Class: TwitterCldr::Resources::UnicodePropertiesImporter

Inherits:
UnicodeImporter show all
Defined in:
lib/twitter_cldr/resources/unicode_properties_importer.rb

Constant Summary collapse

PROPERTIES_BASE_URL = 
'ftp://ftp.unicode.org/Public/UCD/latest/ucd'
PROPERTIES = 
[
  "auxiliary/SentenceBreakProperty",
  "auxiliary/WordBreakProperty",
  "LineBreak"
]

Instance Method Summary collapse

Constructor Details

#initialize(input_path, output_path) ⇒ UnicodePropertiesImporter

Arguments:

input_path  - path to a directory containing the various property files
output_path - output directory for imported YAML files



25
26
27
28
 
# File 'lib/twitter_cldr/resources/unicode_properties_importer.rb', line 25

def initialize(input_path, output_path)
  @input_path  = input_path
  @output_path = output_path
end

Instance Method Details

#import ⇒ Object 



30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
 
# File 'lib/twitter_cldr/resources/unicode_properties_importer.rb', line 30

def import
  FileUtils.mkdir_p(@output_path)

  PROPERTIES.each do |property|
    input_file = property_data_file("#{property}.txt")
    output_file = File.join(@output_path, "#{fix_name(property)}.yml")

    File.open(output_file, "w+") do |f|
      f.write(
        YAML.dump(
          parse_standard_file(input_file).inject({}) do |ret, data|
            name = data[1].strip.to_sym
            ret[name] ||= []
            ret[name] += expand_range(data[0])
            ret
          end.inject({}) do |ret, (key, data)|
            ret[key] = TwitterCldr::Utils::RangeSet.rangify(data)
            ret
          end
        )
      )
    end
  end
end