org.antlr.v4.runtime

Class RuleContext

java.lang.Object

org.antlr.v4.runtime.RuleContext

All Implemented Interfaces:

ParseTree, RuleNode, SyntaxTree, Tree

Direct Known Subclasses:

ParserRuleContext

public class RuleContext
extends Object
implements RuleNode

A rule context is a record of a single rule invocation. We form a stack of these context objects using the parent pointer. A parent pointer of null indicates that the current context is the bottom of the stack. The ParserRuleContext subclass as a children list so that we can turn this data structure into a tree. The root node always has a null pointer and invokingState of -1. Upon entry to parsing, the first invoked rule function creates a context object (a subclass specialized for that rule such as SContext) and makes it the root of a parse tree, recorded by field Parser._ctx. public final SContext s() throws RecognitionException { SContext _localctx = new SContext(_ctx, getState()); <-- create new node enterRule(_localctx, 0, RULE_s); <-- push it ... exitRule(); <-- pop back to _localctx return _localctx; } A subsequent rule invocation of r from the start rule s pushes a new context object for r whose parent points at s and use invoking state is the state with r emanating as edge label. The invokingState fields from a context object to the root together form a stack of rule indication states where the root (bottom of the stack) has a -1 sentinel value. If we invoke start symbol s then call r1, which calls r2, the would look like this: SContext[-1] <- root node (bottom of the stack) R1Context[p] <- p in rule s called r1 R2Context[q] <- q in rule r1 called r2 So the top of the stack, _ctx, represents a call to the current rule and it holds the return address from another rule that invoke to this rule. To invoke a rule, we must always have a current context. The parent contexts are useful for computing lookahead sets and getting error information. These objects are used during parsing and prediction. For the special case of parsers, we use the subclass ParserRuleContext.

See Also:

ParserRuleContext

Field Summary

Fields

Modifier and Type	Field and Description
static ParserRuleContext	EMPTY
int	invokingState What state invoked the rule associated with this context? The "return address" is the followState of invokingState If parent is null, this should be -1 this context object represents the start rule.
RuleContext	parent What context invoked this rule?

Constructor Summary

Constructors

Constructor and Description
RuleContext()
RuleContext(RuleContext parent, int invokingState)

Method Summary

Methods

Modifier and Type	Method and Description
<T> T	accept(ParseTreeVisitor<? extends T> visitor) The ParseTreeVisitor needs a double dispatch method.
int	depth()
int	getAltNumber() For rule associated with this parse tree internal node, return the outer alternative number used to match the input.
ParseTree	getChild(int i) If there are children, get the ith value indexed from 0.
int	getChildCount() How many children are there? If there is none, then this node represents a leaf node.
RuleContext	getParent() The parent of this node.
RuleContext	getPayload() This method returns whatever object represents the data at this note.
RuleContext	getRuleContext()
int	getRuleIndex()
Interval	getSourceInterval() Return an Interval indicating the index in the TokenStream of the first and last token associated with this subtree.
String	getText() Return the combined text of all child nodes.
boolean	isEmpty() A context is empty if there is no invoking state; meaning nobody called current context.
void	setAltNumber(int altNumber) Set the outer alternative number for this context node.
void	setParent(RuleContext parent) Set the parent for this node.
String	toString()
String	toString(List<String> ruleNames)
String	toString(List<String> ruleNames, RuleContext stop)
String	toString(Recognizer<?,?> recog)
String	toString(Recognizer<?,?> recog, RuleContext stop)
String	toStringTree() Print out a whole tree, not just a node, in LISP format {@code (root child1 ..
String	toStringTree(List<String> ruleNames) Print out a whole tree, not just a node, in LISP format (root child1 ..
String	toStringTree(Parser recog) Print out a whole tree, not just a node, in LISP format (root child1 ..

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

EMPTY

public static final ParserRuleContext EMPTY

parent

public RuleContext parent

What context invoked this rule?

invokingState

public int invokingState

What state invoked the rule associated with this context? The "return address" is the followState of invokingState If parent is null, this should be -1 this context object represents the start rule.

Constructor Detail

RuleContext

public RuleContext()

RuleContext

public RuleContext(RuleContext parent,
           int invokingState)

Method Detail

depth

public int depth()

isEmpty

public boolean isEmpty()

A context is empty if there is no invoking state; meaning nobody called current context.

getSourceInterval

public Interval getSourceInterval()

Description copied from interface: SyntaxTree

Return an Interval indicating the index in the TokenStream of the first and last token associated with this subtree. If this node is a leaf, then the interval represents a single token and has interval i..i for token index i.

An interval of i..i-1 indicates an empty interval at position i in the input stream, where 0 <= i <= the size of the input token stream. Currently, the code base can only have i=0..n-1 but in concept one could have an empty interval after EOF.

If source interval is unknown, this returns Interval.INVALID.

As a weird special case, the source interval for rules matched after EOF is unspecified.

Specified by:

getSourceInterval in interface SyntaxTree

getRuleContext

public RuleContext getRuleContext()

Specified by:

getRuleContext in interface RuleNode

getParent

public RuleContext getParent()

Description copied from interface: Tree

The parent of this node. If the return value is null, then this node is the root of the tree.

Specified by:

getParent in interface ParseTree

Specified by:

getParent in interface Tree

getPayload

public RuleContext getPayload()

Description copied from interface: Tree

This method returns whatever object represents the data at this note. For example, for parse trees, the payload can be a Token representing a leaf node or a RuleContext object representing a rule invocation. For abstract syntax trees (ASTs), this is a Token object.

Specified by:

getPayload in interface Tree

getText

public String getText()

Return the combined text of all child nodes. This method only considers tokens which have been added to the parse tree.

Since tokens on hidden channels (e.g. whitespace or comments) are not added to the parse trees, they will not appear in the output of this method.

Specified by:

getText in interface ParseTree

getRuleIndex

public int getRuleIndex()

getAltNumber

public int getAltNumber()

For rule associated with this parse tree internal node, return the outer alternative number used to match the input. Default implementation does not compute nor store this alt num. Create a subclass of ParserRuleContext with backing field and set option contextSuperClass. to set it.

Since:

4.5.3

setAltNumber

public void setAltNumber(int altNumber)

Set the outer alternative number for this context node. Default implementation does nothing to avoid backing field overhead for trees that don't need it. Create a subclass of ParserRuleContext with backing field and set option contextSuperClass.

Since:

4.5.3

setParent

public void setParent(RuleContext parent)

Description copied from interface: ParseTree

Set the parent for this node. This is not backward compatible as it changes the interface but no one was able to create custom nodes anyway so I'm adding as it improves internal code quality. One could argue for a restructuring of the class/interface hierarchy so that setParent, addChild are moved up to Tree but that's a major change. So I'll do the minimal change, which is to add this method.

Specified by:

setParent in interface ParseTree

Since:

4.7. ParseTree.setParent(org.antlr.v4.runtime.RuleContext) comment

getChild

public ParseTree getChild(int i)

Description copied from interface: Tree

If there are children, get the ith value indexed from 0.

Specified by:

getChild in interface ParseTree

Specified by:

getChild in interface Tree

getChildCount

public int getChildCount()

Description copied from interface: Tree

How many children are there? If there is none, then this node represents a leaf node.

Specified by:

getChildCount in interface Tree

accept

public <T> T accept(ParseTreeVisitor<? extends T> visitor)

Description copied from interface: ParseTree

The ParseTreeVisitor needs a double dispatch method.

Specified by:

accept in interface ParseTree

toStringTree

public String toStringTree(Parser recog)

Print out a whole tree, not just a node, in LISP format (root child1 .. childN). Print just a node if this is a leaf. We have to know the recognizer so we can get rule names.

Specified by:

toStringTree in interface ParseTree

toStringTree

public String toStringTree(List<String> ruleNames)

Print out a whole tree, not just a node, in LISP format (root child1 .. childN). Print just a node if this is a leaf.

toStringTree

public String toStringTree()

Description copied from interface: Tree

Print out a whole tree, not just a node, in LISP format (root child1 .. childN). Print just a node if this is a leaf.

Specified by:

toStringTree in interface Tree

toString

public String toString()

Overrides:

toString in class Object

toString

public final String toString(Recognizer<?,?> recog)

toString

public final String toString(List<String> ruleNames)

toString

public String toString(Recognizer<?,?> recog,
              RuleContext stop)

toString

public String toString(List<String> ruleNames,
              RuleContext stop)