README.md.erb

<% $KCODE = "UTF-8" if RUBY_VERSION <= "1.8.7" %>

## twitter-cldr-rb [![Build Status](https://secure.travis-ci.org/twitter/twitter-cldr-rb.png?branch=master)](http://travis-ci.org/twitter/twitter-cldr-rb) [![Code Climate](https://codeclimate.com/github/twitter/twitter-cldr-rb.png)](https://codeclimate.com/github/twitter/twitter-cldr-rb)

TwitterCldr uses Unicode's Common Locale Data Repository (CLDR) to format certain types of text into their
localized equivalents.  Currently supported types of text include dates, times, currencies, decimals, percentages, and symbols.

## Installation

`gem install twitter_cldr`

## Usage

```ruby
require 'twitter_cldr'
```

### Basics

Get a list of all currently supported locales (these are all supported on twitter.com):

```ruby
TwitterCldr.supported_locales             # <%= ellipsize(assert(TwitterCldr.supported_locales.sort[0..5], [:af, :ar, :be, :bg, :bn, :ca])) %>
```

Determine if a locale is supported by TwitterCLDR:

```ruby
TwitterCldr.supported_locale?(:es)        # <%= assert_true(TwitterCldr.supported_locale?(:es)).inspect %>
TwitterCldr.supported_locale?(:xx)        # <%= assert_false(TwitterCldr.supported_locale?(:xx)).inspect %>
```


TwitterCldr patches core Ruby objects like `Fixnum` and `Date` to make localization as straightforward as possible.

### Numbers

`Fixnum`, `Bignum`, and `Float` objects are supported.  Here are some examples:

```ruby
# default formatting with to_s
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 %>

# 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 %>

# decimals
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.

Behind the scenes, these convenience methods are creating instances of `LocalizedNumber`.  You can do the same thing if you're feeling adventurous:

```ruby
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 $") %>

#### More on Currencies

If you're looking for a list of supported currencies, use the `TwitterCldr::Shared::Currencies` class:

```ruby
# all supported currency codes
TwitterCldr::Shared::Currencies.currency_codes             # <%= ellipsize(assert(TwitterCldr::Shared::Currencies.currency_codes.sort[0..3], ["ADP", "AED", "AFA", "AFN"])) %>

# data for a specific currency code
TwitterCldr::Shared::Currencies.for_code("CAD")            # <%= assert(TwitterCldr::Shared::Currencies.for_code("CAD"), {:currency=>:CAD, :name=>"Canadian dollar", :cldr_symbol=>"CA$", :symbol=>"$", :code_points=>[36]}).inspect %>
```

#### Short / Long Decimals

In addition to formatting regular decimals, TwitterCLDR supports short and long decimals.  Short decimals abbreviate the notation for the appropriate power of ten, for example "1M" for 1,000,000 or "2K" for 2,000.  Long decimals include the full notation, for example "1 million" or "2 thousand".  Long and short decimals can be generated using the appropriate `to_` method:

```ruby
2337.localize.to_short_decimal.to_s     # <%= assert(2337.localize.to_short_decimal.to_s, "2K").inspect %>
1337123.localize.to_short_decimal.to_s  # <%= assert(1337123.localize.to_short_decimal.to_s, "1M").inspect %>

2337.localize.to_long_decimal.to_s      # <%= assert(2337.localize.to_long_decimal.to_s, "2 thousand").inspect %>
1337123.localize.to_long_decimal.to_s   # <%= assert(1337123.localize.to_long_decimal.to_s, "1 million").inspect %>
```

### Number Spellout, Ordinalization, and More

TwitterCLDR's rule-based number formatters are capable of transforming integers into their written equivalents. Note that rule-based formatting of decimal numbers is currently not supported for languages other than English.

#### Spellout

For easy spellout formatting, check out the `LocalizedNumber#spellout` method:

```ruby
123.localize.spellout     # <%= assert(123.localize.spellout, "one hundred twenty-three") %>
25_641.localize.spellout  # <%= assert(25_641.localize.spellout, "twenty-five thousand six hundred forty-one") %>
```

As always, you can call `#localize` with a locale symbol:

```ruby
123.localize(:es).spellout     # <%= assert(123.localize(:es).spellout, "ciento veintitrés") %>
25_641.localize(:ru).spellout  # <%= assert(25_641.localize(:ru).spellout, "двадцать пять тысяч шестьсот сорок один") %>
```

#### Ordinalization and More

The available rule-based number formats defined by the CLDR data set vary by language. Some languages support ordinal and cardinal numbers, occasionally with an additional masculine/feminine option, while others do not. You'll need to consult the list of available formats for your language.

Rule-based number formats are categorized by groups, and within groups by rulesets. You'll need to specify both to make use of all the available formats for your language.

To get a list of supported groups, use the `#group_names` method:

```ruby
123.localize(:pt).rbnf.group_names  # <%= assert(123.localize(:pt).rbnf.group_names, ["SpelloutRules", "OrdinalRules"]).inspect %>
```

To get a list of supported rulesets for a group name, use the `#rule_set_names_for_group` method:

```ruby
# <%= assert(123.localize(:pt).rbnf.rule_set_names_for_group("OrdinalRules"), ["digits-ordinal-masculine", "digits-ordinal-feminine", "digits-ordinal"]).inspect %>
123.localize(:pt).rbnf.rule_set_names_for_group("OrdinalRules")
```

Once you've chosen a group and ruleset, you can pass them to the `to_rbnf_s` method:

```ruby
123.localize(:pt).to_rbnf_s("OrdinalRules", "digits-ordinal-feminine")  # <%= assert(123.localize(:pt).to_rbnf_s("OrdinalRules", "digits-ordinal-feminine"), "123a") %>
123.localize(:pt).to_rbnf_s("OrdinalRules", "digits-ordinal-masculine") # <%= assert(123.localize(:pt).to_rbnf_s("OrdinalRules", "digits-ordinal-masculine"), "123o") %>
```

For comparison, here's what English ordinal formatting looks like:

```ruby
123.localize.to_rbnf_s("OrdinalRules", "digits-ordinal")  # <%= assert(123.localize.to_rbnf_s("OrdinalRules", "digits-ordinal"), "123rd") %>
```

For English (and other languages), you can also specify an ordinal spellout:

```ruby
123.localize.to_rbnf_s("SpelloutRules", "spellout-ordinal")  # <%= assert(123.localize.to_rbnf_s("SpelloutRules", "spellout-ordinal"), "one hundred twenty-third") %>
123.localize(:pt).to_rbnf_s("SpelloutRules", "spellout-ordinal-masculine")  # <%= assert(123.localize(:pt).to_rbnf_s("SpelloutRules", "spellout-ordinal-masculine"), "centésimo vigésimo terceiro") %>
```

### Dates and Times

`Time`, and `DateTime` objects are supported.  `Date` objects are supported transiently:

```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_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 %>
Time.now.localize(:es).to_long_s                   # <%= assert(time.localize(:es).to_long_s, "12:20:05 UTC").inspect %>
Time.now.localize(:es).to_medium_s                 # <%= assert(time.localize(:es).to_medium_s, "12:20:05").inspect %>
Time.now.localize(:es).to_short_s                  # <%= assert(time.localize(:es).to_short_s, "12:20").inspect %>

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_short_s      # <%= assert(datetime.localize(:es).to_date.to_short_s, "14/2/14").inspect %>
```

The default CLDR data set only includes 4 date formats, full, long, medium, and short.  See below for a list of additional formats.

Behind the scenes, these convenience methods are creating instances of `LocalizedDate`, `LocalizedTime`, and `LocalizedDateTime`.  You can do the same thing if you're feeling adventurous:

<% assert(TwitterCldr::Localized::LocalizedDateTime.new(datetime, :es).to_short_s, "14/2/14 12:20") %>

```ruby
dt = TwitterCldr::Localized::LocalizedDateTime.new(DateTime.now, :es)
dt.to_short_s  # ...etc
```

#### Additional Date Formats

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"])) %>
DateTime.now.localize(:ja).additional_formats
```

You can use any of the returned formats as the argument to the `to_additional_s` method:

```ruby
# <%= assert(datetime.localize(:ja).to_additional_s("EEEEd"), "14日金曜日").inspect %>
DateTime.now.localize(:ja).to_additional_s("EEEEd")
```

It's important to know that, even though any given format may not be available across locales, TwitterCLDR will do it's best to approximate if no exact match can be found.

##### List of additional date format examples for English:

| Format     | Output                 |
|:-----------|------------------------|
<% datetime.localize(:en).additional_formats.each do |pattern| %>| <%= pattern.ljust(10, " ") %> | <%= datetime.localize(:en).to_additional_s(pattern).ljust(22, " ") %> |<%= "\n" %><% end %>


#### Relative Dates and Times

In addition to formatting full dates and times, TwitterCLDR supports relative time spans via several convenience methods and the `LocalizedTimespan` class.  TwitterCLDR tries to guess the best time unit (eg. days, hours, minutes, etc) based on the length of the time span.  Unless otherwise specified, TwitterCLDR will use the current date and time as the reference point for the calculation.

```ruby
(DateTime.now - 1).localize.ago.to_s        # <%= assert((DateTime.now - 1).localize.ago.to_s, "1 day ago").inspect %>
(DateTime.now - 0.5).localize.ago.to_s      # <%= assert((DateTime.now - 0.5).localize.ago.to_s, "12 hours ago").inspect %>  (i.e. half a day)

(DateTime.now + 1).localize.until.to_s      # <%= assert((DateTime.now + 1).localize.until.to_s, "In 1 day").inspect %>
(DateTime.now + 0.5).localize.until.to_s    # <%= assert((DateTime.now + 0.5).localize.until.to_s, "In 12 hours").inspect %>
```

Specify other locales:

```ruby
(DateTime.now - 1).localize(:de).ago.to_s        # <%= assert((DateTime.now - 1).localize(:de).ago.to_s, "Vor 1 Tag").inspect %>
(DateTime.now + 1).localize(:de).until.to_s      # <%= assert((DateTime.now + 1).localize(:de).until.to_s, "In 1 Tag").inspect %>
```

Force TwitterCLDR to use a specific time unit by including the `:unit` option:

```ruby
(DateTime.now - 1).localize(:de).ago.to_s(:unit => :hour)        # <%= assert((DateTime.now - 1).localize(:de).ago.to_s(:unit => :hour), "Vor 24 Stunden").inspect %>
(DateTime.now + 1).localize(:de).until.to_s(:unit => :hour)      # <%= assert((DateTime.now + 1).localize(:de).until.to_s(:unit => :hour), "In 24 Stunden").inspect %>
```

Specify a different reference point for the time span calculation:

```ruby
# 86400 = 1 day in seconds, 259200 = 3 days in seconds
(Time.now + 86400).localize(:de).ago(:base_time => (Time.now + 259200)).to_s(:unit => :hour)  # <%= assert((Time.now + 86400).localize(:de).ago(:base_time => (Time.now + 259200)).to_s(:unit => :hour), "Vor 48 Stunden").inspect %>
```

Behind the scenes, these convenience methods are creating instances of `LocalizedTimespan`, whose constructor accepts a number of seconds as the first argument.  You can do the same thing if you're feeling adventurous:

```ruby
<% ts = TwitterCldr::Localized::LocalizedTimespan.new(86400, :locale => :de) %>
ts = TwitterCldr::Localized::LocalizedTimespan.new(86400, :locale => :de)
ts.to_s                         # <%= assert(ts.to_s, "In 1 Tag").inspect %>
ts.to_s(:unit => :hour)         # <%= assert(ts.to_s(:unit => :hour), "In 24 Stunden").inspect %>

<% ts = TwitterCldr::Localized::LocalizedTimespan.new(-86400, :locale => :de) %>
ts = TwitterCldr::Localized::LocalizedTimespan.new(-86400, :locale => :de)
ts.to_s                         # <%= assert(ts.to_s, "Vor 1 Tag").inspect %>
ts.to_s(:unit => :hour)         # <%= assert(ts.to_s(:unit => :hour), "Vor 24 Stunden").inspect %>
```

By default, timespans are exact representations of a given unit of elapsed time.  TwitterCLDR also supports approximate timespans which round up to the nearest larger unit.  For example, "44 seconds" remains "44 seconds" while "45 seconds" becomes "1 minute".  To approximate, pass the `:approximate => true` option into `to_s`:

```ruby
TwitterCldr::Localized::LocalizedTimespan.new(44).to_s(:approximate => true)  # <%= assert(TwitterCldr::Localized::LocalizedTimespan.new(44).to_s(:approximate => true), "In 44 seconds").inspect %>
TwitterCldr::Localized::LocalizedTimespan.new(45).to_s(:approximate => true)  # <%= assert(TwitterCldr::Localized::LocalizedTimespan.new(45).to_s(:approximate => true), "In 1 minute").inspect %>
TwitterCldr::Localized::LocalizedTimespan.new(52).to_s(:approximate => true)  # <%= assert(TwitterCldr::Localized::LocalizedTimespan.new(52).to_s(:approximate => true), "In 1 minute").inspect %>
```

### Lists

TwitterCLDR supports formatting lists of strings as you might do in English by using commas, eg: "Apples, cherries, and oranges".  Use the `localize` method on an array followed by a call to `to_sentence`:

```ruby
["apples", "cherries", "oranges"].localize.to_sentence       # <%= assert(["apples", "cherries", "oranges"].localize.to_sentence, "apples, cherries, and oranges").inspect %>
["apples", "cherries", "oranges"].localize(:es).to_sentence  # <%= assert(["apples", "cherries", "oranges"].localize(:es).to_sentence, "apples, cherries y oranges").inspect %>
```

Behind the scenes, these convenience methods are creating instances of `ListFormatter`.  You can do the same thing if you're feeling adventurous:

```ruby
f = TwitterCldr::Formatters::ListFormatter.new(:en)
f.format(["Larry", "Curly", "Moe"])  # <%= assert(TwitterCldr::Formatters::ListFormatter.new(:en).format(["Larry", "Curly", "Moe"]), "Larry, Curly, and Moe").inspect %>

f = TwitterCldr::Formatters::ListFormatter.new(:es)
f.format(["Larry", "Curly", "Moe"])  # <%= assert(TwitterCldr::Formatters::ListFormatter.new(:es).format(["Larry", "Curly", "Moe"]), "Larry, Curly y Moe").inspect %>
```

The TwitterCLDR `ListFormatter` class is smart enough to handle right-to-left (RTL) text and will format the list "backwards" in these cases (note that what looks backwards to English speakers looks frontwards for RTL speakers).  See the section on handling bidirectional text below for more information.

### Plural Rules

Some languages, like English, have "countable" nouns.  You probably know this concept better as "plural" and "singular", i.e. the difference between "strawberry" and "strawberries".  Other languages, like Russian, have three plural forms: one (numbers ending in 1), few (numbers ending in 2, 3, or 4), and many (everything else).  Still other languages like Japanese don't use countable nouns at all.

TwitterCLDR makes it easy to find the plural rules for any numeric value:

```ruby
1.localize(:ru).plural_rule                                # <%= assert(1.localize(:ru).plural_rule, :one).inspect %>
2.localize(:ru).plural_rule                                # <%= assert(2.localize(:ru).plural_rule, :other).inspect %>
5.localize(:ru).plural_rule                                # <%= assert(5.localize(:ru).plural_rule, :many).inspect %>
```

Behind the scenes, these convenience methods use the `TwitterCldr::Formatters::Plurals::Rules` class.  You can do the same thing (and a bit more) if you're feeling adventurous:

```ruby
# get all rules for the default locale
TwitterCldr::Formatters::Plurals::Rules.all                # <%= assert(TwitterCldr::Formatters::Plurals::Rules.all, [:one, :other]).inspect %>

# get all rules for a specific locale
TwitterCldr::Formatters::Plurals::Rules.all_for(:es)       # <%= assert(TwitterCldr::Formatters::Plurals::Rules.all_for(:es), [:one, :other]).inspect %>
TwitterCldr::Formatters::Plurals::Rules.all_for(:ru)       # <%= assert(TwitterCldr::Formatters::Plurals::Rules.all_for(:ru), [:one, :many, :other]).inspect %>

# get the rule for a number in a specific locale
TwitterCldr::Formatters::Plurals::Rules.rule_for(1, :ru)   # <%= assert(TwitterCldr::Formatters::Plurals::Rules.rule_for(1, :ru), :one).inspect %>
TwitterCldr::Formatters::Plurals::Rules.rule_for(2, :ru)   # <%= assert(TwitterCldr::Formatters::Plurals::Rules.rule_for(2, :ru), :other).inspect %>
```

### Plurals

In addition to providing access to plural rules, TwitterCLDR allows you to embed plurals directly in your source code:

```ruby
replacements = {
  :horse_count => 3,
  :horses => {
    :one => "is 1 horse",
    :other => "are %{horse_count} horses"
  }
}

# <%= assert("there %{horse_count:horses} in the barn".localize % { :horse_count => 3, :horses => { :one => "is 1 horse", :other => "are %{horse_count} horses" } }, "there are 3 horses in the barn").inspect %>
"there %{horse_count:horses} in the barn".localize % replacements
```

Because providing a pluralization hash with the correct plural rules can be difficult, you can also embed plurals as a JSON hash into your string:

```ruby
str = 'there %<{ "horse_count": { "one": "is one horse", "other": "are %{horse_count} horses" } }> in the barn'

# <%= assert('there %<{ "horse_count": { "one": "is one horse", "other": "are %{horse_count} horses" } }> in the barn'.localize % { :horse_count => 3 }, "there are 3 horses in the barn").inspect %>
str.localize % { :horse_count => 3 }
```

NOTE: If you're using TwitterCLDR with Rails 3, you may see an error if you try to use the `%` function on a localized string in your views.  Strings in views in Rails 3 are instances of `SafeBuffer`, which patches the `gsub` method that the TwitterCLDR plural formatter relies on.  To fix this issue, simply call `to_str` on any `SafeBuffer` before calling `localize`.  More info [here](https://github.com/rails/rails/issues/1555).  An example:

```ruby
# throws an error in Rails 3 views:
'%<{"count": {"one": "only one", "other": "tons more!"}}'.localize % { :count => 2 }

# works just fine:
'%<{"count": {"one": "only one", "other": "tons more!"}}'.to_str.localize % { :count => 2 }
```

The `LocalizedString` class supports all forms of interpolation and combines support from both Ruby 1.8 and 1.9:

```ruby
# Ruby 1.8
"five euros plus %.3f in tax" % (13.25 * 0.087)

# Ruby 1.9
"five euros plus %.3f in tax" % (13.25 * 0.087)
"there are %{count} horses in the barn" % { :count => "5" }

# with TwitterCLDR
"five euros plus %.3f in tax".localize % (13.25 * 0.087)
"there are %{count} horses in the barn".localize % { :count => "5" }
```

When you pass a Hash as an argument and specify placeholders with `%<foo>d`, TwitterCLDR will interpret the hash values as named arguments and format the string according to the instructions appended to the closing `>`.  In this way, TwitterCLDR supports both Ruby 1.8 and 1.9 interpolation syntax in the same string:

```ruby
"five euros plus %<percent>.3f in %{noun}".localize % { :percent => 13.25 * 0.087, :noun => "tax" }
```

### World Languages

You can use the localize convenience method on language code symbols to get their equivalents in another language:

```ruby
:es.localize(:es).as_language_code                         # <%= assert(:es.localize(:es).as_language_code, "español").inspect %>
:ru.localize(:es).as_language_code                         # <%= assert(:ru.localize(:es).as_language_code, "ruso").inspect %>
```

Behind the scenes, these convenience methods are creating instances of `LocalizedSymbol`.  You can do the same thing if you're feeling adventurous:

```ruby
ls = TwitterCldr::Localized::LocalizedSymbol.new(:ru, :es)
ls.as_language_code  # <%= assert(TwitterCldr::Localized::LocalizedSymbol.new(:ru, :es).as_language_code, "ruso").inspect %>
```

In addition to translating language codes, TwitterCLDR provides access to the full set of supported languages via the `TwitterCldr::Shared::Languages` class:

```ruby
# get all languages for the default locale
TwitterCldr::Shared::Languages.all                                                  # <%= ellipsize(assert(slice_hash(TwitterCldr::Shared::Languages.all, [:"zh-Hant", :vi]), { :"zh-Hant" => "Traditional Chinese", :vi => "Vietnamese" })) %>

# get all languages for a specific locale
TwitterCldr::Shared::Languages.all_for(:es)                                         # <%= ellipsize(assert(slice_hash(TwitterCldr::Shared::Languages.all_for(:es), [:"zh-Hant", :vi]), { :"zh-Hant" => "chino tradicional", :vi => "vietnamita" })) %>

# get a language by its code for the default locale
TwitterCldr::Shared::Languages.from_code(:'zh-Hant')                                # <%= assert(TwitterCldr::Shared::Languages.from_code(:'zh-Hant'), "Traditional Chinese").inspect %>

# get a language from its code for a specific locale
TwitterCldr::Shared::Languages.from_code_for_locale(:'zh-Hant', :es)                # <%= assert(TwitterCldr::Shared::Languages.from_code_for_locale(:'zh-Hant', :es), "chino tradicional").inspect %>

# translate a language from one locale to another
# signature: translate_language(lang, source_locale, destination_locale)
TwitterCldr::Shared::Languages.translate_language("chino tradicional", :es, :en)    # <%= assert(TwitterCldr::Shared::Languages.translate_language("chino tradicional", :es, :en), "Traditional Chinese").inspect %>
TwitterCldr::Shared::Languages.translate_language("Traditional Chinese", :en, :es)  # <%= assert(TwitterCldr::Shared::Languages.translate_language("Traditional Chinese", :en, :es), "chino tradicional").inspect %>
```

### Postal Codes

The CLDR contains postal code validation regexes for a number of countries.

```ruby
# United States
TwitterCldr::Shared::PostalCodes.valid?(:us, "94103")     # <%= assert_true(TwitterCldr::Shared::PostalCodes.valid?(:us, "94103")) %>
TwitterCldr::Shared::PostalCodes.valid?(:us, "9410")      # <%= assert_false(TwitterCldr::Shared::PostalCodes.valid?(:us, "9410")) %>

# England (Great Britain)
TwitterCldr::Shared::PostalCodes.valid?(:gb, "BS98 1TL")  # <%= assert_true(TwitterCldr::Shared::PostalCodes.valid?(:gb, "BS98 1TL")) %>

# Sweden
TwitterCldr::Shared::PostalCodes.valid?(:se, "280 12")    # <%= assert_true(TwitterCldr::Shared::PostalCodes.valid?(:se, "280 12")) %>

# Canada
TwitterCldr::Shared::PostalCodes.valid?(:ca, "V3H 1Z7")   # <%= assert_true(TwitterCldr::Shared::PostalCodes.valid?(:ca, "V3H 1Z7")) %>
```

Get a list of supported territories by using the `#territories` method:

```ruby
TwitterCldr::Shared::PostalCodes.territories  # <%= ellipsize(assert(TwitterCldr::Shared::PostalCodes.territories.sort[0..4], [:ad, :am, :ar, :as, :at])) %>
```

Just want the regex?  No problem:

```ruby
TwitterCldr::Shared::PostalCodes.regex_for_territory(:us)  # <%= assert(TwitterCldr::Shared::PostalCodes.regex_for_territory(:us), /\d{5}([ \-]\d{4})?/).inspect %>
```

### Phone Codes

Look up phone codes by territory:

```ruby
# United States
TwitterCldr::Shared::PhoneCodes.code_for_territory(:us)  # <%= assert(TwitterCldr::Shared::PhoneCodes.code_for_territory(:us), "1").inspect %>

# Perú
TwitterCldr::Shared::PhoneCodes.code_for_territory(:pe)  # <%= assert(TwitterCldr::Shared::PhoneCodes.code_for_territory(:pe), "51").inspect %>

# Egypt
TwitterCldr::Shared::PhoneCodes.code_for_territory(:eg)  # <%= assert(TwitterCldr::Shared::PhoneCodes.code_for_territory(:eg), "20").inspect %>

# Denmark
TwitterCldr::Shared::PhoneCodes.code_for_territory(:dk)  # <%= assert(TwitterCldr::Shared::PhoneCodes.code_for_territory(:dk), "45").inspect %>
```

Get a list of supported territories by using the `#territories` method:

```ruby
TwitterCldr::Shared::PhoneCodes.territories  # <%= ellipsize(assert(TwitterCldr::Shared::PhoneCodes.territories.sort[0..4], [:ac, :ad, :ae, :af, :ag])) %>
```

### Language Codes

Over the years, different standards for language codes have accumulated.  Probably the two most popular are ISO-639 and BCP-47 and their children.  TwitterCLDR provides a way to convert between these codes programmatically.

```ruby
TwitterCldr::Shared::LanguageCodes.convert(:es, :from => :bcp_47, :to => :iso_639_2)  # <%= assert(TwitterCldr::Shared::LanguageCodes.convert(:es, :from => :bcp_47, :to => :iso_639_2), :spa).inspect %>
```

Use the `standards_for` method to get the standards that are available for conversion from a given code.  In the example below, note that the first argument, `:es`, is the correct BCP-47 language code for Spanish, which is the second argument.  The return value comprises all the available conversions:

```ruby
# <%= assert(TwitterCldr::Shared::LanguageCodes.standards_for(:es, :bcp_47), [:bcp_47, :iso_639_1, :iso_639_2, :iso_639_3]).inspect %>
TwitterCldr::Shared::LanguageCodes.standards_for(:es, :bcp_47)
```

Get a list of supported standards for a full English language name:

```ruby
# <%= assert(TwitterCldr::Shared::LanguageCodes.standards_for_language(:Spanish), [:bcp_47, :iso_639_1, :iso_639_2, :iso_639_3]).inspect %>
TwitterCldr::Shared::LanguageCodes.standards_for_language(:Spanish)
```

Get a list of supported languages:

```ruby
TwitterCldr::Shared::LanguageCodes.languages  # <%= ellipsize(assert(TwitterCldr::Shared::LanguageCodes.languages.select { |item| [:Arabic, :German, :Norwegian, :Spanish].include?(item) }.sort, [:Arabic, :German, :Norwegian, :Spanish])) %>
```

Determine valid standards:

```ruby
TwitterCldr::Shared::LanguageCodes.valid_standard?(:iso_639_1)  # <%= assert_true(TwitterCldr::Shared::LanguageCodes.valid_standard?(:iso_639_1)) %>
TwitterCldr::Shared::LanguageCodes.valid_standard?(:blarg)      # <%= assert_false(TwitterCldr::Shared::LanguageCodes.valid_standard?(:blarg)) %>
```

Determine valid codes:

```ruby
TwitterCldr::Shared::LanguageCodes.valid_code?(:es, :bcp_47)     # <%= assert_true(TwitterCldr::Shared::LanguageCodes.valid_code?(:es, :bcp_47)) %>
TwitterCldr::Shared::LanguageCodes.valid_code?(:es, :iso_639_2)  # <%= assert_false(TwitterCldr::Shared::LanguageCodes.valid_code?(:es, :iso_639_2)) %>
```

Convert the full English name of a language into a language code:

```ruby
TwitterCldr::Shared::LanguageCodes.from_language(:Spanish, :iso_639_2)  # <%= assert(TwitterCldr::Shared::LanguageCodes.from_language(:Spanish, :iso_639_2), :spa).inspect %>
```

Convert a language code into it's full English name:

```ruby
TwitterCldr::Shared::LanguageCodes.to_language(:spa, :iso_639_2)  # <%= assert(TwitterCldr::Shared::LanguageCodes.to_language(:spa, :iso_639_2), "Spanish").inspect %>
```

**NOTE**: All of the functions in `TwitterCldr::Shared::LanguageCodes` accept both symbol and string parameters.

### Unicode Regular Expressions

Unicode regular expressions are an extension of the normal regular expression syntax. All of the changes are local to the regex's character class feature and provide support for multi-character strings, Unicode character escapes, set operations (unions, intersections, and differences), and character sets.

#### Changes to Character Classes

Here's a complete list of the operations you can do inside a Unicode regex's character class.

| Regex              | Description                                                                                                         |
|:-------------------|:--------------------------------------------------------------------------------------------------------------------|
|`[a]`               | The set containing 'a'.                                                                                             |
|`[a-z]`             | The set containing 'a' through 'z' and all letters in between, in Unicode order.                                    |
|`[^a-z]`            | The set containing all characters except 'a' through 'z', that is, U+0000 through 'a'-1 and 'z'+1 through U+10FFFF. |
|`[[pat1][pat2]]`    | The union of sets specified by pat1 and pat2.                                                                       |
|`[[pat1]&[pat2]]`   | The intersection of sets specified by pat1 and pat2.                                                                |
|`[[pat1]-[pat2]]`   | The [symmetric difference](http://en.wikipedia.org/wiki/Symmetric_difference) of sets specified by pat1 and pat2.   |
|`[:Lu:] or \p{Lu}`  | The set of characters having the specified Unicode property; in this case, Unicode uppercase letters.               |
|`[:^Lu:] or \P{Lu}` | The set of characters not having the given Unicode property.                                                        |

For a description of available Unicode properties, see [Wikipedia](http://en.wikipedia.org/wiki/Unicode_character_property#General_Category) (click on "[show]").

#### Using Unicode Regexes

Create Unicode regular expressions via the `#compile` method:

```ruby
<% regex = TwitterCldr::Shared::UnicodeRegex.compile("[:Lu:]+") %>
regex = TwitterCldr::Shared::UnicodeRegex.compile("[:Lu:]+")
```

Once compiled, instances of `UnicodeRegex` behave just like normal Ruby regexes and support the `#match` and `#=~` methods:

```ruby
<% assert(regex.match("ABC")[0], "ABC") %>
regex.match("ABC")  # <%= regex.match("ABC").inspect[1..-1] %>
regex =~ "fooABC"   # <%= assert(regex =~ "fooABC", 3) %>
```

Protip: Try to avoid negation in character classes (eg. [^abc] and \P{Lu}) as it tends to negatively affect both performance when constructing regexes as well as matching.

#### Support for Ruby 1.8

Ruby 1.8 does not allow escaped Unicode characters in regular expressions and restricts their maximum length. TwitterCLDR's `UnicodeRegex` class supports escaped unicode characters in Ruby 1.8, but cannot offer a work-around for the length issue. For this reason, Ruby 1.8 users are required to install the oniguruma regex engine and require the oniguruma gem in their projects.

To install oniguruma, run `brew install oniguruma` on MacOS, `[sudo] apt-get install libonig-dev` on Ubuntu (you may need to search for other instructions specific to your platform). Then, install the oniguruma gem via your Gemfile or on your system via `gem install oniguruma`. Once installed, `require oniguruma` somewhere in your project before making use of the `TwitterCldr::Shared::UnicodeRegex` class.

### Text Segmentation

TwitterCLDR currently supports text segmentation by sentence as described in the [Unicode Technical Report #29](http://www.unicode.org/reports/tr29/). The segmentation algorithm makes use of Unicode regular expressions (described above). Because of this, if you're running Ruby 1.8, you'll need to follow the instructions above to install the oniguruma regular expression engine. Segmentation by word, line, and grapheme boundaries could also be supported if someone wants them.

You can break a string into sentences using the `LocalizedString#each_sentence` method:

```ruby
"The. Quick. Brown. Fox.".localize.each_sentence do |sentence|
  puts sentence.to_s  # <%= assert("The. Quick. Brown. Fox.".localize.each_sentence.to_a.map(&:to_s), ["The.", " Quick.", " Brown.", " Fox."]).map { |s| "\"#{s}\"" }.join(", ") %>
end
```

Under the hood, text segmentation is performed by the `BreakIterator` class (name borrowed from ICU). You can use it directly if you're feeling adventurous:

```ruby
<% iterator = TwitterCldr::Shared::BreakIterator.new(:en) %>
iterator = TwitterCldr::Shared::BreakIterator.new(:en)
iterator.each_sentence("The. Quick. Brown. Fox.") do |sentence|
  puts sentence  # <%= assert(iterator.each_sentence("The. Quick. Brown. Fox.").to_a, ["The.", " Quick.", " Brown.", " Fox."]).map { |s| "\"#{s}\"" }.join(", ") %>
end
```

To improve segmentation accuracy, a list of special segmentation exceptions have been created by the ULI (Unicode Interoperability Technical Committee, yikes what a mouthful). They help with special cases like the abbreviations "Mr." and "Ms." where breaks should not occur. ULI rules are enabled by default, but you can disable them via the `:use_uli_exceptions` option:

```ruby
<% iterator = TwitterCldr::Shared::BreakIterator.new(:en, :use_uli_exceptions => false) %>
iterator = TwitterCldr::Shared::BreakIterator.new(:en, :use_uli_exceptions => false)
iterator.each_sentence("I like Ms. Murphy, she's nice.") do |sentence|
  puts sentence  # <%= assert(iterator.each_sentence("I like Ms. Murphy, she's nice.").to_a, ["I like Ms.", " Murphy, she's nice."]).map { |s| "\"#{s}\"" }.join(", ") %>
end
```

### Unicode Data

TwitterCLDR provides ways to retrieve individual code points as well as normalize and decompose Unicode text.

Retrieve data for code points:

```ruby
<% code_point = TwitterCldr::Shared::CodePoint.find(0x1F3E9) %>
code_point = TwitterCldr::Shared::CodePoint.find(0x1F3E9)
code_point.name             # <%= assert(code_point.name, "LOVE HOTEL").inspect %>
code_point.bidi_mirrored    # <%= assert(code_point.bidi_mirrored, "N").inspect %>
code_point.category         # <%= assert(code_point.category, "So").inspect %>
code_point.combining_class  # <%= assert(code_point.combining_class, "0").inspect %>
```

Convert characters to code points:

```ruby
TwitterCldr::Utils::CodePoints.from_string("¿")  # <%= assert(TwitterCldr::Utils::CodePoints.from_string("¿"), [0xBF]).inspect %>
```

Convert code points to characters:

```ruby
TwitterCldr::Utils::CodePoints.to_string([0xBF])  # <%= assert(TwitterCldr::Utils::CodePoints.to_string([0xBF]), "¿").inspect %>
```

#### Normalization

Normalize/decompose a Unicode string (NFD, NFKD, NFC, and NFKC implementations available).  Note that the normalized string will almost always look the same as the original string because most character display systems automatically combine decomposed characters.

```ruby
TwitterCldr::Normalization.normalize("français")  # <%= assert(TwitterCldr::Normalization.normalize("français"), "français").inspect %>
```

Normalization is easier to see in hex:

```ruby
# <%= assert(TwitterCldr::Utils::CodePoints.from_string("español"), [101, 115, 112, 97, 241, 111, 108]).inspect %>
TwitterCldr::Utils::CodePoints.from_string("español")

# <%= assert(TwitterCldr::Utils::CodePoints.from_string(TwitterCldr::Normalization.normalize("español")), [101, 115, 112, 97, 110, 771, 111, 108]).inspect %>
TwitterCldr::Utils::CodePoints.from_string(TwitterCldr::Normalization.normalize("español"))
```

Notice in the example above that the letter "ñ" was transformed from `241` to `110 771`, which represent the "n" and the "˜" respectively.

A few convenience methods also exist for `String` that make it easy to normalize and get code points for strings:

```ruby
# <%= assert("español".localize.code_points, [101, 115, 112, 97, 241, 111, 108]).inspect %>
"español".localize.code_points

# <%= assert("español".localize.normalize.code_points, [101, 115, 112, 97, 110, 771, 111, 108]).inspect %>
"español".localize.normalize.code_points
```

Specify a specific normalization algorithm via the `:using` option.  NFD, NFKD, NFC, and NFKC algorithms are all supported (default is NFD):

```ruby
# <%= assert("español".localize.normalize(:using => :nfkd).code_points, [101, 115, 112, 97, 110, 771, 111, 108]).inspect %>
"español".localize.normalize(:using => :NFKD).code_points
```

#### Casefolding

Casefolding is, generally speaking, the process of converting uppercase characters to lowercase ones so as to make text uniform and therefore easier to search. The canonical example of this is the German double "s". The "ß" character is transformed into "ss" by casefolding.

```ruby
"Hello, World".localize.casefold.to_s  # <%= assert("Hello, World".localize.casefold.to_s, "hello, world") %>
"Weißrussland".localize.casefold.to_s  # <%= assert("Weißrussland".localize.casefold.to_s, "weissrussland") %>
```

Turkic languages make use of the regular and dotted uppercase i characters "I" and "İ". Normal casefolding will convert a dotless uppercase "I" to a lowercase, dotted "i", which is correct in English. Turkic languages however expect the lowercase version of a dotless uppercase "I" to be a lowercase, dotless "ı". Pass the `:t` option to the `casefold` method to force Turkic treatment of "i" characters. By default, the `:t` option is set to true for Turkish and Azerbaijani:

```ruby
"Istanbul".localize.casefold(:t => true).to_s  # <%= assert("Istanbul".localize.casefold(:t => true).to_s, "ıstanbul") %>
"Istanbul".localize(:tr).casefold.to_s         # <%= assert("Istanbul".localize(:tr).casefold.to_s, "ıstanbul") %>
```

### Sorting (Collation)

TwitterCLDR contains an implementation of the [Unicode Collation Algorithm (UCA)](http://unicode.org/reports/tr10/) that provides language-sensitive text sorting capabilities.  Conveniently, all you have to do is use the `sort` method in combination with the familiar `localize` method.  Notice the difference between the default Ruby sort, which simply compares bytes, and the proper language-aware sort from TwitterCLDR in this German example:

```ruby
["Art", "Wasa", "Älg", "Ved"].sort                       # <%= assert(["Art", "Wasa", "Älg", "Ved"].sort, ["Art", "Ved", "Wasa", "Älg"]).inspect %>
["Art", "Wasa", "Älg", "Ved"].localize(:de).sort.to_a    # <%= assert(["Art", "Wasa", "Älg", "Ved"].localize(:de).sort.to_a, ["Älg", "Art", "Ved", "Wasa"]).inspect %>
```

Behind the scenes, these convenience methods are creating instances of `LocalizedArray`, then using the `TwitterCldr::Collation::Collator` class to sort the elements:

```ruby
<% collator = TwitterCldr::Collation::Collator.new(:de) %>
collator = TwitterCldr::Collation::Collator.new(:de)
collator.sort(["Art", "Wasa", "Älg", "Ved"])      # <%= assert(collator.sort(["Art", "Wasa", "Älg", "Ved"]), ["Älg", "Art", "Ved", "Wasa"]).inspect %>
collator.sort!(["Art", "Wasa", "Älg", "Ved"])     # <%= assert(collator.sort!(["Art", "Wasa", "Älg", "Ved"]), ["Älg", "Art", "Ved", "Wasa"]).inspect %>
```

The `TwitterCldr::Collation::Collator` class also provides methods to compare two strings, get sort keys, and calculate collation elements for individual strings:

```ruby
<% collator = TwitterCldr::Collation::Collator.new(:de) %>
collator = TwitterCldr::Collation::Collator.new(:de)
collator.compare("Art", "Älg")           # <%= assert(collator.compare("Art", "Älg"), 1).inspect %>
collator.compare("Älg", "Art")           # <%= assert(collator.compare("Älg", "Art"), -1).inspect %>
collator.compare("Art", "Art")           # <%= assert(collator.compare("Art", "Art"), 0).inspect %>

collator.get_collation_elements("Älg")   # <%= assert(collator.get_collation_elements("Älg"), [[39, 5, 143], [0, 157, 5], [61, 5, 5], [51, 5, 5]]).inspect %>

collator.get_sort_key("Älg")             # <%= assert(collator.get_sort_key("Älg"), [39, 61, 51, 1, 134, 157, 6, 1, 143, 7]).inspect %>
```

**Note**: The TwitterCLDR collator does not currently pass all the collation tests provided by Unicode, but for some strange reasons.  See the [summary](https://gist.github.com/f4ee3bd280a2257c5641) of these discrepancies if you're curious.

### Handling Bidirectional Text

When it comes to displaying text written in both right-to-left (RTL) and left-to-right (LTR) languages, most display systems run into problems.  The trouble is that Arabic or Hebrew text and English text (for example) often get scrambled visually and are therefore difficult to read.  It's not usually the basic ASCII characters like A-Z that get scrambled - it's most often punctuation marks and the like that are confusingly mixed up (they are considered "weak" types by Unicode).

To mitigate this problem, Unicode supports special invisible characters that force visual reordering so that mixed RTL and LTR (called "bidirectional") text renders naturally on the screen.  The Unicode Consortium has developed an algorithm (The Unicode Bidirectional Algorithm, or UBA) that intelligently inserts these control characters where appropriate.  You can make use of the UBA implementation in TwitterCLDR by creating a new instance of `TwitterCldr::Shared::Bidi` using the `from_string` static method, and manipulating it like so:

```ruby
<%
  assert_no_error(lambda do
    bidi = TwitterCldr::Shared::Bidi.from_string("hello نزوة world", :direction => :RTL)
    bidi.reorder_visually!
    bidi.to_s
  end)
%>
bidi = TwitterCldr::Shared::Bidi.from_string("hello نزوة world", :direction => :RTL)
bidi.reorder_visually!
bidi.to_s
```

**Disclaimer**: Google Translate tells me the Arabic in the example above means "fancy", but my confidence is not very high, especially since all the letters are unattached. Apologies to any native speakers :)

### Unicode YAML Support

Ruby 1.8 does not come with great Unicode support, and nowhere is this more apparent then when dumping Unicode characters in YAML.  The Psych gem by @tenderlove is a good replacement and is the default in Ruby 1.9, but requires libyaml and still doesn't handle Unicode characters perfectly.  To mitigate this problem (especially in Ruby 1.8), TwitterCLDR contains an adaptation of the [ya2yaml](https://github.com/afunai/ya2yaml) gem by Akira Funai.  Our changes specifically add better dumping of Ruby symbols.  If you can get Mr. Funai's attention, please gently remind him to merge @camertron's pull request so we can use his gem and not have to maintain a separate version :)  Fortunately, YAML parsing can still be done with the usual `YAML.load` or `YAML.load_file`.

You can make use of TwitterCLDR's YAML dumper by calling `localize` and then `to_yaml` on an `Array`, `Hash`, or `String`:

```ruby
{ :hello => "world" }.localize.to_yaml <% assert_no_error(lambda { { :hello => "world" }.localize.to_yaml }) %>
["hello", "world"].localize.to_yaml <% assert_no_error(lambda { { :hello => "world" }.localize.to_yaml }) %>
"hello, world".localize.to_yaml <% assert_no_error(lambda { "hello, world".localize.to_yaml }) %>
```

Behind the scenes, these convenience methods are using the `TwitterCldr::Shared::YAML` class.  You can do the same thing if you're feeling adventurous:

```ruby
TwitterCldr::Shared::YAML.dump({ :hello => "world" }) <% assert_no_error(lambda { TwitterCldr::Utils::YAML.dump({ :hello => "world" }) }) %>
TwitterCldr::Shared::YAML.dump(["hello", "world"]) <% assert_no_error(lambda { TwitterCldr::Utils::YAML.dump(["hello", "world"]) }) %>
TwitterCldr::Shared::YAML.dump("hello, world") <% assert_no_error(lambda { TwitterCldr::Utils::YAML.dump("hello, world") }) %>
```

## About Twitter-specific Locales

Twitter tries to always use BCP-47 language codes.  Data from the CLDR doesn't always match those codes however, so TwitterCLDR provides a `convert_locale` method to convert between the two.  All functionality throughout the entire gem defers to `convert_locale` before retrieving CLDR data.  `convert_locale` supports Twitter-supported BCP-47 language codes as well as CLDR locale codes, so you don't have to guess which one to use.  Here are a few examples:

```ruby
TwitterCldr.convert_locale(:'zh-cn')          # <%= assert(TwitterCldr.convert_locale(:'zh-cn'), :zh).inspect %>
TwitterCldr.convert_locale(:zh)               # <%= assert(TwitterCldr.convert_locale(:zh), :zh).inspect %>
TwitterCldr.convert_locale(:'zh-tw')          # <%= assert(TwitterCldr.convert_locale(:'zh-tw'), :'zh-Hant').inspect %>
TwitterCldr.convert_locale(:'zh-Hant')        # <%= assert(TwitterCldr.convert_locale(:'zh-Hant'), :'zh-Hant').inspect %>

TwitterCldr.convert_locale(:msa)              # <%= assert(TwitterCldr.convert_locale(:msa), :ms).inspect %>
TwitterCldr.convert_locale(:ms)               # <%= assert(TwitterCldr.convert_locale(:ms), :ms).inspect %>
```

There are a few functions in TwitterCLDR that don't require a locale code, and instead use the default locale by calling `TwitterCldr.locale`.  The `locale` function defers to `FastGettext.locale` when the FastGettext library is available, and falls back on :en (English) when it's not.  (Twitter uses the FastGettext gem to retrieve translations efficiently in Ruby).

```ruby
TwitterCldr.get_locale    # will return <%= assert(TwitterCldr.locale, :en).inspect %>

require 'fast_gettext'
FastGettext.locale = "ru"

TwitterCldr.locale    # will return :ru
```

## Compatibility

TwitterCLDR is fully compatible with Ruby 1.8.7, 1.9.3, 2.0.0, and, experimentally, 2.1.0. We are considering dropping support for Ruby 1.8. If you still need to use TwitterCLDR in a Ruby 1.8 environment, please let us know as soon as possible. Please note that certain TwitterCLDR features require additional dependencies or considerations when run on Ruby 1.8. Refer to the sections above for details.

#### Notes on Ruby 1.8

Numerous TwitterCLDR features have been built with the assumption that they will only ever be used on UTF-8 encoded text, which is mostly due to the need to support Ruby 1.8. For this reason, you may find it necessary to set the global `$KCODE` variable to `"UTF-8"`. Setting this variable tells Ruby what encoding to use when loading source files. TwitterCLDR will **not** set this value for you.

```ruby
$KCODE = "UTF-8"
```

## Requirements

No external requirements.

## Running Tests

`bundle exec rake` will run our basic test suite suitable for development.  To run the full test suite, use `bundle exec rake spec:full`.  The full test suite takes considerably longer to run because it runs against the complete normalization and collation test files from the Unicode Consortium.  The basic test suite only runs normalization and collation tests against a small subset of the complete test file.

Tests are written in RSpec using RR as the mocking framework.

## Test Coverage

You can run the development test coverage suite with `bundle exec rake spec:cov`, or the full suite with `bundle exec rake spec:cov:full`.  TwitterCLDR uses RCov under Ruby 1.8 and Simplecov under Ruby 1.9.

## JavaScript Support

TwitterCLDR currently supports localization of certain textual objects in JavaScript via the twitter-cldr-js gem.  See [http://github.com/twitter/twitter-cldr-js](http://github.com/twitter/twitter-cldr-js) for details.

## Authors

* Cameron C. Dutro: http://github.com/camertron
* Kirill Lashuk: http://github.com/kl-7
* Portions adapted from the ruby-cldr gem by Sven Fuchs: http://github.com/svenfuchs/ruby-cldr

## Links
* ruby-cldr gem: [http://github.com/svenfuchs/ruby-cldr](http://github.com/svenfuchs/ruby-cldr)
* fast_gettext gem: [https://github.com/grosser/fast_gettext](https://github.com/grosser/fast_gettext)
* CLDR homepage: [http://cldr.unicode.org/](http://cldr.unicode.org/)

## License

Copyright 2014 Twitter, Inc.

Licensed under the Apache License, Version 2.0: http://www.apache.org/licenses/LICENSE-2.0