Merge branch 'master' into kl_abbr_numbers_pluralization

Conflicts: lib/twitter_cldr/formatters/numbers/abbreviated/abbreviated_number_formatter.rb resources/locales/bn/numbers.yml resources/locales/de-CH/numbers.yml resources/locales/de/numbers.yml resources/locales/es-CO/numbers.yml resources/locales/es-MX/numbers.yml resources/locales/es-US/numbers.yml resources/locales/es/numbers.yml resources/locales/eu/numbers.yml resources/locales/fr-CA/numbers.yml resources/locales/ga/numbers.yml resources/locales/he/numbers.yml resources/locales/hi/numbers.yml resources/locales/it-CH/numbers.yml resources/locales/it/numbers.yml resources/locales/ko/numbers.yml resources/locales/sr/numbers.yml resources/locales/ur/numbers.yml spec/formatters/numbers/abbreviated/long_decimal_formatter_spec.rb
twitter · Oct 9, 2014 · ca8c9cb · ca8c9cb
2 parents 4ba989c + ca98ed8
commit ca8c9cb
Show file tree

Hide file tree

Showing 105 changed files with 9,392 additions and 5,925 deletions.
diff --git a/README.md.erb b/README.md.erb
@@ -39,18 +39,18 @@ TwitterCldr patches core Ruby objects like `Fixnum` and `Date` to make localizat
 
 ```ruby
 # default formatting with to_s
-1337.localize(:es).to_s                                    # <%= assert(1337.localize(:es).to_s, "1 337").inspect %>
+1337.localize(:es).to_s                                    # <%= assert(1337.localize(:es).to_s, "1.337").inspect %>
 
 # currencies, default USD
-1337.localize(:es).to_currency.to_s                        # <%= assert(1337.localize(:es).to_currency.to_s, "1 337,00 $").inspect %>
-1337.localize(:es).to_currency.to_s(:currency => "EUR")    # <%= assert(1337.localize(:es).to_currency.to_s(:currency => "EUR"), "1 337,00 €").inspect %>
+1337.localize(:es).to_currency.to_s                        # <%= assert(1337.localize(:es).to_currency.to_s, "1.337,00 $").inspect %>
+1337.localize(:es).to_currency.to_s(:currency => "EUR")    # <%= assert(1337.localize(:es).to_currency.to_s(:currency => "EUR"), "1.337,00 €").inspect %>
 
 # percentages
-1337.localize(:es).to_percent.to_s                         # <%= assert(1337.localize(:es).to_percent.to_s, "1 337%").inspect %>
-1337.localize(:es).to_percent.to_s(:precision => 2)        # <%= assert(1337.localize(:es).to_percent.to_s(:precision => 2), "1 337,00%").inspect %>
+1337.localize(:es).to_percent.to_s                         # <%= assert(1337.localize(:es).to_percent.to_s, "1.337 %").inspect %>
+1337.localize(:es).to_percent.to_s(:precision => 2)        # <%= assert(1337.localize(:es).to_percent.to_s(:precision => 2), "1.337,00 %").inspect %>
 
 # decimals
-1337.localize(:es).to_decimal.to_s(:precision => 3)        # <%= assert(1337.localize(:es).to_decimal.to_s(:precision => 3) ,"1 337,000").inspect %>
+1337.localize(:es).to_decimal.to_s(:precision => 3)        # <%= assert(1337.localize(:es).to_decimal.to_s(:precision => 3) ,"1.337,000").inspect %>
 ```
 
 **Note**: The `:precision` option can be used with all these number formatters.
@@ -62,7 +62,7 @@ num = TwitterCldr::Localized::LocalizedNumber.new(1337, :es)
 num.to_currency.to_s  # ...etc
 ```
 
-<% assert(TwitterCldr::Localized::LocalizedNumber.new(1337, :es).to_currency.to_s, "1 337,00 $") %>
+<% assert(TwitterCldr::Localized::LocalizedNumber.new(1337, :es).to_currency.to_s, "1.337,00 $") %>
 
 #### More on Currencies
 
@@ -154,7 +154,7 @@ For English (and other languages), you can also specify an ordinal spellout:
 ```ruby
 DateTime.now.localize(:es).to_full_s               # <%= assert(datetime.localize(:es).to_full_s, "viernes, 14 de febrero de 2014, 12:20:05 (UTC +00:00)").inspect %>
 DateTime.now.localize(:es).to_long_s               # <%= assert(datetime.localize(:es).to_long_s, "14 de febrero de 2014, 12:20:05 UTC").inspect %>
-DateTime.now.localize(:es).to_medium_s             # <%= assert(datetime.localize(:es).to_medium_s, "14/2/2014 12:20:05").inspect %>
+DateTime.now.localize(:es).to_medium_s             # <%= assert(datetime.localize(:es).to_medium_s, "14 de feb. de 2014 12:20:05").inspect %>
 DateTime.now.localize(:es).to_short_s              # <%= assert(datetime.localize(:es).to_short_s, "14/2/14 12:20").inspect %>
 
 Time.now.localize(:es).to_full_s                   # <%= assert(time.localize(:es).to_full_s, "12:20:05 (UTC +00:00)").inspect %>
@@ -164,7 +164,7 @@ Time.now.localize(:es).to_short_s                  # <%= assert(time.localize(:e
 
 DateTime.now.localize(:es).to_date.to_full_s       # <%= assert(datetime.localize(:es).to_date.to_full_s, "viernes, 14 de febrero de 2014").inspect %>
 DateTime.now.localize(:es).to_date.to_long_s       # <%= assert(datetime.localize(:es).to_date.to_long_s, "14 de febrero de 2014").inspect %>
-DateTime.now.localize(:es).to_date.to_medium_s     # <%= assert(datetime.localize(:es).to_date.to_medium_s, "14/2/2014").inspect %>
+DateTime.now.localize(:es).to_date.to_medium_s     # <%= assert(datetime.localize(:es).to_date.to_medium_s, "14 de feb. de 2014").inspect %>
 DateTime.now.localize(:es).to_date.to_short_s      # <%= assert(datetime.localize(:es).to_date.to_short_s, "14/2/14").inspect %>
 ```
 
@@ -184,7 +184,7 @@ dt.to_short_s  # ...etc
 Besides the default date formats, CLDR supports a number of additional ones.  The list of available formats varies for each locale.  To get a full list, use the `additional_formats` method:
 
 ```ruby
-# <%= ellipsize(assert(datetime.localize(:ja).additional_formats.sort[0..11], ["EEEEd", "EHm", "EHms", "Ed", "Ehm", "Ehms", "Gy", "GyMMM", "GyMMMEEEEd", "GyMMMEd", "GyMMMd", "H"])) %>
+# <%= ellipsize(assert(datetime.localize(:ja).additional_formats.sort[0..11], ["E", "EEEEd", "EHm", "EHms", "Ed", "Ehm", "Ehms", "Gy", "GyMMM", "GyMMMEEEEd", "GyMMMEd", "GyMMMd"])) %>
 DateTime.now.localize(:ja).additional_formats
 ```
 

diff --git a/Rakefile b/Rakefile
@@ -96,7 +96,7 @@ end
 # TODO: 'add_locale' task that creates a new directory and runs all necessary 'update' tasks (+ suggests to run those that depend on JRuby)
 
 namespace :update do
-  ICU_JAR = './vendor/icu4j-52_1.jar'
+  ICU_JAR = './vendor/icu4j.jar'
 
   desc 'Import locales resources'
   task :locales_resources, :cldr_path do |_, args|

diff --git a/lib/twitter_cldr/data_readers/number_data_reader.rb b/lib/twitter_cldr/data_readers/number_data_reader.rb
@@ -122,7 +122,7 @@ def valid_type_for?(number, type)
       end
 
       def get_key_for(number)
-        "1#{"0" * (number.to_i.to_s.size - 1)}".to_i
+        "1#{"0" * (number.to_i.abs.to_s.size - 1)}".to_i
       end
 
       def pattern_for_number(pattern, number, decimal)
@@ -163,7 +163,7 @@ def pattern_for_sign(pattern, sign)
         else
           case sign
             when :negative
-              "#{symbols[:minus] || '-'}#{pattern}"
+              "#{symbols[:minus_sign] || '-'}#{pattern}"
             else
               pattern
           end
@@ -178,7 +178,8 @@ def resource
       end
 
       def self.within_abbreviation_range?(number)
-        NUMBER_MIN <= number && number < NUMBER_MAX
+        abs_value = number.abs
+        NUMBER_MIN <= abs_value && abs_value < NUMBER_MAX
       end
     end
   end

diff --git a/lib/twitter_cldr/formatters/numbers/abbreviated/abbreviated_number_formatter.rb b/lib/twitter_cldr/formatters/numbers/abbreviated/abbreviated_number_formatter.rb
@@ -9,7 +9,7 @@ class AbbreviatedNumberFormatter < NumberFormatter
 
       def truncate_number(number, decimal_digits)
         if TwitterCldr::DataReaders::NumberDataReader.within_abbreviation_range?(number)
-          factor = [0, number.to_i.to_s.length - decimal_digits].max
+          factor = [0, number.to_i.abs.to_s.length - decimal_digits].max
           number / (10.0 ** factor)
         else
           number

diff --git a/lib/twitter_cldr/formatters/numbers/number_formatter.rb b/lib/twitter_cldr/formatters/numbers/number_formatter.rb
@@ -7,13 +7,6 @@ module TwitterCldr
   module Formatters
     class NumberFormatter < Formatter
 
-      DEFAULT_SYMBOLS = {
-        :group => ',',
-        :decimal => '.',
-        :plus_sign => '+',
-        :minus_sign => '-'
-      }
-
       attr_reader :data_reader
 
       def initialize(data_reader)

diff --git a/lib/twitter_cldr/formatters/numbers/rbnf.rb b/lib/twitter_cldr/formatters/numbers/rbnf.rb
@@ -18,6 +18,7 @@ module Rbnf
       autoload :RuleGroup,    "twitter_cldr/formatters/numbers/rbnf/rule_group"
       autoload :RuleParser,   "twitter_cldr/formatters/numbers/rbnf/rule_parser"
       autoload :Substitution, "twitter_cldr/formatters/numbers/rbnf/substitution"
+      autoload :Plural,       "twitter_cldr/formatters/numbers/rbnf/plural"
 
       class PrivateRuleSetError < StandardError; end
 
@@ -100,7 +101,7 @@ def rule_set_name_cache
         def rule_set_from_resource(rule_set_data)
           RuleSet.new(
             rule_set_data[:rules].map do |rule|
-              Rule.new(rule[:value], rule[:rule], rule[:radix])
+              Rule.new(rule[:value], rule[:rule], rule[:radix], locale)
             end,
             rule_set_data[:type],
             rule_set_data[:access] || "public"

diff --git a/lib/twitter_cldr/formatters/numbers/rbnf/formatters.rb b/lib/twitter_cldr/formatters/numbers/rbnf/formatters.rb
@@ -16,7 +16,6 @@ class << self
 
           def format(number, rule_set, rule_group, locale)
             rule = rule_set.rule_for(number)
-            # puts "#{number}, #{rule.base_value}: '" + rule.rule_text + "'"
             formatter = formatter_for(rule, rule_set, rule_group, locale)
             result = formatter.format(number, rule)
             keep_soft_hyphens ? result : remove_soft_hyphens(result)
@@ -120,6 +119,10 @@ def semicolon(number, rule, token)
           ""
         end
 
+        def plural(number, rule, token)
+          token.render(number / rule.divisor)
+        end
+
         protected
 
         def invalid_token_error(token)

diff --git a/lib/twitter_cldr/formatters/numbers/rbnf/plural.rb b/lib/twitter_cldr/formatters/numbers/rbnf/plural.rb
@@ -0,0 +1,49 @@
+# encoding: UTF-8
+
+# Copyright 2012 Twitter, Inc
+# http://www.apache.org/licenses/LICENSE-2.0
+
+module TwitterCldr
+  module Formatters
+    module Rbnf
+
+      class Plural
+        def self.from_string(locale, str)
+          # $(cardinal,one{ tysiąc}few{ tysiące}other{ tysięcy})$
+          plural_type, cases = str.gsub(/\$\((.*)\)\$/, '\1').split(',')
+
+          # one{ tysiąc}few{ tysiące}other{ tysięcy}
+          cases = cases.scan(/([\w]+)\{([^}]+)\}/).inject({}) do |ret, case_arr|
+            ret[case_arr.first.to_sym] = case_arr.last
+            ret
+          end
+
+          new(locale, plural_type.to_sym, cases)
+        end
+
+        attr_reader :locale, :plural_type, :cases
+
+        # plural_type = cardinal, etc
+        # cases = hash of form one: "foo", two: "bar"
+        def initialize(locale, plural_type, cases)
+          @locale = locale
+          @plural_type = plural_type
+          @cases = cases
+        end
+
+        def render(number)
+          rule_name = TwitterCldr::Formatters::Plurals::Rules.rule_for(
+            number, locale, plural_type
+          )
+
+          cases[rule_name] || cases[:other]
+        end
+
+        def type
+          :plural
+        end
+      end
+
+    end
+  end
+end
diff --git a/lib/twitter_cldr/formatters/numbers/rbnf/rule.rb b/lib/twitter_cldr/formatters/numbers/rbnf/rule.rb
@@ -14,12 +14,13 @@ class Rule
         PROPER_FRACTION = "0.x"
         NEGATIVE = "-x"
 
-        attr_reader :base_value, :rule_text, :radix
+        attr_reader :base_value, :rule_text, :radix, :locale
 
-        def initialize(base_value, rule_text, radix)
+        def initialize(base_value, rule_text, radix, locale)
           @base_value = base_value
           @rule_text = rule_text
           @radix = radix
+          @locale = locale
         end
 
         def divisor
@@ -76,7 +77,7 @@ def tokens
         private
 
         def inline_substitutions(tokens)
-          parser.parse(tokens)
+          parser.parse(tokens, :locale => locale)
         end
 
         def parser

diff --git a/lib/twitter_cldr/formatters/numbers/rbnf/rule_parser.rb b/lib/twitter_cldr/formatters/numbers/rbnf/rule_parser.rb
@@ -12,6 +12,7 @@ class RuleParser < TwitterCldr::Parsers::Parser
         private
 
         def do_parse(options)
+          @locale = options.fetch(:locale, TwitterCldr.locale)
           switch([])
         end
 
@@ -48,6 +49,16 @@ def right_arrow(list)
           switch(list)
         end
 
+        def plural(list)
+          sub = Plural.from_string(
+            @locale, current_token.value
+          )
+
+          list << sub
+          next_token(:plural)
+          switch(list)
+        end
+
         def plaintext(list)
           add_and_advance(list)
         end
@@ -83,4 +94,4 @@ def descriptor(token)
       end
     end
   end
-end
+end
diff --git a/lib/twitter_cldr/formatters/numbers/rbnf/substitution.rb b/lib/twitter_cldr/formatters/numbers/rbnf/substitution.rb
@@ -27,4 +27,4 @@ def decimal_format
 
     end
   end
-end
+end
diff --git a/lib/twitter_cldr/resources/download.rb b/lib/twitter_cldr/resources/download.rb
@@ -7,7 +7,7 @@ module TwitterCldr
   module Resources
 
     CLDR_URL = 'http://unicode.org/Public/cldr/26/core.zip'
-    ICU4J_URL = 'http://download.icu-project.org/files/icu4j/52.1/icu4j-52_1.jar'
+    ICU4J_URL = 'http://download.icu-project.org/files/icu4j/54.1/icu4j-54_1.jar'
 
     # Use these instead to update collation and tailoring data
     # CLDR_URL = 'http://unicode.org/Public/cldr/23.1/core.zip'
@@ -54,4 +54,4 @@ def download_icu4j_if_necessary(path, url = ICU4J_URL)
     end
 
   end
-end
+end
diff --git a/lib/twitter_cldr/resources/locales_resources_importer.rb b/lib/twitter_cldr/resources/locales_resources_importer.rb
@@ -57,11 +57,15 @@ def import_components
           :merge => true  # fill in the gaps, eg fill in sub-locales like en_GB with en
         }
 
+        locales = Set.new
+
         Cldr::Export.export(export_args) do |component, locale, path|
           add_buddhist_calendar(component, locale, path)
           process_plurals(component, locale, path)
           downcase_territory_codes(component, locale, path)
           deep_symbolize(component, locale, path)
+          locales.add(locale)
+          STDOUT.write "\r#{locale}, #{locales.size} of #{TwitterCldr.supported_locales.size} total"
         end
 
         export_args = {

diff --git a/lib/twitter_cldr/resources/uli/segment_exceptions_importer.rb b/lib/twitter_cldr/resources/uli/segment_exceptions_importer.rb
@@ -12,7 +12,7 @@ module Resources
     module Uli
       class SegmentExceptionsImporter
 
-        URL = "http://unicode.org/uli/trac/export/38/trunk/abbrs/json/%{locale}.json"
+        URL = "http://unicode.org/uli/trac/export/58/trunk/abbrs/json/%{locale}.json"
 
         attr_reader :input_path, :output_path
 

diff --git a/lib/twitter_cldr/tokenizers/numbers/rbnf_tokenizer.rb b/lib/twitter_cldr/tokenizers/numbers/rbnf_tokenizer.rb
@@ -37,6 +37,7 @@ def tokenizer
             TokenRecognizer.new(:open_bracket, /\[/),
             TokenRecognizer.new(:close_bracket, /\]/),
             TokenRecognizer.new(:decimal, /[0#][0#,\.]+/),
+            TokenRecognizer.new(:plural, /\$\(.*\)\$/),
 
             # ending
             TokenRecognizer.new(:semicolon, /;/),
@@ -55,4 +56,4 @@ def tokenizer
 
     end
   end
-end
+end
diff --git a/resources/locales/af/numbers.yml b/resources/locales/af/numbers.yml
@@ -108,3 +108,4 @@
       :percent_sign: "%"
       :plus_sign: "+"
       :superscripting_exponent: "×"
+      :time_separator: ":"
diff --git a/resources/locales/ar/numbers.yml b/resources/locales/ar/numbers.yml
@@ -208,3 +208,4 @@
       :percent_sign: "%"
       :plus_sign: "‎+"
       :superscripting_exponent: "×"
+      :time_separator: ":"
diff --git a/resources/locales/be/numbers.yml b/resources/locales/be/numbers.yml
@@ -108,3 +108,4 @@
       :percent_sign: "%"
       :plus_sign: "+"
       :superscripting_exponent: "×"
+      :time_separator: ":"
diff --git a/resources/locales/bg/numbers.yml b/resources/locales/bg/numbers.yml
@@ -5,7 +5,7 @@
       :currency: 
         :number_system: "latn"
         :patterns: 
-          :default: "#,##0.00 ¤"
+          :default: "#,##0.00 ¤;(#,##0.00 ¤)"
         :unit: 
           :one: "{0} {1}"
           :other: "{0} {1}"
@@ -21,7 +21,7 @@
               :one: "00 хиляди"
               :other: "00 хиляди"
             100000: 
-              :one: "000 Х"
+              :one: "000 хиляди"
               :other: "000 хиляди"
             1000000: 
               :one: "0 милион"
@@ -108,3 +108,4 @@
       :percent_sign: "%"
       :plus_sign: "+"
       :superscripting_exponent: "."
+      :time_separator: ":"
diff --git a/resources/locales/bn/numbers.yml b/resources/locales/bn/numbers.yml
@@ -52,11 +52,11 @@
               :other: "000 ট্রিলিয়ন"
           :short: 
             1000: 
-              :one: "0K"
-              :other: "0K"
+              :one: "0 হাজার"
+              :other: "0 হাজার"
             10000: 
-              :one: "00K"
-              :other: "00K"
+              :one: "00 হাজার"
+              :other: "00 হাজার"
             100000: 
               :one: "0 লাখ"
               :other: "0 লাখ"
@@ -108,3 +108,4 @@
       :percent_sign: "%"
       :plus_sign: "+"
       :superscripting_exponent: "×"
+      :time_separator: ":"
diff --git a/resources/locales/ca/numbers.yml b/resources/locales/ca/numbers.yml
@@ -108,3 +108,4 @@
       :percent_sign: "%"
       :plus_sign: "+"
       :superscripting_exponent: "×"
+      :time_separator: ":"