol type to check
 * @return {@code true} if {@code symbol} can follow the current state in
 * the ATN, otherwise {@code false}.
 */
public boolean isExpectedToken(int symbol) {
    // return getInterpreter().atn.nextTokens(_ctx);
    ATN atn = getInterpreter().atn;
    ParserRuleContext ctx = _ctx;
    ATNState s = atn.states.get(getState());
    IntervalSet following = atn.nextTokens(s);
    if (following.contains(symbol)) {
        return true;
    }
    // System.out.println("following "+s+"="+following);
    if (!following.contains(Token.EPSILON))
        return false;
    while (ctx != null && ctx.invokingState >= 0 && following.contains(Token.EPSILON)) {
        ATNState invokingState = atn.states.get(ctx.invokingState);
        RuleTransition rt = (RuleTransition) invokingState.transition(0);
        following = atn.nextTokens(rt.followState);
        if (following.contains(symbol)) {
            return true;
        }
        ctx = (ParserRuleContext) ctx.parent;
    }
    if (following.contains(Token.EPSILON) && symbol == Token.EOF) {
        return true;
    }
    return false;
}

Example 83 with ATNState

use of org.antlr.v4.runtime.atn.ATNState in project antlr4 by tunnelvisionlabs.

the class ParserInterpreter method parse.

/**
 * Begin parsing at startRuleIndex
 */
public ParserRuleContext parse(int startRuleIndex) {
    RuleStartState startRuleStartState = atn.ruleToStartState[startRuleIndex];
    rootContext = createInterpreterRuleContext(null, ATNState.INVALID_STATE_NUMBER, startRuleIndex);
    if (startRuleStartState.isPrecedenceRule) {
        enterRecursionRule(rootContext, startRuleStartState.stateNumber, startRuleIndex, 0);
    } else {
        enterRule(rootContext, startRuleStartState.stateNumber, startRuleIndex);
    }
    while (true) {
        ATNState p = getATNState();
        switch(p.getStateType()) {
            case ATNState.RULE_STOP:
                // pop; return from rule
                if (_ctx.isEmpty()) {
                    if (startRuleStartState.isPrecedenceRule) {
                        ParserRuleContext result = _ctx;
                        Tuple2<ParserRuleContext, Integer> parentContext = _parentContextStack.pop();
                        unrollRecursionContexts(parentContext.getItem1());
                        return result;
                    } else {
                        exitRule();
                        return rootContext;
                    }
                }
                visitRuleStopState(p);
                break;
            default:
                try {
                    visitState(p);
                } catch (RecognitionException e) {
                    setState(atn.ruleToStopState[p.ruleIndex].stateNumber);
                    getContext().exception = e;
                    getErrorHandler().reportError(this, e);
                    recover(e);
                }
                break;
        }
    }
}

Example 84 with ATNState

use of org.antlr.v4.runtime.atn.ATNState in project antlr4 by tunnelvisionlabs.

the class ParserInterpreter method visitRuleStopState.

protected void visitRuleStopState(ATNState p) {
    RuleStartState ruleStartState = atn.ruleToStartState[p.ruleIndex];
    if (ruleStartState.isPrecedenceRule) {
        Tuple2<ParserRuleContext, Integer> parentContext = _parentContextStack.pop();
        unrollRecursionContexts(parentContext.getItem1());
        setState(parentContext.getItem2());
    } else {
        exitRule();
    }
    RuleTransition ruleTransition = (RuleTransition) atn.states.get(getState()).transition(0);
    setState(ruleTransition.followState.stateNumber);
}

Example 85 with ATNState

use of org.antlr.v4.runtime.atn.ATNState in project antlr4 by tunnelvisionlabs.

the class DefaultErrorStrategy method sync.

/**
 * The default implementation of {@link ANTLRErrorStrategy#sync} makes sure
 * that the current lookahead symbol is consistent with what were expecting
 * at this point in the ATN. You can call this anytime but ANTLR only
 * generates code to check before subrules/loops and each iteration.
 *
 * <p>Implements Jim Idle's magic sync mechanism in closures and optional
 * subrules. E.g.,</p>
 *
 * <pre>
 * a : sync ( stuff sync )* ;
 * sync : {consume to what can follow sync} ;
 * </pre>
 *
 * At the start of a sub rule upon error, {@link #sync} performs single
 * token deletion, if possible. If it can't do that, it bails on the current
 * rule and uses the default error recovery, which consumes until the
 * resynchronization set of the current rule.
 *
 * <p>If the sub rule is optional ({@code (...)?}, {@code (...)*}, or block
 * with an empty alternative), then the expected set includes what follows
 * the subrule.</p>
 *
 * <p>During loop iteration, it consumes until it sees a token that can start a
 * sub rule or what follows loop. Yes, that is pretty aggressive. We opt to
 * stay in the loop as long as possible.</p>
 *
 * <p><strong>ORIGINS</strong></p>
 *
 * <p>Previous versions of ANTLR did a poor job of their recovery within loops.
 * A single mismatch token or missing token would force the parser to bail
 * out of the entire rules surrounding the loop. So, for rule</p>
 *
 * <pre>
 * classDef : 'class' ID '{' member* '}'
 * </pre>
 *
 * input with an extra token between members would force the parser to
 * consume until it found the next class definition rather than the next
 * member definition of the current class.
 *
 * <p>This functionality cost a little bit of effort because the parser has to
 * compare token set at the start of the loop and at each iteration. If for
 * some reason speed is suffering for you, you can turn off this
 * functionality by simply overriding this method as a blank { }.</p>
 */
@Override
public void sync(Parser recognizer) throws RecognitionException {
    ATNState s = recognizer.getInterpreter().atn.states.get(recognizer.getState());
    // If already recovering, don't try to sync
    if (inErrorRecoveryMode(recognizer)) {
        return;
    }
    TokenStream tokens = recognizer.getInputStream();
    int la = tokens.LA(1);
    // try cheaper subset first; might get lucky. seems to shave a wee bit off
    IntervalSet nextTokens = recognizer.getATN().nextTokens(s);
    if (nextTokens.contains(la)) {
        // We are sure the token matches
        nextTokensContext = null;
        nextTokensState = ATNState.INVALID_STATE_NUMBER;
        return;
    }
    if (nextTokens.contains(Token.EPSILON)) {
        if (nextTokensContext == null) {
            // It's possible the next token won't match; information tracked
            // by sync is restricted for performance.
            nextTokensContext = recognizer.getContext();
            nextTokensState = recognizer.getState();
        }
        return;
    }
    switch(s.getStateType()) {
        case ATNState.BLOCK_START:
        case ATNState.STAR_BLOCK_START:
        case ATNState.PLUS_BLOCK_START:
        case ATNState.STAR_LOOP_ENTRY:
            // report error and recover if possible
            if (singleTokenDeletion(recognizer) != null) {
                return;
            }
            throw new InputMismatchException(recognizer);
        case ATNState.PLUS_LOOP_BACK:
        case ATNState.STAR_LOOP_BACK:
            // System.err.println("at loop back: "+s.getClass().getSimpleName());
            reportUnwantedToken(recognizer);
            IntervalSet expecting = recognizer.getExpectedTokens();
            IntervalSet whatFollowsLoopIterationOrRule = expecting.or(getErrorRecoverySet(recognizer));
            consumeUntil(recognizer, whatFollowsLoopIterationOrRule);
            break;
        default:
            // do nothing if we can't identify the exact kind of ATN state
            break;
    }
}