rdf-n3 0.3.0 → 0.3.1

data/.yardopts

@@ -6,7 +6,6 @@
 --markup rdoc
 --readme README.md
 -
-README.md
 History.md
 AUTHORS
 VERSION

data/History.md

@@ -1,5 +1,10 @@
+0.3.1
+-----
+* Add application/turtle, application/x-turtle, text/rdf+n3 and application/rdf+n3 as mime types
+  matching this format, even though only text/turtle and text/n3 are valid.
+
 0.3.0
----------
+-----
 * New Predictive-Parser based N3 Reader, substantially faster than previous Treetop-based parser
 * RDF.rb 0.3.0 compatibility updates
   * Remove literal_normalization and qname_hacks, add back uri_hacks (until 0.3.0)

data/README.md

@@ -69,7 +69,21 @@ results in:
   h = RDF::Query::Variable.new(<#h>)
   g = RDF::Node.new()
   RDF::Statement.new(f, <#loves>, h)
-  
+
+Implementation Notes
+--------------------
+The parser is driven through a rules table contained in lib/rdf/n3/reader/meta.rb. This includes
+branch rules to indicate productions to be taken based on a current production. Terminals are denoted
+through a set of regular expressions used to match each type of terminal.
+
+The meta.rb file is generated from lib/rdf/n3/reader/n3-selectors.n3
+(taken from http://www.w3.org/2000/10/swap/grammar/n3-selectors.n3) which is the result of parsing
+http://www.w3.org/2000/10/swap/grammar/n3.n3 (along with bnf-rules.n3) using cwm using the following command sequence:
+
+    cwm n3.n3 bnf-rules.n3 --think --purge --data > n3-selectors.n3
+
+n3-selectors.n3 is itself used to generate meta.rb using script/build_meta.
+
 Dependencies
 ------------
 * [RDF.rb](http://rubygems.org/gems/rdf) (>= 0.3.0)

data/Rakefile

@@ -12,6 +12,7 @@ begin
     gemspec.homepage = "http://github.com/gkellogg/rdf-n3"
     gemspec.authors = ["Gregg Kellogg"]
     gemspec.add_dependency('rdf', '>= 0.3.0')
+    gemspec.add_development_dependency('open-uri-cached')
     gemspec.add_development_dependency('rspec', '>= 2.1.0')
     gemspec.add_development_dependency('rdf-spec', '>= 0.2.1')
     gemspec.add_development_dependency('rdf-rdfxml', '>= 0.2.1')
@@ -42,32 +43,4 @@ YARD::Rake::YardocTask.new do |t|
   t.files   = %w(lib/**/*.rb README.md History.md AUTHORS VERSION)   # optional
 end
 
-desc "Generate test manifest yaml"
-namespace :spec do
-  task :prepare do
-    $:.unshift(File.join(File.dirname(__FILE__), 'lib'))
-    $:.unshift(File.join(File.dirname(__FILE__), 'spec'))
-    require 'rdf/rdfxml'
-    require 'rdf/n3'
-    require 'rdf_helper'
-    require 'fileutils'
-
-    yaml = File.join(SWAP_DIR, "n3parser.yml")
-    FileUtils.rm_f(yaml)
-    RdfHelper::TestCase.to_yaml(SWAP_TEST, SWAP_DIR, yaml)
-    
-    yaml = File.join(SWAP_DIR, "regression.yml")
-    FileUtils.rm_f(yaml)
-    RdfHelper::TestCase.to_yaml(CWM_TEST, SWAP_DIR, yaml)
-    
-    yaml = File.join(TURTLE_DIR, "manifest.yml")
-    FileUtils.rm_f(yaml)
-    RdfHelper::TestCase.to_yaml(TURTLE_TEST, TURTLE_DIR, yaml)
-    
-    yaml = File.join(TURTLE_DIR, "manifest-bad.yml")
-    FileUtils.rm_f(yaml)
-    RdfHelper::TestCase.to_yaml(TURTLE_BAD_TEST, TURTLE_DIR, yaml)
-  end
-end
-
 task :default => :spec

data/VERSION

@@ -1 +1 @@
-0.3.0
+0.3.1

data/lib/rdf/n3/format.rb

@@ -23,8 +23,12 @@ module RDF::N3
   #
   # @see http://www.w3.org/TR/rdf-testcases/#ntriples
   class Format < RDF::Format
-    content_type     'text/turtle', :extension => :ttl
-    content_type     'text/n3', :extension => :n3
+    content_type     'text/turtle',         :extension => :ttl
+    content_type     'application/turtle',  :extension => :ttl
+    content_type     'application/x-turtle',:extension => :ttl
+    content_type     'text/n3',             :extension => :n3
+    content_type     'text/rdf+n3',         :extension => :n3
+    content_type     'application/rdf+n3',  :extension => :n3
     content_encoding 'utf-8'
 
     reader { RDF::N3::Reader }

data/lib/rdf/n3/reader/bnf-rules.n3

@@ -0,0 +1,134 @@
+#
+# Baccus - Naur Form (BNF) vocabulary
+#
+
+@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
+@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
+@prefix bnf: <http://www.w3.org/2000/10/swap/grammar/bnf#>.
+@prefix : <http://www.w3.org/2000/10/swap/grammar/bnf-rules#>.
+@prefix rul: <http://www.w3.org/2000/10/swap/grammar/bnf-rules#>.
+@prefix n3: <http://www.w3.org/2000/10/swap/grammar/n3#>.
+@prefix list: <http://www.w3.org/2000/10/swap/list#>.
+@prefix doc: <http://www.w3.org/2000/10/swap/pim/doc#>.
+@prefix log: <http://www.w3.org/2000/10/swap/log#>.
+@prefix string: <http://www.w3.org/2000/10/swap/string#>.
+@keywords a, is, of.
+
+
+<> rdfs:comment
+
+"""This set of rules process a BNF graph in its basic
+cfg:mustBeOneOf BNF form and create the branching tables to drive a 
+predictive parser.
+
+See also cfg2bnf.n3 which expands the shothand ontology into the basic
+BNF terms.
+""".
+
+#_____________________________________
+
+
+# Enumerate options:
+
+{ ?x bnf:mustBeOneSequence ?y} => { ?x optionTail ?y }.
+
+{?x optionTail [rdf:first ?y; rdf:rest ?z]} => {
+	?x 	bnf:branch [ bnf:sequence ?y];
+		optionTail ?z.
+  }.
+
+{ ?x bnf:branch [bnf:sequence ?y] } =>  { ?y sequenceTail ?y }.
+
+sequenceTail a log:Chaff.
+optionTail a log:Chaff.
+
+{ ?x sequenceTail [ rdf:rest ?z ] } => { ?x sequenceTail ?z }.
+
+# What productions can follow each other?
+# This is used for working out when to 
+
+{ ?x sequenceTail [ rdf:first ?y; rdf:rest [ rdf:first ?z ]] } =>
+	{ ?y bnf:canPrecede ?z }.
+
+{	?x bnf:branch [
+	    bnf:sequence [
+		list:last ?y]].
+	?x bnf:canPrecede ?z} =>
+	{ ?y bnf:canPrecede ?z }.
+
+{ 	?x bnf:canPrecede ?y.
+	?y bnf:branch [ bnf:sequence () ].
+	?y bnf:canPrecede ?z.
+} => {
+
+	?x bnf:canPrecede ?z.
+}.
+
+
+bnf:eof bnf:canStartWith "@EOFDUMMY". #  @@ kludge
+
+# Have to separate the next three rules or cwm seems to
+# get screwed up and assume there is no solution @@@
+ 
+{ ?x bnf:branch [bnf:sequence [ rdf:first ?y ]].
+  }  => { ?x bnf:TEST ?y }.
+
+{ ?x bnf:TEST ?y .
+  ?y log:rawType log:Literal. }  => { ?x bnf:canStartWithLiteral ?y }.
+
+{ ?x bnf:canStartWithLiteral ?y .
+#  (?y "(.).*")  string:scrape ?c     # Use whole string
+ }  => { ?y bnf:canStartWith ?y }.
+
+#______________________________________________________________
+
+
+
+# Rules for determining branching
+
+# A branch has  a sequence, which is the given BNF production, and
+# one or more conditions, which are the strings on which to consider
+# that branch.  N3 is a langauge in whch the look-ahead often is only
+# one character, and may allways be a constsnt string rather than a 
+# regexp (check).
+
+# A branchTail is a sequnece which a branch could start with
+{ 	?x bnf:branch ?b.
+	?b bnf:sequence ?s.
+} => {
+	?b bnf:branchTail ?s.
+}.
+
+{	?b bnf:branchTail ?s.
+	?s rdf:first [ bnf:branch [ bnf:sequence () ]];
+           rdf:rest ?t
+} => {
+	?b bnf:branchTail ?t.
+}. 
+
+
+{ 	?x bnf:branch ?b.
+	?b bnf:branchTail ?s.
+	?s rdf:first [bnf:canStartWith ?y].
+} => {
+	?x bnf:canStartWith ?y.
+	?b bnf:condition ?y.
+}.
+
+
+
+{ 	?x bnf:branch ?b;
+	    bnf:canPrecede ?z.
+	?z log:rawType log:Literal.
+	?b bnf:sequence ().
+} =>  { ?b bnf:condition ?z}.
+
+{ 	?x bnf:branch ?b;
+	    bnf:canPrecede [bnf:canStartWith ?z].
+	?b bnf:sequence ().
+} =>  { ?b bnf:condition ?z}.
+
+
+
+
+#ends

data/lib/rdf/n3/reader/n3.n3

@@ -0,0 +1,261 @@
+# 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
+# - 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/text/n3>].
+
+
+# __________________________________________________________________
+#
+# 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 (
+		( 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(
+		( "@base" explicituri  )
+		( "@prefix" prefix explicituri  )
+		( "@keywords" [ cfg:commaSeparatedListOf barename ] )
+	).
+
+
+simpleStatement cfg:mustBeOneSequence(( subject propertylist )).
+
+propertylist cfg:mustBeOneSequence (
+		( )
+		( predicate  object objecttail propertylisttail )
+	).
+
+propertylisttail cfg:mustBeOneSequence (
+		( )
+		( ";" propertylist )
+	).
+
+
+objecttail cfg:mustBeOneSequence (
+		( )
+		( ","   object objecttail )
+	).
+
+
+predicate cfg:mustBeOneSequence (
+		( expression )
+		( "@has" expression )
+		( "@is" expression "@of" )
+		( "@a" )
+		( "=" )
+		( "=>" )
+		( "<=" ) 
+	).
+
+subject cfg:mustBeOneSequence ((expression)).
+
+object cfg:mustBeOneSequence ((expression)).
+
+expression cfg:mustBeOneSequence(
+		( pathitem pathtail )
+	).
+
+pathtail cfg:mustBeOneSequence(
+		(  )
+		( "!" expression )
+		( "^" expression )
+	).
+
+
+pathitem cfg:mustBeOneSequence (
+		( symbol )
+		( "{" formulacontent "}" )
+		( quickvariable )
+		( numericliteral )
+		( literal )
+		( "[" propertylist "]"  )
+		(  "("  pathlist ")"  )
+                ( boolean )
+#		( "@this" )	  #  Deprocated.  Was allowed for this log:forAll x
+).
+
+
+boolean cfg:mustBeOneSequence (
+                ( "@true" )
+                ( "@false" )
+) .
+
+pathlist cfg:mustBeOneSequence (() (expression pathlist)).
+
+symbol cfg:mustBeOneSequence (
+		(explicituri)
+		(qname)
+	).
+
+
+numericliteral cfg:mustBeOneSequence (
+          ( integer )
+          ( rational )
+          ( double )
+          ( decimal )
+) .
+
+rational cfg:mustBeOneSequence (( integer "/" unsignedint)).
+
+
+literal cfg:mustBeOneSequence(( string dtlang)).
+
+dtlang cfg:mustBeOneSequence(  ()  ("@" langcode)  ("^^" symbol)).
+
+
+#______________________________________________________________________
+#
+#   TERMINALS
+#
+# "canStartWith" actually gives "a" for the whole class of alpha characters
+# and "0" for any of the digits 0-9.  This is used to build the branching
+# tables.
+# 
+integer 	cfg:matches	"""[-+]?[0-9]+""";
+		cfg:canStartWith 	"0", "-", "+".
+unsignedint 	cfg:matches	"""[0-9]+""";
+		cfg:canStartWith 	"0".
+double	cfg:matches	"""[-+]?[0-9]+(\\.[0-9]+)?([eE][-+]?[0-9]+)""";
+		cfg:canStartWith 	"0", "-", "+".
+decimal	cfg:matches	"""[-+]?[0-9]+\\.[0-9]*""";
+		cfg:canStartWith 	"0", "-", "+".
+
+#numericliteral	cfg:matches	"""[-+]?[0-9]+(\\.[0-9]+)?(e[-+]?[0-9]+)?""";
+#		cfg:canStartWith 	"0", "-", "+".
+
+explicituri 	cfg:matches 	"<[^>]*>";
+		cfg:canStartWith 	"<".
+
+prefix 		cfg:matches  	"([A-Z_a-z\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u02ff\u0370-\u037d\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff][\\-0-9A-Z_a-z\u00b7\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u037d\u037f-\u1fff\u200c-\u200d\u203f-\u2040\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff]*)?:";
+		cfg:canStartWith 	"a", "_", ":".  # @@ etc unicode
+
+qname 		cfg:matches  	"(([A-Z_a-z\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u02ff\u0370-\u037d\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff][\\-0-9A-Z_a-z\u00b7\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u037d\u037f-\u1fff\u200c-\u200d\u203f-\u2040\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff]*)?:)?[A-Z_a-z\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u02ff\u0370-\u037d\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff][\\-0-9A-Z_a-z\u00b7\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u037d\u037f-\u1fff\u200c-\u200d\u203f-\u2040\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff]*";
+		cfg:canStartWith 	"a", "_", ":".  # @@ etc unicode
+
+# ASCII version:
+#barename 	cfg:matches  	"[a-zA-Z_][a-zA-Z0-9_]*";  # subset of qname
+#		cfg:canStartWith 	"a", "_".  # @@ etc
+
+# This is the XML1.1
+barename 	cfg:matches  	"[A-Z_a-z\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u02ff\u0370-\u037d\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff][\\-0-9A-Z_a-z\u00b7\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u037d\u037f-\u1fff\u200c-\u200d\u203f-\u2040\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff]*";
+                cfg:canStartWith 	"a", "_".  # @@ etc .
+
+# as far as I can tell, the regexp should be
+# barename 	cfg:matches  	"[A-Z_a-z\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u02ff\u0370-\u037d\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff][\\-0-9A-Z_a-z\u00b7\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u037d\u037f-\u1fff\u200c-\u200d\u203f-\u2040\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff]*" .
+#
+
+quickvariable 	cfg:matches  	"\\?[A-Z_a-z\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u02ff\u0370-\u037d\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff][\\-0-9A-Z_a-z\u00b7\u00c0-\u00d6\u00d8-\u00f6\u00f8-\u037d\u037f-\u1fff\u200c-\u200d\u203f-\u2040\u2070-\u218f\u2c00-\u2fef\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\U00010000-\U000effff]*";  # ? 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.
+#
+# string3		cfg:matches		"\"\"\"[^\"\\\\]*(?:(?:\\\\.|\"(?!\"\"))[^\"\\\\]*)*\"\"\"".
+# string1		cfg:matches		"\"[^\"\\\\]*(?:\\\\.[^\"\\\\]*)*\"".
+
+string		cfg:matches		"(\"\"\"[^\"\\\\]*(?:(?:\\\\.|\"(?!\"\"))[^\"\\\\]*)*\"\"\")|(\"[^\"\\\\]*(?:\\\\.[^\"\\\\]*)*\")";
+		cfg:canStartWith 	"\"".
+
+
+#ends