Class: RTF::Converter

Inherits:
Object
  • Object
show all
Defined in:
lib/rtf.rb

Class Method Summary collapse

Class Method Details

.rtf2text(str, format = :text) ⇒ Object

crappy



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
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
 
# File 'lib/rtf.rb', line 49

def self.rtf2text str, format=:text
  group = 0
  text = ''
  text << "<html>\n<body>" if format == :html
  group_type = []
  group_tags = []
  RTF::Tokenizer.process(StringIO.new(str)) do |a, b, c|
    add_text = ''
    case a
    when :open_group; group += 1; group_type[group] = nil; group_tags[group] = []
    when :close_group; group_tags[group].reverse.each { |t| text << "</#{t}>" }; group -= 1;
    when :control_word; # ignore
      group_type[group] ||= b
      # maybe change this to use utf8 where possible
      add_text = if b == 'par' || b == 'line' || b == 'page'; "\n"
      elsif b == 'tab' || b == 'cell'; "\t"
      elsif b == 'endash' || b == 'emdash'; "-"
      elsif b == 'emspace' || b == 'enspace' || b == 'qmspace'; " "
      elsif b == 'ldblquote'; '"'
      else ''
      end
      if b == 'b' || b == 'i' and format == :html
        close = c == 0 ? '/' : ''
        text << "<#{close}#{b}>"
        if c == 0
          group_tags[group].delete b
        else
          group_tags[group] << b
        end
      end
      # lot of other ones belong in here. =begin
\bullet   Bullet character.
\lquote   Left single quotation mark.
\rquote   Right single quotation mark.
\ldblquote  Left double quotation mark.
\rdblquote
=end
    when :control_symbol; # ignore
       group_type[group] ||= b
      add_text = ' ' if b == '~' # non-breakable space
      add_text = '-' if b == '_' # non-breakable hypen
    when :text
      add_text = b if group <= 1 or group_type[group] == 'rtlch' && !group_type[0...group].include?('*')
    end
    if format == :html
      text << add_text.gsub(/([<>&"'])/) do
        ent = { '<' => 'lt', '>' => 'gt', '&' => 'amp', '"' => 'quot', "'" => 'apos' }[$1]
        "&#{ent};"
      end
      text << '<br>' if add_text == "\n"
    else
      text << add_text
    end
  end
  text << "</body>\n</html>\n" if format == :html
  text
end