# encoding: UTF-8

# Copyright 2012 Twitter, Inc
# http://www.apache.org/licenses/LICENSE-2.0

require 'spec_helper'

include TwitterCldr::Tokenizers

describe DateTokenizer do
  describe "#tokens" do
    it "should tokenize plaintext segments correctly (i.e. Spanish)" do
      tokenizer = DateTokenizer.new(:locale => :es)
      got       = tokenizer.tokens(:type => :full)
      expected  = [
          { :value => "EEEE", :type => :pattern },
          { :value => ", ", :type => :plaintext },
          { :value => "d", :type => :pattern },
          { :value => " ", :type => :plaintext },
          { :value => "'de'", :type => :plaintext },
          { :value => " ", :type => :plaintext },
          { :value => "MMMM", :type => :pattern },
          { :value => " ", :type => :plaintext },
          { :value => "'de'", :type => :plaintext },
          { :value => " ", :type => :plaintext },
          { :value => "y", :type => :pattern }
      ]
      check_token_list(got, expected)
    end

    it "should tokenize patterns with non-latin characters correctly (i.e. Japanese)" do
      tokenizer = DateTokenizer.new(:locale => :ja)
      got       = tokenizer.tokens(:type => :full)
      expected  = [
          { :value => "y", :type => :pattern },
          { :value => "年", :type => :plaintext },
          { :value => "M", :type => :pattern },
          { :value => "月", :type => :plaintext },
          { :value => "d", :type => :pattern },
          { :value => "日", :type => :plaintext },
          { :value => "EEEE", :type => :pattern }
      ]
      check_token_list(got, expected)
    end

    it "should tokenize composites correctly" do
      # Ensure that buddhist calendar data is present in th locale.
      TwitterCldr.get_locale_resource(:th, :calendars)[:th][:calendars][:buddhist].should_not(
          be_nil, 'buddhist calendar is missing for :th locale (check resources/locales/th/calendars.yml)'
      )

      tokenizer = DateTokenizer.new(:locale => :th, :calendar_type => :buddhist)
      got       = tokenizer.tokens(:type => :long)
      expected  = [
          { :value => "d", :type => :pattern },
          { :value => " ", :type => :plaintext },
          { :value => "MMMM", :type => :pattern },
          { :value => " พ.ศ. ", :type => :plaintext },
          { :to_s => "y + 543", :type => :composite }
      ]
      check_token_list(got, expected)
    end
  end
end