# -*- text -*-
#
# Parsing Expression Grammar declaring a syntax for Parsing Expression
# Grammars, to use in a PEG-based parser generator. This specification
# is self-referential, it uses the grammar described by to describe
# said grammar.

PEG pg::peg::grammar (Grammar)

	# --------------------------------------------------------------------
        # Syntactical constructs

        Grammar         <- SPACE Header Definition+ Final EOF ;

        Header          <- PEG Identifier StartExpr ;
        Definition      <- Attribute? Identifier IS Expression SEMICOLON ;
        Attribute       <- (VOID / LEAF / MATCH) COLON ;
        Expression      <- Sequence (SLASH Sequence)* ;
        Sequence        <- Prefix+ ;
        Prefix          <- (AND / NOT)? Suffix ;
        Suffix          <- Primary (QUESTION / STAR / PLUS)? ;
        Primary         <- ALNUM / ALPHA / Identifier
                        /  OPEN Expression CLOSE
                        /  Literal
                        /  Class
                        /  DOT
                        ;
        Literal         <- APOSTROPH  (!APOSTROPH  Char)* APOSTROPH  SPACE
                        /  DAPOSTROPH (!DAPOSTROPH Char)* DAPOSTROPH SPACE ;
        Class           <- OPENB (!CLOSEB Range)* CLOSEB SPACE ;
        Range           <- Char TO Char / Char ;

        StartExpr       <- OPEN Expression CLOSE ;
void:   Final           <- END SEMICOLON SPACE ;

        # --------------------------------------------------------------------
        # Lexing constructs

        Identifier      <- Ident SPACE ;
match:  Ident           <- ('_' / ':' / <alpha>) ('_' / ':' / <alnum>)* ;
        Char            <- CharSpecial / CharOctalFull / CharOctalPart
                        /  CharUnicode / CharUnescaped
                        ;

match:  CharSpecial     <- "\\" [nrt'"\[\]\\] ;
match:  CharOctalFull   <- "\\" [0-2][0-7][0-7] ;
match:  CharOctalPart   <- "\\" [0-7][0-7]? ;
match:  CharUnicode     <- "\\" 'u' HexDigit (HexDigit (HexDigit HexDigit?)?)? ;
match:  CharUnescaped   <- !"\\" . ;

void:   HexDigit        <- [0-9a-fA-F] ;

void:   TO              <- '-'           ;
void:   OPENB           <- "["           ;
void:   CLOSEB          <- "]"           ;
void:   APOSTROPH       <- "'"           ;
void:   DAPOSTROPH      <- '"'           ;
void:   PEG             <- "PEG"   SPACE ;
void:   IS              <- "<-"    SPACE ;
leaf:   VOID            <- "void"  SPACE ; # Implies that definition has no semantic value.
leaf:   LEAF            <- "leaf"  SPACE ; # Implies that definition has no terminals.
leaf:   MATCH           <- "match" SPACE ; # Implies that semantic value is the matched string,
                                           # not the parse tree from the symbol.
void:   END             <- "END"   SPACE ;
void:   SEMICOLON       <- ";"     SPACE ;
void:   COLON           <- ":"     SPACE ;
void:   SLASH           <- "/"     SPACE ;
leaf:   AND             <- "&"     SPACE ;
leaf:   NOT             <- "!"     SPACE ;
leaf:   QUESTION        <- "?"     SPACE ;
leaf:   STAR            <- "*"     SPACE ;
leaf:   PLUS            <- "+"     SPACE ;
void:   OPEN            <- "("     SPACE ;
void:   CLOSE           <- ")"     SPACE ;
leaf:   DOT             <- "."     SPACE ;
leaf:   ALPHA           <- "<alpha>" SPACE ;
leaf:   ALNUM           <- "<alnum>" SPACE ;

void:   SPACE           <- (" " / "\t" / EOL / COMMENT)* ;
void:   COMMENT         <- '#' (!EOL .)* EOL ;
void:   EOL             <- "\n\r" / "\n" / "\r" ;
void:   EOF             <- !. ;

        # --------------------------------------------------------------------
END;