Class: String

Inherits:
Object
  • Object
show all
Defined in:
lib/lumix/charset.rb,
lib/lumix/concordancer.rb

Constant Summary collapse

NoMatchFound = 
Class.new(Exception)

Instance Method Summary collapse

Instance Method Details

#digest ⇒ Object 



25
26
27
28
29
30
 
# File 'lib/lumix/concordancer.rb', line 25

def digest
  return @digest if @digest
  digest = Digest::MD5.new
  digest.update self
  @digest = digest.hexdigest
end

#find_icu ⇒ Object 



18
19
20
21
22
23
24
25
26
 
# File 'lib/lumix/charset.rb', line 18

def find_icu
  matches = @icu.detect_all(self)
  matches.each do |match|
    if d = icu_return(match.name)
      return d
    end
  end
  return nil
end

#icu_return(cs) ⇒ Object 



28
29
30
31
32
33
 
# File 'lib/lumix/charset.rb', line 28

def icu_return(cs)
  begin
    return Iconv.conv('UTF-8', cs, self)
  rescue
  end
end

#to_utf(default = 'utf-8') ⇒ Object

Raises:



9
10
11
12
13
14
15
16
 
# File 'lib/lumix/charset.rb', line 9

def to_utf(default = 'utf-8')
  @icu ||= ICU::CharDet::Detector.new
  result = icu_return(default) || find_icu
  raise NoMatchFound unless result

  @entities ||= HTMLEntities.new
  @entities.decode(result)
end