chronic 0.4.1 → 0.4.2

data/HISTORY.md

@@ -1,4 +1,11 @@
-# 0.4.1
+# 0.4.2 / 2011-06-07
+
+* Fix MonthRepeater for fetching past month offsets when current day is
+  later than the last day of a past month (ie on 29th of March when parsing
+  `last month` Chronic would return March instead of February. Now Chronic
+  returns the last day of the past month)
+
+# 0.4.1 / 2011-06-05
 
 * Fix MiniDate ranges for parsing seasons (Thomas Walpole)
 
@@ -27,7 +34,7 @@
 * Fix undefined variable in RepeaterHour (Ryan Garver)
 * Added support for parsing 'Fourty' as another mis-spelling (Lee Reilly)
 * Added ordinal format support: ie 'February 14th, 2004' (Jeff Felchner)
-* Fix dates when working with daylight saving times Mike Mangino
+* Fix dates when working with daylight saving times (Mike Mangino)
 
 # 0.3.0 / 2010-10-22
 

data/Rakefile

@@ -54,14 +54,6 @@ task :coverage do
   sh "open coverage/index.html"
 end
 
-require 'rake/rdoctask'
-Rake::RDocTask.new do |rdoc|
-  rdoc.rdoc_dir = 'rdoc'
-  rdoc.title = "#{name} #{version}"
-  rdoc.rdoc_files.include('README*')
-  rdoc.rdoc_files.include('lib/**/*.rb')
-end
-
 desc "Open an irb session preloaded with this library"
 task :console do
   sh "irb -Ilib -rchronic"

data/chronic.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name              = 'chronic'
-  s.version           = '0.4.1'
+  s.version           = '0.4.2'
   s.rubyforge_project = 'chronic'
 
   s.summary     = "Natural language date/time parsing."

data/lib/chronic.rb

@@ -1,33 +1,82 @@
-#=============================================================================
+require 'time'
+require 'date'
+
+# Parse natural language dates and times into Time or {Chronic::Span} objects
 #
-#  Name:       Chronic
-#  Author:     Tom Preston-Werner
-#  Purpose:    Parse natural language dates and times into Time or
-#              Chronic::Span objects
+# @example
+#   require 'chronic'
 #
-#=============================================================================
-
+#   Time.now   #=> Sun Aug 27 23:18:25 PDT 2006
+#
+#   Chronic.parse('tomorrow')
+#     #=> Mon Aug 28 12:00:00 PDT 2006
+#
+#   Chronic.parse('monday', :context => :past)
+#     #=> Mon Aug 21 12:00:00 PDT 2006
+#
+#   Chronic.parse('this tuesday 5:00')
+#     #=> Tue Aug 29 17:00:00 PDT 2006
+#
+#   Chronic.parse('this tuesday 5:00', :ambiguous_time_range => :none)
+#     #=> Tue Aug 29 05:00:00 PDT 2006
+#
+#   Chronic.parse('may 27th', :now => Time.local(2000, 1, 1))
+#     #=> Sat May 27 12:00:00 PDT 2000
+#
+#   Chronic.parse('may 27th', :guess => false)
+#     #=> Sun May 27 00:00:00 PDT 2007..Mon May 28 00:00:00 PDT 2007
+#
+# @author Tom Preston-Werner, Lee Jarvis
 module Chronic
-  VERSION = "0.4.1"
+  VERSION = "0.4.2"
 
   class << self
+
+    # @return [Boolean] true when debug mode is enabled
     attr_accessor :debug
+
+    # @example
+    #   require 'chronic'
+    #   require 'active_support/time'
+    #
+    #   Time.zone = 'UTC'
+    #   Chronic.time_class = Time.zone
+    #   Chronic.parse('June 15 2006 at 5:54 AM')
+    #     # => Thu, 15 Jun 2006 05:45:00 UTC +00:00
+    #
+    # @return [Time] The time class Chronic uses internally
     attr_accessor :time_class
+
+    # The current Time Chronic is using to base from
+    #
+    # @example
+    #   Time.now #=> 2011-06-06 14:13:43 +0100
+    #   Chronic.parse('yesterday') #=> 2011-06-05 12:00:00 +0100
+    #
+    #   now = Time.local(2025, 12, 24)
+    #   Chronic.parse('tomorrow', :now => now) #=> 2025-12-25 12:00:00 +0000
+    #
+    # @return [Time, nil]
+    attr_accessor :now
   end
 
   self.debug = false
   self.time_class = Time
 end
 
-require 'time'
-require 'date'
-
 require 'chronic/chronic'
 require 'chronic/handlers'
 require 'chronic/mini_date'
 require 'chronic/tag'
 require 'chronic/span'
 require 'chronic/token'
+require 'chronic/grabber'
+require 'chronic/pointer'
+require 'chronic/scalar'
+require 'chronic/ordinal'
+require 'chronic/separator'
+require 'chronic/time_zone'
+require 'chronic/numerizer'
 
 require 'chronic/repeater'
 require 'chronic/repeaters/repeater_year'
@@ -47,42 +96,6 @@ require 'chronic/repeaters/repeater_minute'
 require 'chronic/repeaters/repeater_second'
 require 'chronic/repeaters/repeater_time'
 
-require 'chronic/grabber'
-require 'chronic/pointer'
-require 'chronic/scalar'
-require 'chronic/ordinal'
-require 'chronic/separator'
-require 'chronic/time_zone'
-
-require 'chronic/numerizer'
-
-# class Time
-#   def self.construct(year, month = 1, day = 1, hour = 0, minute = 0, second = 0)
-#     # extra_seconds = second > 60 ? second - 60 : 0
-#     # extra_minutes = minute > 59 ? minute - 59 : 0
-#     # extra_hours = hour > 23 ? hour - 23 : 0
-#     # extra_days = day >
-#
-#     if month > 12
-#       if month % 12 == 0
-#         year += (month - 12) / 12
-#         month = 12
-#       else
-#         year += month / 12
-#         month = month % 12
-#       end
-#     end
-#
-#     base = Time.local(year, month)
-#     puts base
-#     offset = ((day - 1) * 24 * 60 * 60) + (hour * 60 * 60) + (minute * 60) + second
-#     puts offset.to_s
-#     date = base + offset
-#     puts date
-#     date
-#   end
-# end
-
 class Time
   def self.construct(year, month = 1, day = 1, hour = 0, minute = 0, second = 0)
     if second >= 60

data/lib/chronic/chronic.rb

@@ -11,76 +11,76 @@ 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>)
+    # @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.
     #
-    #     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 [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`
     #
-    # [<tt>:now</tt>]
-    #     Time (defaults to 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
     #
-    #     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.
-    #
-    # [<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 = {})
+    #     `: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={})
       @text = text
-      options = DEFAULT_OPTIONS.merge specified_options
+      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 InvalidArgumentException, "#{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 InvalidArgumentException, "Invalid context, :past/:future only"
+      end
 
+      options[:text] = text
       options[:now] ||= Chronic.time_class.now
-      @now = options[:now]
-
-      # put the text into a normal format to ease scanning
-      text = pre_normalize(text)
+      Chronic.now = options[: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
@@ -89,52 +89,66 @@ module Chronic
       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)
+    #
+    # @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) #: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
+      text = text.to_s.downcase
+      text.gsub!(/['"\.,]/, '')
+      text.gsub!(/\bsecond (of|day|month|hour|minute|second)\b/, '2nd \1')
+      text = numericize_numbers(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!(/\bbefore now\b/, 'past')
+      text.gsub!(/\bnow\b/, 'this second')
+      text.gsub!(/\b(ago|before)\b/, 'past')
+      text.gsub!(/\bthis past\b/, 'last')
+      text.gsub!(/\bthis last\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)
       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:
+    #
+    # @param [Span] span
+    # @return [Time]
+    def guess(span)
       return nil if span.nil?
       if span.width > 1
         span.begin + (span.width / 2)
@@ -142,16 +156,133 @@ module Chronic
         span.begin
       end
     end
+
+    # 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]
+
+      @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([: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([: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)
+        ],
+
+        # 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 InvalidArgumentException, "Unknown endian option '#{endian}'"
+      end
+
+      @definitions
+    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|
+        tokens = tok.scan(tokens, options)
+      end
+      tokens.select { |token| token.tagged? }
+    end
+
+    def tokens_to_span(tokens, options) #:nodoc:
+      definitions = definitions(options)
+
+      (definitions[:date] + definitions[:endian]).each do |handler|
+        if handler.match(tokens, definitions)
+          puts "-date" if Chronic.debug
+          good_tokens = tokens.select { |o| !o.get_tag Separator }
+          return Handlers.send(handler.handler_method, good_tokens, options)
+        end
+      end
+
+      definitions[:anchor].each do |handler|
+        if handler.match(tokens, definitions)
+          puts "-anchor" if Chronic.debug
+          good_tokens = tokens.select { |o| !o.get_tag Separator }
+          return Handlers.send(handler.handler_method, good_tokens, options)
+        end
+      end
+
+      definitions[:arrow].each do |handler|
+        if handler.match(tokens, definitions)
+          puts "-arrow" if Chronic.debug
+          tags = [SeparatorAt, SeparatorSlashOrDash, SeparatorComma]
+          good_tokens = tokens.reject { |o| tags.any? { |t| o.get_tag(t) } }
+          return Handlers.send(handler.handler_method, good_tokens, options)
+        end
+      end
+
+      definitions[:narrow].each do |handler|
+        if handler.match(tokens, definitions)
+          puts "-narrow" if Chronic.debug
+          good_tokens = tokens.select { |o| !o.get_tag Separator }
+          return Handlers.send(handler.handler_method, tokens, options)
+        end
+      end
+
+      puts "-none" if Chronic.debug
+      return nil
+    end
+
   end
 
   # Internal exception
-  class ChronicPain < Exception #:nodoc:
-
+  class ChronicPain < Exception
   end
 
   # This exception is raised if an invalid argument is provided to
   # any of Chronic's methods
   class InvalidArgumentException < Exception
-
   end
 end