rley 0.0.02

data/lib/rley/parser/parse_state.rb

@@ -0,0 +1,54 @@
+module Rley # This module is used as a namespace
+  module Parser # This module is used as a namespace
+
+    class ParseState
+      attr_reader(:dotted_rule)
+
+      # the position in the input that matches the beginning of the rhs
+      # of the production.
+      attr_reader(:origin)
+
+      def initialize(aDottedRule, theOrigin)
+        @dotted_rule = valid_dotted_rule(aDottedRule)
+        @origin = theOrigin
+      end
+
+      # Equality comparison. A parse state behaves as a value object.
+      def ==(other)
+        return true if self.object_id == other.object_id
+
+        if (dotted_rule == other.dotted_rule) && (origin == other.origin)
+          result = true
+        else
+          result = false
+        end
+
+        return result
+      end
+      
+      # Returns true if the dot is at the end of the rhs of the production.
+      # In other words, the complete rhs matches the input.
+      def complete?()
+        return dotted_rule.reduce_item?
+      end
+      
+      # Next expected symbol in the production
+      def next_symbol()
+        return dotted_rule.next_symbol
+      end
+      
+      private
+      
+      # Return the validated dotted item(rule)
+      def valid_dotted_rule(aDottedRule)
+        fail StandardError, 'Dotted item cannot be nil' if aDottedRule.nil?
+        
+        return aDottedRule
+      end
+
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/parser/parsing.rb

@@ -0,0 +1,101 @@
+require_relative 'chart'
+
+module Rley # This module is used as a namespace
+  module Parser # This module is used as a namespace
+
+    class Parsing
+      attr_reader(:chart)
+      
+      # The sequence of input token to parse
+      attr_reader(:tokens)
+
+      def initialize(startDottedRule, theTokens)
+        @tokens = theTokens.dup
+        @chart = Chart.new(startDottedRule, tokens.size)
+      end
+      
+      # Return true if the parse was successful (= input tokens
+      # followed the syntax specified by the grammar)
+      def success?()
+        # Success can be detected as follows:
+        # The last chart entry has a parse state
+        # that involves the start production and
+        # has a dot positioned at the end of its rhs.
+        
+        start_dotted_rule = chart.start_dotted_rule
+        start_production = start_dotted_rule.production
+        last_chart_entry = chart.state_sets.last
+        candidate_states = last_chart_entry.states_for(start_production)
+        found = candidate_states.find(&:complete?)
+        
+        return ! found.nil?
+      end
+
+
+      # Push a parse state (dotted item + origin) to the
+      # chart entry with given index if it isn't yet in the chart entry.
+      def push_state(aDottedItem, anOrigin, aChartIndex)
+        fail StandardError, 'Dotted item may not be nil' if aDottedItem.nil?
+        chart.push_state(aDottedItem, anOrigin, aChartIndex)
+      end
+
+
+      # This method is called when a parse state for chart entry at position 
+      # 'pos' expects a terminal as next symbol.
+      # If the input token matches the terminal symbol then:
+      # Retrieve all parse states for chart entry at 'aPosition'
+      # that have the given terminal as next symbol.
+      # For each s of the above states, push to chart entry aPosition + 1 
+      # a new state like: <next dotted rule, s.origin, aPosition + 1>
+      # In other words, we place the dotted rules in the next state set 
+      # such that the dot appears after terminal.
+      # @param Terminal [Terminal] a terminal symbol that 
+      #   immediately follows a dot 
+      # @param aPosition [Fixnum] position in the input token sequence.
+      # @param nextMapping [Proc or Lambda] code to evaluate in order to
+      #   determine the "next" dotted rule for a given one.
+      def scanning(aTerminal, aPosition, &nextMapping)
+        curr_token = tokens[aPosition]
+        
+        if curr_token.terminal == aTerminal
+          states = states_expecting(aTerminal, aPosition)
+          states.each do |s|
+            next_item = nextMapping.call(s.dotted_rule)
+            push_state(next_item, s.origin, aPosition + 1)
+          end
+        end
+      end
+
+
+
+      # This method is called when a parse state at chart entry reaches the end
+      # of a production.
+      # For every state in chart[aPosition] that is complete (i.e. of the form:
+      #   { dotted_rule: X -> γ •, origin: j}),
+      # Find states s in chart[j] of the form {dotted_rule: Y -> α • X β, origin: i}
+      #   In other words, rules that predicted the non-terminal X.
+      # For each s, add to chart[aPosition] a state of the form
+      #   { dotted_rule: Y → α X • β, origin: i})
+      def completion(aState, aPosition, &nextMapping)
+        curr_origin = aState.origin
+        curr_lhs = aState.dotted_rule.lhs
+        states = states_expecting(curr_lhs, curr_origin)
+        states.each do |s|
+          next_item = nextMapping.call(s.dotted_rule)
+          push_state(next_item, s.origin, aPosition)
+        end
+      end
+
+
+      # The list of ParseState from the chart entry at given position
+      # that expect the given terminal
+      def states_expecting(aTerminal, aPosition)
+        return chart[aPosition].states_expecting(aTerminal)
+      end
+
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/parser/state_set.rb

@@ -0,0 +1,47 @@
+require 'forwardable' # Delegation
+
+module Rley # This module is used as a namespace
+  module Parser # This module is used as a namespace
+  
+    class StateSet
+      extend Forwardable
+      def_delegators :states, :empty?, :size, :first, :each
+      
+      # The set of parse states
+      attr_reader(:states)
+      
+    
+      def initialize()
+        @states = []
+      end
+      
+      # Append the given state (if it isn't yet in the set) 
+      # to the list of states
+      # @param aState [ParseState] the state to push.
+      def push_state(aState)
+        @states << aState unless include?(aState)
+      end
+      
+      # The list of ParseState that expect the given terminal
+      def states_expecting(aTerminal)
+        return states.select { |s| s.dotted_rule.next_symbol == aTerminal }
+      end
+      
+      # The list of ParseState that involve the given production
+      def states_for(aProduction)
+        return states.select { |s| s.dotted_rule.production == aProduction }
+      end
+      
+      private
+      
+      def include?(aState)
+        # TODO: make it better than linear search
+        return states.include?(aState)
+      end
+
+    end # class
+  
+  end # module
+end # module
+
+# End of file

data/lib/rley/parser/token.rb

@@ -0,0 +1,21 @@
+require_relative '../syntax/grammar'
+require_relative 'dotted_item'
+
+module Rley # This module is used as a namespace
+  module Parser # This module is used as a namespace
+
+    class Token
+      attr_reader(:lexeme)
+      attr_reader(:terminal)
+
+      def initialize(theLexeme, aTerminal)
+        @lexeme = theLexeme
+        @terminal = aTerminal
+      end
+
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/grammar.rb

@@ -0,0 +1,59 @@
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+
+    # A grammar specifies the syntax of a language.
+    #   Formally, a grammar has:
+    #   One start symbol,
+    #   One or more other production rules,
+    #   Each production has a rhs that is a sequence of grammar symbols.
+    #   Grammar symbols are categorized into
+    #   -terminal symbols
+    #   -non-terminal symbols
+    class Grammar
+      # A non-terminal symbol that represents all the possible strings
+      # in the language.
+      attr_reader(:start_symbol)
+
+      # The list of production rules for the language.
+      attr_reader(:rules)
+
+      # The list of grammar symbols in the language.
+      attr_reader(:symbols)
+
+      # @param theProduction [Array of Production]
+      def initialize(theProductions)
+        @rules = []
+        @symbols = []
+        valid_productions = validate_productions(theProductions)
+        # TODO: use topological sorting
+        @start_symbol = valid_productions[0].lhs
+        valid_productions.each { |prod| add_production(prod) }
+      end
+
+      private
+
+      # Validation method. Return the validated list of productions
+      def validate_productions(theProductions)
+        msg = 'A grammar must have at least one production'
+        fail StandardError, msg if theProductions.nil? || theProductions.empty?
+        return theProductions
+      end
+
+      def add_production(aProduction)
+        @rules << aProduction
+        the_lhs = aProduction.lhs
+        @symbols << the_lhs unless @symbols.include? the_lhs
+
+        # TODO: remove quadratic execution time
+        aProduction.rhs.members.each do |symb|
+          next if symbols.include? symb
+          @symbols << symb
+        end
+      end
+
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/grm_symbol.rb

@@ -0,0 +1,18 @@
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+
+    # Abstract class for grammar symbols.
+    # A grammar symbol is an element that appears in grammar rules.
+    class GrmSymbol
+      # The name of the grammar symbol
+      attr_reader(:name)
+
+      def initialize(aName)
+        @name = aName.dup
+      end
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/literal.rb

@@ -0,0 +1,20 @@
+require_relative 'terminal' # Load superclass
+
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+
+    # A literal is terminal symbol that matches a lexical pattern
+    class Literal < Terminal
+      # The exact text representation of the word.
+      attr_reader(:pattern)
+
+      def initialize(aName, aPattern)
+        super(aName)
+        @pattern = aPattern
+      end
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/non_terminal.rb

@@ -0,0 +1,18 @@
+require_relative 'grm_symbol' # Load superclass
+
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+
+    # A non-terminal symbol (sometimes called a syntactic variable) represents
+    # a composition of terminal or non-terminal symbols
+    class NonTerminal < GrmSymbol
+
+      def initialize(aName)
+        super(aName)
+      end
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/production.rb

@@ -0,0 +1,42 @@
+require_relative 'symbol_seq'
+
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+
+    # In a context-free grammar, a production is a rule in which
+    # its left-hand side (LHS) consists solely of a non-terminal symbol
+    # and the right-hand side (RHS) consists of a sequence of symbols.
+    # The symbols in RHS can be either terminal or non-terminal symbols.
+    # The rule stipulates that the LHS is equivalent to the RHS,
+    # in other words every occurrence of the LHS can be substituted to
+    # corresponding RHS.
+    # Implementation note: the object id of the production is taken as its LHS.
+    class Production
+      # The right-hand side (rhs) consists of a sequence of grammar symbols
+      attr_reader(:rhs)
+
+      # The left-hand side of the rule. It must be a non-terminal symbol
+      attr_reader(:lhs)
+
+      # Provide common alternate names to lhs and rhs accessors
+
+      alias :body :rhs
+      alias :head :lhs
+
+      def initialize(aNonTerminal, theSymbols)
+        @lhs = aNonTerminal
+        @rhs = SymbolSeq.new(theSymbols)
+      end
+
+      # Is the rhs empty?
+      # @ return true if the rhs has no members.
+      def empty?()
+        return rhs.empty?
+      end
+
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/symbol_seq.rb

@@ -0,0 +1,36 @@
+require 'forwardable'
+
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+
+    # A symbol sequence is a suite of grammar symbols
+    class SymbolSeq
+      extend Forwardable
+      def_delegators :@members, :empty?, :size, :[]
+
+      # The sequence of symbols
+      attr_reader(:members)
+
+      def initialize(theSymbols)
+        @members = theSymbols.dup
+      end
+
+      # Equality operator.
+      def ==(other)
+        return true if other.object_id == self.object_id
+
+        case other
+        when SymbolSeq then result = other.members == self.members
+        when Array then result = other == self.members
+        else
+          fail StandardError, "Cannot compare a SymbolSeq with a #{other.class}"
+        end
+
+        return result
+      end
+    end # class
+
+  end # module
+end # module
+
+# End of file

data/lib/rley/syntax/terminal.rb

@@ -0,0 +1,18 @@
+require_relative 'grm_symbol' # Load superclass
+
+module Rley # This module is used as a namespace
+  module Syntax # This module is used as a namespace
+  
+    # A terminal symbol represents a class of words in the language 
+    # defined the grammar.
+    class Terminal < GrmSymbol
+      
+      def initialize(aName)
+        super(aName)
+      end
+    end # class
+  
+  end # module
+end # module
+
+# End of file