pseudo_date 0.1.0

data/Manifest

@@ -0,0 +1,11 @@
+Manifest
+README.mdown
+Rakefile
+lib/core_extensions/object.rb
+lib/core_extensions/string.rb
+lib/pseudo_date.rb
+lib/pseudo_date/parser.rb
+lib/pseudo_date/pseudo_date.rb
+pseudo_date.gemspec
+test/test_helper.rb
+test/test_parser.rb

data/README.mdown

@@ -0,0 +1,69 @@
+# Pseudo Date
+
+## What is a Pseudo Date?
+
+It's a date but not really.  A PseudoDate object has a day, month, and year but it does not require all of them like the built in ruby date classes.  This allows you to parse obscure date strings that may or may not be complete.
+
+## What Is This For?
+
+PseudoDate was created to parse odd dates in odd formats and attempt to extract as much information from them as possible.  It's especially handy when you're trying to convert a date string that has come from an OCR'd source.
+
+## Assumptions
+
+As with all parsing, one needs to make assumptions.  The main assumption made here is that all dates will be in the past.
+
+## Other Notes
+
+PseudoDate stores date attributes in strings instead of integers to avoid losing the preceding '0' on various attributes.  This was a decision made when first creating the class because of the way things were being output in the project it was created for.
+
+## Compatability
+
+PseudoDates are not really compatible with other built-in date/time objects.  They do support some of the basic methods for abstracting numbers though.  PseudoDates that are of exact precision can be turned into ruby date objects.
+
+	>> p = PseudoDate.new('19850625')
+	 => #<PseudoDate:0x10190eff0 @month="06", @date_hash={:day=>"25", :month=>"06", :year=>"1985"}, year"1985", day"25"
+	>> p.precision
+	 => "exact" 
+	>> p.year
+	 => "1985" 
+	>> p.month
+	 => "06" 
+	>> p.day
+	 => "25"
+	>> p.to_date
+	 => #<Date: 4892483/2,0,2299161>
+
+## Examples
+
+	>> PseudoDate.new('19850625').to_hash
+	 => {:day=>"25", :month=>"06", :year=>"1985"}
+	
+	>> PseudoDate.new('1985-25-06').to_hash
+	 => {:day=>"25", :month=>"06", :year=>"1985"}
+	
+	>> PseudoDate.new('06-25-1985').to_hash
+	 => {:day=>"25", :month=>"06", :year=>"1985"}
+	
+	>> PseudoDate.new('25-06-1985').to_hash
+	 => {:day=>"25", :month=>"06", :year=>"1985"}
+	
+	>> PseudoDate.new('06/25/1985').to_hash
+	 => {:day=>"25", :month=>"06", :year=>"1985"}
+	
+	>> PseudoDate.new('06/1985').to_hash
+	 => {:day=>"00", :month=>"06", :year=>"1985"}
+	
+	>> PseudoDate.new('85').to_hash
+	 => {:day=>"00", :month=>"00", :year=>"1985"}
+	
+	>> PseudoDate.new('1985').to_hash
+	 => {:day=>"00", :month=>"00", :year=>"1985"}
+	
+	>> PseudoDate.new('Jun 25, 1985').to_hash
+	 => {:day=>"25", :month=>"06", :year=>"1985"}
+
+## Patches, Bugfixes, Additions...
+
+Feel free to fork and add stuff as you see fit.  This was written quickly to solve some problems so it could certainly use some more structure and organization.
+
+If you add something or fix something, make sure you run the very basic tests and add any new ones as required by your changes.

data/Rakefile

@@ -0,0 +1,13 @@
+require 'rubygems'
+require 'rake'
+require 'echoe'
+
+Echoe.new('pseudo_date', '0.1.0') do |p|
+  p.description = 'Date parser and container for partial or incomplete dates.'
+  p.url = 'http://github.com/PatrickTulskie/pseudo_date'
+  p.author = 'Patrick Tulskie'
+  p.email = 'PatrickTulskie@gmail.com'
+  p.ignore_pattern = ['tmp/*', 'script/*', 'lib/main.rb']
+end
+
+Dir["#{File.dirname(__FILE__)}/tasks/*.rake"].sort.each { |ext| load ext }

data/lib/core_extensions/object.rb

@@ -0,0 +1,7 @@
+class Object
+  
+  def is_numeric?(n) 
+    n.nil? ? false : Integer(n) rescue false
+  end
+  
+end

data/lib/core_extensions/string.rb

@@ -0,0 +1,11 @@
+class String
+  
+  def to_date_hash
+    Parser.parse(self)
+  end
+    
+  def to_pseudo_date
+    PseudoDate.new(self)
+  end
+  
+end

data/lib/pseudo_date/parser.rb

@@ -0,0 +1,90 @@
+require 'date'
+class Parser
+  
+  def self.parse(input)
+    date_hash = {}
+    # Minor Pre Cleanup
+    input.strip!; input.gsub!('~','')
+    
+    date = input.split(/\/|-/).length < 3 ? nil : Date.parse(input) rescue nil
+    if date
+      date_hash = { :year => date.year.to_s, :month => date.month.to_s, :day => date.day.to_s }
+    else
+      year, month, day = parse_string(input)
+      date_hash = { :year => year, :month => month, :day => day }
+    end
+    
+    # Post parsing cleanup
+    date_hash.each do |key, value|
+      date_hash[key] = if value.nil?
+        key.to_s == 'year' ? '0000' : '00'
+      else
+        date_hash[key] = value.to_s.strip
+      end
+    end
+    
+    # Cleanup the single digit values
+    unless date_hash.empty?
+      date_hash.each do |key,value|
+        date_hash[key] = "0#{value}" if value.to_s.length == 1
+      end
+    end
+    
+    # Two character years
+    if date_hash[:year].length == 2
+      date_hash[:year] = date_hash[:year].to_i > Date.today.year.to_s.slice(2..4).to_i ? "19#{date_hash[:year]}" : "20#{date_hash[:year]}"
+    end
+    
+    # Attempt to correct some known OCR issues
+    if date_hash[:year].to_s.match('00') && date_hash[:year] != '0000'
+      date_hash[:year] = "2#{date_hash[:year].slice(1..3)}"
+    end
+    
+    return date_hash.empty? ? nil : date_hash
+  end
+  
+  private
+  
+  def self.parse_string(input)
+    day, month, year = "00", "00", "0000"
+    if input.match('/') # 02/25/2008
+      date_array = input.split('/')
+      if date_array.length == 3
+        begin
+          parsed_date = Date.parse(self)
+          month, day, year = parsed_date.month, parsed_date.day, parsed_date.year
+        rescue
+          month, day, year = date_array
+        end
+      elsif date_array.length == 2
+        month, year = date_array
+      end
+    elsif input.length == 8 && is_numeric?(input) # 20080225
+      year, month, day = input.slice(0..3), input.slice(4..5), input.slice(6..7)
+    elsif input.match('-') # 1985-09-25 or 02-25-2008
+      date_array = input.split('-')
+      year = date_array.select{ |part| part.length == 4 }.first
+      unless year.nil? || date_array.length != 3
+        if date_array.first == year
+          month = date_array.last
+          day = date_array[1]
+        else
+          month = date_array.first
+          day = date_array[1]
+        end
+        month, day = [day, month] if month.to_i > 12 && month.to_i > day.to_i
+      end
+    elsif input.length == 4 # 2004
+      year = input.to_s if (input.slice(0..1) == '19' || input.slice(0..1) == '20')
+    elsif input.length == 2 # 85
+      year = (input.to_i > Date.today.year.to_s.slice(2..4).to_i) ? "19#{input}" : "20#{input}"
+    elsif input.match(/\w/) # Jun 23, 2004
+      begin
+        d = Date.parse(input)
+        year, month, day = d.year.to_s, d.month.to_s, d.day.to_s
+      rescue; end
+    end
+    return [year, month, day]
+  end
+  
+end

data/lib/pseudo_date/pseudo_date.rb

@@ -0,0 +1,75 @@
+class PseudoDate
+  attr_accessor :year, :day, :month, :date_hash
+  
+  def initialize(input)
+    @date_hash = if input.is_a?(Hash)
+      input.to_hash.symbolize_keys
+      # But it's already a hash!  Well, sometimes silly people pass a Mash object into this without realizing it.
+    else
+      input = '00000000' if input.to_s.strip == '19000000'
+      @date_hash = input.to_s.strip.to_date_hash
+    end
+    if @date_hash
+      @year = @date_hash[:year].to_s.match('1900') ? '0000' : @date_hash[:year].to_s.strip
+      @month = @date_hash[:month].to_s.strip
+      @day = @date_hash[:day].to_s.strip
+    else
+      @year = @month = @day = nil
+    end
+    if @year && @year.match('~') && @year.length == 3
+      @year = @year.to_i
+      @year = @year > Date.today.year.to_s.gsub('20','').to_i ? "19#{@year}" : "20#{@year}"
+    end
+    correct_digits
+    @year.to_s.gsub!('~','')
+  end
+  
+  def precision
+    correct_digits
+    if @year.nil? || (@year.to_s == '0000' && @month.to_s == '00') || (@year.to_s == "8888")
+      "invalid"
+    elsif self.to_date
+      'exact'
+    elsif @month != '00' && @day == '00'
+      'year_month'
+    elsif @month == '00' && @day == '00'
+      'year'
+    else
+      'invalid'
+    end
+  end
+  
+  def valid?
+    !(@date_hash.nil? || @date_hash.empty?)
+  end
+  
+  def to_date
+    self.valid? ? Date.parse("#{@year}-#{@month}-#{@day}") : nil rescue nil
+  end
+  
+  def to_s
+    return "" unless self.valid?
+    
+    case self.precision
+    when 'invalid'; ""
+    when 'weak_year'; ""
+    when 'exact'; "#{month}/#{day}/#{year}"
+    when 'year_month'; "#{month}/#{year}"
+    when 'year'; year
+    else; ''
+    end
+  end
+  
+  def to_hash
+    @date_hash
+  end
+  
+  private
+  
+  def correct_digits
+    @year = '0000' if @year.to_s.strip.length == 0
+    @month = '00' if @month.to_s.strip.length == 0
+    @day = '00' if @day.to_s.strip.length == 0
+  end
+  
+end

data/lib/pseudo_date.rb

@@ -0,0 +1,4 @@
+require 'pseudo_date/pseudo_date'
+require 'pseudo_date/parser'
+require 'core_extensions/string'
+require 'core_extensions/object'

data/pseudo_date.gemspec

@@ -0,0 +1,31 @@
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = %q{pseudo_date}
+  s.version = "0.1.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 1.2") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Patrick Tulskie"]
+  s.date = %q{2010-01-08}
+  s.description = %q{Date parser and container for partial or incomplete dates.}
+  s.email = %q{PatrickTulskie@gmail.com}
+  s.extra_rdoc_files = ["README.mdown", "lib/core_extensions/object.rb", "lib/core_extensions/string.rb", "lib/pseudo_date.rb", "lib/pseudo_date/parser.rb", "lib/pseudo_date/pseudo_date.rb"]
+  s.files = ["Manifest", "README.mdown", "Rakefile", "lib/core_extensions/object.rb", "lib/core_extensions/string.rb", "lib/pseudo_date.rb", "lib/pseudo_date/parser.rb", "lib/pseudo_date/pseudo_date.rb", "test/test_helper.rb", "test/test_parser.rb", "pseudo_date.gemspec"]
+  s.homepage = %q{http://github.com/PatrickTulskie/pseudo_date}
+  s.rdoc_options = ["--line-numbers", "--inline-source", "--title", "Pseudo_date", "--main", "README.mdown"]
+  s.require_paths = ["lib"]
+  s.rubyforge_project = %q{pseudo_date}
+  s.rubygems_version = %q{1.3.5}
+  s.summary = %q{Date parser and container for partial or incomplete dates.}
+  s.test_files = ["test/test_helper.rb", "test/test_parser.rb"]
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
+    else
+    end
+  else
+  end
+end

data/test/test_helper.rb

@@ -0,0 +1,4 @@
+require 'rubygems'
+require 'test/unit'
+require 'shoulda'
+require File.dirname(__FILE__) + '/../lib/pseudo_date'

data/test/test_parser.rb

@@ -0,0 +1,137 @@
+require File.dirname(__FILE__) + '/test_helper.rb'
+
+class TestParser < Test::Unit::TestCase
+  
+  context "Date formats" do
+    
+    setup do
+      @day = '25'
+      @month = '06'
+      @year = '1985'
+      @string_date = 'Jun 25, 1985'
+    end
+    
+    # 19850625
+    context "yearmonthday" do
+      should 'be exact precision' do
+        assert_equal PseudoDate.new("#{@year}#{@month}#{@day}").precision, 'exact'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("#{@year}#{@month}#{@day}")
+        assert_equal @day, pd.day
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 1985-25-06
+    context "year-day-month" do
+      should 'be exact precision' do
+        assert_equal PseudoDate.new("#{@year}-#{@day}-#{@month}").precision, 'exact'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("#{@year}-#{@day}-#{@month}")
+        assert_equal @day, pd.day
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 06-25-1985
+    context "month-day-year" do
+      should 'be exact precision' do
+        assert_equal PseudoDate.new("#{@month}-#{@day}-#{@year}").precision, 'exact'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("#{@month}-#{@day}-#{@year}")
+        assert_equal @day, pd.day
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 25-06-1985
+    context "day-month-year" do
+      should 'be exact precision' do
+        assert_equal PseudoDate.new("#{@day}-#{@month}-#{@year}").precision, 'exact'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("#{@day}-#{@month}-#{@year}")
+        assert_equal @day, pd.day
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 06/25/1985
+    context "month/day/year" do
+      should 'be exact precision' do
+        assert_equal PseudoDate.new("#{@month}/#{@day}/#{@year}").precision, 'exact'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("#{@month}/#{@day}/#{@year}")
+        assert_equal @day, pd.day
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 06/1985
+    context "month/year" do
+      should 'be partial precision' do
+        assert_equal PseudoDate.new("#{@month}/#{@year}").precision, 'year_month'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("#{@month}/#{@year}")
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 85
+    context "two digit year" do
+      should 'be year precision' do
+        assert_equal PseudoDate.new("85").precision, 'year'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new("85")
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # 1985
+    context "four digit year" do
+      should 'be year precision' do
+        assert_equal PseudoDate.new(@year).precision, 'year'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new(@year)
+        assert_equal @year, pd.year
+      end
+    end
+    
+    # Jun 25, 1985
+    context "string date" do
+      should 'be exact precision' do
+        assert_equal PseudoDate.new(@string_date).precision, 'exact'
+      end
+      
+      should 'match original input' do
+        pd = PseudoDate.new(@string_date)
+        assert_equal @day, pd.day
+        assert_equal @month, pd.month
+        assert_equal @year, pd.year
+      end
+    end
+  
+  end
+  
+end

metadata

@@ -0,0 +1,76 @@
+--- !ruby/object:Gem::Specification 
+name: pseudo_date
+version: !ruby/object:Gem::Version 
+  version: 0.1.0
+platform: ruby
+authors: 
+- Patrick Tulskie
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-01-08 00:00:00 -05:00
+default_executable: 
+dependencies: []
+
+description: Date parser and container for partial or incomplete dates.
+email: PatrickTulskie@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- README.mdown
+- lib/core_extensions/object.rb
+- lib/core_extensions/string.rb
+- lib/pseudo_date.rb
+- lib/pseudo_date/parser.rb
+- lib/pseudo_date/pseudo_date.rb
+files: 
+- Manifest
+- README.mdown
+- Rakefile
+- lib/core_extensions/object.rb
+- lib/core_extensions/string.rb
+- lib/pseudo_date.rb
+- lib/pseudo_date/parser.rb
+- lib/pseudo_date/pseudo_date.rb
+- test/test_helper.rb
+- test/test_parser.rb
+- pseudo_date.gemspec
+has_rdoc: true
+homepage: http://github.com/PatrickTulskie/pseudo_date
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --line-numbers
+- --inline-source
+- --title
+- Pseudo_date
+- --main
+- README.mdown
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "1.2"
+  version: 
+requirements: []
+
+rubyforge_project: pseudo_date
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: Date parser and container for partial or incomplete dates.
+test_files: 
+- test/test_helper.rb
+- test/test_parser.rb