parse 0.0.1.1 → 0.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 7d04d59e0d0c1eadc1d912f109697f28b18eb6bc
-  data.tar.gz: c0207dac3109de5a81e8265e0e339138392f76a9
+  metadata.gz: 321f7a0116ae940b829a45faedb2fe731b3ada97
+  data.tar.gz: 6599e616962df9eac3ff27f31f4d9f11418270ab
 SHA512:
-  metadata.gz: 2047276d42adf38a1b31af30d8610ed1daae0254765c4051ed381a15da44504b345f90cf83f57fc83bdfab84dd89700dc21793f8eecd06c5f31148453ae85d25
-  data.tar.gz: 392dc0723a262777f89086320a34c9ae50eec4197fce5f174ac087cd91b29538c660bb96a4d7fe0b39f108dc853ce32956aa59eae7a44aed91bf47f39a913365
+  metadata.gz: a1d0fc63cabe88dcea9ae7f1049c99288782ffe4ebcaa575a2ae30b22f5f84aa4466223158b152d07d24118af84792eac77e40425c62690f25fe36692f9bafb9
+  data.tar.gz: 8faa4572ebe3ae57960e983596c98324c25188e576d1cd5d1a15d41561ae0818cbda6d564d3c603f6ef043081ca8cd550034bbcda8d7e072180a524dac4e1794

data/CHANGELOG

@@ -1,3 +1,15 @@
+0.1.0
+
+* Breaking changes
+
+  * Not parsed as numbers anymore: "10_14" or "10_400" or "-1_2e3_5" or "-$123_456.7" -- very unlikely to see in the wild
+  * Not parsed as numbers anymore: '1,2' or '-1,2' or '1,20' or '1,22.0' -- too hard to distinguish from CSV
+
+* Enhancements
+
+  * Using proper versioning scheme - every version of algorithm involves a minor bump!
+  * Recover from YAML parsing errors in 1.9
+
 0.0.1.1 / 2014-02-06
 
 * Bug fixes

data/README.md

@@ -4,10 +4,12 @@ Detect and convert short strings into integers, floats, dates, times, booleans,
 
 ## Note on versions
 
-You can always use `Parse.parse`. It will always point to the most recent version of the algorithm (currently `Parse.ver0_0_1`).
+You can always use `Parse.parse`. It will always point to the most recent version of the algorithm (currently `Parse.ver0_1_0`).
 
 If the algorithm changes and you need the old version, you can reference it by its version number. For example, `Parse.ver0_0_1`.
 
+Since almost any change to the algorithm is a breaking change, there are going to be lots of minor version bumps (as opposed to patches).
+
 ## Usage
 
 You get the idea:
@@ -39,26 +41,24 @@ More esoteric stuff:
     Parse.parse("-")                             #=> nil
     Parse.parse("?")                             #=> nil
     Parse.parse("-8e-05")                        #=> -8.0e-05
-    Parse.parse("-1_2.5e-1_3")                   #=> -1.25e-12
+    Parse.parse("-12.5e-13")                   #=> -1.25e-12
     Parse.parse("05753")                         #=> 5753
-    Parse.parse("15_000")                        #=> 15000
-    Parse.parse("15_00_0")                       #=> 15000
+    Parse.parse("15000")                        #=> 15000
     Parse.parse("15.0")                          #=> 15.0
     Parse.parse("15,000.0")                      #=> 15000.0
-    Parse.parse("15_000.0")                      #=> 15000.0
-    Parse.parse("15_00_0.0")                     #=> 15000.0
+    Parse.parse("15000.0")                      #=> 15000.0
     Parse.parse("0015")                          #=> 15
     Parse.parse("0015.0")                        #=> 15.0
-    Parse.parse("0_015.0")                       #=> 15.0
+    Parse.parse("0015.0")                       #=> 15.0
     Parse.parse("0x15")                          #=> 21
     Parse.parse("0o15")                          #=> 13
     Parse.parse("8e-05")                         #=> 8.0e-05
-    Parse.parse("1_2.5e-1_3")                    #=> 1.25e-12
+    Parse.parse("12.5e-13")                    #=> 1.25e-12
     Parse.parse("0$123.4")                       #=> 123.4
     Parse.parse("$15,000")                       #=> 15000
     Parse.parse("0$15,000")                      #=> 15000
-    Parse.parse("$123_456")                      #=> 123456
-    Parse.parse("$123_456.7")                    #=> 123456.7
+    Parse.parse("$123456")                      #=> 123456
+    Parse.parse("$123456.7")                    #=> 123456.7
     Parse.parse("10,000,000")                    #=> 10000000
     Parse.parse("10,000,000.00")                 #=> 10000000.0
     Parse.parse("$10,000,000.00")                #=> 10000000.0
@@ -66,15 +66,13 @@ More esoteric stuff:
     Parse.parse("$010,000,000.00")               #=> 10000000.0
     Parse.parse("-15")                           #=> -15
     Parse.parse("-15,000")                       #=> -15000
-    Parse.parse("-15_000")                       #=> -15000
-    Parse.parse("-15_00_0")                      #=> -15000
+    Parse.parse("-15000")                       #=> -15000
     Parse.parse("-15.0")                         #=> -15.0
     Parse.parse("-15,000.0")                     #=> -15000.0
-    Parse.parse("-15_000.0")                     #=> -15000.0
-    Parse.parse("-15_00_0.0")                    #=> -15000.0
+    Parse.parse("-15000.0")                     #=> -15000.0
+    Parse.parse("-15000.0")                    #=> -15000.0
     Parse.parse("00-15")                         #=> -15
     Parse.parse("00-15.0")                       #=> -15.0
-    Parse.parse("0_0-15.0")                      #=> "0_0-15.0"
     Parse.parse("-0x15")                         #=> -21
     Parse.parse("-0o15")                         #=> -13
     Parse.parse("-$123.4")                       #=> -123.4
@@ -82,10 +80,10 @@ More esoteric stuff:
     Parse.parse("0($123.4)")                     #=> -123.4
     Parse.parse("-$15,000")                      #=> -15000
     Parse.parse("($15,000)")                     #=> -15000
-    Parse.parse("-$123_456")                     #=> -123456
-    Parse.parse("($123_456)")                    #=> -123456
-    Parse.parse("-$123_456.7")                   #=> -123456.7
-    Parse.parse("($123_456.7)")                  #=> -123456.7
+    Parse.parse("-$123,456")                     #=> -123456
+    Parse.parse("($123,456)")                    #=> -123456
+    Parse.parse("-$123,456.7")                   #=> -123456.7
+    Parse.parse("($123,456.7)")                  #=> -123456.7
     Parse.parse("-10,000,000")                   #=> -10000000
     Parse.parse("(10,000,000)")                  #=> -10000000
     Parse.parse("-10,000,000.00")                #=> -10000000.0

data/lib/parse.rb

@@ -1,102 +1,24 @@
 require "parse/version"
+require 'parse/algorithm'
+require 'parse/algorithm/ver0_0_1'
+require 'parse/algorithm/ver0_1_0'
 
 require 'date'
 require 'yaml'
 require 'safe_yaml/load'
+require 'active_support/core_ext'
 
 module Parse
-  # only need to deal with stuff not caught by YAML or JSON
-  NULL = [ '', '-', '?', 'N/A', 'n/a', 'NULL', 'null', '#REF!', '#NAME?', 'NIL', 'nil', 'NA', 'na', '#VALUE!', '#NULL!'] # from bigml's list
-  NAN = [ 'NaN' ]
-  INFINITY = [ '#DIV/0', 'Infinity' ]
-  NEG_INFINITY = [ '-Infinity' ]
-  DATE = {
-    euro: ['%d-%m-%Y', '%d-%m-%y'],
-    us:   ['%m-%d-%Y', '%m-%d-%y'],
-  }
-
   def self.parse(raw, options = nil)
-    ver0_0_1 raw, options
+    ver0_1_0 raw, options
   end
 
-  # @private
-  # use YAML to parse stuff like '1.5'
-  # ruby's yaml is 1.1, which means it does weird stuff with '001' (fixed in 1.2, which jruby has)
-  def self.ver0_0_1(raw, options = nil)
-    return raw unless raw.is_a? String
-    
-    memo = raw.strip
-  
-    return nil if NULL.include? memo
-    return 1.0/0 if INFINITY.include? memo
-    return -1.0/0 if NEG_INFINITY.include? memo
-    return 0.0/0 if NAN.include? memo
-    
-    if options and options[:date]
-      yyyy, yy = DATE.fetch options[:date]
-      memo.sub!(/0+/, '')
-      memo.gsub! '/', '-'
-      if memo =~ /\d{4,}/ # yyyy
-        return Date.strptime(memo, yyyy)
-      else
-        return Date.strptime(memo, yy)
-      end
-    end
-
-    not_numeric = nil
-    not_numeric ||= memo =~ /,\d{1,2},/ # comma not used for thousands, like 10,20,30
-    not_numeric ||= memo =~ /\..*,/ # comma following a period, like 1.0,2
-    not_numeric ||= memo =~ /\A[^(+\-\$0-9%]/ # starts with letter or smth
-    possible_numeric = !not_numeric
-    accounting_negative = nil
-    percentage = nil
-
-    if possible_numeric
-      accounting_negative = memo =~ /\A[0$]*\([0$]*/
-      percentage = memo.end_with?('%')
-      memo.sub! /%\z/, '' if percentage
-      memo.delete!('()') if accounting_negative # accounting negative
-      # in yaml 1.1, anything starting with zero is treated as octal... in 1.2, it's 0o
-      memo.sub!(/0+/, '') if memo =~ /\A[+\-]?0+[+\-\$]?[1-9]+/ # leading zeros
-      memo.delete!('$') if memo =~ /\A[+\-]?0*\$/
-      if memo.include?(',')
-        a, b = memo.split('.', 2)
-        a.delete! ','
-        memo = b ? [a, b].join('.') : a
-      end
-    end
-
-    not_safe_for_yaml = nil
-    not_safe_for_yaml ||= memo.include?('#')
-    not_safe_for_yaml ||= not_numeric && memo =~ /\A[\d,]+\z/ #1,2,3, maybe a csv
-    safe_for_yaml = !not_safe_for_yaml
-
-    if safe_for_yaml
-      begin
-        memo = SafeYAML.load memo
-      rescue
-        $stderr.puts "#{memo.inspect} => #{$!}"
-      end
-    end
-
-    if possible_numeric
-      case memo
-      when /\A[+\-]?[\d._]+[eE][+\-]?[\d._]+\z/
-        # scientific notation
-        memo = memo.to_f
-      when /\A[+\-]?0o/
-        # octal per yaml 1.2
-        memo = memo.to_i 8
-      end
-    end
-    
-    if memo.is_a?(String)
-      # compress whitespace
-      memo.gsub! /\s+/, ' '
-    end
+  def self.ver0_1_0(*args)
+    Algorithm::Ver0_1_0.new(*args).result
+  end
 
-    memo = memo / 100.0 if percentage
-    memo = -memo if accounting_negative
-    memo
+  def self.ver0_0_1(*args)
+    Algorithm::Ver0_0_1.new(*args).result
   end
+
 end

data/lib/parse/algorithm.rb

@@ -0,0 +1,8 @@
+module Parse
+  module Algorithm
+    # only need to deal with stuff not caught by YAML or JSON
+    NAN = [ 'NaN' ]
+    INFINITY = [ '#DIV/0', 'Infinity' ]
+    NEG_INFINITY = [ '-Infinity' ]
+  end
+end

data/lib/parse/algorithm/ver0_0_1.rb

@@ -0,0 +1,99 @@
+module Parse
+  module Algorithm
+    class Ver0_0_1
+      NULL = [ '', '-', '?', 'N/A', 'n/a', 'NULL', 'null', '#REF!', '#NAME?', 'NIL', 'nil', 'NA', 'na', '#VALUE!', '#NULL!'] # from bigml's list
+      DATE = {
+        euro: ['%d-%m-%Y', '%d-%m-%y'],
+        us:   ['%m-%d-%Y', '%m-%d-%y'],
+      }
+
+      attr_reader :raw
+      attr_reader :options
+      def initialize(raw, options = nil)
+        @raw = raw
+        @options = options
+      end
+
+      # @private
+      # use YAML to parse stuff like '1.5'
+      # ruby's yaml is 1.1, which means it does weird stuff with '001' (fixed in 1.2, which jruby has)
+      def result
+        return raw unless raw.is_a? String
+        
+        memo = raw.strip
+      
+        return nil if NULL.include? memo
+        return 1.0/0 if INFINITY.include? memo
+        return -1.0/0 if NEG_INFINITY.include? memo
+        return 0.0/0 if NAN.include? memo
+        
+        if options and options[:date]
+          yyyy, yy = DATE.fetch options[:date]
+          memo.sub!(/0+/, '')
+          memo.gsub! '/', '-'
+          if memo =~ /\d{4,}/ # yyyy
+            return Date.strptime(memo, yyyy)
+          else
+            return Date.strptime(memo, yy)
+          end
+        end
+
+        not_numeric = nil
+        not_numeric ||= memo =~ /,\d{1,2},/ # comma not used for thousands, like 10,20,30
+        not_numeric ||= memo =~ /\..*,/ # comma following a period, like 1.0,2
+        not_numeric ||= memo =~ /\A[^(+\-\$0-9%]/ # starts with letter or smth
+        possible_numeric = !not_numeric
+        accounting_negative = nil
+        percentage = nil
+
+        if possible_numeric
+          accounting_negative = memo =~ /\A[0$]*\([0$]*/
+          percentage = memo.end_with?('%')
+          memo.sub! /%\z/, '' if percentage
+          memo.delete!('()') if accounting_negative # accounting negative
+          # in yaml 1.1, anything starting with zero is treated as octal... in 1.2, it's 0o
+          memo.sub!(/0+/, '') if memo =~ /\A[+\-]?0+[+\-\$]?[1-9]+/ # leading zeros
+          memo.delete!('$') if memo =~ /\A[+\-]?0*\$/
+          if memo.include?(',')
+            a, b = memo.split('.', 2)
+            a.delete! ','
+            memo = b ? [a, b].join('.') : a
+          end
+        end
+
+        not_safe_for_yaml = nil
+        not_safe_for_yaml ||= memo.include?('#')
+        not_safe_for_yaml ||= not_numeric && memo =~ /\A[\d,]+\z/ #1,2,3, maybe a csv
+        safe_for_yaml = !not_safe_for_yaml
+
+        if safe_for_yaml
+          begin
+            memo = SafeYAML.load memo
+          rescue
+            $stderr.puts "#{memo.inspect} => #{$!}"
+          end
+        end
+
+        if possible_numeric
+          case memo
+          when /\A[+\-]?[\d._]+[eE][+\-]?[\d._]+\z/
+            # scientific notation
+            memo = memo.to_f
+          when /\A[+\-]?0o/
+            # octal per yaml 1.2
+            memo = memo.to_i 8
+          end
+        end
+        
+        if memo.is_a?(String)
+          # compress whitespace
+          memo.gsub! /\s+/, ' '
+        end
+
+        memo = memo / 100.0 if percentage
+        memo = -memo if accounting_negative
+        memo
+      end
+    end
+  end
+end

data/lib/parse/algorithm/ver0_1_0.rb

@@ -0,0 +1,149 @@
+module Parse
+  module Algorithm
+    class Ver0_1_0
+      NULL = [ '', '-', '?', 'N/A', 'n/a', 'NULL', 'null', '#REF!', '#NAME?', 'NIL', 'nil', 'NA', 'na', '#VALUE!', '#NULL!', '00/00/00', '0000-00-00'] # from bigml's list
+      REGION_DATE_FORMAT = {
+        euro: ['%d-%m-%Y', '%d-%m-%y'],
+        us:   ['%m-%d-%Y', '%m-%d-%y'],
+        iso:  ['%Y-%m-%d', '%y-%m-%d'], # second one is silly
+      }
+      DATE_DETECT = {
+        %r{\A0*[12]\d\d\d[\-/](?:(?:0[1-9])|(?:1[0-2]))[\-/][1-9]\d\z} => :iso, # $1 will be delimiter
+      }
+      EMPTY_OPTIONS = {}
+
+      attr_reader :raw
+      attr_reader :options
+      def initialize(raw, options = nil)
+        @raw = raw
+        @options = options || EMPTY_OPTIONS
+      end
+
+      def result
+        return raw unless raw.is_a? String
+        
+        memo = raw.strip
+      
+        return nil if NULL.include? memo
+        return 1.0/0 if INFINITY.include? memo
+        return -1.0/0 if NEG_INFINITY.include? memo
+        return 0.0/0 if NAN.include? memo
+
+        date_region = if options[:date]
+          options[:date]
+        else
+          catch :hit do
+            DATE_DETECT.each do |pattern, date_region|
+              # binding.pry if memo.include?('2011-')
+              if memo =~ pattern
+                throw :hit, date_region
+              end
+            end
+            nil
+          end
+        end
+
+        if date_region.nil? and options[:type] == Date
+          date_region = :iso
+        end
+
+        if date_region
+          yyyy, yy = REGION_DATE_FORMAT.fetch date_region
+          is_yyyy = memo =~ /[1-9]\d\d\d/
+          memo.sub! /\A0+/, ''
+          memo.gsub! '/', '-'
+          if is_yyyy
+            if memo.length < 10 and date_region == :iso
+              return Date.parse(memo)
+            else
+              return Date.strptime(memo, yyyy)
+            end
+          else
+            return Date.strptime(memo, yy)
+          end
+        end
+
+        possible_numeric = nil
+        not_numeric = nil
+        certain_numeric = nil
+        if [Numeric, Integer, Float].include?(options[:type])
+          certain_numeric = true
+          possible_numeric = true
+          not_numeric = false
+        else
+          # not_numeric ||= memo =~ /[1-9][^)\d_,%.eE]/ # has a dash in the middle
+          not_numeric ||= memo.include?('_')
+          not_numeric ||= memo =~ %r{[1-9][/-]\d}
+          not_numeric ||= memo =~ /,\d{1,2}(?:[.\D]|\z)/
+          not_numeric ||= memo.scan(/[^\d_,%.eE]/).length > memo.scan(/[\d_,%.eE]/).length
+          not_numeric ||= memo =~ /\A[^(+\-\$0-9%]/ # starts with letter or smth
+          possible_numeric = !not_numeric
+        end
+        accounting_negative = nil
+        percentage = nil
+        if possible_numeric
+          accounting_negative = memo =~ /\A[0$]*\([0$]*/
+          percentage = memo.end_with?('%')
+          memo.sub! /%\z/, '' if percentage
+          memo.delete!('()') if accounting_negative # accounting negative
+          # in yaml 1.1, anything starting with zero is treated as octal... in 1.2, it's 0o
+          memo.sub!(/0+/, '') if memo =~ /\A[+\-]?0+[+\-\$]?[1-9]+/ # leading zeros
+          memo.delete!('$') if memo =~ /\A[+\-]?0*\$/
+          memo.sub!('D', 'e') if memo =~ /\A[+\-]?[\d.]+D[+\-]?[\d.]+\z/ # fortran double precision
+          if memo.include?(',')
+            a, b = memo.split('.', 2)
+            a.delete! ','
+            memo = b ? [a, b].join('.') : a
+          end
+        end
+
+        if certain_numeric
+          memo.gsub! /[a-z]/i, ''
+        end
+
+        not_safe_for_yaml = nil
+        not_safe_for_yaml ||= memo =~ /\A(on|off)\z/i
+        not_safe_for_yaml ||= memo.include?('#')
+        not_safe_for_yaml ||= memo =~ /\A[@&,]/
+        not_safe_for_yaml ||= not_numeric && memo.start_with?('0')
+        not_safe_for_yaml ||= not_numeric && memo =~ /\A[^{\[]*\d[,_]/ #1,2,3, maybe a csv
+
+        safe_for_yaml = !not_safe_for_yaml
+
+        if safe_for_yaml
+          begin
+            memo = SafeYAML.load memo
+          rescue Exception # Psych::SyntaxError will blow up plain rescue in 1.9.3
+            $stderr.puts "#{memo.inspect} => #{$!}"
+          end
+        end
+
+        if possible_numeric
+          case memo
+          when /\A[+\-]?[\d._]+[eE][+\-]?[\d._]+\z/
+            # scientific notation
+            memo = memo.to_f
+          when /\A[+\-]?0o/
+            # octal per yaml 1.2
+            memo = memo.to_i 8
+          end
+        end
+        
+        if memo.is_a?(String)
+          # compress whitespace
+          memo.gsub! /\s+/, ' '
+        end
+
+        memo = memo / 100.0 if percentage
+        memo = -memo if accounting_negative
+        memo
+      rescue
+        if options and options[:ignore_error]
+          # nothing to see here
+        else
+          raise "#{memo.inspect} => #{$!}"
+        end
+      end
+    end
+  end
+end

data/lib/parse/version.rb

@@ -1,3 +1,3 @@
 module Parse
-  VERSION = "0.0.1.1"
+  VERSION = "0.1.0"
 end

data/parse.gemspec

@@ -18,12 +18,13 @@ Gem::Specification.new do |spec|
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ["lib"]
 
-  spec.add_runtime_dependency 'safe_yaml'
+  spec.add_runtime_dependency 'safe_yaml', '>=1'
+  spec.add_runtime_dependency 'activesupport'
 
   spec.add_development_dependency "bundler", "~> 1.5"
   spec.add_development_dependency "rake"
   spec.add_development_dependency "rspec"
   spec.add_development_dependency 'multi_json'
-  spec.add_development_dependency 'activesupport'
   spec.add_development_dependency 'pry'
+  # spec.add_development_dependency 'twitter_cldr'
 end

data/spec/parse_spec.rb

@@ -1,8 +1,8 @@
 require 'spec_helper'
 
 describe Parse do
-  it "should parse with version 0.0.1 of the algorithm" do
+  it "should parse with version 0.0.2 of the algorithm" do
     v = "  1990-04-03  "
-    expect(Parse.parse(v)).to eq(Parse.ver0_0_1(v))
+    expect(Parse.parse(v)).to eq(Parse.ver0_1_0(v))
   end
 end

data/spec/parse_ver0_0_1_spec.rb

@@ -1,6 +1,6 @@
 require 'spec_helper'
 
-describe Parse do
+describe Parse::Algorithm::Ver0_0_1 do
   {
     ''                                 => nil,
     'nil'                              => nil,

data/spec/parse_ver0_1_0_spec.rb

@@ -0,0 +1,277 @@
+require 'spec_helper'
+require 'twitter_cldr'
+
+describe Parse::Algorithm::Ver0_1_0 do
+  same = [
+    '1,2',
+    '1,20',
+    '1,2.0',
+    '-1,2',
+    '-1,20',
+    '-1,2.0',
+    '01,2',
+    '01,20',
+    '01,2.0',
+    '15_000',
+    '15_00_0',
+    '15_000.0',
+    '15_00_0.0',
+    '0_015.0', # just weird
+    '1_2.5e-1_3',
+    '-1_2.5e-1_3',
+    '$123_456',
+    '$123_456.7',
+    '-15_000',
+    '-15_00_0',
+    '-15_000.0',
+    '-15_00_0.0',
+    '0_0-15.0', # just weird
+    '-$123_456',
+    '($123_456)',
+    '-$123_456.7',
+    '($123_456.7)',
+    '10_14_A',
+    '10_14',
+    '10_140',
+  ]
+  same.each do |v|
+    it "parses #{v.inspect} as itself (yaml=#{SafeYAML.load(v).inspect})" do
+      expect(Parse.ver0_1_0(v)).to eq(v)
+    end
+  end
+
+  a = {
+
+    "@ foo"                            => "@ foo",
+    ", foo"                            => ", foo",
+    "044-1-276-000"                    => "044-1-276-000",
+
+    ['1 BEDROOMS', { type: Numeric } ] => 1,
+    '1 BEDROOMS'                       => '1 BEDROOMS',
+
+    [ '2.4 SQFT', { type: Numeric } ]  => 2.4,
+    '2.4 SQFT'                         => '2.4 SQFT',
+
+    ['000', { date: :us, ignore_error: true}] => nil,
+    ['7/7/2004', {date: :us}]           => Date.new(2004,7,7),
+    "999 HOLY CROSS ROAD, COLCHESTER, VT 05446" => "999 HOLY CROSS ROAD, COLCHESTER, VT 05446",
+
+    '00020110628'                      => 20110628,
+    '0002011-06-28'                    => Date.new(2011,6,28),
+    '0002011/06/28'                    => Date.new(2011,6,28),
+    ['00020110628', {date: :iso}]      => Date.new(2011,6,28),
+    ['00020110628', {type: Date}]      => Date.new(2011,6,28),
+
+    '00019800628'                      => 19800628,
+    '0001980-06-28'                    => Date.new(1980,6,28),
+    '0001980/06/28'                    => Date.new(1980,6,28),
+    ['00019800628', {date: :iso}]      => Date.new(1980,6,28),
+    ['00019800628', {type: Date}]      => Date.new(1980,6,28),
+
+    '00030000628'                      => 30000628,
+    '0003000-06-28'                    => '0003000-06-28',
+    '0003000/06/28'                    => '0003000/06/28',
+    ['00030000628', {date: :iso}]      => Date.new(3000,6,28),
+    ['00030000628', {type: Date}]      => Date.new(3000,6,28),
+
+    ['', {type: Numeric}]              => nil,
+
+    # fortran double precision
+    '0.225120000000000D+06'            => 0.22512e6,
+    '0.341913000000000D+07'            => 0.341913e7,
+    '0.2500000E-01'                    => 0.25e-1,
+    '3.1D0'                            => 3.1,
+    '-2.D0'                            => -2.0,
+
+    '8e-05'                            => 8e-5,
+    '8e+4'                             => 8e4,
+    '8.0e+4'                           => 8.0e4,
+    '8e-4'                             => 8e-4,
+    '8.0e-4'                           => 8.0e-4,
+    '-8e+4'                            => -8e4,
+    '-8.0e+4'                          => -8.0e4,
+    '-8e-4'                            => -8e-4,
+    '-8.0e-4'                          => -8.0e-4,
+    '8E+4'                             => 8e4,
+    '8.0E+4'                           => 8.0e4,
+    '8E-4'                             => 8e-4,
+    '8.0E-4'                           => 8.0e-4,
+    '-8E+4'                            => -8e4,
+    '-8.0E+4'                          => -8.0e4,
+    '-8E-4'                            => -8e-4,
+    '-8.0E-4'                          => -8.0e-4,
+
+    # http://dojotoolkit.org/reference-guide/1.9/dojo/number.html
+    # '1,000,000.00'                     => 1_000_000.0,
+    # '1.000.000,00'                     => 1_000_000.0, # german
+    # '1 000 000,00'                     => 1_000_000.0, # french
+    # '10,00,000.00'                     => 1_000_000.0, # indian
+
+    '060-10-01'                        => '60-10-01',
+    'OFF'                              => 'OFF',
+    'ON'                               => 'ON',
+
+    '& P4'                             => '& P4',
+
+    # EVERYTHING BELOW IS SAME AS 0.0.1
+    
+    ''                                 => nil,
+    'nil'                              => nil,
+    '15'                               => 15,
+    '15,000'                           => 15_000,
+    '15.0'                             => 15.0,
+    '15,000.0'                         => 15_000.0,
+    '0015'                             => 15,   # not octal
+    '0015.0'                           => 15.0,  # not octal
+    '0x15'                             => 0x15, # hex
+    '0o15'                             => 015,  # octal
+    '8e-05'                            => 8e-05,
+    '12.5e-13'                       => 12.5e-13,
+    '-12.5e-13'                      => -12.5e-13,
+    '$123.4'                           => 123.4,
+    '0$123.4'                           => 123.4,
+    '$15,000'                          => 15_000,
+    '0$15,000'                          => 15_000,
+    '10,000,000'                       => 10_000_000,
+    '10,000,000.00'                    => 10_000_000.0,
+    '$10,000,000.00'                    => 10_000_000.0,
+    '0$10,000,000.00'                    => 10_000_000.0,
+    '$010,000,000.00'                    => 10_000_000.0,
+
+    '-15'                              => -15,
+    '-15,000'                          => -15_000,
+    '-15.0'                            => -15.0,
+    '-15,000.0'                        => -15_000.0,
+    '00-15'                            => -15,   # not octal
+    '00-15.0'                          => -15.0,  # not octal
+    '-0x15'                            => -0x15, # hex
+    '-0o15'                            => -015,  # octal
+    '-8e-05'                           => -8e-05,
+    '-$123.4'                          => -123.4,
+    '($123.4)'                         => -123.4,
+    '0($123.4)'                        => -123.4,
+    '-$15,000'                         => -15_000,
+    '($15,000)'                        => -15_000,
+    '-$123456'                        => -123_456,
+    '($123456)'                       => -123_456,
+    '-$123456.7'                      => -123_456.7,
+    '($123456.7)'                     => -123_456.7,
+    '-$123,456'                        => -123_456,
+    '($123,456)'                       => -123_456,
+    '-$123,456.7'                      => -123_456.7,
+    '($123,456.7)'                     => -123_456.7,
+    '-10,000,000'                      => -10_000_000,
+    '(10,000,000)'                     => -10_000_000,
+    '-10,000,000.00'                   => -10_000_000.0,
+    '(10,000,000.00)'                  => -10_000_000.0,
+    '-10000000'                        => -10_000_000,
+    '(10000000)'                       => -10_000_000,
+    '-10000000.00'                     => -10_000_000.0,
+    '(10000000.00)'                    => -10_000_000.0,
+    '1,200'                            => 1_200,
+    '1,200.0'                          => 1_200.0,
+    '1.0,2'                            => '1.0,2',
+    '1.0,2.0'                          => '1.0,2.0',
+    '-1,200'                           => -1_200,
+    '-1,200.0'                         => -1_200.0,
+    '-1.0,2'                           => '-1.0,2',
+    '-1.0,2.0'                         => '-1.0,2.0',
+    '01,200'                           => 1_200,
+    '01,200.0'                         => 1_200.0,
+    '01.0,2'                           => '01.0,2',
+    '01.0,2.0'                         => '01.0,2.0',
+
+    '05753'                            => 5753,
+    'true'                             => true,
+    'yes'                              => true,
+    'false'                            => false,
+    'no'                               => false,
+    '#DIV/0'                           => (1.0/0),
+    '#NAME?'                           => nil,
+    'Inf'                              => 'Inf',
+    'Infinity'                         => (1.0/0),
+    '-Infinity'                        => -(1.0/0),
+    'NaN'                              => 0.0/0, # need the dot
+    '.NaN'                             => 0.0/0,  # NaN
+    '-.inf'                            => -(1.0/0), # -Infinity
+    '-'                                => nil, # per bigml
+    '?'                                => nil,
+    '1982-01-01'                       => Date.new(1982,1,1),
+    '2010-05-05 13:42:16 Z'            => Time.parse('2010-05-05 13:42:16 Z'),
+    '2010-05-05 13:42:16 -02:00'       => Time.parse('2010-05-05 13:42:16 -02:00'),
+    ":not_a_symbol"                    => ':not_a_symbol',
+    '#hello'                           => '#hello',
+    "\n#hello\n#world"                 => '#hello #world',
+    "hello\nworld"                     => 'hello world', # whitespace compression
+
+    '0%'                               => 0.0,
+    '100%'                             => 1.0,
+    '50%'                              => 0.5,
+    '5%'                               => 0.05,
+    '00000%'                           => 0.0,
+    '0000100%'                         => 1.0,
+    '000050%'                          => 0.5,
+    '00005%'                           => 0.05,
+
+    ['12/25/82', {date: :us}]          => Date.new(1982,12,25),
+    ['12/25/1982', {date: :us}]        => Date.new(1982,12,25),
+    ['25/12/82', {date: :euro}]        => Date.new(1982,12,25),
+    ['25/12/1982', {date: :euro}]      => Date.new(1982,12,25),
+    ['12-25-82', {date: :us}]          => Date.new(1982,12,25),
+    ['12-25-1982', {date: :us}]        => Date.new(1982,12,25),
+    ['25-12-82', {date: :euro}]        => Date.new(1982,12,25),
+    ['25-12-1982', {date: :euro}]      => Date.new(1982,12,25),
+
+    '12/25/82'                         => '12/25/82',
+
+    ',1'                               => ',1', # not a csv parser
+    ',1,'                              => ',1,', # not a csv parser
+    '1,2,3'                            => '1,2,3', # not a csv parser
+    '[1,2,3]'                          => [1,2,3],
+    YAML.dump('a' => 1)                => { 'a' => 1 },
+    YAML.dump(a: 1)                    => { ':a' => 1 }, # doesn't parse symbols
+    YAML.dump('a' => 1, 5 => "c\n3")   => { 'a' => 1, 5 => "c\n3" },
+    MultiJson.dump(a: 1)               => { 'a' => 1 }, # json always loses symbols
+    MultiJson.dump(a: 1, 5 => "c\n3")  => { 'a' => 1, '5' => "c\n3" },
+  }
+
+  # TwitterCldr.supported_locales.each do |locale|
+  #   1.upto(9).map do |power|
+  #     num = (rand * (10 ** power)).round(4)
+  #     # a[[num.localize(locale).to_s, {locale: locale}]] = num
+  #     a[[num.localize(locale).to_s, { locale: locale }]] = num
+  #     # a[num.localize(locale).to_currency.to_s] = num
+  #   end
+  # end
+
+  # and next dates!
+  # Time.now.localize(:es).to_full_s
+
+  a.each do |input, expected|
+    input = Array.wrap input
+    locale = if input[1].is_a?(Hash)
+      input[1][:locale]
+    end
+    it "#{locale ? "(#{locale}) " : nil}parses #{input[0].inspect} as #{expected.inspect}" do
+      got = Parse.ver0_1_0(*input)
+      # $lines << [ "Parse.parse(#{input.inspect})".ljust(45), "#=> #{got.inspect}" ].join
+      if expected.is_a?(Float) and expected.nan?
+        expect(got.nan?).to eq(true)
+      elsif expected.is_a?(Float) and got.is_a?(Float)
+        expect(got.round(8)).to eq(expected.round(8))
+      else
+        expect(got).to eq(expected)
+      end
+
+      input_with_spaces = [ "\t" + input[0] + "\t", input[1] ]
+      got_with_spaces = Parse.ver0_1_0(*input_with_spaces)
+      if expected.is_a?(Float) and expected.nan?
+        expect(got.nan?).to eq(true)
+      elsif expected.is_a?(Float) and got.is_a?(Float)
+        expect(got.round(8)).to eq(expected.round(8))
+      else
+        expect(got_with_spaces).to eq(expected)
+      end
+    end
+  end
+end

metadata

@@ -1,17 +1,31 @@
 --- !ruby/object:Gem::Specification
 name: parse
 version: !ruby/object:Gem::Version
-  version: 0.0.1.1
+  version: 0.1.0
 platform: ruby
 authors:
 - Seamus Abshere
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-02-07 00:00:00.000000000 Z
+date: 2014-03-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: safe_yaml
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '1'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '1'
+- !ruby/object:Gem::Dependency
+  name: activesupport
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -80,20 +94,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-- !ruby/object:Gem::Dependency
-  name: activesupport
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
 - !ruby/object:Gem::Dependency
   name: pry
   requirement: !ruby/object:Gem::Requirement
@@ -125,10 +125,14 @@ files:
 - README.md
 - Rakefile
 - lib/parse.rb
+- lib/parse/algorithm.rb
+- lib/parse/algorithm/ver0_0_1.rb
+- lib/parse/algorithm/ver0_1_0.rb
 - lib/parse/version.rb
 - parse.gemspec
 - spec/parse_spec.rb
 - spec/parse_ver0_0_1_spec.rb
+- spec/parse_ver0_1_0_spec.rb
 - spec/spec_helper.rb
 homepage: https://github.com/seamusabshere/parse
 licenses:
@@ -150,7 +154,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.2.1
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: Detect and convert short strings into integers, floats, dates, times, booleans,
@@ -158,4 +162,5 @@ summary: Detect and convert short strings into integers, floats, dates, times, b
 test_files:
 - spec/parse_spec.rb
 - spec/parse_ver0_0_1_spec.rb
+- spec/parse_ver0_1_0_spec.rb
 - spec/spec_helper.rb