skeem 0.0.3 → 0.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 670bb82d2fecac647938c03d92b8c910f9385a49
-  data.tar.gz: 3ac772b051b36ded30aefe62c6745239c9a1941f
+  metadata.gz: b820c8eb3683e1da04e9c28e41a4de2517894573
+  data.tar.gz: 31dfc0fa971d27e165a73653abb824fdc39d00b6
 SHA512:
-  metadata.gz: 6b8488389ba2a11fa1664bb6f3e2a40c752a16045061f516360cce2744e9e05e2746d823f8854fa40f23fe24183105384863f19156a7b425a86d95071d2ad9c4
-  data.tar.gz: 1c6029799408c83027072cdfca72f9d86dd6bbf5519af149fbb3736e4224f223f553c1463e56fd0461b0e977ae91a2be14ab2a8f4355e9a1ae5449094167a08b
+  metadata.gz: 333f74805437ad0199755a3bdb6bb8ffe8197c5d988979cf0bf8ec95dfd751fcdc423dcaa38c26adb892b982f62505a8bcb25ce62a0dade1c026ab05de0b3a5d
+  data.tar.gz: cd5852bda6bb381855a2ef9a64e3d6875da172ede1c6f520d5b97c697e5a484096cc93cd9874a95311b866ff4b248841ed6d1fd0d802e30dc0e3be60e38d5d08

data/CHANGELOG.md

@@ -1,3 +1,11 @@
+## [0.0.4] - 2018-08-29
+### Added
+- File `s_expr_nodes.rb` with initial implementation of `SExprTerminalNode` classes.
+
+### Changed
+- Class`Tokenizer` converts literal into Ruby "native" objects
+
+
 ## [0.0.3] - 2018-08-25
 ### Added
 - File `grammar.rb` with minimalist grammar.

data/lib/skeem/node_builder.rb

@@ -0,0 +1,47 @@
+require 'stringio'
+require_relative 'regex_repr'
+
+module SrlRuby
+  # The purpose of a ASTBuilder is to build piece by piece an AST
+  # (Abstract Syntax Tree) from a sequence of input tokens and
+  # visit events produced by walking over a GFGParsing object.
+  # Uses the Builder GoF pattern.
+  # The Builder pattern creates a complex object
+  # (say, a parse tree) from simpler objects (terminal and non-terminal
+  # nodes) and using a step by step approach.
+  class ASTBuilder < Rley::ParseRep::ASTBaseBuilder
+    Terminal2NodeClass = {}.freeze
+
+    # Create a new AST builder instance.
+    # @param theTokens [Array<Token>] The sequence of input tokens.
+    def initialize(theTokens)
+      super(theTokens)
+    end
+
+    # Notification that the parse tree construction is complete.
+    def done!
+      super
+    end
+
+    protected
+
+    def terminal2node
+      Terminal2NodeClass
+    end
+
+
+    # Overriding method.
+    # Factory method for creating a node object for the given
+    # input token.
+    # @param _production [Rley::Syntax::Production]
+    # @param _terminal [Rley::Syntax::Terminal] Terminal symbol associated with the token
+    # @param aTokenPosition [Integer] Position of token in the input stream
+    # @param aToken [Rley::Lexical::Token] The input token
+    def new_leaf_node(_production, _terminal, aTokenPosition, aToken)
+      node = Rley::PTree::TerminalNode.new(aToken, aTokenPosition)
+
+      return node
+    end
+  end # class
+end # module
+# End of file

data/lib/skeem/s_expr_nodes.rb

@@ -0,0 +1,86 @@
+# Classes that implement nodes of Abstract Syntax Trees (AST) representing
+# Skeem parse results.
+
+module Skeem
+  # Abstract class. Root of class hierarchy needed for Interpreter
+  # design pattern
+  SExprTerminalNode = Struct.new(:token, :value, :position) do
+    def initialize(aToken, aPosition)
+      self.token = aToken
+      self.position = aPosition
+      init_value(aToken.lexeme)
+    end
+
+    # This method can be overriden
+    def init_value(aValue)
+      self.value = aValue.dup
+    end
+
+    def symbol()
+      token.terminal
+    end
+
+    def interpret()
+      return value
+    end
+    
+    def done!()
+      # Do nothing
+    end
+
+    # Part of the 'visitee' role in Visitor design pattern.
+    # @param aVisitor[ParseTreeVisitor] the visitor
+    def accept(aVisitor)
+      aVisitor.visit_terminal(self)
+    end
+  end
+
+  class SExprBooleanNode < SExprTerminalNode
+  end # class
+  
+  class SExprNumberNode < SExprTerminalNode
+  
+  class SExprRealNode < SExprNumberNode
+  end # class
+  
+  class SExprIntegerNode < SExprRealNode
+  end # class
+  
+  class SExprStringNode < SExprTerminalNode
+  end # class
+  
+=begin
+  class SExprCompositeNode
+    attr_accessor(:children)
+    attr_accessor(:symbol)
+    attr_accessor(:position)
+
+    def initialize(aSymbol, aPosition)
+      @symbol = aSymbol
+      @children = []
+      @position = aPosition
+    end
+
+    # Part of the 'visitee' role in Visitor design pattern.
+    # @param aVisitor[ParseTreeVisitor] the visitor
+    def accept(aVisitor)
+      aVisitor.visit_nonterminal(self)
+    end
+    
+    def done!()
+      # Do nothing
+    end
+
+    alias subnodes children
+  end # class
+
+  class SExprUnaryOpNode < SExprCompositeNode
+    def initialize(aSymbol, aPosition)
+      super(aSymbol, aPosition)
+    end
+
+    alias members children
+  end # class
+=end
+end # module
+# End of file

data/lib/skeem/tokenizer.rb

@@ -24,12 +24,12 @@ module Skeem
       '(' => 'LPAREN',
       ')' => 'RPAREN'
     }.freeze
-    
+
     # Here are all the SRL keywords (in uppercase)
     @@keywords = %w[
       BEGIN
       DEFINE
-    ].map { |x| [x, x] } .to_h      
+    ].map { |x| [x, x] } .to_h
 
     class ScanError < StandardError; end
 
@@ -74,14 +74,13 @@ module Skeem
       elsif (lexeme = scanner.scan(/#(?:\.)(?=\s|[|()";]|$)/)) # Single char occurring alone
         token = build_token('PERIOD', lexeme)
       elsif (lexeme = scanner.scan(/#(?:t|f|true|false)(?=\s|[|()";]|$)/))
-        token = build_token('BOOLEAN', lexeme) # normalized lexeme
+        token = build_token('BOOLEAN', lexeme)
       elsif (lexeme = scanner.scan(/[+-]?[0-9]+(?=\s|[|()";]|$)/))
         token = build_token('INTEGER', lexeme) # Decimal radix
       elsif (lexeme = scanner.scan(/[+-]?[0-9]+\.[0-9]+(?:(?:e|E)[+-]?[0-9]+)?/))
         token = build_token('REAL', lexeme)
       elsif (lexeme = scanner.scan(/"(?:\\"|[^"])*"/)) # Double quotes literal?
-        unquoted = lexeme.gsub(/(^")|("$)/, '')
-        token = build_token('STRING_LIT', unquoted)
+        token = build_token('STRING_LIT', lexeme)
       elsif (lexeme = scanner.scan(/[a-zA-Z!$%&*\/:<=>?@^_~][a-zA-Z0-9!$%&*+-.\/:<=>?@^_~+-]*/))
         keyw = @@keywords[lexeme.upcase]
         tok_type = keyw ? keyw : 'IDENTIFIER'
@@ -107,11 +106,12 @@ module Skeem
       return token
     end
 
-    def build_token(aSymbolName, aLexeme)
+    def build_token(aSymbolName, aLexeme, aFormat = :default)
       begin
+        value = convert_to(aLexeme, aSymbolName, aFormat)
         col = scanner.pos - aLexeme.size - @line_start + 1
         pos = Position.new(@lineno, col)
-        token = SToken.new(aLexeme, aSymbolName, pos)
+        token = SToken.new(value, aSymbolName, pos)
       rescue StandardError => exc
         puts "Failing with '#{aSymbolName}' and '#{aLexeme}'"
         raise exc
@@ -120,6 +120,65 @@ module Skeem
       return token
     end
 
+    def convert_to(aLexeme, aSymbolName, aFormat)
+      case aSymbolName
+      when 'BOOLEAN'
+        value = to_boolean(aLexeme, aFormat)
+      when 'INTEGER'
+        value = to_integer(aLexeme, aFormat)
+      when 'REAL'
+        value = to_real(aLexeme, aFormat)
+      when 'STRING_LIT'
+        value = to_string(aLexeme, aFormat)
+      when 'SYMBOL'
+        value = to_string(aLexeme, aFormat)        
+      else
+        value = aLexeme
+      end
+
+      return value
+    end
+
+    def to_boolean(aLexeme, aFormat)
+      result = (aLexeme =~ /^#t/) ? true : false
+    end
+
+    def to_integer(aLexeme, aFormat)
+      case aFormat
+      when :default, :base10
+        value = aLexeme.to_i
+      end
+
+      return value
+    end
+
+    def to_real(aLexeme, aFormat)
+      case aFormat
+      when :default
+        value = aLexeme.to_f
+      end
+
+      return value
+    end
+    
+    def to_string(aLexeme, aFormat)
+      case aFormat
+      when :default
+        value = aLexeme.gsub(/(^")|("$)/, '')
+      end
+
+      return value
+    end  
+
+    def to_symbol(aLexeme, aFormat)
+      case aFormat
+      when :default
+        value = aLexeme
+      end
+
+      return value
+    end     
+    
     def skip_whitespaces
       pre_pos = scanner.pos
 

data/lib/skeem/version.rb

@@ -1,3 +1,3 @@
 module Skeem
-  VERSION = '0.0.3'.freeze
+  VERSION = '0.0.4'.freeze
 end

data/spec/skeem/tokenizer_spec.rb

@@ -43,10 +43,10 @@ module Skeem
       it 'should tokenize boolean constants' do
         tests = [
           # couple [raw input, expected]
-          ['#t', '#t'],
-          [' #f', '#f'],
-          ['#true ', '#true'],
-          [' #false', '#false']
+          ['#t', true],
+          [' #f', false],
+          ['#true ', true],
+          [' #false', false]
         ]
 
         tests.each do |(input, prediction)|
@@ -62,11 +62,11 @@ module Skeem
       it 'should tokenize integers in default radix 10' do
         tests = [
           # couple [raw input, expected]
-          ['0', '0'],
-          [' 3', '3'],
-          ['+3 ', '+3'],
-          ['-3', '-3'],
-          ['-1234', '-1234']
+          ['0', 0],
+          [' 3', 3],
+          ['+3 ', +3],
+          ['-3', -3],
+          ['-1234', -1234]
         ]
 
         tests.each do |(input, prediction)|
@@ -82,9 +82,9 @@ module Skeem
       it 'should tokenize real numbers' do
         tests = [
           # couple [raw input, expected]
-          ["\t\t3.45e+6", '3.45e+6'],
-          ['+3.45e+6', '+3.45e+6'],
-          ['-3.45e+6', '-3.45e+6']
+          ["\t\t3.45e+6", 3.45e+6],
+          ['+3.45e+6', +3.45e+6],
+          ['-3.45e+6', -3.45e+6]
         ]
 
         tests.each do |(input, prediction)|
@@ -96,6 +96,7 @@ module Skeem
       end
     end # context
 
+    # TODO
     context 'String recognition:' do
       it 'should tokenize strings' do
         examples = [
@@ -114,13 +115,13 @@ module Skeem
       end
     end # context
 
-=begin
-For later:
-"Another example:\ntwo lines of text"
-"Here's text \
-containing just one line"
-"\x03B1; is named GREEK SMALL LETTER ALPHA."
-=end
+
+# For later:
+# "Another example:\ntwo lines of text"
+# "Here's text \
+# containing just one line"
+# "\x03B1; is named GREEK SMALL LETTER ALPHA."
+
 
     context 'Identifier recognition:' do
       it 'should tokenize identifier' do
@@ -141,7 +142,7 @@ containing just one line"
         end
       end
     end # context
-
+=begin
     context 'Scanning Scheme sample code' do
       it 'should read examples from lis.py page' do
         # Shallow tokenizer testing
@@ -201,5 +202,6 @@ SCHEME
         match_expectations(subject, predicted)
       end
     end # context
+=end
   end # describe
 end # module

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: skeem
 version: !ruby/object:Gem::Version
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors:
 - Dimitri Geshef
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-08-26 00:00:00.000000000 Z
+date: 2018-08-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rley
@@ -86,7 +86,9 @@ files:
 - appveyor.yml
 - lib/skeem.rb
 - lib/skeem/grammar.rb
+- lib/skeem/node_builder.rb
 - lib/skeem/parser.rb
+- lib/skeem/s_expr_nodes.rb
 - lib/skeem/stoken.rb
 - lib/skeem/tokenizer.rb
 - lib/skeem/version.rb