antelope 0.1.1 → 0.1.2

data/examples/simple.output

@@ -6,30 +6,6 @@ Productions:
   r → l 
   $start → e $ 
 
-Productions, Again:
-  0/n0: $start(0) → e(0:1) $
-  1/n1: e(0:1) → l(0:2) "=" r(8:10)
-  2/n1: e(0:1) → r(0:3)
-  3/n1: l(0:2) → IDENT
-  4/n1: l(0:2) → "*" r(5:9)
-  5/n1: r(0:3) → l(0:2)
-  12/n1: r(5:9) → l(5:2)
-  13/n1: l(5:2) → IDENT
-  14/n1: l(5:2) → "*" r(5:9)
-  18/n1: r(8:10) → l(8:2)
-  19/n1: l(8:2) → IDENT
-  20/n1: l(8:2) → "*" r(5:9)
-
-FOLLOW:
-  $start(0): {}
-  e(0:1): {$}
-  l(0:2): {"=", $}
-  r(0:3): {$}
-  r(5:9): {"=", $}
-  l(5:2): {}
-  l(8:2): {$}
-  r(8:10): {$}
-
 Precedence:
   --- highest
   nonassoc 1:
@@ -38,44 +14,7 @@ Precedence:
     {$}
   --- lowest
 
-Table:
   State 0:
-    e     : state (1)
-    l     : state (2)
-    r     : state (3)
-    IDENT : state (4)
-    STAR  : state (5)
-  State 1:
-    $     : state (7)
-  State 2:
-    EQUALS: state (8)
-    $     : reduce (5)
-  State 3:
-    $     : reduce (2)
-  State 4:
-    $     : reduce (3)
-  State 5:
-    r     : state (9)
-    l     : state (2)
-    IDENT : state (4)
-    STAR  : state (5)
-  State 6:
-    $     : reduce (5)
-  State 7:
-    $     : accept (0)
-  State 8:
-    r     : state (10)
-    l     : state (2)
-    IDENT : state (4)
-    STAR  : state (5)
-  State 9:
-    $     : reduce (4)
-  State 10:
-    $     : reduce (1)
-  
-
-State 0:
-  rules:
     0/n0: $start(0) →  • e(0:1) $
       {}
     1/n1: e(0:1) →  • l(0:2) "=" r(8:10)
@@ -88,105 +27,92 @@ State 0:
       {}
     5/n1: r(0:3) →  • l(0:2)
       {}
+    transitions:
+      e: State 1
+      l: State 2
+      r: State 3
+      IDENT: State 4
+      STAR: State 5
 
-  transitions:
-    e    : State 1
-    l    : State 2
-    r    : State 3
-    IDENT: State 4
-    STAR : State 5
-
-State 1:
-  rules:
+  State 1:
     6/n0: $start → e • $
       {}
+    transitions:
+      $: State 7
 
-  transitions:
-    $: State 7
-
-State 2:
-  rules:
+  State 2:
     7/n1: e → l • "=" r
       {}
     15/n1: r → l • 
       {$}
+    transitions:
+      EQUALS: State 8
+    reductions:
+      $: Rule 5
 
-  transitions:
-    EQUALS: State 8
-
-State 3:
-  rules:
+  State 3:
     9/n1: e → r • 
       {$}
+    reductions:
+      $: Rule 2
 
-  transitions:
-
-State 4:
-  rules:
+  State 4:
     10/n1: l → IDENT • 
       {$}
+    reductions:
+      $: Rule 3
 
-  transitions:
-
-State 5:
-  rules:
+  State 5:
     11/n1: l → "*" • r
       {}
-    12/n1: r(5:9) →  • l(5:2)
+    12/n1: r(5:9) →  • l(5:6)
       {}
-    13/n1: l(5:2) →  • IDENT
+    13/n1: l(5:6) →  • IDENT
       {}
-    14/n1: l(5:2) →  • "*" r(5:9)
+    14/n1: l(5:6) →  • "*" r(5:9)
       {}
+    transitions:
+      r: State 9
+      l: State 6
+      IDENT: State 4
+      STAR: State 5
 
-  transitions:
-    r    : State 9
-    l    : State 2
-    IDENT: State 4
-    STAR : State 5
-
-State 6:
-  rules:
+  State 6:
     15/n1: r → l • 
       {$}
+    reductions:
+      $: Rule 5
 
-  transitions:
-
-State 7:
-  rules:
+  State 7:
     16/n0: $start → e $ • 
       {}
+    accepting:
+      $: Rule 0
 
-  transitions:
-
-State 8:
-  rules:
+  State 8:
     17/n1: e → l "=" • r
       {}
-    18/n1: r(8:10) →  • l(8:2)
+    18/n1: r(8:10) →  • l(8:6)
       {}
-    19/n1: l(8:2) →  • IDENT
+    19/n1: l(8:6) →  • IDENT
       {}
-    20/n1: l(8:2) →  • "*" r(5:9)
+    20/n1: l(8:6) →  • "*" r(5:9)
       {}
+    transitions:
+      r: State 10
+      l: State 6
+      IDENT: State 4
+      STAR: State 5
 
-  transitions:
-    r    : State 10
-    l    : State 2
-    IDENT: State 4
-    STAR : State 5
-
-State 9:
-  rules:
+  State 9:
     21/n1: l → "*" r • 
       {$}
+    reductions:
+      $: Rule 4
 
-  transitions:
-
-State 10:
-  rules:
+  State 10:
     22/n1: e → l "=" r • 
       {$}
-
-  transitions:
+    reductions:
+      $: Rule 1
 

data/lib/antelope/ace/compiler.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "rubygems/requirement"
 
 module Antelope
@@ -93,6 +95,13 @@ module Antelope
         @options  = { :terminals => [], :prec => [], :extra => {} }
       end
 
+      # Pretty inspect.
+      #
+      # @return [String]
+      def inspect
+        "#<#{self.class} state=#{@state.inspect} options=#{@options.inspect}>"
+      end
+
       # Runs the compiler on the input tokens.  For each token,
       # it calls `compile_<type>` with `<type>` being the first
       # element of the token, with the remaining part of the array
@@ -279,7 +288,7 @@ module Antelope
 
         unless required_version =~ antelope_version
           raise IncompatibleVersionError,
-            "Grammar requires #{args[0]}, " \
+            "Grammar requires #{required}, " \
             "have #{Antelope::VERSION}"
         end
       end

data/lib/antelope/ace/errors.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
 
@@ -7,6 +9,12 @@ module Antelope
     class Error < Antelope::Error
     end
 
+    # Used primarily in the {Compiler}, this is raised when the
+    # version requirement of the Ace file doesn't match the running
+    # version of Ace.
+    class IncompatibleVersionError < Error
+    end
+
     # Used primarily in the {Scanner}, this is raised when an input
     # is malformed.  The message should contain a snippet of the input
     # which caused the error.

data/lib/antelope/ace/grammar/generation.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Grammar
@@ -34,7 +36,10 @@ module Antelope
                      modifiers  = DEFAULT_MODIFIERS)
           mods = modifiers.map(&:last).
             map  { |x| x.new(self) }
-          mods.map(&:call)
+          mods.each do |mod|
+            puts "Running mod #{mod.class}..."
+            mod.call
+          end
           hash = Hash[modifiers.map(&:first).zip(mods)]
           # This is when we'd generate
 

data/lib/antelope/ace/grammar/loading.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Grammar

data/lib/antelope/ace/grammar/precedences.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "set"
 
 module Antelope
@@ -28,7 +30,7 @@ module Antelope
           set = Set.new([token, :_])
 
           precedence.
-            select { |pr| set.intersect?(pr.tokens) }.
+            select { |pr| set.intersection(pr.tokens).any? }.
             first
         end
 

data/lib/antelope/ace/grammar/production.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Grammar

data/lib/antelope/ace/grammar/productions.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Grammar

data/lib/antelope/ace/grammar/terminals.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Grammar

data/lib/antelope/ace/grammar.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "antelope/ace/grammar/terminals"
 require "antelope/ace/grammar/productions"
 require "antelope/ace/grammar/precedences"

data/lib/antelope/ace/precedence.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
 

data/lib/antelope/ace/scanner/first.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Scanner

data/lib/antelope/ace/scanner/second.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Scanner
@@ -86,7 +88,7 @@ module Antelope
         #
         # @return [Boolean] if it matched.
         def scan_second_rule_part
-          if @scanner.scan(/([A-Za-z._-]+)(?!\:)/)
+          if @scanner.scan(/([A-Za-z._-]+)(?!\:|[A-Za-z._-])/)
             tokens << [:part, @scanner[1]]
           end
         end

data/lib/antelope/ace/scanner/third.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Scanner

data/lib/antelope/ace/scanner.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "strscan"
 require "antelope/ace/scanner/first"
 require "antelope/ace/scanner/second"

data/lib/antelope/ace/token/epsilon.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Token
@@ -7,9 +9,9 @@ module Antelope
       # reduce to nothing.
       class Epsilon < Token
         # Initialize.  Technically takes no arguments.  Sets
-        # the name of the token to be `:epsilon`.
+        # the name of the token to be `:$empty`.
         def initialize(*)
-          super :epsilon
+          super :"$empty"
         end
 
         # (see Token#epsilon?)

data/lib/antelope/ace/token/error.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Token

data/lib/antelope/ace/token/nonterminal.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Token

data/lib/antelope/ace/token/terminal.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Ace
     class Token

data/lib/antelope/ace/token.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "antelope/ace/token/nonterminal"
 require "antelope/ace/token/terminal"
 require "antelope/ace/token/epsilon"

data/lib/antelope/ace.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "antelope/ace/errors"
 require "antelope/ace/scanner"
 require "antelope/ace/compiler"

data/lib/antelope/cli.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "thor"
 
 module Antelope

data/lib/antelope/errors.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
 
   # Every error in antelope inherits this error class.

data/lib/antelope/generation/constructor/first.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Generation
     class Constructor

data/lib/antelope/generation/constructor/follow.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Generation
     class Constructor
@@ -71,6 +73,9 @@ module Antelope
           # This is going to be the output set.
           set = Set.new
 
+          productions = grammar.states.map(&:rules).flatten.
+            inject(Set.new, :merge)
+
           productions.each do |rule|
             items = rule.right
 

data/lib/antelope/generation/constructor/nullable.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Generation
     class Constructor

data/lib/antelope/generation/constructor.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "set"
 require "antelope/generation/constructor/nullable"
 require "antelope/generation/constructor/first"

data/lib/antelope/generation/errors.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Generation
 

data/lib/antelope/generation/recognizer/rule.rb

@@ -136,6 +136,15 @@ module Antelope
           !succ?
         end
 
+        # The complete opposite of {#final?} - it checks to see if
+        # this is the first rule, as in no rule can exist before this
+        # one; i.e. the position is zero.
+        #
+        # @return [Boolean]
+        def start?
+          position.zero?
+        end
+
         # Compares this rule to another object.  If the other object
         # is not a rule, it delegates the comparison.  Otherwise, it
         # converts both this and the other rule into arrays and

data/lib/antelope/generation/recognizer/state.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "forwardable"
 require "securerandom"
 

data/lib/antelope/generation/recognizer.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "antelope/generation/recognizer/rule"
 require "antelope/generation/recognizer/state"
 
@@ -31,6 +33,7 @@ module Antelope
       def initialize(grammar)
         @grammar = grammar
         @states = Set.new
+        @map = {}
       end
 
       # Runs the recognizer.  After all states have been created, it
@@ -94,6 +97,7 @@ module Antelope
                 ns << succ
                 compute_closure(ns)
                 states << ns
+                @map[succ] = ns
                 ns
               end
 
@@ -129,8 +133,9 @@ module Antelope
       # @param rule [Rule]
       # @yield [rule]
       # @return [State]
-      def find_state_for(rule)
-        states.find { |state| state.include?(rule) } or yield(rule)
+      def find_state_for(rule, &block)
+        #states.find { |state| state.include?(rule) } or yield(rule)
+        @map.fetch(rule) { block.call(rule) }
       end
 
       # Changes the IDs of the states into a more friendly format.

data/lib/antelope/generation/tableizer.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 module Antelope
   module Generation
 
@@ -96,8 +98,7 @@ module Antelope
 
             case result
             when 0
-              p v, terminal, @rules[rule_part[1]].precedence
-              raise UnresolvableConflictError,
+              $stderr.puts \
                 "Could not determine move for #{on} in state #{state}"
             when 1
               @table[state][on] = rule_part

data/lib/antelope/generation.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "antelope/generation/errors"
 require "antelope/generation/constructor"
 require "antelope/generation/recognizer"

data/lib/antelope/generator/output.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "pp"
 
 module Antelope

data/lib/antelope/generator/ruby.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "pp"
 
 module Antelope

data/lib/antelope/generator/templates/output.erb

@@ -5,16 +5,6 @@ Productions:
 %  end
 % end
 
-Productions, Again:
-% constructor.productions.each do |production|
-  <%= production.to_s(false) %>
-% end
-
-FOLLOW:
-% constructor.instance_variable_get(:@follows).each do |k, v|
-  <%= k %>: {<%= v.map(&:to_s).join(", ") %>}
-% end
-
 Precedence:
   --- highest
 % grammar.precedence.each do |pr|
@@ -23,28 +13,34 @@ Precedence:
 % end
   --- lowest
 
-Table:
 % len = tableizer.table.flatten.map(&:keys).flatten.map(&:size).max
+% states = grammar.states.to_a
 % tableizer.table.each_with_index do |v, i|
+%   state = states[i]
   State <%= i %>:
-%   v.each do |token, action|
-    <%= "%-#{len}s" % token %>: <%= action[0] %> (<%= action[1] %>)
-%   end
-% end
-  <%# PP.pp(Hash[tableizer.table.each_with_index.to_a.map(&:reverse)], _erbout) %>
-
-% grammar.states.each do |state|
-State <%= state.id %>:
-  rules:
 %   state.rules.each do |rule|
     <%= rule %>
       {<%= rule.lookahead.to_a.join(", ") %>}
 %   end
-
-  transitions:
-%   max = state.transitions.keys.map(&:length).max || 0
-%   state.transitions.each do |on, to|
-    <%= "%-#{max}s" % on %>: State <%= to.id %>
+%   transitions = v.each.select { |_, a| a[0] == :state      }
+%   reductions  = v.each.select { |_, a| a[0] == :reduce     }
+%   accepting   = v.each.select { |_, a| a[0] == :accept     }
+%   conflicts   = v.each.select { |_, (a, b)| a.is_a?(Array) }
+%   thing = [:transitions, :reductions, :accepting]
+%   num_type = {transitions: "State", reductions: "Rule", accepting: "Rule"}
+%   h = Hash[thing.zip([transitions, reductions, accepting])]
+%   h.each do |key, value|
+%     next unless value.any?
+    <%= key %>:
+%     value.each do |token, (_, name)|
+      <%= token %>: <%= num_type[key] %> <%= name %>
+%     end
+%   end
+%   if conflicts.any?
+    conflicts:
+%     conflicts.each do |token, (first, second)|
+      <%= token %>: <%= first.join(" ") %>/<%= second.join(" ") %>
+%     end
 %   end
 
 % end

data/lib/antelope/generator/templates/ruby.erb

@@ -48,7 +48,7 @@ def parse(input)
     when :reduce
       production = PRODUCTIONS[action.last]
       removing   = stack.pop(production[1])
-      value = production[2].call(*removing.map(&:first))
+      value = instance_exec(*removing.map(&:first), &production[2])
       goto  = ACTION_TABLE[stack.last.last][production[0]]
       stack.push([value, goto.last])
     when :state

data/lib/antelope/generator.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require "antelope/generator/output"
 require "antelope/generator/ruby"
 require "erb"