Examples with Token org.antlr.v4.runtime.Token used on opensource projects

Example 6 with Token

use of org.antlr.v4.runtime.Token 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 7 with Token

use of org.antlr.v4.runtime.Token in project antlr4 by antlr.

the class LL1Analyzer method LOOK.

/**
	 * Compute set of tokens that can follow {@code s} in the ATN in the
	 * specified {@code ctx}.
	 *
	 * <p>If {@code ctx} is {@code null} and the end of the rule containing
	 * {@code s} is reached, {@link Token#EPSILON} is added to the result set.
	 * If {@code ctx} is not {@code null} and the end of the outermost rule is
	 * reached, {@link Token#EOF} is added to the result set.</p>
	 *
	 * @param s the ATN state
	 * @param stopState the ATN state to stop at. This can be a
	 * {@link BlockEndState} to detect epsilon paths through a closure.
	 * @param ctx the complete parser context, or {@code null} if the context
	 * should be ignored
	 *
	 * @return The set of tokens that can follow {@code s} in the ATN in the
	 * specified {@code ctx}.
	 */
public IntervalSet LOOK(ATNState s, ATNState stopState, RuleContext ctx) {
    IntervalSet r = new IntervalSet();
    // ignore preds; get all lookahead
    boolean seeThruPreds = true;
    PredictionContext lookContext = ctx != null ? PredictionContext.fromRuleContext(s.atn, ctx) : null;
    _LOOK(s, stopState, lookContext, r, new HashSet<ATNConfig>(), new BitSet(), seeThruPreds, true);
    return r;
}

Example 8 with Token

use of org.antlr.v4.runtime.Token in project antlr4 by antlr.

the class LexerATNSimulator method execATN.

protected int execATN(CharStream input, DFAState ds0) {
    //System.out.println("enter exec index "+input.index()+" from "+ds0.configs);
    if (debug) {
        System.out.format(Locale.getDefault(), "start state closure=%s\n", ds0.configs);
    }
    if (ds0.isAcceptState) {
        // allow zero-length tokens
        captureSimState(prevAccept, input, ds0);
    }
    int t = input.LA(1);
    // s is current/from DFA state
    DFAState s = ds0;
    while (true) {
        // while more work
        if (debug) {
            System.out.format(Locale.getDefault(), "execATN loop starting closure: %s\n", s.configs);
        }
        // As we move src->trg, src->trg, we keep track of the previous trg to
        // avoid looking up the DFA state again, which is expensive.
        // If the previous target was already part of the DFA, we might
        // be able to avoid doing a reach operation upon t. If s!=null,
        // it means that semantic predicates didn't prevent us from
        // creating a DFA state. Once we know s!=null, we check to see if
        // the DFA state has an edge already for t. If so, we can just reuse
        // it's configuration set; there's no point in re-computing it.
        // This is kind of like doing DFA simulation within the ATN
        // simulation because DFA simulation is really just a way to avoid
        // computing reach/closure sets. Technically, once we know that
        // we have a previously added DFA state, we could jump over to
        // the DFA simulator. But, that would mean popping back and forth
        // a lot and making things more complicated algorithmically.
        // This optimization makes a lot of sense for loops within DFA.
        // A character will take us back to an existing DFA state
        // that already has lots of edges out of it. e.g., .* in comments.
        DFAState target = getExistingTargetState(s, t);
        if (target == null) {
            target = computeTargetState(input, s, t);
        }
        if (target == ERROR) {
            break;
        }
        // end of the token.
        if (t != IntStream.EOF) {
            consume(input);
        }
        if (target.isAcceptState) {
            captureSimState(prevAccept, input, target);
            if (t == IntStream.EOF) {
                break;
            }
        }
        t = input.LA(1);
        // flip; current DFA target becomes new src/from state
        s = target;
    }
    return failOrAccept(prevAccept, input, s.configs, t);
}

Example 9 with Token

use of org.antlr.v4.runtime.Token in project antlr4 by antlr.

the class ParseTreePatternMatcher method matchImpl.

// ---- SUPPORT CODE ----
/**
	 * Recursively walk {@code tree} against {@code patternTree}, filling
	 * {@code match.}{@link ParseTreeMatch#labels labels}.
	 *
	 * @return the first node encountered in {@code tree} which does not match
	 * a corresponding node in {@code patternTree}, or {@code null} if the match
	 * was successful. The specific node returned depends on the matching
	 * algorithm used by the implementation, and may be overridden.
	 */
protected ParseTree matchImpl(ParseTree tree, ParseTree patternTree, MultiMap<String, ParseTree> labels) {
    if (tree == null) {
        throw new IllegalArgumentException("tree cannot be null");
    }
    if (patternTree == null) {
        throw new IllegalArgumentException("patternTree cannot be null");
    }
    // x and <ID>, x and y, or x and x; or could be mismatched types
    if (tree instanceof TerminalNode && patternTree instanceof TerminalNode) {
        TerminalNode t1 = (TerminalNode) tree;
        TerminalNode t2 = (TerminalNode) patternTree;
        ParseTree mismatchedNode = null;
        // both are tokens and they have same type
        if (t1.getSymbol().getType() == t2.getSymbol().getType()) {
            if (t2.getSymbol() instanceof TokenTagToken) {
                // x and <ID>
                TokenTagToken tokenTagToken = (TokenTagToken) t2.getSymbol();
                // track label->list-of-nodes for both token name and label (if any)
                labels.map(tokenTagToken.getTokenName(), tree);
                if (tokenTagToken.getLabel() != null) {
                    labels.map(tokenTagToken.getLabel(), tree);
                }
            } else if (t1.getText().equals(t2.getText())) {
            // x and x
            } else {
                // x and y
                if (mismatchedNode == null) {
                    mismatchedNode = t1;
                }
            }
        } else {
            if (mismatchedNode == null) {
                mismatchedNode = t1;
            }
        }
        return mismatchedNode;
    }
    if (tree instanceof ParserRuleContext && patternTree instanceof ParserRuleContext) {
        ParserRuleContext r1 = (ParserRuleContext) tree;
        ParserRuleContext r2 = (ParserRuleContext) patternTree;
        ParseTree mismatchedNode = null;
        // (expr ...) and <expr>
        RuleTagToken ruleTagToken = getRuleTagToken(r2);
        if (ruleTagToken != null) {
            ParseTreeMatch m = null;
            if (r1.getRuleContext().getRuleIndex() == r2.getRuleContext().getRuleIndex()) {
                // track label->list-of-nodes for both rule name and label (if any)
                labels.map(ruleTagToken.getRuleName(), tree);
                if (ruleTagToken.getLabel() != null) {
                    labels.map(ruleTagToken.getLabel(), tree);
                }
            } else {
                if (mismatchedNode == null) {
                    mismatchedNode = r1;
                }
            }
            return mismatchedNode;
        }
        // (expr ...) and (expr ...)
        if (r1.getChildCount() != r2.getChildCount()) {
            if (mismatchedNode == null) {
                mismatchedNode = r1;
            }
            return mismatchedNode;
        }
        int n = r1.getChildCount();
        for (int i = 0; i < n; i++) {
            ParseTree childMatch = matchImpl(r1.getChild(i), patternTree.getChild(i), labels);
            if (childMatch != null) {
                return childMatch;
            }
        }
        return mismatchedNode;
    }
    // if nodes aren't both tokens or both rule nodes, can't match
    return tree;
}

Example 10 with Token

use of org.antlr.v4.runtime.Token in project antlr4 by antlr.

the class ParseTreePatternMatcher method tokenize.

public List<? extends Token> tokenize(String pattern) {
    // split pattern into chunks: sea (raw input) and islands (<ID>, <expr>)
    List<Chunk> chunks = split(pattern);
    // create token stream from text and tags
    List<Token> tokens = new ArrayList<Token>();
    for (Chunk chunk : chunks) {
        if (chunk instanceof TagChunk) {
            TagChunk tagChunk = (TagChunk) chunk;
            // add special rule token or conjure up new token from name
            if (Character.isUpperCase(tagChunk.getTag().charAt(0))) {
                Integer ttype = parser.getTokenType(tagChunk.getTag());
                if (ttype == Token.INVALID_TYPE) {
                    throw new IllegalArgumentException("Unknown token " + tagChunk.getTag() + " in pattern: " + pattern);
                }
                TokenTagToken t = new TokenTagToken(tagChunk.getTag(), ttype, tagChunk.getLabel());
                tokens.add(t);
            } else if (Character.isLowerCase(tagChunk.getTag().charAt(0))) {
                int ruleIndex = parser.getRuleIndex(tagChunk.getTag());
                if (ruleIndex == -1) {
                    throw new IllegalArgumentException("Unknown rule " + tagChunk.getTag() + " in pattern: " + pattern);
                }
                int ruleImaginaryTokenType = parser.getATNWithBypassAlts().ruleToTokenType[ruleIndex];
                tokens.add(new RuleTagToken(tagChunk.getTag(), ruleImaginaryTokenType, tagChunk.getLabel()));
            } else {
                throw new IllegalArgumentException("invalid tag: " + tagChunk.getTag() + " in pattern: " + pattern);
            }
        } else {
            TextChunk textChunk = (TextChunk) chunk;
            ANTLRInputStream in = new ANTLRInputStream(textChunk.getText());
            lexer.setInputStream(in);
            Token t = lexer.nextToken();
            while (t.getType() != Token.EOF) {
                tokens.add(t);
                t = lexer.nextToken();
            }
        }
    }
    //		System.out.println("tokens="+tokens);
    return tokens;
}