Class: Yosina::Transliterators::HistoricalHirakatas::Transliterator

Inherits:
BaseTransliterator show all
Defined in:
lib/yosina/transliterators/historical_hirakatas.rb

Overview

Transliterator for historical hiragana/katakana conversion

Instance Method Summary collapse

Constructor Details

#initialize(options = {}) ⇒ Transliterator

Initialize the transliterator with options

Parameters:

  • options (Hash) (defaults to: {}) —

    Configuration options

Options Hash (options):

  • :hiraganas (String) —

    “simple” (default), “decompose”, or “skip”

  • :katakanas (String) —

    “simple” (default), “decompose”, or “skip”

  • :voiced_katakanas (String) —

    “decompose” or “skip” (default)



46
47
48
49
50
51
 
# File 'lib/yosina/transliterators/historical_hirakatas.rb', line 46

def initialize(options = {})
  super()
  @hiraganas = (options[:hiraganas] || :simple).to_sym
  @katakanas = (options[:katakanas] || :simple).to_sym
  @voiced_katakanas = (options[:voiced_katakanas] || :skip).to_sym
end

Instance Method Details

#call(input_chars) ⇒ Enumerable<Char>

Convert historical hiragana/katakana characters to modern equivalents

Parameters:

  • input_chars (Enumerable<Char>) —

    The characters to transliterate

Returns:

  • (Enumerable<Char>) —

    The transliterated characters



57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
 
# File 'lib/yosina/transliterators/historical_hirakatas.rb', line 57

def call(input_chars)
  Chars.enum do |y|
    offset = 0
    pending = nil
    input_chars.each do |char|
      if char.sentinel?
        offset = emit_char(y, pending, offset) if pending
        pending = nil
        y << char
        break
      end

      if pending.nil?
        pending = char
        next
      end

      if char.c == COMBINING_DAKUTEN
        # Check if pending char could be a decomposed voiced base
        decomposed = VOICED_HISTORICAL_KANA_DECOMPOSED_MAPPINGS[pending.c]
        if @voiced_katakanas == :skip || decomposed.nil?
          y << pending.with_offset(offset)
          offset += pending.c.length
          pending = char
          next
        end
        y << Char.new(c: U, offset: offset, source: pending)
        offset += U.length
        y << char.with_offset(offset)
        offset += char.c.length
        y << Char.new(c: decomposed, offset: offset, source: pending)
        offset += decomposed.length
        pending = nil
        next
      end

      offset = emit_char(y, pending, offset)
      pending = char
    end
    # Flush any remaining pending char
    emit_char(y, pending, offset) if pending
  end
end