chronic 0.4.1 → 0.6.0

data/lib/chronic/chronic.rb

@@ -11,147 +11,306 @@ module Chronic
 
   class << self
 
-    # Parses a string containing a natural language date or time. If the parser
-    # can find a date or time, either a Time or Chronic::Span will be returned
-    # (depending on the value of <tt>:guess</tt>). If no date or time can be found,
-    # +nil+ will be returned.
+    # Parses a string containing a natural language date or time
     #
-    # Options are:
+    # If the parser can find a date or time, either a Time or Chronic::Span
+    # will be returned (depending on the value of `:guess`). If no
+    # date or time can be found, `nil` will be returned
     #
-    # [<tt>:context</tt>]
-    #     <tt>:past</tt> or <tt>:future</tt> (defaults to <tt>:future</tt>)
+    # @param [String] text The text to parse
     #
-    #     If your string represents a birthday, you can set <tt>:context</tt> to <tt>:past</tt>
-    #     and if an ambiguous string is given, it will assume it is in the
-    #     past. Specify <tt>:future</tt> or omit to set a future context.
+    # @option opts [Symbol] :context (:future)
+    #   * If your string represents a birthday, you can set `:context` to
+    #     `:past` and if an ambiguous string is given, it will assume it is
+    #     in the past. Specify `:future` or omit to set a future context.
     #
-    # [<tt>:now</tt>]
-    #     Time (defaults to Time.now)
+    # @option opts [Object] :now (Time.now)
+    #   * By setting `:now` to a Time, all computations will be based off of
+    #     that time instead of `Time.now`. If set to nil, Chronic will use
+    #     `Time.now`
     #
-    #     By setting <tt>:now</tt> to a Time, all computations will be based off
-    #     of that time instead of Time.now. If set to nil, Chronic will use Time.now.
+    # @option opts [Boolean] :guess (true)
+    #   * By default, the parser will guess a single point in time for the
+    #     given date or time. If you'd rather have the entire time span
+    #     returned, set `:guess` to `false` and a {Chronic::Span} will
+    #     be returned
     #
-    # [<tt>:guess</tt>]
-    #     +true+ or +false+ (defaults to +true+)
-    #
-    #     By default, the parser will guess a single point in time for the
-    #     given date or time. If you'd rather have the entire time span returned,
-    #     set <tt>:guess</tt> to +false+ and a Chronic::Span will be returned.
-    #
-    # [<tt>:ambiguous_time_range</tt>]
-    #     Integer or <tt>:none</tt> (defaults to <tt>6</tt> (6am-6pm))
-    #
-    #     If an Integer is given, ambiguous times (like 5:00) will be
+    # @option opts [Integer] :ambiguous_time_range (6)
+    #   * If an Integer is given, ambiguous times (like 5:00) will be
     #     assumed to be within the range of that time in the AM to that time
-    #     in the PM. For example, if you set it to <tt>7</tt>, then the parser will
-    #     look for the time between 7am and 7pm. In the case of 5:00, it would
-    #     assume that means 5:00pm. If <tt>:none</tt> is given, no assumption
-    #     will be made, and the first matching instance of that time will
-    #     be used.
+    #     in the PM. For example, if you set it to `7`, then the parser
+    #     will look for the time between 7am and 7pm. In the case of 5:00, it
+    #     would assume that means 5:00pm. If `:none` is given, no
+    #     assumption will be made, and the first matching instance of that
+    #     time will be used
     #
-    # [<tt>:endian_precedence</tt>]
-    #     Array (defaults to <tt>[:middle, :little]</tt>)
-    #
-    #     By default, Chronic will parse "03/04/2011" as the fourth day
+    # @option opts [Array] :endian_precedence ([:middle, :little])
+    #   * By default, Chronic will parse "03/04/2011" as the fourth day
     #     of the third month. Alternatively you can tell Chronic to parse
     #     this as the third day of the fourth month by altering the
-    #     <tt>:endian_precedence</tt> to <tt>[:little, :middle]</tt>.
-    def parse(text, specified_options = {})
-      @text = text
-      options = DEFAULT_OPTIONS.merge specified_options
+    #     `:endian_precedence` to `[:little, :middle]`
+    #
+    # @option opts [Integer] :ambiguous_year_future_bias (50)
+    #   * When parsing two digit years (ie 79) unlike Rubys Time class,
+    #     Chronic will attempt to assume the full year using this figure.
+    #     Chronic will look x amount of years into the future and past. If
+    #     the two digit year is `now + x years` it's assumed to be the
+    #     future, `now - x years` is assumed to be the past
+    #
+    # @return [Time, Chronic::Span, nil]
+    def parse(text, opts={})
+      options = DEFAULT_OPTIONS.merge opts
 
       # ensure the specified options are valid
-      (specified_options.keys-DEFAULT_OPTIONS.keys).each {|key| raise(InvalidArgumentException, "#{key} is not a valid option key.")}
+      (opts.keys - DEFAULT_OPTIONS.keys).each do |key|
+        raise ArgumentError, "#{key} is not a valid option key."
+      end
 
-      [:past, :future, :none].include?(options[:context]) || raise(InvalidArgumentException, "Invalid value ':#{options[:context]}' for :context specified. Valid values are :past and :future.")
+      unless [:past, :future, :none].include?(options[:context])
+        raise ArgumentError, "Invalid context, :past/:future only"
+      end
 
-      options[:now] ||= Chronic.time_class.now
-      @now = options[:now]
-
-      # put the text into a normal format to ease scanning
-      text = pre_normalize(text)
+      options[:text] = text
+      Chronic.now = options[:now] || Chronic.time_class.now
 
       # tokenize words
-      @tokens = tokenize(text, options)
+      tokens = tokenize(text, options)
 
       if Chronic.debug
-        puts "+---------------------------------------------------"
-        puts "| " + @tokens.to_s
-        puts "+---------------------------------------------------"
+        puts "+#{'-' * 51}\n| #{tokens}\n+#{'-' * 51}"
       end
 
-      span = tokens_to_span(@tokens, options)
+      span = tokens_to_span(tokens, options)
 
-      if options[:guess]
-        guess span
-      else
-        span
+      if span
+        options[:guess] ? guess(span) : span
       end
     end
 
-    # Clean up the specified input text by stripping unwanted characters,
-    # converting idioms to their canonical form, converting number words
-    # to numbers (three => 3), and converting ordinal words to numeric
+    # Clean up the specified text ready for parsing
+    #
+    # Clean up the string by stripping unwanted characters, converting
+    # idioms to their canonical form, converting number words to numbers
+    # (three => 3), and converting ordinal words to numeric
     # ordinals (third => 3rd)
-    def pre_normalize(text) #:nodoc:
-      normalized_text = text.to_s.downcase
-      normalized_text.gsub!(/['"\.,]/, '')
-      normalized_text.gsub!(/\bsecond (of|day|month|hour|minute|second)\b/, '2nd \1')
-      normalized_text = numericize_numbers(normalized_text)
-      normalized_text.gsub!(/ \-(\d{4})\b/, ' tzminus\1')
-      normalized_text.gsub!(/([\/\-\,\@])/) { ' ' + $1 + ' ' }
-      normalized_text.gsub!(/\b0(\d+:\d+\s*pm?\b)/, '\1')
-      normalized_text.gsub!(/\btoday\b/, 'this day')
-      normalized_text.gsub!(/\btomm?orr?ow\b/, 'next day')
-      normalized_text.gsub!(/\byesterday\b/, 'last day')
-      normalized_text.gsub!(/\bnoon\b/, '12:00')
-      normalized_text.gsub!(/\bmidnight\b/, '24:00')
-      normalized_text.gsub!(/\bbefore now\b/, 'past')
-      normalized_text.gsub!(/\bnow\b/, 'this second')
-      normalized_text.gsub!(/\b(ago|before)\b/, 'past')
-      normalized_text.gsub!(/\bthis past\b/, 'last')
-      normalized_text.gsub!(/\bthis last\b/, 'last')
-      normalized_text.gsub!(/\b(?:in|during) the (morning)\b/, '\1')
-      normalized_text.gsub!(/\b(?:in the|during the|at) (afternoon|evening|night)\b/, '\1')
-      normalized_text.gsub!(/\btonight\b/, 'this night')
-      normalized_text.gsub!(/\b\d+:?\d*[ap]\b/,'\0m')
-      normalized_text.gsub!(/(\d)([ap]m|oclock)\b/, '\1 \2')
-      normalized_text.gsub!(/\b(hence|after|from)\b/, 'future')
-      normalized_text
+    #
+    # @example
+    #   Chronic.pre_normalize('first day in May')
+    #     #=> "1st day in may"
+    #
+    #   Chronic.pre_normalize('tomorrow after noon')
+    #     #=> "next day future 12:00"
+    #
+    #   Chronic.pre_normalize('one hundred and thirty six days from now')
+    #     #=> "136 days future this second"
+    #
+    # @param [String] text The string to normalize
+    # @return [String] A new string ready for Chronic to parse
+    def pre_normalize(text)
+      text = text.to_s.downcase
+      text.gsub!(/['"\.]/, '')
+      text.gsub!(/,/, ' ')
+      text.gsub!(/\bsecond (of|day|month|hour|minute|second)\b/, '2nd \1')
+      text = Numerizer.numerize(text)
+      text.gsub!(/ \-(\d{4})\b/, ' tzminus\1')
+      text.gsub!(/([\/\-\,\@])/) { ' ' + $1 + ' ' }
+      text.gsub!(/\b0(\d+:\d+\s*pm?\b)/, '\1')
+      text.gsub!(/\btoday\b/, 'this day')
+      text.gsub!(/\btomm?orr?ow\b/, 'next day')
+      text.gsub!(/\byesterday\b/, 'last day')
+      text.gsub!(/\bnoon\b/, '12:00')
+      text.gsub!(/\bmidnight\b/, '24:00')
+      text.gsub!(/\bnow\b/, 'this second')
+      text.gsub!(/\b(?:ago|before(?: now)?)\b/, 'past')
+      text.gsub!(/\bthis (?:last|past)\b/, 'last')
+      text.gsub!(/\b(?:in|during) the (morning)\b/, '\1')
+      text.gsub!(/\b(?:in the|during the|at) (afternoon|evening|night)\b/, '\1')
+      text.gsub!(/\btonight\b/, 'this night')
+      text.gsub!(/\b\d+:?\d*[ap]\b/,'\0m')
+      text.gsub!(/(\d)([ap]m|oclock)\b/, '\1 \2')
+      text.gsub!(/\b(hence|after|from)\b/, 'future')
+      text
     end
 
-    # Convert number words to numbers (three => 3)
-    def numericize_numbers(text) #:nodoc:
+    # Convert number words to numbers (three => 3, fourth => 4th)
+    #
+    # @see Numerizer.numerize
+    # @param [String] text The string to convert
+    # @return [String] A new string with words converted to numbers
+    def numericize_numbers(text)
+      warn "Chronic.numericize_numbers will be deprecated in version 0.7.0. Please use Chronic::Numerizer.numerize instead"
       Numerizer.numerize(text)
     end
 
-    def tokenize(text, options) #:nodoc:
-      tokens = text.split(' ').map { |word| Token.new(word) }
-      [Repeater, Grabber, Pointer, Scalar, Ordinal, Separator, TimeZone].each do |tok|
-        tokens = tok.scan(tokens, options)
-      end
-      tokens.delete_if { |token| !token.tagged? }
-    end
-
     # Guess a specific time within the given span
-    def guess(span) #:nodoc:
-      return nil if span.nil?
+    #
+    # @param [Span] span
+    # @return [Time]
+    def guess(span)
       if span.width > 1
         span.begin + (span.width / 2)
       else
         span.begin
       end
     end
-  end
 
-  # Internal exception
-  class ChronicPain < Exception #:nodoc:
+    # List of {Handler} definitions. See {parse} for a list of options this
+    # method accepts
+    #
+    # @see parse
+    # @return [Hash] A Hash of Handler definitions
+    def definitions(options={})
+      options[:endian_precedence] ||= [:middle, :little]
 
-  end
+      @definitions ||= {
+        :time => [
+          Handler.new([:repeater_time, :repeater_day_portion?], nil)
+        ],
+
+        :date => [
+          Handler.new([:repeater_day_name, :repeater_month_name, :scalar_day, :repeater_time, :separator_slash_or_dash?, :time_zone, :scalar_year], :handle_rdn_rmn_sd_t_tz_sy),
+          Handler.new([:scalar_year, :separator_slash_or_dash, :scalar_month, :separator_slash_or_dash, :scalar_day, :repeater_time, :time_zone], :handle_sy_sm_sd_t_tz),
+          Handler.new([:repeater_month_name, :scalar_day, :scalar_year], :handle_rmn_sd_sy),
+          Handler.new([:repeater_month_name, :ordinal_day, :scalar_year], :handle_rmn_od_sy),
+          Handler.new([:repeater_month_name, :scalar_day, :scalar_year, :separator_at?, 'time?'], :handle_rmn_sd_sy),
+          Handler.new([:repeater_month_name, :ordinal_day, :scalar_year, :separator_at?, 'time?'], :handle_rmn_od_sy),
+          Handler.new([:repeater_month_name, :scalar_day, :separator_at?, 'time?'], :handle_rmn_sd),
+          Handler.new([:repeater_time, :repeater_day_portion?, :separator_on?, :repeater_month_name, :scalar_day], :handle_rmn_sd_on),
+          Handler.new([:repeater_month_name, :ordinal_day, :separator_at?, 'time?'], :handle_rmn_od),
+          Handler.new([:ordinal_day, :repeater_month_name, :scalar_year, :separator_at?, 'time?'], :handle_od_rmn_sy),
+          Handler.new([:ordinal_day, :repeater_month_name, :separator_at?, 'time?'], :handle_od_rmn),
+          Handler.new([:repeater_time, :repeater_day_portion?, :separator_on?, :repeater_month_name, :ordinal_day], :handle_rmn_od_on),
+          Handler.new([:repeater_month_name, :scalar_year], :handle_rmn_sy),
+          Handler.new([:scalar_day, :repeater_month_name, :scalar_year, :separator_at?, 'time?'], :handle_sd_rmn_sy),
+          Handler.new([:scalar_year, :separator_slash_or_dash, :scalar_month, :separator_slash_or_dash, :scalar_day, :separator_at?, 'time?'], :handle_sy_sm_sd),
+          Handler.new([:scalar_month, :separator_slash_or_dash, :scalar_year], :handle_sm_sy)
+        ],
+
+        # tonight at 7pm
+        :anchor => [
+          Handler.new([:grabber?, :repeater, :separator_at?, :repeater?, :repeater?], :handle_r),
+          Handler.new([:grabber?, :repeater, :repeater, :separator_at?, :repeater?, :repeater?], :handle_r),
+          Handler.new([:repeater, :grabber, :repeater], :handle_r_g_r)
+        ],
+
+        # 3 weeks from now, in 2 months
+        :arrow => [
+          Handler.new([:scalar, :repeater, :pointer], :handle_s_r_p),
+          Handler.new([:pointer, :scalar, :repeater], :handle_p_s_r),
+          Handler.new([:scalar, :repeater, :pointer, 'anchor'], :handle_s_r_p_a)
+        ],
 
-  # This exception is raised if an invalid argument is provided to
-  # any of Chronic's methods
-  class InvalidArgumentException < Exception
+        # 3rd week in march
+        :narrow => [
+          Handler.new([:ordinal, :repeater, :separator_in, :repeater], :handle_o_r_s_r),
+          Handler.new([:ordinal, :repeater, :grabber, :repeater], :handle_o_r_g_r)
+        ]
+      }
+
+      endians = [
+        Handler.new([:scalar_month, :separator_slash_or_dash, :scalar_day, :separator_slash_or_dash, :scalar_year, :separator_at?, 'time?'], :handle_sm_sd_sy),
+        Handler.new([:scalar_day, :separator_slash_or_dash, :scalar_month, :separator_slash_or_dash, :scalar_year, :separator_at?, 'time?'], :handle_sd_sm_sy)
+      ]
+
+      case endian = Array(options[:endian_precedence]).first
+      when :little
+        @definitions[:endian] = endians.reverse
+      when :middle
+        @definitions[:endian] = endians
+      else
+        raise ArgumentError, "Unknown endian option '#{endian}'"
+      end
 
+      @definitions
+    end
+
+    # Construct a time Object
+    #
+    # @return [Time]
+    def construct(year, month = 1, day = 1, hour = 0, minute = 0, second = 0)
+      if second >= 60
+        minute += second / 60
+        second = second % 60
+      end
+
+      if minute >= 60
+        hour += minute / 60
+        minute = minute % 60
+      end
+
+      if hour >= 24
+        day += hour / 24
+        hour = hour % 24
+      end
+
+      # determine if there is a day overflow. this is complicated by our crappy calendar
+      # system (non-constant number of days per month)
+      day <= 56 || raise("day must be no more than 56 (makes month resolution easier)")
+      if day > 28
+        # no month ever has fewer than 28 days, so only do this if necessary
+        leap_year_month_days = [31, 29, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]
+        common_year_month_days = [31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]
+        days_this_month = Date.leap?(year) ? leap_year_month_days[month - 1] : common_year_month_days[month - 1]
+        if day > days_this_month
+          month += day / days_this_month
+          day = day % days_this_month
+        end
+      end
+
+      if month > 12
+        if month % 12 == 0
+          year += (month - 12) / 12
+          month = 12
+        else
+          year += month / 12
+          month = month % 12
+        end
+      end
+
+      Chronic.time_class.local(year, month, day, hour, minute, second)
+    end
+
+    private
+
+    def tokenize(text, options)
+      text = pre_normalize(text)
+      tokens = text.split(' ').map { |word| Token.new(word) }
+      [Repeater, Grabber, Pointer, Scalar, Ordinal, Separator, TimeZone].each do |tok|
+        tok.scan(tokens, options)
+      end
+      tokens.select { |token| token.tagged? }
+    end
+
+    def tokens_to_span(tokens, options)
+      definitions = definitions(options)
+
+      definitions.each do |type, handlers|
+        handlers.each do |handler|
+          next unless handler.match(tokens, definitions)
+
+          good_tokens = case type
+          when :date, :endian, :anchor
+            tokens.reject { |o| o.get_tag Separator }
+          when :arrow
+            tokens.reject { |o| o.get_tag(SeparatorAt) || o.get_tag(SeparatorSlashOrDash) || o.get_tag(SeparatorComma) }
+          else
+            tokens
+          end
+
+          if handler.handler_method
+            return handler.invoke(type, good_tokens, options)
+          end
+
+        end
+      end
+
+      puts "-none" if Chronic.debug
+      return nil
+    end
+
+  end
+
+  # Internal exception
+  class ChronicPain < Exception
   end
 end

data/lib/chronic/grabber.rb

@@ -1,11 +1,20 @@
 module Chronic
-  class Grabber < Tag #:nodoc:
+  class Grabber < Tag
+
+    # Scan an Array of {Token}s and apply any necessary Grabber tags to
+    # each token
+    #
+    # @param [Array<Token>] tokens Array of tokens to scan
+    # @param [Hash] options Options specified in {Chronic.parse}
+    # @return [Array] list of tokens
     def self.scan(tokens, options)
-      tokens.each_index do |i|
-        if t = scan_for_all(tokens[i]) then tokens[i].tag(t); next end
+      tokens.each do |token|
+        if t = scan_for_all(token) then token.tag(t); next end
       end
     end
 
+    # @param [Token] token
+    # @return [Grabber, nil]
     def self.scan_for_all(token)
       scan_for token, self,
       {

data/lib/chronic/handler.rb

@@ -0,0 +1,90 @@
+module Chronic
+  class Handler
+
+    # @return [Array] A list of patterns
+    attr_reader :pattern
+
+    # @return [Symbol] The method which handles this list of patterns.
+    #   This method should exist inside the {Handlers} module
+    attr_reader :handler_method
+
+    # @param [Array]  pattern A list of patterns to match tokens against
+    # @param [Symbol] handler_method The method to be invoked when patterns
+    #   are matched. This method should exist inside the {Handlers} module
+    def initialize(pattern, handler_method)
+      @pattern = pattern
+      @handler_method = handler_method
+    end
+
+    # @param [Array] tokens
+    # @param [Hash]  definitions
+    # @return [Boolean]
+    # @see Chronic.tokens_to_span
+    def match(tokens, definitions)
+      token_index = 0
+
+      @pattern.each do |element|
+        name = element.to_s
+        optional = name[-1, 1] == '?'
+        name = name.chop if optional
+
+        case element
+        when Symbol
+          if tags_match?(name, tokens, token_index)
+            token_index += 1
+            next
+          else
+            if optional
+              next
+            else
+              return false
+            end
+          end
+        when String
+          return true if optional && token_index == tokens.size
+
+          if definitions.key?(name.to_sym)
+            sub_handlers = definitions[name.to_sym]
+          else
+            raise ChronicPain, "Invalid subset #{name} specified"
+          end
+
+          sub_handlers.each do |sub_handler|
+            return true if sub_handler.match(tokens[token_index..tokens.size], definitions)
+          end
+        else
+          raise ChronicPain, "Invalid match type: #{element.class}"
+        end
+      end
+
+      return false if token_index != tokens.size
+      return true
+    end
+
+    def invoke(type, tokens, options)
+      if Chronic.debug
+        puts "-#{type}"
+        puts "Handler: #{@handler_method}"
+      end
+
+      Handlers.send(@handler_method, tokens, options)
+    end
+
+    # @param [Handler]  The handler to compare
+    # @return [Boolean] True if these handlers match
+    def ==(other)
+      @pattern == other.pattern
+    end
+
+    private
+
+    def tags_match?(name, tokens, token_index)
+      klass = Chronic.const_get(name.to_s.gsub(/(?:^|_)(.)/) { $1.upcase })
+
+      if tokens[token_index]
+        !tokens[token_index].tags.select { |o| o.kind_of?(klass) }.empty?
+      end
+    end
+
+  end
+end