Class: GIGO::CharDet::EscCharSetProber

Inherits:
CharSetProber show all
Defined in:
lib/gigo/rchardet/escprober.rb

Instance Attribute Summary

Attributes inherited from CharSetProber

#active

Instance Method Summary collapse

Methods inherited from CharSetProber

#filter_high_bit_only, #filter_with_english_letters, #filter_without_english_letters, #get_state

Constructor Details

#initialize ⇒ EscCharSetProber

Returns a new instance of EscCharSetProber.



31
32
33
34
35
36
37
38
39
40
 
# File 'lib/gigo/rchardet/escprober.rb', line 31

def initialize
  super()
  @_mCodingSM = [ 
  	CodingStateMachine.new(HZSMModel),
  	CodingStateMachine.new(ISO2022CNSMModel),
  	CodingStateMachine.new(ISO2022JPSMModel),
  	CodingStateMachine.new(ISO2022KRSMModel)
  ]
  reset
end

Instance Method Details

#feed(aBuf) ⇒ Object 



65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
 
# File 'lib/gigo/rchardet/escprober.rb', line 65

def feed(aBuf)
  for c in aBuf.split('')
  	for codingSM in @_mCodingSM
  	  next unless codingSM
  	  next unless codingSM.active
  	  codingState = codingSM.next_state(c)
  	  if codingState == EError
  	    codingSM.active = false
  	    @_mActiveSM -= 1
  	    if @_mActiveSM <= 0
  	      @_mState = ENotMe
  	      return get_state
  	    end
  	  elsif codingState == EItsMe
  	    @_mState = EFoundIt
  	    @_mDetectedCharset = codingSM.get_coding_state_machine
  	    return get_state
  	  end
  	end
  end

  return get_state
end

#get_charset_name ⇒ Object 



53
54
55
 
# File 'lib/gigo/rchardet/escprober.rb', line 53

def get_charset_name
  return @_mDetectedCharset
end

#get_confidence ⇒ Object 



57
58
59
60
61
62
63
 
# File 'lib/gigo/rchardet/escprober.rb', line 57

def get_confidence
  if @_mDetectedCharset
   return 0.99
  else
   return 0.00
  end
end

#reset ⇒ Object 



42
43
44
45
46
47
48
49
50
51
 
# File 'lib/gigo/rchardet/escprober.rb', line 42

def reset
  super
  for codingSM in @_mCodingSM
  	next if not codingSM
  	codingSM.active = true
  	codingSM.reset
  end
  @_mActiveSM = @_mCodingSM.length
  @_mDetectedCharset = nil
end