Class: TwitterCldr::Collation::Collator

Inherits:
Object
  • Object
show all
Defined in:
lib/twitter_cldr/collation/collator.rb

Overview

Collator uses fractional collation elements table form CLDR to generate sort keys for Unicode strings as well as compare and sort such strings by generated sort keys.

Constant Summary collapse

FRACTIONAL_UCA_SHORT_RESOURCE = 
'collation/FractionalUCA_SHORT.txt'

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(locale = nil) ⇒ Collator

Returns a new instance of Collator.



18
19
20
21
 
# File 'lib/twitter_cldr/collation/collator.rb', line 18

def initialize(locale = nil)
  @locale = TwitterCldr.convert_locale(locale) if locale
  @trie   = load_trie
end

Instance Attribute Details

#locale ⇒ Object

Returns the value of attribute locale.



16
17
18
 
# File 'lib/twitter_cldr/collation/collator.rb', line 16

def locale
  @locale
end

Class Method Details

.default_fce_trie ⇒ Object

Loads and memoizes the default Fractional Collation Elements trie.



138
139
140
 
# File 'lib/twitter_cldr/collation/collator.rb', line 138

def default_fce_trie
  @default_fce_trie ||= TwitterCldr::Collation::TrieBuilder.load_trie(FRACTIONAL_UCA_SHORT_RESOURCE).lock
end

.tailored_fce_trie(locale) ⇒ Object 



142
143
144
 
# File 'lib/twitter_cldr/collation/collator.rb', line 142

def tailored_fce_trie(locale)
  tailored_fce_tries_cache[locale]
end

Instance Method Details

#compare(string_a, string_b) ⇒ Object 



32
33
34
 
# File 'lib/twitter_cldr/collation/collator.rb', line 32

def compare(string_a, string_b)
  string_a == string_b ? 0 : get_sort_key(string_a) <=> get_sort_key(string_b)
end

#get_collation_elements(string_or_code_points) ⇒ Object 



40
41
42
43
44
45
46
 
# File 'lib/twitter_cldr/collation/collator.rb', line 40

def get_collation_elements(string_or_code_points)
  integer_code_points = get_normalized_code_points(string_or_code_points)

  result = []
  result.concat(code_point_collation_elements(integer_code_points)) until integer_code_points.empty?
  result
end

#get_sort_key(string_or_code_points) ⇒ Object 



36
37
38
 
# File 'lib/twitter_cldr/collation/collator.rb', line 36

def get_sort_key(string_or_code_points)
  TwitterCldr::Collation::SortKeyBuilder.build(get_collation_elements(string_or_code_points))
end

#sort(strings) ⇒ Object 



23
24
25
 
# File 'lib/twitter_cldr/collation/collator.rb', line 23

def sort(strings)
  strings.map{ |s| [s, get_sort_key(s)] }.sort{ |a, b| a[1] <=> b[1] }.map(&:first)
end

#sort!(strings) ⇒ Object 



27
28
29
30
 
# File 'lib/twitter_cldr/collation/collator.rb', line 27

def sort!(strings)
  sort_keys = Hash.new { |hash, string| hash[string] = get_sort_key(string) }
  strings.replace(strings.sort_by { |s| sort_keys[s] })
end