activefacts-cql 1.7.1

data/lib/activefacts/cql/parser/Language/Mandarin.treetop

@@ -0,0 +1,304 @@
+# Encoding: UTF-8
+#
+#       ActiveFacts CQL Parser for Mandarin
+#
+# Copyright (c) 2009 Clifford Heath. Read the LICENSE file.
+#
+module ActiveFacts
+  module CQL
+    grammar Mandarin
+
+      # >>>>>>>>>>>>>>>>>>>> Object Types <<<<<<<<<<<<<<<<<<<<
+      # The pattern to introduce a Value Type
+      rule written_as
+        s est s 'écrit' S 'en' s
+      end
+
+      # The pattern to introduce an Entity Type
+      rule identified_by
+	'identifié' s 'par' s
+      end
+
+      rule basetype_expression
+        is s i:( independent s )? identification
+      end
+
+      # The pattern to introduce an Entity Subtype
+      rule subtype_prefix
+        est s un s ('type'/'genre') s 'de' S
+      end
+
+      rule subtype_expression
+        subtype_prefix i:( independent s )? supertype_list ident:identification?
+      end
+
+      # The pattern to introduce an objectified fact type with implicit identification
+      rule is_where
+	existe s i:(independent s)? si
+	{ def independent; !i.empty?; end }
+      end
+
+      rule in_which	# Introduce an objectification step
+	where /	        # Old syntax
+	dans s laquel   # preferred syntax
+      end
+
+      # Units conversion keyword
+      rule conversion
+	converts s a:(approximately s)? to s
+	{
+	  def approximate?
+	    !a.empty?
+	  end
+	}
+      end
+
+      # >>>>>>>>>>>>>>>>>>>> Constraints <<<<<<<<<<<<<<<<<<<<
+      # External presence constraint syntax:
+      rule each_occurs_in_clauses
+	s chaque s ('combinaison' S)? role_list s existe s quantifier 'fois' S
+	  'parmis' S ce S 'qui' S 'suit' s ':' s
+	  clauses_list s c:context_note? ';'
+	{
+	  def role_list_ast
+	    role_list.ast
+	  end
+	  def quantifier_ast
+	    quantifier.ast
+	  end
+	  def clauses_ast
+	    clauses_list.ast
+	  end
+	}
+      end
+
+      rule either_or
+      s soit s r1:clauses s 'ou' S r2:clauses
+	s c:context_note? enforcement ';'
+      {
+	  def role_list_ast
+	    nil
+	  end
+	  def quantifier_ast
+	    Compiler::Quantifier.new(1, nil)
+	  end
+	  def clauses_ast
+	    [r1.ast, r2.ast]
+	  end
+	}
+      end
+
+      # Used for exclusion (at most one) and mandatory exclusion (exactly one)
+      rule for_each_how_many
+	'pour' S 'chaque' S role_list s
+	  quantifier s 'des' S 'suivants' S 's\'applique' s ':' s
+	  clauses_list s c:context_note? ';'
+	{
+	  def role_list_ast
+	    role_list.ast
+	  end
+	  def quantifier_ast
+	    quantifier.ast
+	  end
+	  def clauses_ast
+	    clauses_list.ast
+	  end
+	}
+      end
+
+      rule either_or_not_both
+	s soit s r1:clauses s 'ou' S r2:clauses 'mais' S 'pas' S 'les' S 'deux'
+	  s c:context_note? enforcement ';'
+	{
+	  def role_list_ast
+	    nil
+	  end
+	  def quantifier_ast
+	    Compiler::Quantifier.new(1, 1)
+	  end
+	  def clauses_ast
+	    [r1.ast, r2.ast]
+	  end
+	}
+      end
+
+      rule a_only_if_b
+	s clauses s only_if r2:clauses s c:context_note? enforcement ';'
+      end
+
+      rule only_if
+	que s si s
+      end
+
+      rule if_b_then_a
+	s 'si' S clauses s 'puis' S r2:clauses s c:context_note? enforcement ';'
+      end
+
+      rule if_and_only_if
+        s clauses s tail:( 'si' S 'et' S ('seulement'/'uniquement'/'que') S 'si' S clauses s)+
+	  c:context_note? enforcement ';'
+      end
+
+      # During the prescan we need to know where terms in a role list finish.
+      # This rule matches any non-term expressions that may follow a role list.
+      rule role_list_constraint_followers
+	existe
+      end
+
+      # >>>>>>>>>>>>>>>>>>>> Quantifiers <<<<<<<<<<<<<<<<<<<<
+      rule quantifier
+	(
+	  ( chaque s					# "each"
+	    { def value; [1, nil]; end }
+	  / un s 'certain' 'e'? ) s			# "some"
+	    { def value; nil; end }
+	  / ce s					# "that"
+	    { def value; nil; end }
+	  / un s					# "one"; masculine or feminine
+	    { def value; [1, 1]; end }
+	  / 'pas' s 'de' s				# REVISIT: Example: "Personne n'a pas de Casier-judiciaire"
+	    { def value; [0, 0]; end }
+	  / exactement s quantity
+	    { def value; q = quantity.value; [q, q]; end }
+	  / au s moins s quantity most:( et s au s plus q:quantity )? # At least (or at most)
+	    { def value; [ quantity.value, most.empty? ? nil : most.q.value ]; end }
+	  / au s plus s quantity			# At most
+	    { def value; [ nil, quantity.value ]; end }
+	  / entre s numeric_range s			# within numeric_range
+	    { def value; numeric_range.value; end }
+	  / soit s tous s soit s aucun			# either all or none
+	    { def value; [ -1, 1 ]; end }
+	)
+	{
+	  def ast
+	    v = value
+	    Compiler::Quantifier.new(v[0], v[1])
+	  end
+	}
+      end
+
+      rule quantity
+        un s                     { def value; 1; end }
+        / number s                { def value; number.value; end }
+      end
+
+      # >>>>>>>>>>>>>>>>>>>> Context Notes <<<<<<<<<<<<<<<<<<<<
+      rule as_agreed_by
+	s 'comme' S 'convenu' S d:('le' s date s)? 'par' S agents
+        { def value; [ d.empty? ? nil : d.date.value, agents.value ]; end }
+      end
+
+      rule date
+        s d:(!(by/')') .)+
+        { def value; d.text_value.strip; end }
+      end
+
+      rule agents
+        s h:agent s t:(',' s !context_type agent s)*
+        {
+	  def value; [h.text_value] + t.elements.map{|e| e.agent.text_value }; end
+	  def node_type; :linking; end
+	}
+      end
+
+      rule agent
+        id (s id)*
+      end
+
+      rule negative_prefix
+	s 'ce' S "n'est" s 'pas' s 'le' s 'cas' s 'que'
+      end
+
+      rule agg_of de end 
+      rule agg_in dans end 
+      rule restricted_to 'limité' s 'à' !alphanumeric s end
+      rule any un !alphanumeric end
+
+      # >>>>>>>>>>>>>>>>>>>> Internal vocabulary <<<<<<<<<<<<<<<<<<<<
+      rule all 'tous' !alphanumeric end
+      rule at 'a' !alphanumeric end
+      rule both 'les deux' !alphanumeric end 
+      rule ce ('cette'/'cet'/'ce') !alphanumeric end	# ce is masculine, cette feminine, cet is ce before a noun
+      rule chaque 'chaque' !alphanumeric end
+      rule converts 'convertit' !alphanumeric end 
+      rule de 'de' !alphanumeric end
+      rule dans 'dans' !alphanumeric end
+      rule laquel 'laquel' 'le'? !alphanumeric end
+      rule descending 'descendant' !alphanumeric end
+      rule each chaque end
+      rule either 'soit' !alphanumeric end 
+      rule entity 'entité' !alphanumeric end 
+      rule est 'est' !alphanumeric end 
+      rule exactement 'exactement' !alphanumeric end
+      rule existe 'existe' !alphanumeric end
+      rule from 'à partir de' !alphanumeric end		# REVISIT: google translate
+      rule includes 'comprend' !alphanumeric end
+      rule matches 'matches' !alphanumeric end
+      rule moins 'moins' !alphanumeric end
+      rule none 'aucun' !alphanumeric end
+      rule not 'not' !alphanumeric end			# REVISIT: Used in constraints
+      rule plus 'plus' !alphanumeric end
+      rule some 'quelques' !alphanumeric end
+      rule un 'un' 'e'? !alphanumeric end	# un is masculine, une is feminine
+      rule que 'que' !alphanumeric end
+
+      # >>>>>>>>>>>>>>>>>>>> External vocabulary <<<<<<<<<<<<<<<<<<<<
+      rule according_to 'selon' !alphanumeric end 
+      rule acyclic 'acyclique' !alphanumeric end
+      rule alias 'alias' !alphanumeric end 
+      rule and 'et' !alphanumeric end
+      rule antisymmetric 'antisymmetric' !alphanumeric end
+      rule approximately 'environ' !alphanumeric end		# REVISIT: google translate
+      rule as 'comme' !alphanumeric end
+      rule as_opposed_to 'au' S 'lieu' S 'de' !alphanumeric end
+      rule asymmetric 'asymétrique' !alphanumeric end
+      rule au 'au' !alphanumeric end
+      rule because ('parce' s 'que' / 'car') !alphanumeric end
+      rule but 'mais' !alphanumeric end 
+      rule by 'par' !alphanumeric end 
+      rule definitely 'définitivement' !alphanumeric end
+      rule entre 'entre' !alphanumeric end
+      rule ephemera 'éphémère' !alphanumeric end 
+      rule existe 'existe' !alphanumeric end
+      rule false 'faux' !alphanumeric end 
+      rule feminine 'féminin' !alphanumeric end 
+      rule identified ('identifié') !alphanumeric end 
+      rule if 'si' !alphanumeric end
+      rule import 'import' !alphanumeric end			# REVISIT: translation?
+      rule independent 'indépendant' !alphanumeric end
+      rule stronglyintransitive 'stronglyintransitif' !alphanumeric end
+      rule intransitive 'intransitif' !alphanumeric end
+      rule irreflexive 'irréflexive' !alphanumeric end
+      rule is est end			# Called from ObjectTypes.treetop in "is identified by"
+      rule its ('sa' / 'son') !alphanumeric end
+      rule masculine 'masculin' !alphanumeric end 
+      rule maybe 'peut-être' !alphanumeric end			# REVISIT: eventuellement = possibly?
+      rule only 'que' !alphanumeric end			# REVISIT: Used in constraints
+      rule or 'ou' !alphanumeric end 
+      rule ordering_prefix by s (ascending/descending)? s end
+      rule otherwise 'sinon' !alphanumeric end
+      rule partitioned 'partitionné' !alphanumeric end 
+      rule personal 'personelle' !alphanumeric end 
+      rule radix_point ',' end
+      rule reflexive 'réflexive' !alphanumeric end
+      rule returning 'retour' !alphanumeric end
+      rule separate 'distincte' !alphanumeric end 
+      rule si 'si' !alphanumeric end
+      rule so_that 'pour' s 'que' !alphanumeric end
+      rule soit 'soit' !alphanumeric end
+      rule static 'statique' !alphanumeric end
+      rule symmetric 'symétrique' !alphanumeric end
+      rule that que end
+      rule then 'puis' !alphanumeric end
+      rule to 'à' !alphanumeric end
+      rule to_avoid ('pour' s 'empecher' s 'de' / 'pour' s 'ne' s 'pas') !alphanumeric end
+      rule transient 'transitoires' !alphanumeric end
+      rule transitive 'transitif' !alphanumeric end
+      rule true 'vrai' !alphanumeric end 
+      rule vocabulary 'vocabulaire' !alphanumeric end 
+      rule where 'où' !alphanumeric end 
+      rule who 'qui' !alphanumeric end 
+
+    end
+  end
+end

data/lib/activefacts/cql/parser/LexicalRules.treetop

@@ -0,0 +1,253 @@
+#
+#       ActiveFacts CQL Parser.
+#       Various lexical rules for CQL.
+#
+# Copyright (c) 2009 Clifford Heath. Read the LICENSE file.
+#
+module ActiveFacts
+  module CQL
+    grammar LexicalRules
+
+      rule range
+        (numeric_range / string_range)
+	{
+	  def node_type; :literal; end
+	}
+      end
+
+      rule numeric_range
+        number s tail:( '..' s end:number? s )?
+        {
+          def value
+            if !tail.empty?
+              last = tail.end.value unless tail.end.empty?
+              [ number.value, last ]
+            else
+              number.value
+            end
+          end
+        }
+        / '..' s number s
+        {
+          def value
+            [ nil, number.value ]
+          end
+        }
+      end
+
+      rule string_range
+        string s tail:( '..' s end:string? s )?
+        {
+          # Ranges require the original text of the string, not the content:
+          def value
+            first = string.text_value
+            if !tail.empty?
+              last = tail.end.text_value unless tail.end.empty?
+              [ first, last ]
+            else
+              first
+            end
+          end
+        }
+        / '..' s string s
+        {
+          def value
+            [ nil, string.value ]
+          end
+        }
+      end
+
+      rule url
+	# url_scheme ':' (user ( ':' !(port '/') password )? '@' )? hostname ( ':' port )? '/' path query? fragment?
+	( !(white / ';') .)+
+	{
+	  def node_type; :literal; end
+	}
+      end
+
+      rule literal
+        ( boolean_literal
+        / string
+        / number
+        ) s
+        {
+          def value
+            elements[0].value
+          end
+	  def node_type; :literal; end
+        }
+      end
+
+      rule boolean_literal
+        ( true { def value; true; end }
+        / false { def value; false; end }
+        ) !alphanumeric
+        {
+          def value; elements[0].value end
+        }
+      end
+
+      rule string
+        "'" (string_char)* "'"
+        {
+          def value
+            text_value
+            eval(text_value.sub(/\A'(.*)'\Z/,'"\1"'))
+          end
+        }
+      end
+
+      rule number
+        ( real /
+          fractional_real /
+          hexnumber /
+          octalnumber
+        ) !alphanumeric
+        {
+          def value
+            eval(text_value)
+          end
+	  def node_type; :literal; end
+        }
+      end
+
+      # All purely lexical rules from here down, no-one looks at the structure, just the text_value:
+
+      rule string_char
+        ( '\\' [befntr\\']
+        / '\\' [0-7] [0-7] [0-7]
+	/ '\\' [\r]* [\n] [\r]*
+        / '\\0'
+        / '\\x' [0-9A-Fa-f] [0-9A-Fa-f]
+        / '\\u' [0-9A-Fa-f] [0-9A-Fa-f] [0-9A-Fa-f] [0-9A-Fa-f]
+        / (![\'\\\0-\x07\x0A-\x1F] .)
+        )
+      end
+
+      rule real
+        [-+]? [1-9] [0-9]* fraction? exponent?
+      end
+
+      rule fractional_real
+        [-+]? '0' fraction exponent?
+      end
+
+      rule fraction
+        radix_point [0-9]+
+      end
+
+      rule exponent
+        ( [Ee] [-+]? [0-9]+ )
+      end
+
+      rule hexnumber
+        '0x' [0-9A-Fa-f]+
+      end
+
+      rule octalnumber
+        '0' [0-7]*
+      end
+
+      rule mul_op
+        '/' / '%' / '*'
+      end
+
+      rule id
+        alpha alphanumeric*
+        { def value; text_value; end }
+      end
+
+      rule alpha
+	[[:alpha:]_]
+      end
+
+      rule alphanumeric
+        alpha / [0-9]
+      end
+
+      rule s    # Optional space
+        S?
+      end
+
+      rule S    # Mandatory space
+        (white / comment_to_eol / comment_c_style)+
+      end
+
+      rule white
+        [ \t\n\r]+
+      end
+
+      rule comment_to_eol
+        '//' (!"\n" .)*
+	{
+	  def node_type; :comment; end
+	}
+      end
+
+      rule comment_c_style
+        '/*' (!'*/' . )* '*/'
+	{
+	  def node_type; :comment; end
+	}
+      end
+
+      rule regular_expression
+	'/' !'/' regular_expression_contents '/'
+	{
+	  def contents
+	    regular_expression_contents.text_value
+	  end
+	}
+      end
+
+      rule regular_expression_contents
+	regular_expression_alternate ( '|' regular_expression_alternate )*
+      end
+
+      rule regular_expression_alternate
+	regular_expression_sequence
+      end
+
+      rule regular_expression_sequence
+	regular_expression_atom*
+      end
+
+      rule regular_expression_atom
+	(
+	  '[' character_classes ']'
+	  / regular_expression_group
+	  / ![*+?()|/] string_char
+	) regular_expression_multiplicity?
+      end
+
+      rule character_classes
+	character_class+
+      end
+
+      rule character_class
+	!']' string_char '-' !']' string_char
+	/ '-'
+	/ !']' string_char
+      end
+
+      rule regular_expression_multiplicity
+	'*' / '+' / '?'
+      end
+
+      rule regular_expression_group
+	'('
+	  regular_expression_group_extension?
+	  regular_expression_contents
+	')'
+      end
+
+      rule regular_expression_group_extension
+	'?' (
+	  '<' ( !'>' .)+ '>'	# A tag for a regular expression group
+	# REVISIT: Add more group extensions as needed
+	)
+      end
+
+    end
+  end
+end