# Notation3 in Notation3
# Context Free Grammar without tokenization
#
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
@prefix cfg: <http://www.w3.org/2000/10/swap/grammar/bnf#>.
@prefix rul: <http://www.w3.org/2000/10/swap/grammar/bnf-rules#>.
@prefix : <http://www.w3.org/2000/10/swap/grammar/n3#>.
@prefix n3: <http://www.w3.org/2000/10/swap/grammar/n3#>.
@prefix list: <http://www.w3.org/2000/10/swap/list#>.
@prefix string: <http://www.w3.org/2000/10/swap/string#>.
@keywords a, is, of.


# Issues:
# - string token regexp not right  FIXED
# - tokenizing rules in general: whitespace are not defined in n3.n3
#   and it would be nice for the *entire* syntax description to be in RDF.
# - encoding really needs specifying
# - @keywords affects tokenizing
# - Use of dot for !
# - comments (tokenizer deals with)
# - We assume ASCII, in fact should use not notNameChars for i18n

# tokenizing:
# Absorb anything until end of regexp, then stil white space
#  period followed IMMEDIATELY by an opener or name char is taken as "!".
#  Except after a "." used instead of in those circumstances,
#	ws may be inserted between tokens.
#  WS MUST be inserted between tokens where ambiguity would arise.
#  (possible ending characters of one and beginning characters overlap)
#

<> cfg:syntaxFor [ cfg:internetMediaType 
		<http://www.w3.org/2003/mediatypes#application/n3>].

# <> rdfsem:semanticsFor ""   .....

# __________________________________________________________________
#
# The N3 Full Grammar


language a cfg:Language;
	cfg:document	document;
	cfg:whiteSpace	"@@@@@".


document a rul:Used;
	cfg:mustBeOneSequence(
	
		(
#			[ cfg:zeroOrMore declaration ]
#			[ cfg:zeroOrMore universal ]
#			[ cfg:zeroOrMore existential ]
			statements_optional
			cfg:eof
		)
	).

statements_optional cfg:mustBeOneSequence (() ( statement "." statements_optional ) ).

# Formula does NOT need period on last statement

formulacontent cfg:mustBeOneSequence (
		( ) 
		(
#		 [ cfg:zeroOrMore declaration ]
#		 [ cfg:zeroOrMore  universal ]
#		 [ cfg:zeroOrMore existential ]
		statementlist
	)).


statementlist cfg:mustBeOneSequence (
		( )
		( statement statementtail )
	).

statementtail cfg:mustBeOneSequence (
		( )
		( "." statementlist )
	).


statement  cfg:mustBeOneSequence (
		(declaration)
		(universal)
		(existential)
		(simpleStatement)
	).

universal cfg:mustBeOneSequence (
		(
			"@forAll"
			[ cfg:commaSeparatedListOf symbol ]
		)).

existential cfg:mustBeOneSequence(
		(	 "@forSome" 
			 [ cfg:commaSeparatedListOf symbol ]
		)).


declaration cfg:mustBeOneSequence(
		( "@prefix" qname explicituri  )
		( "@keywords" [ cfg:commaSeparatedListOf barename ] )
	).


simpleStatement cfg:mustBeOneSequence(( subject propertylist )).

propertylist cfg:mustBeOneSequence (
		( )
		( verb  object objecttail propertylisttail )
	).

propertylisttail cfg:mustBeOneSequence (
		( )
		( ";" propertylist )
	).


objecttail cfg:mustBeOneSequence (
		( )
		( ","   object objecttail )
	).


verb cfg:mustBeOneSequence (
		( path )
		( "@has" path )
		( "@is" path "@of" )
		( "@a" )
		( "=" )
		( "=>" )
		( "<=" ) 
	).

# prop cfg:mustBeOneSequence ((node)).

subject cfg:mustBeOneSequence ((path)).

object cfg:mustBeOneSequence ((path)).

path cfg:mustBeOneSequence(
		( node pathtail )
	).

pathtail cfg:mustBeOneSequence(
		(  )
		( "!" path )
		( "^" path )
	).


node cfg:mustBeOneSequence (
		( symbol )
		( "{" formulacontent "}" )
		( variable )
		( numericliteral )
		( literal )
		( "[" propertylist "]"  )
		(  "("  pathlist ")"  )
		( "@this" )	  #  Deprocated.  Was allowed for this log:forAll x
).


pathlist cfg:mustBeOneSequence (() (path pathlist)).

symbol cfg:mustBeOneSequence (
		(explicituri)
		(qname)
	).


literal cfg:mustBeOneSequence(( string dtlang)).

dtlang cfg:mustBeOneSequence(  ()  ("@" langcode)  ("^^" symbol)).


#______________________________________________________________________
#
#   TERMINALS

numericliteral	cfg:matches	"""[-+]?[0-9]+(\\.[0-9]+)?(e[-+]?[0-9]+)?""";
		cfg:canStartWith 	"0", "-", "+".

explicituri 	cfg:matches 	"<[^>]*>";
		cfg:canStartWith 	"<".

qname 		cfg:matches  	"(([a-zA-Z_][a-zA-Z0-9_]*)?:)?([a-zA-Z_][a-zA-Z0-9_]*)?";
		cfg:canStartWith 	"a", "_", ":".  # @@ etc unicode

barename 	cfg:matches  	"[a-zA-Z_][a-zA-Z0-9_]*";  # subset of qname
		cfg:canStartWith 	"a", "_".  # @@ etc

variable 	cfg:matches  	"\\?[a-zA-Z_][a-zA-Z0-9_]*";  # ? barename
		cfg:canStartWith 	"?".  #

# Maybe dtlang should just be part of string regexp?
# Whitespace is not allowed

# was: "[a-zA-Z][a-zA-Z0-9]*(-[a-zA-Z0-9]+)?";
langcode	cfg:matches  	"[a-z]+(-[a-z0-9]+)*"; # http://www.w3.org/TR/rdf-testcases/#language
		cfg:canStartWith 	"a".


#               raw regexp single quoted would be   "([^"]|(\\"))*"
# See:
# 	$ PYTHONPATH=$SWAP python
# 	>>> import tokenize 
# 	>>> import notation3
# 	>>> print notation3.stringToN3(tokenize.Double3)
# 	"[^\"\\\\]*(?:(?:\\\\.|\"(?!\"\"))[^\"\\\\]*)*\"\"\""
# 	>>> print notation3.stringToN3(tokenize.Double)
# 	"[^\"\\\\]*(?:\\\\.[^\"\\\\]*)*\""
# After that we have to prefix with one or three opening \"  which
# the python regexp doesn't have them.
#
# string3		cfg:matches		"\"\"\"[^\"\\\\]*(?:(?:\\\\.|\"(?!\"\"))[^\"\\\\]*)*\"\"\"".
# string1		cfg:matches		"\"[^\"\\\\]*(?:\\\\.[^\"\\\\]*)*\"".

string		cfg:matches		"(\"\"\"[^\"\\\\]*(?:(?:\\\\.|\"(?!\"\"))[^\"\\\\]*)*\"\"\")|(\"[^\"\\\\]*(?:\\\\.[^\"\\\\]*)*\")";
		cfg:canStartWith 	"\"".

#____________________________________________________

#  Axioms reducing the shortcut CFG terms to cfg:musBeOneSequence.

{ ?x cfg:zeroOrMore ?y } => {?x cfg:mustBeOneSequence ( () (?y ?x) ) }.


{ ?x cfg:commaSeparatedPeriodTerminatedListOf ?y } =>
{
	?x cfg:mustBeOneSequence (
		( "." )
		( ?y [cfg:CSLTail ?y]  )
	)
}.

{ ?x cfg:CSLTail ?y } =>
{
	?x cfg:mustBeOneSequence (
		( "." )
		( "," ?y ?x )
	)
}.

# Without the period

{ ?x cfg:commaSeparatedListOf ?y } =>
{
	?x cfg:mustBeOneSequence (
		(  )
		( ?y [cfg:CSLTail2 ?y]  )
	)
}.

{ ?x cfg:CSLTail2 ?y } =>
{
	?x cfg:mustBeOneSequence (
		( )
		( "," ?y ?x )
	)
}.


#  labelling of things which do not have explicit URIs:

{ ?x cfg:zeroOrMore [ cfg:label ?y].
	( ?y "_s" ) string:concatenation ?str } => { ?x cfg:label ?str }.

{ ?x cfg:commaSeparatedPeriodTerminatedListOf [ cfg:label ?y].
	( ?y "_csl" ) string:concatenation ?str } => { ?x cfg:label ?str }.

{ ?x cfg:CSLTail [ cfg:label ?y].
	( ?y "_necsl" ) string:concatenation ?str } => { ?x cfg:label ?str }.


#ends