cskit 1.0.1 → 1.1.0

data/lib/cskit/parsers/bible/bible_tokenizer.rb

@@ -0,0 +1,32 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+    module Bible
+
+      class BibleTokenizer < CSKit::Parsers::Tokenizer
+        PATTERNS = {
+          left_paren:  /\A\(/,
+          right_paren: /\A\)/,
+          dash:        /\A-/,
+          colon:       /\A:/,
+          semicolon:   /\A;/,
+          comma:       /\A,/,
+          to:          /\Ato/,
+          cardinality: /\A(1st|2nd|3rd|4th)/,
+          number:      /\A\d+/,
+          text:        /\A[^\s\(\);:,]+/,
+          space:       /\A[\s\t]+/
+        }
+
+        private
+
+        def patterns
+          PATTERNS
+        end
+      end
+
+    end
+  end
+end
+

data/lib/cskit/parsers/parser.rb

@@ -0,0 +1,68 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+    class ParserError < StandardError; end
+
+    class Parser
+      def initialize(citation_text)
+        @citation_text = citation_text
+        @token_stream = get_token_stream
+        @current = token_stream.next
+      end
+
+      def parse
+        result = entry_point
+
+        unless eos?
+          raise ParserError, "Expected end of input but more input is available "\
+            "at position #{current.position}"
+        end
+
+        result
+      end
+
+      def entry_point
+        raise NotImplementedError,
+          "`#{__method__} must be defined in derived classes"
+      end
+
+      private
+
+      def get_token_stream
+        raise NotImplementedError,
+          "`#{__method__} must be defined in derived classes"
+      end
+
+      attr_reader :citation_text, :token_stream, :current
+
+      def eos?
+        token_stream.peek
+        false
+      rescue StopIteration
+        true
+      end
+
+      def eos_token
+        @eos_token ||= Token.new(:eos, nil, citation_text.size)
+      end
+
+      def next_token(*token_types)
+        if !token_types.include?(current.type)
+          raise ParserError, "Expected #{token_types.join(', ')} but got "\
+            "#{current.type} ('#{current.value}') at position #{current.position}"
+        end
+
+        if eos?
+          if current.type == :eos
+            raise(ParserError, 'Unexpected end of input')
+          else
+            @current = eos_token
+          end
+        else
+          @current = token_stream.next
+        end
+      end
+    end
+  end
+end

data/lib/cskit/parsers/science_health.rb

@@ -0,0 +1,10 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+    module ScienceHealth
+      autoload :ScienceHealthParser,    'cskit/parsers/science_health/science_health_parser'
+      autoload :ScienceHealthTokenizer, 'cskit/parsers/science_health/science_health_tokenizer'
+    end
+  end
+end

data/lib/cskit/parsers/science_health/science_health_parser.rb

@@ -0,0 +1,201 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+    module ScienceHealth
+
+      DEFAULT_CARDINALITY = 1
+
+      Citation = Struct.new(:page, :lines) do
+        def to_s
+          "#{page}:#{lines.map(&:to_s).join(", ")}"
+        end
+
+        def to_hash
+          {
+            page: page,
+            lines: lines.map(&:to_hash)
+          }
+        end
+      end
+
+      Line = Struct.new(:start, :finish, :starter, :terminator) do
+        def to_s
+          str = if finish
+            "#{start}-#{finish}"
+          else
+            start.to_s
+          end
+
+          str << ' (only)' if only?
+          str << " #{start_fragment}" if start_fragment
+          str
+        end
+
+        def to_hash
+          {
+            start: start,
+            finish: finish,
+            starter: starter ? starter.to_hash : nil,
+            terminator: terminator ? terminator.to_hash : nil
+          }
+        end
+      end
+
+      class Positional
+        attr_reader :cardinality, :fragment
+
+        def initialize(cardinality, fragment)
+          @cardinality = cardinality
+          @fragment = fragment
+        end
+
+        def to_s
+          card_s = case cardinality
+            when 1 then '1st'
+            when 2 then '2nd'
+            when 3 then '3rd'
+          end
+
+          if cardinality
+            "#{card_s} #{fragment}"
+          else
+            fragment
+          end
+        end
+
+        def to_hash
+          {
+            cardinality: cardinality,
+            fragment: fragment
+          }
+        end
+      end
+
+      class Starter < Positional
+      end
+
+      class FragmentTerminator < Positional
+      end
+
+      class OnlyTerminator
+        def self.instance
+          @instance ||= send(:new)
+        end
+
+        def to_hash
+          { only: true }
+        end
+
+        private def initialize
+        end
+      end
+
+
+      class ScienceHealthParser < CSKit::Parsers::Parser
+        def entry_point
+          page
+        end
+
+        private
+
+        def get_token_stream
+          ScienceHealthTokenizer.new(citation_text).each_token.lazy
+        end
+
+        def page
+          page_num = page_number
+          next_token(:colon)
+          llist = line_list
+
+          Citation.new(page_num, llist)
+        end
+
+        def page_number
+          current.value.tap { next_token(:page_number, :number) }
+        end
+
+        def line_list
+          [].tap do |list|
+            loop do
+              list << line
+
+              case current.type
+                when :comma
+                  next_token(:comma)
+                else
+                  break
+              end
+
+              break if eos?
+            end
+          end
+        end
+
+        def line
+          start = current.value.tap { next_token(:number) }.to_i
+          finish = start
+          starter = nil
+
+          if current.type == :dash
+            next_token(:dash)
+            finish = current.value.tap { next_token(:number) }.to_i
+          end
+
+          starter = line_starter
+          terminator = line_terminator
+
+          Line.new(start, finish, starter, terminator)
+        end
+
+        def line_starter
+          case current.type
+            when :text, :cardinality
+              card = cardinality
+              fragment = current.value
+              next_token(:text, :colon, :comma)
+              Starter.new(card, fragment)
+          end
+        end
+
+        def line_terminator
+          if current.type == :left_paren
+            next_token(:left_paren)
+
+            terminator = if current.type == :to
+              fragment_terminator
+            else
+              only_terminator
+            end
+
+            next_token(:right_paren)
+            terminator
+          end
+        end
+
+        def fragment_terminator
+          next_token(:to)
+
+          card = cardinality
+          fragment = current.value
+          next_token(:text, :colon, :comma)
+
+          FragmentTerminator.new(card, fragment)
+        end
+
+        def only_terminator
+          next_token(:only)
+          OnlyTerminator.instance
+        end
+
+        def cardinality
+          if current.type == :cardinality
+            current.value.tap { next_token(:cardinality) }.to_i
+          else
+            DEFAULT_CARDINALITY
+          end
+        end
+      end
+    end
+  end
+end

data/lib/cskit/parsers/science_health/science_health_tokenizer.rb

@@ -0,0 +1,33 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+    module ScienceHealth
+
+      class ScienceHealthTokenizer < CSKit::Parsers::Tokenizer
+        PATTERNS = {
+          left_paren:  /\A\(/,
+          right_paren: /\A\)/,
+          dash:        /\A-/,
+          colon:       /\A:/,
+          comma:       /\A,/,
+          to:          /\Ato/,
+          only:        /\Aonly(?=\))/,
+          cardinality: /\A(1st|2nd|3rd|4th)/,
+          page_number: /\A(vii|viii|ix|x|xi|xii)(?=:)/,  # must precede a colon
+          number:      /\A\d+/,
+          text:        /\A[^\s\(\):,]+/,
+          space:       /\A[\s\t]+/
+        }
+
+        private
+
+        def patterns
+          PATTERNS
+        end
+      end
+
+    end
+  end
+end
+

data/lib/cskit/parsers/token.rb

@@ -0,0 +1,17 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+
+    class Token
+      attr_reader :type, :value, :position
+
+      def initialize(type, value, position)
+        @type = type
+        @value = value
+        @position = position
+      end
+    end
+
+  end
+end

data/lib/cskit/parsers/tokenizer.rb

@@ -0,0 +1,43 @@
+# encoding: UTF-8
+
+module CSKit
+  module Parsers
+    # base class for tokenizers
+    class Tokenizer
+      attr_reader :citation
+
+      def initialize(citation)
+        @citation = citation
+      end
+
+      def each_token
+        return to_enum(__method__) unless block_given?
+
+        text = citation.dup
+        pos = 0
+
+        until text.empty?
+          patterns.each_pair do |token_type, pattern|
+            if match = pattern.match(text)
+              unless token_type == :space
+                yield Token.new(token_type, match[0], pos)
+              end
+
+              text[0...match[0].size] = ''
+              pos += match[0].size
+
+              break
+            end
+          end
+        end
+      end
+
+      private
+
+      def patterns
+        raise NotImplementedError,
+          "`#{__method__}' must be implemented by derived classes"
+      end
+    end
+  end
+end

data/lib/cskit/readers.rb

@@ -2,8 +2,8 @@
 
 module CSKit
   module Readers
-    autoload :ScienceHealthReader, "cskit/readers/science_health_reader"
-    autoload :BibleReader,         "cskit/readers/bible_reader"
-    autoload :Reading,             "cskit/readers/reading"
+    autoload :ScienceHealthReader, 'cskit/readers/science_health_reader'
+    autoload :BibleReader,         'cskit/readers/bible_reader'
+    autoload :Reading,             'cskit/readers/reading'
   end
-end
+end

data/lib/cskit/readers/bible_reader.rb

@@ -34,7 +34,7 @@ module CSKit
         end
       end
 
-      protected
+      private
 
       def map_verse_texts_for(chapter, book_name)
         result = []
@@ -50,7 +50,7 @@ module CSKit
       end
 
       def convert_book_name(book_name)
-        volume.unabbreviate_book_name(book_name).downcase.gsub(" ", "_")
+        volume.unabbreviate_book_name(book_name).downcase.gsub(' ', '_')
       end
 
     end