Class: ChupaText::UTF8Converter

Inherits:
Object
  • Object
show all
Defined in:
lib/chupa-text/utf8-converter.rb

Instance Method Summary collapse

Constructor Details

#initialize(string, max_size: nil) ⇒ UTF8Converter


19
20
21
22
# File 'lib/chupa-text/utf8-converter.rb', line 19

def initialize(string, max_size: nil)
  @string = string
  @max_size = max_size
end

Instance Method Details

#convertObject


24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
# File 'lib/chupa-text/utf8-converter.rb', line 24

def convert
  encoding = @string.encoding
  case encoding
  when Encoding::UTF_8
    bom_size, bom_encoding = detect_bom
    if bom_size
      utf8_string = @string.byteslice(bom_size,
                                      @string.bytesize - bom_size)
    else
      utf8_string = @string
    end
    return truncate(utf8_string)
  when Encoding::ASCII_8BIT
    return truncate(@string) if @string.ascii_only?
  else
    utf8_string = @string.encode(Encoding::UTF_8,
                                 invalid: :replace,
                                 undef: :replace,
                                 replace: "")
    return truncate(utf8_string)
  end

  bom_size, bom_encoding = detect_bom
  if bom_encoding
    string_without_bom = @string.byteslice(bom_size,
                                           @string.bytesize - bom_size)
    utf8_string = string_without_bom.encode(Encoding::UTF_8,
                                            bom_encoding,
                                            invalid: :replace,
                                            undef: :replace,
                                            replace: "")
    return truncate(utf8_string)
  end

  guessed_encoding = guess_encoding
  if guessed_encoding
    truncate(@string.encode(Encoding::UTF_8,
                            guessed_encoding,
                            invalid: :replace,
                            undef: :replace,
                            replace: ""))
  else
    if @max_size
      utf8_string = @string.byteslice(0, @max_size)
    else
      utf8_string = @string.dup
    end
    utf8_string.force_encoding(Encoding::UTF_8)
    utf8_string.scrub!("")
    utf8_string.gsub!(/\p{Control}+/, "")
    utf8_string
  end
end