lrama 0.5.7 → 0.5.9

data/lib/lrama/lexer/token/tag.rb

@@ -0,0 +1,8 @@
+module Lrama
+  class Lexer
+    class Token
+      class Tag < Token
+      end
+    end
+  end
+end

data/lib/lrama/lexer/token/user_code.rb

@@ -0,0 +1,14 @@
+module Lrama
+  class Lexer
+    class Token
+      class UserCode < Token
+        attr_accessor :references
+
+        def initialize(s_value: nil, alias_name: nil)
+          super
+          self.references = []
+        end
+      end
+    end
+  end
+end

data/lib/lrama/lexer/token.rb

@@ -1,84 +1,26 @@
-require 'lrama/lexer/token/type'
-
 module Lrama
   class Lexer
-    class Token
+    class Token < Struct.new(:s_value, :alias_name, keyword_init: true)
 
       attr_accessor :line, :column, :referred
-      # For User_code
-      attr_accessor :references
 
       def to_s
         "#{super} line: #{line}, column: #{column}"
       end
 
       def referred_by?(string)
-        [self.s_value, self.alias].include?(string)
+        [self.s_value, self.alias_name].include?(string)
       end
 
       def ==(other)
-        self.class == other.class && self.type == other.type && self.s_value == other.s_value
+        self.class == other.class && self.s_value == other.s_value
       end
-
-      def numberize_references(lhs, rhs)
-        self.references.map! {|ref|
-          ref_name = ref[1]
-          if ref_name.is_a?(::String) && ref_name != '$'
-            value =
-              if lhs.referred_by?(ref_name)
-                '$'
-              else
-                index = rhs.find_index {|token| token.referred_by?(ref_name) }
-
-                if index
-                  index + 1
-                else
-                  raise "'#{ref_name}' is invalid name."
-                end
-              end
-            [ref[0], value, ref[2], ref[3], ref[4]]
-          else
-            ref
-          end
-        }
-      end
-
-      @i = 0
-      @types = []
-
-      def self.define_type(name)
-        type = Type.new(id: @i, name: name.to_s)
-        const_set(name, type)
-        @types << type
-        @i += 1
-      end
-
-      # Token types
-      define_type(:P_expect)         # %expect
-      define_type(:P_define)         # %define
-      define_type(:P_printer)        # %printer
-      define_type(:P_error_token)    # %error-token
-      define_type(:P_lex_param)      # %lex-param
-      define_type(:P_parse_param)    # %parse-param
-      define_type(:P_initial_action) # %initial-action
-      define_type(:P_union)          # %union
-      define_type(:P_token)          # %token
-      define_type(:P_type)           # %type
-      define_type(:P_nonassoc)       # %nonassoc
-      define_type(:P_left)           # %left
-      define_type(:P_right)          # %right
-      define_type(:P_precedence)     # %precedence
-      define_type(:P_prec)           # %prec
-      define_type(:User_code)        # { ... }
-      define_type(:Tag)              # <int>
-      define_type(:Number)           # 0
-      define_type(:Ident_Colon)      # k_if:, k_if  : (spaces can be there)
-      define_type(:Ident)            # api.pure, tNUMBER
-      define_type(:Named_Ref)        # [foo]
-      define_type(:Semicolon)        # ;
-      define_type(:Bar)              # |
-      define_type(:String)           # "str"
-      define_type(:Char)             # '+'
     end
   end
 end
+
+require 'lrama/lexer/token/char'
+require 'lrama/lexer/token/ident'
+require 'lrama/lexer/token/parameterizing'
+require 'lrama/lexer/token/tag'
+require 'lrama/lexer/token/user_code'

data/lib/lrama/lexer.rb

@@ -3,6 +3,7 @@ require "lrama/lexer/token"
 
 module Lrama
   class Lexer
+    attr_reader :head_line, :head_column
     attr_accessor :status
     attr_accessor :end_symbol
 
@@ -24,6 +25,8 @@ module Lrama
       %precedence
       %prec
       %error-token
+      %empty
+      %code
     )
 
     def initialize(text)
@@ -63,8 +66,6 @@ module Lrama
         when @scanner.scan(/\/\//)
           @scanner.scan_until(/\n/)
           newline
-        when @scanner.scan(/%empty/)
-          # noop
         else
           break
         end
@@ -80,18 +81,20 @@ module Lrama
         return [@scanner.matched, @scanner.matched]
       when @scanner.scan(/#{PERCENT_TOKENS.join('|')}/)
         return [@scanner.matched, @scanner.matched]
+      when @scanner.scan(/[\?\+\*]/)
+        return [@scanner.matched, @scanner.matched]
       when @scanner.scan(/<\w+>/)
-        return [:TAG, build_token(type: Token::Tag, s_value: @scanner.matched)]
+        return [:TAG, setup_token(Lrama::Lexer::Token::Tag.new(s_value: @scanner.matched))]
       when @scanner.scan(/'.'/)
-        return [:CHARACTER, build_token(type: Token::Char, s_value: @scanner.matched)]
+        return [:CHARACTER, setup_token(Lrama::Lexer::Token::Char.new(s_value: @scanner.matched))]
       when @scanner.scan(/'\\\\'|'\\b'|'\\t'|'\\f'|'\\r'|'\\n'|'\\v'|'\\13'/)
-        return [:CHARACTER, build_token(type: Token::Char, s_value: @scanner.matched)]
+        return [:CHARACTER, setup_token(Lrama::Lexer::Token::Char.new(s_value: @scanner.matched))]
       when @scanner.scan(/"/)
         return [:STRING, %Q("#{@scanner.scan_until(/"/)})]
       when @scanner.scan(/\d+/)
         return [:INTEGER, Integer(@scanner.matched)]
       when @scanner.scan(/([a-zA-Z_.][-a-zA-Z0-9_.]*)/)
-        token = build_token(type: Token::Ident, s_value: @scanner.matched)
+        token = setup_token(Lrama::Lexer::Token::Ident.new(s_value: @scanner.matched))
         type =
           if @scanner.check(/\s*(\[\s*[a-zA-Z_.][-a-zA-Z0-9_.]*\s*\])?\s*:/)
             :IDENT_COLON
@@ -100,7 +103,7 @@ module Lrama
           end
         return [type, token]
       else
-        raise
+        raise ParseError, "Unexpected token: #{@scanner.peek(10).chomp}."
       end
     end
 
@@ -115,13 +118,13 @@ module Lrama
         when @scanner.scan(/}/)
           if nested == 0 && @end_symbol == '}'
             @scanner.unscan
-            return [:C_DECLARATION, build_token(type: Token::User_code, s_value: code, references: [])]
+            return [:C_DECLARATION, setup_token(Lrama::Lexer::Token::UserCode.new(s_value: code))]
           else
             code += @scanner.matched
             nested -= 1
           end
         when @scanner.check(/#{@end_symbol}/)
-          return [:C_DECLARATION, build_token(type: Token::User_code, s_value: code, references: [])]
+          return [:C_DECLARATION, setup_token(Lrama::Lexer::Token::UserCode.new(s_value: code))]
         when @scanner.scan(/\n/)
           code += @scanner.matched
           newline
@@ -136,7 +139,7 @@ module Lrama
           code += @scanner.getch
         end
       end
-      raise
+      raise ParseError, "Unexpected code: #{code}."
     end
 
     private
@@ -155,13 +158,9 @@ module Lrama
       end
     end
 
-    def build_token(type:, s_value:, **options)
-      token = Token.new(type: type, s_value: s_value)
+    def setup_token(token)
       token.line = @head_line
       token.column = @head_column
-      options.each do |attr, value|
-        token.public_send("#{attr}=", value)
-      end
 
       token
     end

data/lib/lrama/option_parser.rb

@@ -58,10 +58,10 @@ module Lrama
         o.separator 'Tuning the Parser:'
         o.on('-S', '--skeleton=FILE', 'specify the skeleton to use') {|v| @options.skeleton = v }
         o.on('-t', 'reserved, do nothing') { }
+        o.on('--debug', 'display debugging outputs of internal parser') {|v| @options.debug = true }
         o.separator ''
         o.separator 'Output:'
         o.on('-H', '--header=[FILE]', 'also produce a header file named FILE') {|v| @options.header = true; @options.header_file = v }
-        o.on('-h=[FILE]', 'also produce a header file named FILE (deprecated)') {|v| @options.header = true; @options.header_file = v }
         o.on('-d', 'also produce a header file') { @options.header = true }
         o.on('-r', '--report=THINGS', Array, 'also produce details on the automaton') {|v| @report = v }
         o.on('--report-file=FILE', 'also produce details on the automaton output to a file named FILE') {|v| @options.report_file = v }
@@ -74,7 +74,7 @@ module Lrama
         o.separator ''
         o.separator 'Other options:'
         o.on('-V', '--version', "output version information and exit") {|v| puts "lrama #{Lrama::VERSION}"; exit 0 }
-        o.on('--help', "display this help and exit") {|v| puts o; exit 0 }
+        o.on('-h', '--help', "display this help and exit") {|v| puts o; exit 0 }
         o.separator ''
         o.parse!(argv)
       end
@@ -109,7 +109,7 @@ module Lrama
     def validate_trace(trace)
       list = %w[
         none locations scan parse automaton bitsets
-        closure grammar resource sets muscles tools
+        closure grammar rules resource sets muscles tools
         m4-early m4 skeleton time ielr cex all
       ]
       h = {}

data/lib/lrama/options.rb

@@ -4,7 +4,8 @@ module Lrama
     attr_accessor :skeleton, :header, :header_file,
                   :report_file, :outfile,
                   :error_recovery, :grammar_file,
-                  :report_file, :trace_opts, :report_opts, :y
+                  :report_file, :trace_opts, :report_opts, :y,
+                  :debug
 
     def initialize
       @skeleton = "bison/yacc.c"

data/lib/lrama/output.rb

@@ -349,6 +349,15 @@ module Lrama
       end
     end
 
+    # b4_percent_code_get
+    def percent_code(name)
+      @grammar.percent_codes.select do |percent_code|
+        percent_code.id.s_value == name
+      end.map do |percent_code|
+        percent_code.code.s_value
+      end.join
+    end
+
     private
 
     def eval_template(file, path)