BNF for Junicon.jj

/*************************************************************************
 * Tokens
 ************************************************************************/

//========================================================================
// Tokens: White space
//========================================================================

<DEFAULT> SKIP : {
<WHITESPACE: " " | "\t" | "\f">
}

   

<DEFAULT> SPECIAL : {
<NEWLINE: "\n" | "\r" | "\r\n">
}

   

//========================================================================
// Tokens: Comments
//========================================================================

<DEFAULT> TOKEN : {
<ANNOTATION_COMMENT_PREFIX: "#@<">
| <#REST_OF_LINE: (~["\n","\r"])* ("\n" | "\r" | "\r\n")?>
}

   

<DEFAULT> SPECIAL : {
<SINGLE_LINE_COMMENT: "#" (~["\n","\r"])* ("\n" | "\r" | "\r\n")?>
}

   

//========================================================================
// Tokens: Reserved words
//========================================================================

<DEFAULT> TOKEN : {
<ABSTRACT: "abstract">
| <BREAK: "break">
| <BY: "by">
| <CASE: "case">
| <CLASS: "class">
| <CREATE: "create">
| <THREAD: "thread">
| <CRITICAL: "critical">
| <DEFAULT_token: "default">
| <DO: "do">
| <ELSE: "else">
| <END: "end">
| <EVERY: "every">
| <FAIL: "fail">
| <GLOBAL: "global">
| <IF: "if">
| <IMPORT: "import">
| <IN: "in">
| <INITIAL: "initial">
| <INITIALLY: "initially">
| <INVOCABLE: "invocable">
| <LINK: "link">
| <LOCAL: "local">
| <METHOD: "method">
| <NEW: "new">
| <NEXT: "next">
| <NOT: "not">
| <NULL_LITERAL: "null">
| <OF: "of">
| <PACKAGE: "package">
| <PROCEDURE: "procedure">
| <RECORD: "record">
| <REPEAT: "repeat">
| <RETURN: "return">
| <STATIC: "static">
| <SUSPEND: "suspend">
| <THEN: "then">
| <TO: "to">
| <UNTIL: "until">
| <WHILE: "while">
}

   

//========================================================================
// Tokens: Separators
//========================================================================

<DEFAULT> TOKEN : {
<LPAREN: "(">
| <RPAREN: ")">
| <LBRACE: "{">
| <RBRACE: "}">
| <LBRACKET: "[">
| <RBRACKET: "]">
| <SEMICOLON: ";">
| <COMMA: ",">
| <DOTDOTDOT: "...">
| <DOT: ".">
| <ANNOTATION: "@<">
| <AT: "@">
| <COMPR: "[:">
| <COMPREND: ":]">
| <COLON: ":">
| <COLONCOLON: "::">
| <LBRACE_EBCDIC: "$("> : {
| <RBRACE_EBCDIC: "$)"> : {
| <LBRACKET_EBCDIC: "$<"> : {
| <RBRACKET_EBCDIC: "$>"> : {
}

   

//========================================================================
// Tokens: Operators, grouped by precedence
//========================================================================

<DEFAULT> TOKEN : {
<AND: "&">
| <QMARK: "?">
| <ASSIGN: ":=">
| <REVASSIGN: "<-">
| <REVSWAP: "<->">
| <SWAP: ":=:">
| <PMATCH: "??">
| <POR: ".|">
| <BAR: "|">
| <PAND: "&&">
| <GT: ">">
| <LT: "<">
| <EQ: "==">
| <LE: "<=">
| <GE: ">=">
| <EQUALS: "=">
| <LSHIFT: "<<">
| <RSHIFT: ">>">
| <SLE: "<<=">
| <EQUIV: "===">
| <SGE: ">>=">
| <NMNE: "~=">
| <SNE: "~==">
| <NEQUIV: "~===">
| <PIMDASSN: "$$">
| <PASSNONMATCH: "->">
| <SND: "@>">
| <SNDBK: "@>>">
| <RCV: "<@">
| <RCVBK: "<<@">
| <CONCAT: "||">
| <LCONCAT: "|||">
| <PLUS: "+">
| <MINUS: "-">
| <INCR: "++">
| <DECR: "--">
| <STAR: "*">
| <SLASH: "/">
| <PERCENT: "%">
| <INTER: "**">
| <CARET: "^">
| <BANG: "!">
| <BACKSLASH: "\\">
| <TILDE: "~">
| <PSETCUR: ".$">
| <PIPE: "|>">
| <FUTURE: "|>>">
| <COEXPR: "|<>">
| <FIRSTCLASS: "<>">
| <PARALLEL: "|<>|">
| <BACKQUOTE: "`">
| <PCOLON: "+:">
| <MCOLON: "-:">
| <DOLLAR: "$">
| <AUGMOD: "%:=">
| <AUGAND: "&:=">
| <AUGSTAR: "*:=">
| <AUGINTER: "**:=">
| <AUGPLUS: "+:=">
| <AUGUNION: "++:=">
| <AUGMINUS: "-:=">
| <AUGDIFF: "--:=">
| <AUGSLASH: "/:=">
| <AUGNMLT: "<:=">
| <AUGSLT: "<<:=">
| <AUGSLE: "<<=:=">
| <AUGNMLE: "<=:=">
| <AUGNMEQ: "=:=">
| <AUGSEQ: "==:=">
| <AUGEQUIV: "===:=">
| <AUGNMGT: ">:=">
| <AUGNMGE: ">=:=">
| <AUGSGT: ">>:=">
| <AUGSGE: ">>=:=">
| <AUGQMARK: "?:=">
| <AUGAT: "@:=">
| <AUGCARET: "^:=">
| <AUGCONCAT: "||:=">
| <AUGNEQUIV: "~===:=">
| <AUGSNE: "~==:=">
| <AUGNMNE: "~=:=">
| <AUGLCONCAT: "|||:=">
}

   

//========================================================================
// Tokens: Numeric literals
//========================================================================

<DEFAULT> TOKEN : {
<INTEGER_LITERAL: <DECIMAL_LITERAL>>
| <RADIX_LITERAL: <DECIMAL_LITERAL> ["r","R"] (["0"-"9","a"-"f","A"-"F"])+>
| <#DECIMAL_LITERAL: (["0"-"9"])+>
| <REAL_LITERAL: (["0"-"9"])+ "." (["0"-"9"])* (<EXPONENT>)? | "." (["0"-"9"])+ (<EXPONENT>)? | (["0"-"9"])+ <EXPONENT>>
| <#EXPONENT: ["e","E"] (["+","-"])? (["0"-"9"])+>
}

   

<DEFAULT> MORE : {
"\"" : WithinQuote
}

   

<WithinQuote> SKIP : {
<"_" <NEWLINE> (<WHITESPACE>)*> : WithinQuote
}

   

<WithinQuote> TOKEN : {
<STRING_LITERAL: "\""> : DEFAULT
}

   

<WithinQuote> MORE : {
<~["\"","\\"] | <ESCAPE_LITERAL>>
}

   

//====================
// Single quote -- multiline continuations are as for double quotes.
//====================

<DEFAULT> MORE : {
"\'" : WithinSingleQuote
}

   

<WithinSingleQuote> SKIP : {
<"_" <NEWLINE> (<WHITESPACE>)*> : WithinSingleQuote
}

   

<WithinSingleQuote> TOKEN : {
<SINGLE_QUOTE_LITERAL: "\'"> : DEFAULT
}

   

<WithinSingleQuote> MORE : {
<~["\'","\\"] | <ESCAPE_LITERAL>>
}

   

//====================
// Quote regex dependencies
//====================

<DEFAULT> TOKEN : {
<#OLD_SINGLE_QUOTE_LITERAL: "\'" (~["\'","\\"] | <ESCAPE_LITERAL>)* "\'">
| <#OLD_STRING_LITERAL: "\"" (~["\"","\\"] | <ESCAPE_LITERAL>)* "\"">
| <#ESCAPE_LITERAL: "\\" (<HEX_ESCAPE> | <OCTAL_ESCAPE> | <CONTROL_ESCAPE> | ["!"-"~"," ","\t","\n","\r","\f"])>
| <#ESCAPE_LITERAL_NO_NL: "\\" (<HEX_ESCAPE> | <OCTAL_ESCAPE> | <CONTROL_ESCAPE> | ["!"-"~"," ","\t","\f"])>
| <#HEX_ESCAPE: ["x","X"] <HEX_DIGITS>>
| <#OCTAL_ESCAPE: ["0"-"7"] ["0"-"7"]>
| <#CONTROL_ESCAPE: "^" ["!"-"~"]>
| <#HEX_DIGITS: ["0"-"9","a"-"f","A"-"F"]>
}

   

//========================================================================
// Tokens: Big literals (Block quotes, or Big quotes)
//========================================================================

<DEFAULT> TOKEN : {
<BIG_LITERAL: "{<" (~[">","}","\\"] | ">" ~["}","\\"] | ~[">","\\"] "}" | ">" <ESCAPE_LITERAL> | <ESCAPE_LITERAL> ("}")?)* ([">","}"])? ">}">
}

   

//========================================================================
// Tokens: Identifiers
//========================================================================

<DEFAULT> TOKEN : {
<IDENTIFIER: <IDENTIFIER_BASE>>
| <#IDENTIFIER_BASE: <ALPHA> (<ALPHA> | <DIGIT>)*>
| <#ALPHA: ["A"-"Z","a"-"z","_"]>
| <#LETTER: ["$","A"-"Z","_","a"-"z","\u00c0"-"\u00d6","\u00d8"-"\u00f6","\u00f8"-"\u00ff","\u0100"-"\u1fff","\u3040"-"\u318f","\u3300"-"\u337f","\u3400"-"\u3d2d","\u4e00"-"\u9fff","\uf900"-"\ufaff"]>
| <#DIGIT: ["0"-"9","\u0660"-"\u0669","\u06f0"-"\u06f9","\u0966"-"\u096f","\u09e6"-"\u09ef","\u0a66"-"\u0a6f","\u0ae6"-"\u0aef","\u0b66"-"\u0b6f","\u0be7"-"\u0bef","\u0c66"-"\u0c6f","\u0ce6"-"\u0cef","\u0d66"-"\u0d6f","\u0e50"-"\u0e59","\u0ed0"-"\u0ed9","\u1040"-"\u1049"]>
}

   

/*************************************************************************
 * LANGUAGE GRAMMAR STARTS HERE
 ************************************************************************/

//========================================================================
// Grammar
//========================================================================

/****
Tags for the generated XML syntax tree are as follows:

identifier		# Atoms
literal    @isInteger @isReal @isRadix (numbers)
literal	   @isQuote ("") @isSingleQuote ('') @isBigLiteral {< >}
literal    @isWord (null)
andKeyword &i
methodref  i.i::i | ((Cast) i).i::i	# i::i called packageref in Unicon
packagref  i.i::i		# type instead of reference in expression above
dotname    i.i	# introduced atom when transforming commands, new allocation

operator   @isAugment @isBoolean	# Operators
delimiter  @id ,;
keyword	   reserved word
assign     x := y
product    x & y

program			# Program, everything is expression
expression
type            # type, declaration, or atom used in expression
                # e.g., declaration/ identifier | dotname
		#       atom/ identifier | dotname
declaration @isGlobalVariable, @isLocalVariable, @isParameter, @isClassOrMethod
comment    #
newline

tuple      (e,e) | (e)	# Grouping
list       [e,e] | [e]	@isEmpty=true if empty list or map
block      {e;e} | {e}	@isClosure=true if body of closure
set        {e,e}
map        [e:e, ] | [e:e; ]
angleTuple  | 
comprehension [: e :]	# list comprehension

			# Expressions
			#     where e=expr, i=identifier, k=keyword, op=operator
annotation @<i i=e ...> e	# Scoped annotations are a hybrid of XML and Java annotations
	   @</i>		# Can omit end tag for non-strict XML
           @<i i=e .../> e	# Can use abbreviated empty-element tags
           @<i(i=e,...)> e	# Can also use Java style attributes
           @<i(e)> e
directive  $i e e # preprocessor
dollar	   $i(e,) # super invoke
keyvalue   e:e
slice      e..e
operation  e op e | op e   @isUnary @isBinary 
qualified  i:i:i | i:i=i | i=i | i:i	# parameter declaration
statement  k ...		# declaration, or control construct
subscript  [e,e]		# when appears in index operation
closure    (e,e) -> {e}		# alternative: {(e,e) -> e}

			# Declarations
dotname    i.i			# used in declarations, e.g., import and package
enum       e,e | e e (open)	# used in declarations

iterator   (i in e)	# Primaries
cast       (type) e
invoke     oe(,)[,]		# where oe is (alloc | atom | group)
				# Restricted to e(e) after transformation
index	   e[e]			# Only introduced after transformation
objref     oe(,)[,].f(,)[,]	# where f is atom subset (identifier | methodref | "initially")
atom       @emptyInner between or leading , or ;    @emptyTrailing    @allEmpty
group
command    dotname|id|methodref expr ...  # treated as dotname(expr,...)
				# where expr is limited primary expression
				#	that starts with identifier or literal
allocation new e		# where e is type

EXAMPLE:  (x=1) =>
          expr/group/list/{expr/operation/{expr/atom/identifier,operator=,expr/atom/literal}}

Concrete syntax tags: identifier, literal, keyword, operator, delimiter.


****/

//========================================================================
// Program structuring syntax follows.
//========================================================================

//====
// Junicon allows optional trailing semicolons in declarations.
//====

//========================================================================
// Procedures.
//========================================================================

//========================================================================
// Classes.
//========================================================================

//============================================================================
// Declared names (variables and parameters), Types, and Atoms in expressions.
//============================================================================

//====
// Parameter can be param[:type][: or =literal]
//====

//========================================================================
// Parameterized closure.
//========================================================================

//========================================================================
// Block statements and bounded expressions.
//========================================================================

//====
// ExprSequence is to be non-empty, and ends with an expression or innerEmpty.
//	If it would be empty, test and use allEmpty instead.
// Invariant: {AllEmpty} or {x;TrailingEmpty} or {x;y} where x may be InnerEmpty
//====
// We pad leading ";", trailing ";", and ";;" with an empty atom.
// We indicate an empty ExprSequence with AllEmpty().
// A trailing Empty() mimics Icon sequence semantics which ends with ;null.
// Transform must later detect last expr after semicolon to be unbounded.
//====

//====
// Blockstatement ends with semicolon.
//====

//====
// BlockExpr need not have semicolon after it.
//====

//========================================================================
// Control constructs.
//========================================================================

//========================================================================
// Expressions.
//========================================================================

//====
// Cascade down for precedence, low to high.
// Higher precedence operators bind more tightly, i.e., are executed first.
//====

//====
// Iterative production is left-associative.
//====

//====
// Recursive production is right-associative.
//====

//====
// Transform must detect Lhs of assign.
//====

//====
// Operations are only allowed to the left of a statement, e.g., if.
// Anything to the right of a statement will bind together.
//====

//====
// CaretExpression is right associative via right recursion.
//====

//====
// UnaryExpression is right associative via right recursion.
//====

//====
// Transform must detect that unary Not applies to boundedExpression()
//====

//========================================================================
// Primary expressions.
//========================================================================

//====
// InvokeExpr includes invoke as well as standalone atom and group.
//====

//====
// ObjectExpr and FieldExpression in Java 8 use: [Arguments()] (ArraySuffix())*
//====

//====
// Exprlist is to be non-empty, and ends with an expression or innerEmpty.
//	If it would be empty, test and use allEmpty instead.
// Invariant: (AllEmpty) or (x,TrailingEmpty) or (x,y) where x may be Empty
//====
// We pad leading ",", trailing ",", and ",," with Empty().
// We indicate an empty ExprList with AllEmpty().
//====

//====
// Methodref: x.y::f | ((Cast) x).y::f(z)
//====

//========================================================================
// Tuple, map, list, and set.
//========================================================================

//========================================================================
// Iterators.
//========================================================================

//========================================================================
// Annotations.
//========================================================================

//========================================================================
// Grouped literals and keywords
//========================================================================

/*************************************************************************
 * TERMINALS START HERE
 ************************************************************************/

//========================================================================
// Terminals: Reserved Words
//========================================================================

//========================================================================
// Terminals: Delimiters
//========================================================================

//========================================================================
// Terminals: Operators
//========================================================================

//========================================================================
// Terminals: Identifiers
//========================================================================

//========================================================================
// Terminals: Literals
//========================================================================

//========================================================================
// Terminals: Annotations
//========================================================================