xref: /Universal-ctags/peg/kotlin.peg (revision b7514e1ec672a8580bea041544e9259a169f09c3)

# Copyright (c) 2021 Jan Dolinár
#
# This source code is released for free distribution under the terms of the
# GNU General Public License version 2.
#
# This file contains description of Kotlin grammar based on the ANTLR grammar
# from https://github.com/Kotlin/kotlin-grammar-gpl2.
#
# Notes:
#   1. This file was originally derived from files (reference grammar files) in https://github.com/Kotlin/kotlin-spec
#   2. Kotlin/kotlin-spec is distributed under Apache2 license
#   3. I have requested the original developers to "re-license or dual-license" (see https://github.com/Kotlin/kotlin-spec/issues/70)
#   4. The request was accepted and the original developer made a new repository https://github.com/Kotlin/kotlin-grammar-gpl2
#      for redistributing the reference grammar files under GPLv2 license.
#   5. This file is now derived from Kotlin/kotlin-grammar-gpl2 repository.
#   6. Unlike many other files in Universal Ctags, this source code is released strictly under GPLv2
#      and should not be relicensed to later versions of GPL.



%prefix "pkotlin"

%auxil	"struct parserCtx *"

%earlysource {
    #include "general.h"
}

%header {
	struct parserCtx;
}

%source {
#include "kotlin_pre.h"
}

file <- shebangLine? NL* fileAnnotation* _* packageHeader* _* importList* _* (filePart / _ / unparsable)* EOF
filePart <- (topLevelObject / (statement _* semi)) {resetFailure(auxil, $0s);}
unparsable <- [^\n]+ NL* {reportFailure(auxil, $0s);}

### Converted from peg/kotlin/KotlinParser.g4

# options
# // SECTION: general
#kotlinFile <- shebangLine? NL* fileAnnotation* _* packageHeader _* importList _* topLevelObject* EOF
#script <- shebangLine? NL* fileAnnotation* _* packageHeader _* importList _* (statement _* semi)* EOF
shebangLine <- ShebangLine _* NL+
fileAnnotation <- (AT_NO_WS / AT_PRE_WS) FILE NL* COLON _* NL* (LSQUARE _* unescapedAnnotation+ _* RSQUARE / unescapedAnnotation) _* NL*
packageHeader <- PACKAGE {PUSH_KIND(auxil, K_PACKAGE);} _ <identifier> {makeKotlinTag(auxil, $1, $1s, true);} _* semi?
importList <- importHeader+
importHeader <- IMPORT _ identifier (DOT MULT / importAlias)? _* semi? _*
importAlias <- _ AS _ simpleIdentifier
topLevelObject <- declaration _* semis?
typeAlias <- modifiers? _* TYPE_ALIAS {PUSH_KIND(auxil, K_TYPEALIAS);} (_ / NL)* <simpleIdentifier> {makeKotlinTag(auxil, $1, $1s, false);} _* (__* typeParameters)? __* ASSIGNMENT __* type
declaration <- classDeclaration / objectDeclaration / functionDeclaration / propertyDeclaration / typeAlias

# // SECTION: classes
classDeclaration <- modifiers? (CLASS {PUSH_KIND(auxil, K_CLASS);} / (FUN __*)? INTERFACE {PUSH_KIND(auxil, K_INTERFACE);}) _ NL* <simpleIdentifier> {makeKotlinTag(auxil, $1, $1s, true);} (__* typeParameters)? (__* primaryConstructor)? (__* COLON __* delegationSpecifiers)? (__* typeConstraints)? (__* classBody / __* enumClassBody)? {POP_SCOPE(auxil);}
primaryConstructor <- (modifiers? CONSTRUCTOR __*)? classParameters
classBody <- LCURL __* classMemberDeclarations __* RCURL
classParameters <- LPAREN __* (classParameter (__* COMMA __* classParameter)* (__* COMMA)?)? __* RPAREN
classParameter <- (modifiers? _* VAL {PUSH_KIND(auxil, K_CONSTANT);} / modifiers? _* VAR {PUSH_KIND(auxil, K_VARIABLE);} / modifiers? {PUSH_KIND(auxil, K_IGNORE);} _*)? __* <simpleIdentifier> {makeKotlinTag(auxil, $1, $1s, true);} _* COLON __* type (__* ASSIGNMENT __* expression)? {POP_SCOPE(auxil);}
delegationSpecifiers <- annotatedDelegationSpecifier (__* COMMA __* annotatedDelegationSpecifier)*
delegationSpecifier <- constructorInvocation / explicitDelegation / userType / functionType
constructorInvocation <- userType _* valueArguments
annotatedDelegationSpecifier <- annotation* __* delegationSpecifier
explicitDelegation <- (userType / functionType) __* BY __* expression
typeParameters <- LANGLE __* typeParameter (__* COMMA __* typeParameter)* (__* COMMA)? __* RANGLE
typeParameter <- typeParameterModifiers? __* simpleIdentifier (__* COLON __* type)?
typeConstraints <- WHERE __* typeConstraint (__* COMMA __* typeConstraint)*
typeConstraint <- annotation* simpleIdentifier __* COLON __* type

# // SECTION: classMembers
classMemberDeclarations <- (classMemberDeclaration semis?)*
classMemberDeclaration <- secondaryConstructor / anonymousInitializer / companionObject / declaration
anonymousInitializer <- INIT __* block
companionObject <- modifiers? COMPANION __* OBJECT {PUSH_KIND(auxil, K_OBJECT);} <(__* simpleIdentifier)?> {makeKotlinTag(auxil, $1e-$1s != 0 ? $1 : "Companion", $1s, true);} (__* COLON __* delegationSpecifiers)? (__* classBody)? {POP_SCOPE(auxil);}
functionValueParameters <- LPAREN __* (functionValueParameter (__* COMMA __* functionValueParameter)* (__* COMMA)?)? __* RPAREN
functionValueParameter <- parameterModifiers? _* parameter (__* ASSIGNMENT __* expression)?
functionDeclaration <- modifiers? _* FUN {PUSH_KIND(auxil, K_METHOD);} _* (__* typeParameters)? _* (__* receiverTypeAndDot)? __* <simpleIdentifier> {makeKotlinTag(auxil, $1, $1s, true);} __* functionValueParameters _* (__* COLON __* type)? _* (__* typeConstraints)? _* (__* functionBody)? {POP_SCOPE(auxil);}
functionBody <- block / ASSIGNMENT __* expression
variableDeclaration <- annotation* __* <simpleIdentifier> {makeKotlinTag(auxil, $1, $1s, false);} (__* COLON __* type)?
multiVariableDeclaration <- LPAREN __* variableDeclaration _* (__* COMMA __* variableDeclaration)* _* (__* COMMA)? __* RPAREN
propertyDeclaration <- modifiers? _* (VAL {PUSH_KIND(auxil, K_CONSTANT);} / VAR {PUSH_KIND(auxil, K_VARIABLE);}) _ (__* typeParameters)? (__* receiverTypeAndDot)? (__* (multiVariableDeclaration / variableDeclaration)) (__* typeConstraints)? (__* (ASSIGNMENT __* expression / propertyDelegate))? (semi? _* setter (NL* semi? _* getter)? / semi? _* getter (NL* semi? _* setter)?)?
propertyDelegate <- BY __* expression
# TODO: better handling of empty getters and setters?
getter <- (modifiers _*)? GET __* LPAREN __* RPAREN (__* COLON __* type)? __* functionBody / (modifiers _*)? GET !(_* [^;\r\n])
setter <- (modifiers _*)? SET __* LPAREN __* parameterWithOptionalType (__* COMMA)? __* RPAREN (__* COLON __* type)? __* functionBody / (modifiers _*)? SET !(_* [^;\r\n])
parametersWithOptionalType <- LPAREN __* (parameterWithOptionalType (__* COMMA __* parameterWithOptionalType)* (__* COMMA)?)? __* RPAREN
parameterWithOptionalType <- parameterModifiers? simpleIdentifier __* (COLON __* type)?
parameter <- simpleIdentifier __* COLON __* type
objectDeclaration <- modifiers? _* OBJECT {PUSH_KIND(auxil, K_OBJECT);} __* <simpleIdentifier> {makeKotlinTag(auxil, $1, $1s, true);} (__* COLON __* delegationSpecifiers)? (__* classBody)? {POP_SCOPE(auxil);}
secondaryConstructor <- modifiers? CONSTRUCTOR __* functionValueParameters (__* COLON __* constructorDelegationCall)? __* block?
constructorDelegationCall <- THIS __* valueArguments / SUPER __* valueArguments

# // SECTION: enumClasses
enumClassBody <- LCURL __* enumEntries? (__* SEMICOLON __* classMemberDeclarations)? __* RCURL
enumEntries <- enumEntry (__* COMMA __* enumEntry)* __* COMMA?
enumEntry <- (modifiers __*)? simpleIdentifier (__* valueArguments)? (__* classBody)?

# // SECTION: types
type <- typeModifiers? ( functionType / nullableType / parenthesizedType / typeReference)
typeReference <- userType / DYNAMIC
nullableType <- (typeReference / parenthesizedType) __* quest+
quest <- !elvis (QUEST_WS / QUEST_NO_WS)
userType <- simpleUserType (__* DOT __* simpleUserType)*
simpleUserType <- simpleIdentifier (__* typeArguments)?
typeProjection <- typeProjectionModifiers? type / MULT
typeProjectionModifiers <- typeProjectionModifier+
typeProjectionModifier <- varianceModifier __* / annotation
functionType <- (receiverType __* DOT __*)? functionTypeParameters __* ARROW __* type
functionTypeParameters <- LPAREN __* (parameter / type)? _* (__* COMMA __* (parameter / type))* _* (__* COMMA)? __* RPAREN
parenthesizedType <- LPAREN __* type __* RPAREN
receiverType <- (typeModifiers _*)? (nullableType / parenthesizedType / typeReference)
# parenthesizedUserType <- LPAREN __* userType __* RPAREN / LPAREN __* parenthesizedUserType __* RPAREN
receiverTypeAndDot <- (typeModifiers _*)? (nullableType __* DOT __* / parenthesizedType __* DOT __* / (simpleUserType __* DOT __*)+)

# // SECTION: statements
#statements <- (statement (semis statement)*)? semis?
statements <- (statement _* (semis _* statement _*)*)? _* semis?
statement <- (label / annotation)* ( declaration / assignment / loopStatement / expression)
label <- simpleIdentifier (AT_POST_WS / AT_NO_WS) __*
controlStructureBody <- block / statement
block <- LCURL __* statements __* RCURL
loopStatement <- forStatement / whileStatement / doWhileStatement
forStatement <- FOR __* LPAREN _* annotation* _* (variableDeclaration / multiVariableDeclaration) _ IN _ inside_expression _* RPAREN __* (controlStructureBody)?
whileStatement <- WHILE __* LPAREN _* inside_expression _* RPAREN __* controlStructureBody / WHILE __* LPAREN _* expression _* RPAREN __* SEMICOLON
doWhileStatement <- DO __* controlStructureBody? __* WHILE __* LPAREN _* expression _* RPAREN
assignment <- directlyAssignableExpression _* ASSIGNMENT __* expression / assignableExpression _* assignmentAndOperator __* expression
semi <- (_* (SEMICOLON / NL) _*) NL*
semis <- (_* (SEMICOLON / NL) _*)+

# // SECTION: expressions
expression <- disjunction
disjunction <- conjunction (__* DISJ __* conjunction)*
conjunction <- equality (__* CONJ __* equality)*
equality <- comparison (_* equalityOperator __* comparison _*)*
comparison <- genericCallLikeComparison (_* comparisonOperator __* genericCallLikeComparison _*)*
genericCallLikeComparison <- infixOperation (_* callSuffix)*
infixOperation <- elvisExpression (_* inOperator __* elvisExpression / _* isOperator __* type)*
elvisExpression <- infixFunctionCall (__* elvis __* infixFunctionCall)*
elvis <- QUEST_NO_WS COLON
infixFunctionCall <- rangeExpression (_* simpleIdentifier __* rangeExpression)*
rangeExpression <- additiveExpression (_* RANGE __* additiveExpression)*
additiveExpression <- multiplicativeExpression (_* additiveOperator __* multiplicativeExpression)*
multiplicativeExpression <- asExpression (_* multiplicativeOperator __* asExpression)*
asExpression <- prefixUnaryExpression (__* asOperator __* type)*
prefixUnaryExpression <- (unaryPrefix _*)* postfixUnaryExpression
unaryPrefix <- annotation / label / prefixUnaryOperator __*
postfixUnaryExpression <- primaryExpression (_* postfixUnarySuffix)+ / primaryExpression
postfixUnarySuffix <- postfixUnaryOperator / typeArguments / callSuffix / indexingSuffix / navigationSuffix
directlyAssignableExpression <- postfixUnaryExpression _* assignableSuffix / postfixUnaryExpression / simpleIdentifier / parenthesizedDirectlyAssignableExpression
parenthesizedDirectlyAssignableExpression <- LPAREN __* inside_directlyAssignableExpression __* RPAREN
assignableExpression <- prefixUnaryExpression / parenthesizedAssignableExpression
parenthesizedAssignableExpression <- LPAREN __* inside_assignableExpression __* RPAREN
assignableSuffix <- navigationSuffix / typeArguments / indexingSuffix
indexingSuffix <- LSQUARE __* inside_expression (__* COMMA __* inside_expression)* (__* COMMA)? __* RSQUARE
navigationSuffix <- __* memberAccessOperator __* (simpleIdentifier / parenthesizedExpression / CLASS)
callSuffix <- typeArguments? _* valueArguments? _* annotatedLambda / typeArguments? _* valueArguments
annotatedLambda <- annotation* _* label? __* lambdaLiteral
typeArguments <- LANGLE __* typeProjection (__* COMMA __* typeProjection)* (__* COMMA)? __* RANGLE
valueArguments <- LPAREN __* RPAREN / LPAREN __* inside_valueArgument (__* COMMA __* inside_valueArgument)* (__* COMMA)? __* RPAREN
#valueArgument <- annotation? __* (simpleIdentifier __* ASSIGNMENT __*)? MULT? __* expression
primaryExpression <- thisExpression / superExpression / ifExpression / whenExpression / tryExpression / jumpExpression / parenthesizedExpression/ callableReference / stringLiteral / functionLiteral / objectLiteral / collectionLiteral  / simpleIdentifier / literalConstant
parenthesizedExpression <- LPAREN __* inside_expression __* RPAREN
collectionLiteral <- LSQUARE __* inside_expression (__* COMMA __* inside_expression)* (__* COMMA)? __* RSQUARE / LSQUARE __* RSQUARE
literalConstant <- BooleanLiteral / CharacterLiteral / NullLiteral / RealLiteral / UnsignedLiteral / LongLiteral / HexLiteral / BinLiteral / IntegerLiteral
stringLiteral <- multiLineStringLiteral / lineStringLiteral
lineStringLiteral <- QUOTE_OPEN (lineStringExpression / lineStringContent)* QUOTE_CLOSE
#lineStringLiteral <- QUOTE_OPEN (lineStringExpression / EscapedIdentifier / UniCharacterLiteral / stringChar)* QUOTE_CLOSE
multiLineStringLiteral <- TRIPLE_QUOTE_OPEN (multiLineStringExpression / multiLineStringContent / MultiLineStringQuote)* TRIPLE_QUOTE_CLOSE
#multiLineStringLiteral <- TRIPLE_QUOTE_OPEN (multiLineStringExpression / MultiLineStringQuote / EscapedIdentifier / UniChracterLiteral / stringChar)* TRIPLE_QUOTE_CLOSE
lineStringContent <- LineStrText / LineStrEscapedChar / LineStrRef
lineStringExpression <- LineStrExprStart __* expression __* RCURL
multiLineStringContent <- MultiLineStrRef / MultiLineStrText / MultiLineStringQuote
multiLineStringExpression <- MultiLineStrExprStart __* expression __* RCURL

inside_expression <- inside_disjunction
inside_disjunction <- inside_conjunction (__* DISJ __* inside_conjunction)*
inside_conjunction <- inside_equality (__* CONJ __* inside_equality)*
inside_equality <- inside_comparison ((_ / NL)* equalityOperator __* inside_comparison (_ / NL)*)*
inside_comparison <- inside_genericCallLikeComparison ((_ / NL)* comparisonOperator __* inside_genericCallLikeComparison (_ / NL)*)*
inside_genericCallLikeComparison <- inside_infixOperation ((_ / NL)* callSuffix)*
inside_infixOperation <- inside_elvisExpression ((_ / NL)* inOperator __* inside_elvisExpression / (_ / NL)* isOperator __* type)*
inside_elvisExpression <- inside_infixFunctionCall (__* elvis __* inside_infixFunctionCall)*
inside_infixFunctionCall <- inside_rangeExpression ((_ / NL)* simpleIdentifier __* inside_rangeExpression)*
inside_rangeExpression <- inside_additiveExpression ((_ / NL)* RANGE __* inside_additiveExpression)*
inside_additiveExpression <- inside_multiplicativeExpression ((_ / NL)* additiveOperator __* inside_multiplicativeExpression)*
inside_multiplicativeExpression <- inside_asExpression ((_ / NL)* multiplicativeOperator __* inside_asExpression)*
inside_asExpression <- inside_prefixUnaryExpression (__* asOperator __* type)*
inside_prefixUnaryExpression <- (inside_unaryPrefix (_ / NL)*)* inside_postfixUnaryExpression
inside_unaryPrefix <- annotation / label / prefixUnaryOperator __*
inside_postfixUnaryExpression <- primaryExpression ((_ / NL)* inside_postfixUnarySuffix)+ / primaryExpression
inside_postfixUnarySuffix <- postfixUnaryOperator / typeArguments / callSuffix / indexingSuffix / navigationSuffix
inside_directlyAssignableExpression <- inside_postfixUnaryExpression (_ / NL)* assignableSuffix / inside_postfixUnaryExpression / simpleIdentifier / parenthesizedDirectlyAssignableExpression
inside_assignableExpression <- inside_prefixUnaryExpression / parenthesizedAssignableExpression
inside_valueArgument <- annotation? __* (simpleIdentifier __* ASSIGNMENT __*)? MULT? __* inside_expression


#characterLiteral <- "'" (UniCharacterLiteral / EscapedIdentifier / [^\n\r'\\]) "'"
#stringChar <- [^"]

lambdaLiteral <- LCURL {PUSH_KIND(auxil, K_METHOD); makeKotlinTag(auxil, "<lambda>", $0s, true);} __* statements __* RCURL {POP_SCOPE(auxil);} / LCURL {PUSH_KIND(auxil, K_METHOD); makeKotlinTag(auxil, "<lambda>", 8, true);} __* lambdaParameters? __* ARROW __* statements __* RCURL {POP_SCOPE(auxil);}
lambdaParameters <- lambdaParameter (__* COMMA __* lambdaParameter)* (__* COMMA)?
lambdaParameter <- variableDeclaration / multiVariableDeclaration (__* COLON __* type)?
anonymousFunction <- FUN {PUSH_KIND(auxil, K_METHOD); makeKotlinTag(auxil, "<anonymous>", $0s, true);} (__* type __* DOT)? __* parametersWithOptionalType (__* COLON __* type)? (__* typeConstraints)? (__* functionBody)? {POP_SCOPE(auxil);}
functionLiteral <- lambdaLiteral / anonymousFunction
objectLiteral <- OBJECT __* COLON __* delegationSpecifiers __* classBody / OBJECT __* classBody
thisExpression <- THIS_AT / THIS !(Letter / UnicodeDigit)
superExpression <- SUPER_AT / SUPER (LANGLE __* type __* RANGLE)? (AT_NO_WS simpleIdentifier)?
ifExpression <- IF __* LPAREN __* expression __* RPAREN __* controlStructureBody? __* SEMICOLON? __* ELSE __* (controlStructureBody / SEMICOLON) / IF __* LPAREN __* expression __* RPAREN __* (controlStructureBody / SEMICOLON)
whenSubject <- LPAREN (annotation* __* VAL __* variableDeclaration __* ASSIGNMENT __*)? expression RPAREN
whenExpression <- WHEN __* whenSubject? __* LCURL __* (whenEntry __*)* __* RCURL
whenEntry <- whenCondition (__* COMMA __* whenCondition)* (__* COMMA)? __* ARROW __* controlStructureBody semi? / ELSE __* ARROW __* controlStructureBody semi?
whenCondition <- expression / rangeTest / typeTest
rangeTest <- inOperator __* expression
typeTest <- isOperator __* type
tryExpression <- TRY __* block ((__* catchBlock)+ (__* finallyBlock)? / __* finallyBlock)
catchBlock <- CATCH __* LPAREN _* (annotation _*)* simpleIdentifier _* COLON _* type (__* COMMA)? _* RPAREN __* block
finallyBlock <- FINALLY __* block
jumpExpression <- THROW __* expression / (RETURN_AT / RETURN) _* expression? / CONTINUE_AT / CONTINUE / BREAK_AT / BREAK
callableReference <- (receiverType? __* COLONCOLON __* (simpleIdentifier / CLASS))
assignmentAndOperator <- ADD_ASSIGNMENT / SUB_ASSIGNMENT / MULT_ASSIGNMENT / DIV_ASSIGNMENT / MOD_ASSIGNMENT
equalityOperator <- EQEQEQ / EQEQ / EXCL_EQEQ / EXCL_EQ
comparisonOperator <- LE / GE / LANGLE / RANGLE
inOperator <- IN / NOT_IN
isOperator <- IS / NOT_IS
additiveOperator <- ADD / SUB
multiplicativeOperator <- MULT / DIV / MOD
asOperator <- AS_SAFE / AS
prefixUnaryOperator <- INCR / DECR / SUB / ADD / excl
postfixUnaryOperator <- INCR / DECR / EXCL_NO_WS excl
excl <- EXCL_WS / EXCL_NO_WS
memberAccessOperator <- DOT / safeNav / COLONCOLON
safeNav <- QUEST_NO_WS DOT

# // SECTION: modifiers
modifiers <- (annotation / modifier)+
parameterModifiers <- (annotation / parameterModifier)+
modifier <- (classModifier / memberModifier / visibilityModifier / functionModifier / propertyModifier / inheritanceModifier / parameterModifier / platformModifier) __*
typeModifiers <- typeModifier+
typeModifier <- annotation / SUSPEND __*
classModifier <- ENUM / SEALED / ANNOTATION / DATA / INNER
memberModifier <- OVERRIDE / LATEINIT
visibilityModifier <- PUBLIC / PRIVATE / INTERNAL / PROTECTED
varianceModifier <- IN / OUT
typeParameterModifiers <- typeParameterModifier+
typeParameterModifier <- reificationModifier __* / varianceModifier __* / annotation
functionModifier <- TAILREC / OPERATOR / INFIX / INLINE / EXTERNAL / SUSPEND
propertyModifier <- CONST
inheritanceModifier <- ABSTRACT / FINAL / OPEN
parameterModifier <- VARARG / NOINLINE / CROSSINLINE
reificationModifier <- REIFIED
platformModifier <- EXPECT / ACTUAL

# // SECTION: annotations
annotation <- (singleAnnotation / multiAnnotation) __*
singleAnnotation <- annotationUseSiteTarget __* unescapedAnnotation / (AT_NO_WS / AT_PRE_WS) unescapedAnnotation
multiAnnotation <- annotationUseSiteTarget __* LSQUARE unescapedAnnotation+ RSQUARE / (AT_NO_WS / AT_PRE_WS) LSQUARE unescapedAnnotation+ RSQUARE
annotationUseSiteTarget <- (AT_NO_WS / AT_PRE_WS) (FIELD / PROPERTY / GET / SET / RECEIVER / PARAM / SETPARAM / DELEGATE) __* COLON
unescapedAnnotation <- constructorInvocation / userType

# // SECTION: identifiers
simpleIdentifier <- !(hardKeyword !(Letter / '_' / UnicodeDigit)) Identifier / ABSTRACT / ANNOTATION / BY / CATCH / COMPANION / CONSTRUCTOR / CROSSINLINE / DATA / DYNAMIC / ENUM / EXTERNAL / FINAL / FINALLY / GET / IMPORT / INFIX / INIT / INLINE / INNER / INTERNAL / LATEINIT / NOINLINE / OPEN / OPERATOR / OUT / OVERRIDE / PRIVATE / PROTECTED / PUBLIC / REIFIED / SEALED / TAILREC / SET / VARARG / WHERE / FIELD / PROPERTY / RECEIVER / PARAM / SETPARAM / DELEGATE / FILE / EXPECT / ACTUAL / CONST / SUSPEND
identifier <- simpleIdentifier (__* DOT simpleIdentifier)*

hardKeyword <- AS / BREAK / CLASS / CONTINUE / DO / ELSE / FOR / FUN / IF / IN / INTERFACE / IS / NullLiteral / OBJECT / PACKAGE / RETURN / SUPER / THIS / THROW / TRY / TYPE_ALIAS / TYPEOF / VAL / VAR / WHEN / WHILE / BooleanLiteral


### Converted from peg/kotlin/KotlinLexer.g4


# // SECTION: lexicalGeneral
ShebangLine <- '#!' [^\r\n]*
DelimitedComment <- '/*' (DelimitedComment / !'*/' .)* '*/'
LineComment <- '//' [^\r\n]*
#WS <- [\u0020\u0009\u000C]
#NL <- '\n' / '\r' '\n'?
Hidden <- DelimitedComment / LineComment / WS

# // SECTION: separatorsAndOperations
#RESERVED <- '...'
DOT <- '.'
COMMA <- ','
LPAREN <- '('
RPAREN <- ')'
LSQUARE <- '['
RSQUARE <- ']'
LCURL <- '{'
# /*
# * When using another programming language (not Java) to generate a parser,
# * please replace this code with the corresponding code of a programming language you are using.
# */
RCURL <- '}'
MULT <- '*'
MOD <- '%'
DIV <- '/'
ADD <- '+'
SUB <- '-'
INCR <- '++'
DECR <- '--'
CONJ <- '&&'
DISJ <- '||'
EXCL_WS <- '!' Hidden
EXCL_NO_WS <- '!'
COLON <- ':'
SEMICOLON <- ';'
ASSIGNMENT <- '=' !'='
ADD_ASSIGNMENT <- '+='
SUB_ASSIGNMENT <- '-='
MULT_ASSIGNMENT <- '*='
DIV_ASSIGNMENT <- '/='
MOD_ASSIGNMENT <- '%='
ARROW <- '->'
#DOUBLE_ARROW <- '=>'
RANGE <- '..'
COLONCOLON <- '::'
#DOUBLE_SEMICOLON <- ';;'
#HASH <- '#'
AT_NO_WS <- '@'
AT_POST_WS <- '@' (Hidden / NL)
AT_PRE_WS <- (Hidden / NL) '@'
#AT_BOTH_WS <- (Hidden / NL) '@' (Hidden / NL)
QUEST_WS <- '?' Hidden
QUEST_NO_WS <- '?'
LANGLE <- '<'
RANGLE <- '>'
LE <- '<='
GE <- '>='
EXCL_EQ <- '!='
EXCL_EQEQ <- '!=='
AS_SAFE <- 'as?'
EQEQ <- '=='
EQEQEQ <- '==='
#SINGLE_QUOTE <- '\''

# // SECTION: keywords
RETURN_AT <- 'return@' Identifier
CONTINUE_AT <- 'continue@' Identifier
BREAK_AT <- 'break@' Identifier
THIS_AT <- 'this@' Identifier
SUPER_AT <- 'super@' Identifier
FILE <- 'file' !(Letter / UnicodeDigit)
FIELD <- 'field' !(Letter / UnicodeDigit)
PROPERTY <- 'property' !(Letter / UnicodeDigit)
GET <- 'get' !(Letter / UnicodeDigit)
SET <- 'set' !(Letter / UnicodeDigit)
RECEIVER <- 'receiver' !(Letter / UnicodeDigit)
PARAM <- 'param' !(Letter / UnicodeDigit)
SETPARAM <- 'setparam' !(Letter / UnicodeDigit)
DELEGATE <- 'delegate' !(Letter / UnicodeDigit)
PACKAGE <- 'package' !(Letter / UnicodeDigit)
IMPORT <- 'import' !(Letter / UnicodeDigit)
CLASS <- 'class' !(Letter / UnicodeDigit)
INTERFACE <- 'interface' !(Letter / UnicodeDigit)
FUN <- 'fun' !(Letter / UnicodeDigit)
OBJECT <- 'object' !(Letter / UnicodeDigit)
VAL <- 'val' !(Letter / UnicodeDigit)
VAR <- 'var' !(Letter / UnicodeDigit)
TYPE_ALIAS <- 'typealias' !(Letter / UnicodeDigit)
CONSTRUCTOR <- 'constructor' !(Letter / UnicodeDigit)
BY <- 'by' !(Letter / UnicodeDigit)
COMPANION <- 'companion' !(Letter / UnicodeDigit)
INIT <- 'init' !(Letter / UnicodeDigit)
THIS <- 'this' !(Letter / UnicodeDigit)
SUPER <- 'super' !(Letter / UnicodeDigit)
TYPEOF <- 'typeof' !(Letter / UnicodeDigit)
WHERE <- 'where' !(Letter / UnicodeDigit)
IF <- 'if' !(Letter / UnicodeDigit)
ELSE <- 'else' !(Letter / UnicodeDigit)
WHEN <- 'when' !(Letter / UnicodeDigit)
TRY <- 'try' !(Letter / UnicodeDigit)
CATCH <- 'catch' !(Letter / UnicodeDigit)
FINALLY <- 'finally' !(Letter / UnicodeDigit)
FOR <- 'for' !(Letter / UnicodeDigit)
DO <- 'do' !(Letter / UnicodeDigit)
WHILE <- 'while' !(Letter / UnicodeDigit)
THROW <- 'throw' !(Letter / UnicodeDigit)
RETURN <- 'return' !(Letter / UnicodeDigit)
CONTINUE <- 'continue' !(Letter / UnicodeDigit)
BREAK <- 'break' !(Letter / UnicodeDigit)
AS <- 'as' !(Letter / UnicodeDigit)
IS <- 'is' !(Letter / UnicodeDigit)
IN <- 'in' !(Letter / UnicodeDigit)
NOT_IS <- '!is' !(Letter / UnicodeDigit)
NOT_IN <- '!in' !(Letter / UnicodeDigit)
OUT <- 'out' !(Letter / UnicodeDigit)
DYNAMIC <- 'dynamic' !(Letter / UnicodeDigit)

# // SECTION: lexicalModifiers
PUBLIC <- 'public' !(Letter / UnicodeDigit)
PRIVATE <- 'private' !(Letter / UnicodeDigit)
PROTECTED <- 'protected' !(Letter / UnicodeDigit)
INTERNAL <- 'internal' !(Letter / UnicodeDigit)
ENUM <- 'enum' !(Letter / UnicodeDigit)
SEALED <- 'sealed' !(Letter / UnicodeDigit)
ANNOTATION <- 'annotation' !(Letter / UnicodeDigit)
DATA <- 'data' !(Letter / UnicodeDigit)
INNER <- 'inner' !(Letter / UnicodeDigit)
TAILREC <- 'tailrec' !(Letter / UnicodeDigit)
OPERATOR <- 'operator' !(Letter / UnicodeDigit)
INLINE <- 'inline' !(Letter / UnicodeDigit)
INFIX <- 'infix' !(Letter / UnicodeDigit)
EXTERNAL <- 'external' !(Letter / UnicodeDigit)
SUSPEND <- 'suspend' !(Letter / UnicodeDigit)
OVERRIDE <- 'override' !(Letter / UnicodeDigit)
ABSTRACT <- 'abstract' !(Letter / UnicodeDigit)
FINAL <- 'final' !(Letter / UnicodeDigit)
OPEN <- 'open' !(Letter / UnicodeDigit)
CONST <- 'const' !(Letter / UnicodeDigit)
LATEINIT <- 'lateinit' !(Letter / UnicodeDigit)
VARARG <- 'vararg' !(Letter / UnicodeDigit)
NOINLINE <- 'noinline' !(Letter / UnicodeDigit)
CROSSINLINE <- 'crossinline' !(Letter / UnicodeDigit)
REIFIED <- 'reified' !(Letter / UnicodeDigit)
EXPECT <- 'expect' !(Letter / UnicodeDigit)
ACTUAL <- 'actual' !(Letter / UnicodeDigit)

# // SECTION: literals
DecDigit <- [0-9]
DecDigitNoZero <- [1-9]
DecDigitOrSeparator <- DecDigit / '_'
DecDigits <- DecDigit DecDigitOrSeparator* / DecDigit
DoubleExponent <- [eE] [-+]? DecDigits
RealLiteral <- FloatLiteral / DoubleLiteral
FloatLiteral <- DoubleLiteral [fF] / DecDigits [fF]
DoubleLiteral <- DecDigits? '.' DecDigits DoubleExponent? / DecDigits DoubleExponent
IntegerLiteral <- DecDigitNoZero DecDigitOrSeparator* / DecDigit
#IntegerLiteral <- DecDigitNoZero DecDigitOrSeparator* DecDigit / DecDigit
HexDigit <- [0-9a-fA-F]
HexDigitOrSeparator <- HexDigit / '_'
HexLiteral <- '0' [xX] HexDigit HexDigitOrSeparator* / '0' [xX] HexDigit
BinDigit <- [01]
BinDigitOrSeparator <- BinDigit / '_'
BinLiteral <- '0' [bB] BinDigit BinDigitOrSeparator* / '0' [bB] BinDigit
UnsignedLiteral <- (HexLiteral / BinLiteral / IntegerLiteral) [uU] [lL]?
LongLiteral <- (HexLiteral / BinLiteral / IntegerLiteral) [lL]
BooleanLiteral <- 'true'/ 'false'
NullLiteral <- 'null'
CharacterLiteral <- '\'' (EscapeSeq / [^\n\r'\\]) '\''

# // SECTION: lexicalIdentifiers
#UnicodeDigit <- UNICODE_CLASS_ND
Identifier <- '`' [^`\r\n]+ '`' / (Letter / '_') (Letter / '_' / UnicodeDigit)*
IdentifierOrSoftKey <- Identifier / ABSTRACT / ANNOTATION / BY / CATCH / COMPANION / CONSTRUCTOR / CROSSINLINE / DATA / DYNAMIC / ENUM / EXTERNAL / FINAL / FINALLY / IMPORT / INFIX / INIT / INLINE / INNER / INTERNAL / LATEINIT / NOINLINE / OPEN / OPERATOR / OUT / OVERRIDE / PRIVATE / PROTECTED / PUBLIC / REIFIED / SEALED / TAILREC / VARARG / WHERE / GET / SET / FIELD / PROPERTY / RECEIVER / PARAM / SETPARAM / DELEGATE / FILE / EXPECT / ACTUAL / CONST / SUSPEND
FieldIdentifier <- '$' IdentifierOrSoftKey
UniCharacterLiteral <- '\\' 'u' HexDigit HexDigit HexDigit HexDigit
EscapedIdentifier <- '\\' ('t' / 'b' / 'r' / 'n' / '\'' / '"' / '\\' / '$')
EscapeSeq <- UniCharacterLiteral / EscapedIdentifier

# // SECTION: characters
Letter <- [\u0041-\u005A\u0061-\u007A\u00AA\u00B5\u00BA\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02C1\u02C6-\u02D1\u02E0-\u02E4\u02EC\u02EE\u0370-\u0374\u0376-\u0377\u037A-\u037D\u0386\u0388-\u038A\u038C\u038E-\u03A1\u03A3-\u03F5\u03F7-\u0481\u048A-\u0527\u0531-\u0556\u0559\u0561-\u0587\u05D0-\u05EA\u05F0-\u05F2\u0620-\u064A\u066E-\u066F\u0671-\u06D3\u06D5\u06E5-\u06E6\u06EE-\u06EF\u06FA-\u06FC\u06FF\u0710\u0712-\u072F\u074D-\u07A5\u07B1\u07CA-\u07EA\u07F4-\u07F5\u07FA\u0800-\u0815\u081A\u0824\u0828\u0840-\u0858\u08A0\u08A2-\u08AC\u0904-\u0939\u093D\u0950\u0958-\u0961\u0971-\u0977\u0979-\u097F\u0985-\u098C\u098F-\u0990\u0993-\u09A8\u09AA-\u09B0\u09B2\u09B6-\u09B9\u09BD\u09CE\u09DC-\u09DD\u09DF-\u09E1\u09F0-\u09F1\u0A05-\u0A0A\u0A0F-\u0A10\u0A13-\u0A28\u0A2A-\u0A30\u0A32-\u0A33\u0A35-\u0A36\u0A38-\u0A39\u0A59-\u0A5C\u0A5E\u0A72-\u0A74\u0A85-\u0A8D\u0A8F-\u0A91\u0A93-\u0AA8\u0AAA-\u0AB0\u0AB2-\u0AB3\u0AB5-\u0AB9\u0ABD\u0AD0\u0AE0-\u0AE1\u0B05-\u0B0C\u0B0F-\u0B10\u0B13-\u0B28\u0B2A-\u0B30\u0B32-\u0B33\u0B35-\u0B39\u0B3D\u0B5C-\u0B5D\u0B5F-\u0B61\u0B71\u0B83\u0B85-\u0B8A\u0B8E-\u0B90\u0B92-\u0B95\u0B99-\u0B9A\u0B9C\u0B9E-\u0B9F\u0BA3-\u0BA4\u0BA8-\u0BAA\u0BAE-\u0BB9\u0BD0\u0C05-\u0C0C\u0C0E-\u0C10\u0C12-\u0C28\u0C2A-\u0C33\u0C35-\u0C39\u0C3D\u0C58-\u0C59\u0C60-\u0C61\u0C85-\u0C8C\u0C8E-\u0C90\u0C92-\u0CA8\u0CAA-\u0CB3\u0CB5-\u0CB9\u0CBD\u0CDE\u0CE0-\u0CE1\u0CF1-\u0CF2\u0D05-\u0D0C\u0D0E-\u0D10\u0D12-\u0D3A\u0D3D\u0D4E\u0D60-\u0D61\u0D7A-\u0D7F\u0D85-\u0D96\u0D9A-\u0DB1\u0DB3-\u0DBB\u0DBD\u0DC0-\u0DC6\u0E01-\u0E30\u0E32-\u0E33\u0E40-\u0E46\u0E81-\u0E82\u0E84\u0E87-\u0E88\u0E8A\u0E8D\u0E94-\u0E97\u0E99-\u0E9F\u0EA1-\u0EA3\u0EA5\u0EA7\u0EAA-\u0EAB\u0EAD-\u0EB0\u0EB2-\u0EB3\u0EBD\u0EC0-\u0EC4\u0EC6\u0EDC-\u0EDF\u0F00\u0F40-\u0F47\u0F49-\u0F6C\u0F88-\u0F8C\u1000-\u102A\u103F\u1050-\u1055\u105A-\u105D\u1061\u1065-\u1066\u106E-\u1070\u1075-\u1081\u108E\u10A0-\u10C5\u10C7\u10CD\u10D0-\u10FA\u10FC-\u1248\u124A-\u124D\u1250-\u1256\u1258\u125A-\u125D\u1260-\u1288\u128A-\u128D\u1290-\u12B0\u12B2-\u12B5\u12B8-\u12BE\u12C0\u12C2-\u12C5\u12C8-\u12D6\u12D8-\u1310\u1312-\u1315\u1318-\u135A\u1380-\u138F\u13A0-\u13F4\u1401-\u166C\u166F-\u167F\u1681-\u169A\u16A0-\u16EA\u16EE-\u16F0\u1700-\u170C\u170E-\u1711\u1720-\u1731\u1740-\u1751\u1760-\u176C\u176E-\u1770\u1780-\u17B3\u17D7\u17DC\u1820-\u1877\u1880-\u18A8\u18AA\u18B0-\u18F5\u1900-\u191C\u1950-\u196D\u1970-\u1974\u1980-\u19AB\u19C1-\u19C7\u1A00-\u1A16\u1A20-\u1A54\u1AA7\u1B05-\u1B33\u1B45-\u1B4B\u1B83-\u1BA0\u1BAE-\u1BAF\u1BBA-\u1BE5\u1C00-\u1C23\u1C4D-\u1C4F\u1C5A-\u1C7D\u1CE9-\u1CEC\u1CEE-\u1CF1\u1CF5-\u1CF6\u1D00-\u1DBF\u1E00-\u1F15\u1F18-\u1F1D\u1F20-\u1F45\u1F48-\u1F4D\u1F50-\u1F57\u1F59\u1F5B\u1F5D\u1F5F-\u1F7D\u1F80-\u1FB4\u1FB6-\u1FBC\u1FBE\u1FC2-\u1FC4\u1FC6-\u1FCC\u1FD0-\u1FD3\u1FD6-\u1FDB\u1FE0-\u1FEC\u1FF2-\u1FF4\u1FF6-\u1FFC\u2071\u207F\u2090-\u209C\u2102\u2107\u210A-\u2113\u2115\u2119-\u211D\u2124\u2126\u2128\u212A-\u212D\u212F-\u2139\u213C-\u213F\u2145-\u2149\u214E\u2160-\u2188\u2C00-\u2C2E\u2C30-\u2C5E\u2C60-\u2CE4\u2CEB-\u2CEE\u2CF2-\u2CF3\u2D00-\u2D25\u2D27\u2D2D\u2D30-\u2D67\u2D6F\u2D80-\u2D96\u2DA0-\u2DA6\u2DA8-\u2DAE\u2DB0-\u2DB6\u2DB8-\u2DBE\u2DC0-\u2DC6\u2DC8-\u2DCE\u2DD0-\u2DD6\u2DD8-\u2DDE\u2E2F\u3005-\u3007\u3021-\u3029\u3031-\u3035\u3038-\u303C\u3041-\u3096\u309D-\u309F\u30A1-\u30FA\u30FC-\u30FF\u3105-\u312D\u3131-\u318E\u31A0-\u31BA\u31F0-\u31FF\u3400\u4DB5\u4E00\u9FCC\uA000-\uA48C\uA4D0-\uA4FD\uA500-\uA60C\uA610-\uA61F\uA62A-\uA62B\uA640-\uA66E\uA67F-\uA697\uA6A0-\uA6EF\uA717-\uA71F\uA722-\uA788\uA78B-\uA78E\uA790-\uA793\uA7A0-\uA7AA\uA7F8-\uA801\uA803-\uA805\uA807-\uA80A\uA80C-\uA822\uA840-\uA873\uA882-\uA8B3\uA8F2-\uA8F7\uA8FB\uA90A-\uA925\uA930-\uA946\uA960-\uA97C\uA984-\uA9B2\uA9CF\uAA00-\uAA28\uAA40-\uAA42\uAA44-\uAA4B\uAA60-\uAA76\uAA7A\uAA80-\uAAAF\uAAB1\uAAB5-\uAAB6\uAAB9-\uAABD\uAAC0\uAAC2\uAADB-\uAADD\uAAE0-\uAAEA\uAAF2-\uAAF4\uAB01-\uAB06\uAB09-\uAB0E\uAB11-\uAB16\uAB20-\uAB26\uAB28-\uAB2E\uABC0-\uABE2\uAC00\uD7A3\uD7B0-\uD7C6\uD7CB-\uD7FB\uF900-\uFA6D\uFA70-\uFAD9\uFB00-\uFB06\uFB13-\uFB17\uFB1D\uFB1F-\uFB28\uFB2A-\uFB36\uFB38-\uFB3C\uFB3E\uFB40-\uFB41\uFB43-\uFB44\uFB46-\uFBB1\uFBD3-\uFD3D\uFD50-\uFD8F\uFD92-\uFDC7\uFDF0-\uFDFB\uFE70-\uFE74\uFE76-\uFEFC\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFFBE\uFFC2-\uFFC7\uFFCA-\uFFCF\uFFD2-\uFFD7\uFFDA-\uFFDC]
UnicodeDigit <- [\u0030-\u0039\u0660-\u0669\u06F0-\u06F9\u07C0-\u07C9\u0966-\u096F\u09E6-\u09EF\u0A66-\u0A6F\u0AE6-\u0AEF\u0B66-\u0B6F\u0BE6-\u0BEF\u0C66-\u0C6F\u0CE6-\u0CEF\u0D66-\u0D6F\u0E50-\u0E59\u0ED0-\u0ED9\u0F20-\u0F29\u1040-\u1049\u1090-\u1099\u17E0-\u17E9\u1810-\u1819\u1946-\u194F\u19D0-\u19D9\u1A80-\u1A89\u1A90-\u1A99\u1B50-\u1B59\u1BB0-\u1BB9\u1C40-\u1C49\u1C50-\u1C59\uA620-\uA629\uA8D0-\uA8D9\uA900-\uA909\uA9D0-\uA9D9\uAA50-\uAA59\uABF0-\uABF9\uFF10-\uFF19]

# // SECTION: strings
QUOTE_OPEN <- '"' !'""'
TRIPLE_QUOTE_OPEN <- '"""'
QUOTE_CLOSE <- '"'
LineStrRef <- FieldIdentifier
#LineStrText <- [^\\"$]+ / '$'
LineStrText <- [^\\"$]+ / '$'
LineStrEscapedChar <- EscapedIdentifier / UniCharacterLiteral
LineStrExprStart <- '${'
TRIPLE_QUOTE_CLOSE <- '"""""'/ '""""' / '"""'
MultiLineStringQuote <- '""' !'"' / '"' !'""'
MultiLineStrRef <- FieldIdentifier
#MultiLineStrText <- !('"' / '$')+ / '$'
MultiLineStrText <- [^"$]+ / '$'
MultiLineStrExprStart <- '${'

_ <- (WS / DelimitedComment / LineComment)+
__ <- ([ \t\f\r\n] / DelimitedComment / LineComment)+
WS <- [ \t\f]
NL <- _* ('\n' / '\r' '\n'?) _*
EOF <- !.

%%
#include "kotlin_post.h"