citeproc-ruby 0.0.1

data/lib/citeproc/formatter.rb

@@ -0,0 +1,38 @@
+module CiteProc
+  
+  module Format
+    def self.default; CiteProc::Format::Default.new; end
+  end
+  
+  class Formatter
+  
+    def format(*args)
+      @format ||= CiteProc.default_format
+      args.empty? ? @format : apply(args[0], args[1])
+    end
+
+    def format=(format)
+      @format = Format.const_get(format.to_s.split(/[\s_-]+/).map(&:capitalize).join).new
+    rescue Exception => e
+      CiteProc.log :warn, "failed to set format to #{ format.inspect }", e
+    end
+    
+    def apply(input='', attributes={})
+      return input if attributes.nil? || input.nil? || input.empty?
+
+      format.input = input
+      
+      CSL::Nodes.formatting_attributes.each do |attribute|
+        method_id = ['set', attribute.gsub(/-/, '_')].join('_')
+        
+        if !attributes[attribute].nil? && format.respond_to?(method_id)
+          format.send(method_id, attributes[attribute])
+        end
+      end
+      
+      format.finalize
+    end
+  
+  end
+  
+end

data/lib/citeproc/item.rb

@@ -0,0 +1,53 @@
+require 'observer'
+
+module CiteProc
+
+  class Item
+    include Comparable
+    include Observable
+    include Support::Attributes
+    
+    attr_fields Variable.fields
+    attr_fields %w{ locator label suppress-author author-only prefix suffix }
+    
+    def initialize(attributes={}, filter=nil)
+      self.merge!(attributes)
+      yield self if block_given?
+    end
+    
+    def self.filter(attributes, filter)
+      # TODO
+    end
+    
+    # @see CSL::Nodes::Group
+    alias :access :[]
+    def [](key)
+      value = access(key)
+      changed
+      notify_observers(key, value)
+      value
+    end
+    
+    def merge!(other)
+      other = other.attributes unless other.is_a?(Hash)
+      other.each_pair { |key, value| self.attributes[key] = Variable.parse(value, key) }
+      self
+    end
+
+    def reverse_merge!(other)
+      other = other.attributes unless other.is_a?(Hash)
+      other.each_pair { |key, value| self.attributes[key] ||= Variable.parse(value, key) }
+      self
+    end
+    
+    def to_s
+      self.attributes.inspect
+    end
+    
+    
+    def <=>(other)
+      self.attributes <=> other.attributes
+    end
+  end
+
+end

data/lib/citeproc/name.rb

@@ -0,0 +1,284 @@
+module CiteProc
+
+
+  # == Name Variables
+  #
+  # When present in the item data, CSL name variables must be delivered as a
+  # list of JavaScript arrays, with one array for each name represented by the
+  # variable. Simple personal names are composed of family and given elements,
+  # containing respectively the family and given name of the individual.
+  #
+  # { "author" : [
+  #     { "family" : "Doe", "given" : "Jonathan" },
+  #     { "family" : "Roe", "given" : "Jane" }
+  #   ],
+  #   "editor" : [
+  #     { "family" : "Saunders",
+  #       "given" : "John Bertrand de Cusance Morant" }
+  #   ]
+  # }
+  #
+  # Institutional and other names that should always be presented literally
+  # (such as "The Artist Formerly Known as Prince", "Banksy", or "Ramses IV")
+  # should be delivered as a single literal element in the name array:
+  # 
+  # { "author" : [
+  #     { "literal" : "Society for Putting Things on Top of Other Things" }
+  #   ]
+  # }
+  #
+  # If the name is spelled using a 'byzantine' alphabet (i.e., latin or
+  # cyrillic) its sort and display order is computed according to the given
+  # arguments.
+  #
+  class Name < Variable
+
+    # Based on the regular expression in citeproc-js
+    ROMANESQUE = /^[a-zA-Z\u0080-\u017f\u0400-\u052f\u0386-\u03fb\u1f00-\u1ffe\.,\s'\u0027\u02bc\u2019-]*$/
+    
+    Variable.name_fields.each { |field| Variable.types[field] = Name }
+    
+    attr_fields %w{ given family literal suffix dropping-particle
+      non-dropping-particle comma-suffix static-ordering parse-names }
+
+    [[:last, :family], [:first, :given]].each do |m|
+      alias_method m[0], m[1]
+    end
+    
+    def defaults
+      Hash['form', 'long', 'name-as-sort-order', 'false', 'demote-non-dropping-particle', 'display-and-sort']
+    end
+    
+    def options
+      @options ||= defaults
+    end
+    
+    def merge_options(options)
+      options.each_pair { |key, value| self.options[key] = value unless value.nil? }
+    end
+    
+    def parse!(argument)
+      return super unless argument.is_a?(String)
+      parse_name!(argument)
+    end
+    
+    def merge!(argument)
+      if argument['parse-names'] && argument.delete('parse-names') != 'false'
+        parse_family!(argument.delete('family'))
+        parse_given!(argument.delete('given'))
+      end
+      
+      argument.map { |key, value| self[key] = value }
+    end
+    
+    def given
+      initialize? ? to_initials(self['given']) : self['given']
+    end
+    
+    def to_initials(name)
+      return name if name.nil?
+      
+      name.split(/\s+|\.\s*/).map do |token|
+        token.split(/-/).map do |part|
+          # Keep all-lowercase names; otherwise keep only upper case letters
+          part.match(/^[[:lower:]]+$/) ? part.center(part.length + 2) : part.scan(/[[:upper:]]/).join.capitalize + options['initialize-with']
+        end.join(options['initialize-with-hyphen'] == 'false' ? '' : '-' ).gsub(/\s+-/, '-')
+      end.join
+    end
+    
+    # Parses a string and sets :family, :given, :suffix, and :particles
+    # correspondingly.
+    #
+    # * non-dropping-particle: A string at the beginning of the family field
+    #   consisting of spaces and lowercase roman or Cyrillic characters will
+    #   be treated as a non-dropping-particle. The particles preceding some
+    #   names should be treated as part of the last name, depending on the
+    #   cultural heritage and personal preferences of the individual. To
+    #   suppress parsing and treat such particles as part of the family name
+    #   field, enclose the family name field content in double-quotes
+    # * dropping-particle: A string at the end of the given name field
+    #   consisting of spaces and lowercase roman or Cyrillic characters will
+    #   be treated as a dropping-particle.
+    # * suffix: Content following a comma in the given name field will be
+    #   parse out as a name suffix. Modern typographical convention does not
+    #   place a comma between suffixes such as "Jr." and the last name, when
+    #   rendering the name in normal order: "John Doe Jr." If an individual
+    #   prefers that the traditional comma be used in rendering their name,
+    #   the comma can be force by placing a exclamation mark after the comma.
+    #
+    def parse_name!(string)
+      return if string.nil?
+
+      tokens = string.split(/,\s+/)
+
+      parse_family!(tokens[0])
+      parse_given!(tokens[1])
+      
+      self
+    end
+    
+    # @see parse
+    def parse_family!(string)
+      return if string.nil?
+      
+      tokens = string.scan(/^['"](.+)['"]$|^([[:lower:]\s]+)?([[:upper:]][[:alpha:]\s]*)$/).first
+            
+      if tokens.nil?
+        self['family'] = string
+      else
+        self['family'] = tokens[0] || tokens[2] || string
+        self['non-dropping-particle'] = tokens[1].gsub(/^\s+|\s+$/, '') unless tokens[1].nil?
+      end
+      
+      self
+    end
+    
+    # @see parse
+    def parse_given!(string)
+      return if string.nil?
+
+      tokens = string.scan(/^((?:[[:upper:]][[:alpha:]\.]*\s*)+)([[:lower:]\s]+)?(?:,!?\s([[:alpha:]\.\s]+))?$/).first
+
+      if tokens.nil?
+        self['given'] = string
+      else
+        self['given'] = (tokens[0] || string).gsub(/^\s+|\s+$/, '')
+        self['dropping-particle'] = tokens[1] unless tokens[1].nil?
+        self['suffix'] = tokens[2] unless tokens[2].nil?
+        self['comma-suffix'] = 'true' if string.match(/,!/)
+      end
+      
+      self
+    end
+    
+    def personal?
+      self.family?
+    end
+    
+    def romanesque?
+      (self['given'].nil? || self['given'].match(ROMANESQUE)) && (self['family'].nil? || self['family'].match(ROMANESQUE))
+    end
+    
+    alias :byzantine? :romanesque?
+        
+    def comma_suffix
+      self.comma_suffix? && self.suffix? ? comma : nil
+    end
+    
+    def comma
+      options['sort-separator'] || ', '
+    end
+    
+    def delimiter
+      romanesque? ? ' ' : ''
+    end
+    
+    def value
+      self
+    end
+    
+    #
+    # CSL1.0 Warning
+    #
+    # @returns true if, using the current options, the name's given name is
+    # to be displayed using initials. Takes into account whether or not the
+    # family name is set (if not, given name should not be turned to initials,
+    # this is *not* sepcified in CSL 1.0).
+    #
+    def initialize?
+      options.has_key?('initialize-with') && family? && romanesque?
+    end
+    
+    def static_order?
+      static_ordering? || !romanesque?
+    end
+    
+    def sort_order?
+      ['all', 'true', 'yes', 'always'].include?(options['name-as-sort-order'])
+    end
+ 
+    def numeric?
+      false
+    end
+       
+    # @returns a list of strings, representing a given order of the individual
+    # tokens when displaying the name.
+    def display_order(opts={})
+      merge_options(opts)
+
+      case
+      when literal?
+        return %w{ literal }
+      
+      when static_order?
+        return %w{ family given }
+        
+      when options['form'] != 'short' && !sort_order?
+        return %w{ given dropping-particle non-dropping-particle family comma-suffix suffix }
+
+      when options['form'] != 'short' && sort_order? && ['never', 'sort-only'].include?(options['demote-non-dropping-particle'])
+        return %w{ non-dropping-particle family comma given dropping-particle comma suffix }
+    
+      when options['form'] != 'short' && sort_order? && options['demote-non-dropping-particle'] == 'display-and-sort'
+        return %w{ family comma given dropping-particle non-dropping-particle comma suffix }
+    
+      else # options['form'] == 'short'
+        return %w{ non-dropping-particle family}
+      end
+      
+    end
+    
+    # @returns a list of strings, representing the order of precedence of the
+    # individual tokens when sorting the name.
+    def sort_order(opts={})
+      merge_options(opts)
+
+      case
+      when literal?
+        return %w{ literal }
+        
+      when options['demote-non-dropping-particle'] == 'never'
+        return %w{ non-dropping-particle+family dropping-particle given suffix }
+      else
+        return %w{ family non-dropping-particle+dropping-particle given suffix }
+      end
+    end
+    
+    # @returns a string representing the name according to the given set of
+    # display order options.
+    def display(options={})
+      normalize((display_order(options).map { |token| send(token.tr('-', '_')) }.compact.join(delimiter)))
+    end
+    
+    def normalize(string)
+      string.squeeze(' ').gsub(/^[\s,]+|[\s,]+$|\s(,)/, '\1').squeeze(',')
+    end
+    
+    def to_s
+      display
+    end
+    
+    def literal_as_sort_order
+      literal.gsub(/^(the|an?|der|die|das|eine?|l[ae])\s+/i, '')
+    end
+    
+    def <=>(other)
+
+      tests = self.sort_order.zip(other.sort_order).map do |pair|
+        this, that = pair.map { |token| token.gsub(/[\s-]+/,'_').gsub(/literal/, 'literal_sort_order') }          
+
+        this = this.split(/\+/).map { |token| self.send(token) }.join.downcase
+        that = that.split(/\+/).map { |token| other.send(token) }.join.downcase
+
+        # TODO should we ignore '' here?
+        this <=> that
+      end
+
+      tests = tests.reject(&:nil?)
+      zeroes = tests.take_while(&:zero?)
+
+      zeroes.length != tests.length ? tests[zeroes.length] : zeroes.empty? ? nil : 0
+    end
+  end
+
+end

data/lib/citeproc/processor.rb

@@ -0,0 +1,166 @@
+require 'forwardable'
+
+module CiteProc
+
+  class Processor
+    extend Forwardable
+    
+    attr_reader :style, :formatter
+    
+    def initialize
+      @formatter = Formatter.new
+      yield self if block_given?
+    end
+
+    def self.process(items, options = {})
+      return '' if items.nil? || items.empty?
+      
+      processor = Processor.new do |p|
+        p.style = options[:style] || CSL.default_style
+        p.locale = options[:locale] || CSL.default_locale
+        p.format = options[:format] || :default
+        p.import(items)
+      end
+
+      if options[:mode].to_s.match(/cit(e|ation)/i)
+        processor.cite(:all).map(&:last)
+      else
+        processor.bibliography.data.join
+      end
+    end
+    
+    def self.cite(items, options = {})
+      process(items, options.merge(:mode => 'citation'))
+    end
+
+    
+    def style=(resource)
+      @style = resource.is_a?(CSL::Style) ? resource : CSL::Style.new(resource)
+    end
+
+    def format(*args); @formatter.format(*args); end
+    
+    def format=(format); @formatter.format = format; end
+        
+    def locale=(locale)
+      @locale = locale.is_a?(CSL::Locale) ? locale : CSL::Locale.new(locale)
+    end
+    
+    def locale
+      @locale ||= CSL.default_locale
+    end
+    
+    def_delegators :locale, :language, :language=, :region, :region=
+    
+    # @returns the abbreviations, a self-recording hash.
+    def abbreviations
+      @abbreviations ||= new_abbreviations
+    end
+    
+    alias :transfrom :abbreviations
+    
+    def abbreviations=(abbreviations)
+      @abbreviations = new_abbreviations
+      add_abbreviations(abbreviations)
+    end
+    
+    def add_abbreviations(abbreviations)
+      abbreviations.keys.each do |list|
+        abbreviations[list].keys.each do |category|
+          abbreviations[list][category].each_pair do |long, short|
+            self.abbreviations[list] ||= new_self_recording_hash
+            self.abbreviations[list][category][long] = short
+          end
+        end
+      end
+    end
+        
+    def abbreviate(category, name, list='default')
+      self.abbreviations[list][category][name]
+    end
+    
+    def items
+      @items ||= {}
+    end
+    
+    def import(items)
+      # TODO assign default ids if no id
+      items = to_a(items)
+      items.each do |item|
+        item = Item.new(item)
+        self.items[item['id'].to_s] = item
+      end
+    end
+    
+    def bibliography(selector = :all)
+      data = items.values.select(&Selector.new(selector)).map { |i| { 'id' => i.id } }
+      data = CitationData.new(data).populate!(items)
+      
+      data = style.bibliography.process(data, self)
+      Bibliography.new(data)
+    end
+    
+
+    #
+    # @param data Symbol :all / or id of item
+    # @param data String  id of item
+    # @param data Array list of ids or citation data
+    # @param data Hash citation data or citation items
+    #
+    # @returns a list of lists; [[1, 'Doe, 2000, p. 1'], ...]
+    #
+    def cite(data)
+      data = extract_citation_data(data)
+
+      data.populate!(items)
+      citation = style.citation.render(data, self)
+      
+      [[register(citation), citation]]
+    end
+
+    def nocite(ids, options={})
+      @bibliography + to_a(ids).map { |id| items[id] }
+    end
+
+    alias :make_bibliography :bibliography
+    alias :update_items :cite
+    alias :update_uncited_items :nocite
+    
+        
+    private
+    
+    def register(id)
+      1
+    end
+    
+    def to_a(attribute)
+      attribute.is_a?(Array) ? attribute : [attribute]
+    end
+    
+    # @returns a citation data object
+    def extract_citation_data(argument)
+      case
+      when argument == :all
+        argument = items.keys.map { |id| { 'id' => id } }
+              
+      when items.has_key?(argument.to_s)
+        argument = { 'id' => argument.to_s }
+
+      when argument.is_a?(Array) && items.has_key?(argument.first.to_s)
+        argument = argument.map { |id| { 'id' => id } }
+      
+      end
+
+      CitationData.new(argument)
+    end
+    
+    def new_abbreviations
+      { 'default' => new_self_recording_hash }
+    end
+    
+    def new_self_recording_hash
+      Hash.new { |h,k| h[k] = Hash.new { |h,k| h[k] = k } }
+    end
+  end
+  
+end