use of org.antlr.v4.runtime.atn.BlockStartState in project antlr4 by tunnelvisionlabs.
the class TestATNInterpreter method checkMatchedAlt.
public void checkMatchedAlt(LexerGrammar lg, final Grammar g, String inputString, int expected) {
ATN lexatn = createATN(lg, true);
LexerATNSimulator lexInterp = new LexerATNSimulator(lexatn);
IntegerList types = getTokenTypesViaATN(inputString, lexInterp);
System.out.println(types);
g.importVocab(lg);
ParserATNFactory f = new ParserATNFactory(g);
ATN atn = f.createATN();
IntTokenStream input = new IntTokenStream(types);
System.out.println("input=" + input.types);
ParserInterpreterForTesting interp = new ParserInterpreterForTesting(g, input);
ATNState startState = atn.ruleToStartState[g.getRule("a").index];
if (startState.transition(0).target instanceof BlockStartState) {
startState = startState.transition(0).target;
}
DOTGenerator dot = new DOTGenerator(g);
System.out.println(dot.getDOT(atn.ruleToStartState[g.getRule("a").index]));
Rule r = g.getRule("e");
if (r != null)
System.out.println(dot.getDOT(atn.ruleToStartState[r.index]));
int result = interp.matchATN(input, startState);
assertEquals(expected, result);
}
use of org.antlr.v4.runtime.atn.BlockStartState in project antlr4 by tunnelvisionlabs.
the class ParserATNFactory method optional.
/**
* From {@code (A)?} build either:
*
* <pre>
* o--A->o
* | ^
* o---->|
* </pre>
*
* or, if {@code A} is a block, just add an empty alt to the end of the
* block
*/
@NotNull
@Override
public Handle optional(@NotNull GrammarAST optAST, @NotNull Handle blk) {
BlockStartState blkStart = (BlockStartState) blk.left;
ATNState blkEnd = blk.right;
preventEpsilonOptionalBlocks.add(Tuple.create(currentRule, blkStart, blkEnd));
boolean greedy = ((QuantifierAST) optAST).isGreedy();
// no way to express SLL restriction
blkStart.sll = false;
blkStart.nonGreedy = !greedy;
epsilon(blkStart, blk.right, !greedy);
optAST.atnState = blk.left;
return blk;
}
use of org.antlr.v4.runtime.atn.BlockStartState 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.isNil() && 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;
}
use of org.antlr.v4.runtime.atn.BlockStartState in project antlr4 by antlr.
the class TestATNInterpreter method checkMatchedAlt.
public void checkMatchedAlt(LexerGrammar lg, final Grammar g, String inputString, int expected) {
ATN lexatn = createATN(lg, true);
LexerATNSimulator lexInterp = new LexerATNSimulator(lexatn, new DFA[] { new DFA(lexatn.modeToStartState.get(Lexer.DEFAULT_MODE)) }, null);
IntegerList types = getTokenTypesViaATN(inputString, lexInterp);
// System.out.println(types);
g.importVocab(lg);
ParserATNFactory f = new ParserATNFactory(g);
ATN atn = f.createATN();
TokenStream input = new MockIntTokenStream(types);
// System.out.println("input="+input.types);
ParserInterpreterForTesting interp = new ParserInterpreterForTesting(g, input);
ATNState startState = atn.ruleToStartState[g.getRule("a").index];
if (startState.transition(0).target instanceof BlockStartState) {
startState = startState.transition(0).target;
}
DOTGenerator dot = new DOTGenerator(g);
// System.out.println(dot.getDOT(atn.ruleToStartState[g.getRule("a").index]));
Rule r = g.getRule("e");
// if ( r!=null ) System.out.println(dot.getDOT(atn.ruleToStartState[r.index]));
int result = interp.matchATN(input, startState);
assertEquals(expected, result);
}
use of org.antlr.v4.runtime.atn.BlockStartState in project antlr4 by antlr.
the class SwiftTarget method serializeTojson.
// added by janyou -->
public String serializeTojson(ATN atn) {
JsonObjectBuilder builder = Json.createObjectBuilder();
builder.add("version", ATNDeserializer.SERIALIZED_VERSION);
builder.add("uuid", ATNDeserializer.SERIALIZED_UUID.toString());
// convert grammar type to ATN const to avoid dependence on ANTLRParser
builder.add("grammarType", atn.grammarType.ordinal());
builder.add("maxTokenType", atn.maxTokenType);
// states
int nedges = 0;
Map<IntervalSet, Integer> setIndices = new HashMap<IntervalSet, Integer>();
List<IntervalSet> sets = new ArrayList<IntervalSet>();
JsonArrayBuilder statesBuilder = Json.createArrayBuilder();
IntegerList nonGreedyStates = new IntegerList();
IntegerList precedenceStates = new IntegerList();
for (ATNState s : atn.states) {
JsonObjectBuilder stateBuilder = Json.createObjectBuilder();
if (s == null) {
// might be optimized away
statesBuilder.addNull();
continue;
}
int stateType = s.getStateType();
stateBuilder.add("stateType", stateType);
// stateBuilder.add("stateNumber",s.stateNumber);
stateBuilder.add("ruleIndex", s.ruleIndex);
if (s instanceof DecisionState && ((DecisionState) s).nonGreedy) {
nonGreedyStates.add(s.stateNumber);
}
if (s instanceof RuleStartState && ((RuleStartState) s).isLeftRecursiveRule) {
precedenceStates.add(s.stateNumber);
}
if (s.getStateType() == ATNState.LOOP_END) {
stateBuilder.add("detailStateNumber", ((LoopEndState) s).loopBackState.stateNumber);
} else if (s instanceof BlockStartState) {
stateBuilder.add("detailStateNumber", ((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;
if (!setIndices.containsKey(st.set)) {
sets.add(st.set);
setIndices.put(st.set, sets.size() - 1);
}
}
}
statesBuilder.add(stateBuilder);
}
builder.add("states", statesBuilder);
// non-greedy states
JsonArrayBuilder nonGreedyStatesBuilder = Json.createArrayBuilder();
for (int i = 0; i < nonGreedyStates.size(); i++) {
nonGreedyStatesBuilder.add(nonGreedyStates.get(i));
}
builder.add("nonGreedyStates", nonGreedyStatesBuilder);
// precedence states
JsonArrayBuilder precedenceStatesBuilder = Json.createArrayBuilder();
for (int i = 0; i < precedenceStates.size(); i++) {
precedenceStatesBuilder.add(precedenceStates.get(i));
}
builder.add("precedenceStates", precedenceStatesBuilder);
JsonArrayBuilder ruleToStartStateBuilder = Json.createArrayBuilder();
int nrules = atn.ruleToStartState.length;
for (int r = 0; r < nrules; r++) {
JsonObjectBuilder stateBuilder = Json.createObjectBuilder();
ATNState ruleStartState = atn.ruleToStartState[r];
stateBuilder.add("stateNumber", ruleStartState.stateNumber);
if (atn.grammarType == ATNType.LEXER) {
// if (atn.ruleToTokenType[r] == Token.EOF) {
// //data.add(Character.MAX_VALUE);
// stateBuilder.add("ruleToTokenType",-1);
// }
// else {
// //data.add(atn.ruleToTokenType[r]);
stateBuilder.add("ruleToTokenType", atn.ruleToTokenType[r]);
// }
}
ruleToStartStateBuilder.add(stateBuilder);
}
builder.add("ruleToStartState", ruleToStartStateBuilder);
JsonArrayBuilder modeToStartStateBuilder = Json.createArrayBuilder();
int nmodes = atn.modeToStartState.size();
if (nmodes > 0) {
for (ATNState modeStartState : atn.modeToStartState) {
modeToStartStateBuilder.add(modeStartState.stateNumber);
}
}
builder.add("modeToStartState", modeToStartStateBuilder);
JsonArrayBuilder nsetsBuilder = Json.createArrayBuilder();
int nsets = sets.size();
// data.add(nsets);
builder.add("nsets", nsets);
for (IntervalSet set : sets) {
JsonObjectBuilder setBuilder = Json.createObjectBuilder();
boolean containsEof = set.contains(Token.EOF);
if (containsEof && set.getIntervals().get(0).b == Token.EOF) {
// data.add(set.getIntervals().size() - 1);
setBuilder.add("size", set.getIntervals().size() - 1);
} else {
// data.add(set.getIntervals().size());
setBuilder.add("size", set.getIntervals().size());
}
setBuilder.add("containsEof", containsEof ? 1 : 0);
JsonArrayBuilder IntervalsBuilder = Json.createArrayBuilder();
for (Interval I : set.getIntervals()) {
JsonObjectBuilder IntervalBuilder = Json.createObjectBuilder();
if (I.a == Token.EOF) {
if (I.b == Token.EOF) {
continue;
} else {
IntervalBuilder.add("a", 0);
// data.add(0);
}
} else {
IntervalBuilder.add("a", I.a);
// data.add(I.a);
}
IntervalBuilder.add("b", I.b);
IntervalsBuilder.add(IntervalBuilder);
}
setBuilder.add("Intervals", IntervalsBuilder);
nsetsBuilder.add(setBuilder);
}
builder.add("IntervalSet", nsetsBuilder);
// builder.add("nedges",nedges);
JsonArrayBuilder allTransitionsBuilder = Json.createArrayBuilder();
for (ATNState s : atn.states) {
if (s == null) {
// might be optimized away
continue;
}
if (s.getStateType() == ATNState.RULE_STOP) {
continue;
}
JsonArrayBuilder transitionsBuilder = Json.createArrayBuilder();
for (int i = 0; i < s.getNumberOfTransitions(); i++) {
JsonObjectBuilder transitionBuilder = Json.createObjectBuilder();
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;
}
transitionBuilder.add("src", src);
transitionBuilder.add("trg", trg);
transitionBuilder.add("edgeType", edgeType);
transitionBuilder.add("arg1", arg1);
transitionBuilder.add("arg2", arg2);
transitionBuilder.add("arg3", arg3);
transitionsBuilder.add(transitionBuilder);
}
allTransitionsBuilder.add(transitionsBuilder);
}
builder.add("allTransitionsBuilder", allTransitionsBuilder);
int ndecisions = atn.decisionToState.size();
// data.add(ndecisions);
JsonArrayBuilder decisionToStateBuilder = Json.createArrayBuilder();
for (DecisionState decStartState : atn.decisionToState) {
// data.add(decStartState.stateNumber);
decisionToStateBuilder.add(decStartState.stateNumber);
}
builder.add("decisionToState", decisionToStateBuilder);
//
// LEXER ACTIONS
//
JsonArrayBuilder lexerActionsBuilder = Json.createArrayBuilder();
if (atn.grammarType == ATNType.LEXER) {
// data.add(atn.lexerActions.length);
for (LexerAction action : atn.lexerActions) {
JsonObjectBuilder lexerActionBuilder = Json.createObjectBuilder();
lexerActionBuilder.add("actionType", action.getActionType().ordinal());
// data.add(action.getActionType().ordinal());
switch(action.getActionType()) {
case CHANNEL:
int channel = ((LexerChannelAction) action).getChannel();
lexerActionBuilder.add("a", channel);
lexerActionBuilder.add("b", 0);
break;
case CUSTOM:
int ruleIndex = ((LexerCustomAction) action).getRuleIndex();
int actionIndex = ((LexerCustomAction) action).getActionIndex();
lexerActionBuilder.add("a", ruleIndex);
lexerActionBuilder.add("b", actionIndex);
break;
case MODE:
int mode = ((LexerModeAction) action).getMode();
lexerActionBuilder.add("a", mode);
lexerActionBuilder.add("b", 0);
break;
case MORE:
lexerActionBuilder.add("a", 0);
lexerActionBuilder.add("b", 0);
break;
case POP_MODE:
lexerActionBuilder.add("a", 0);
lexerActionBuilder.add("b", 0);
break;
case PUSH_MODE:
mode = ((LexerPushModeAction) action).getMode();
lexerActionBuilder.add("a", mode);
lexerActionBuilder.add("b", 0);
break;
case SKIP:
lexerActionBuilder.add("a", 0);
lexerActionBuilder.add("b", 0);
break;
case TYPE:
int type = ((LexerTypeAction) action).getType();
lexerActionBuilder.add("a", type);
lexerActionBuilder.add("b", 0);
break;
default:
String message = String.format(Locale.getDefault(), "The specified lexer action type %s is not valid.", action.getActionType());
throw new IllegalArgumentException(message);
}
lexerActionsBuilder.add(lexerActionBuilder);
}
}
builder.add("lexerActions", lexerActionsBuilder);
// 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 out of range.");
// }
//
// int value = (data.get(i) + 2) & 0xFFFF;
// data.set(i, value);
// }
JsonObject data = builder.build();
// System.out.print(data.toString());
return data.toString();
}
Aggregations