Class: TwitterCldr::Segmentation::PossibleWord

Inherits:
Object
  • Object
show all
Defined in:
lib/twitter_cldr/segmentation/possible_word.rb

Constant Summary collapse

POSSIBLE_WORD_LIST_MAX =

list size, limited by the maximum number of words in the dictionary that form a nested sequence.

20

Instance Method Summary collapse

Constructor Details

#initialize ⇒ PossibleWord

Returns a new instance of PossibleWord.



13
14
15
16
17
 
# File 'lib/twitter_cldr/segmentation/possible_word.rb', line 13

def initialize
  @lengths = []
  @count = nil
  @offset = -1
end

Instance Method Details

#accept_marked(cursor) ⇒ Object

select the currently marked candidate, point after it in the text, and invalidate self



46
47
48
49
 
# File 'lib/twitter_cldr/segmentation/possible_word.rb', line 46

def accept_marked(cursor)
  cursor.position = @offset + @lengths[@mark]
  @lengths[@mark]
end

#back_up(cursor) ⇒ Object

back up from the current candidate to the next shorter one; return true if that exists and point the text after it



53
54
55
56
57
58
59
60
61
 
# File 'lib/twitter_cldr/segmentation/possible_word.rb', line 53

def back_up(cursor)
  if @current > 0
    @current -= 1
    cursor.position = @offset + @lengths[@current]
    return true
  end

  false
end

#candidates(cursor, dictionary, end_pos) ⇒ Object

fill the list of candidates if needed, select the longest, and return the number found



20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
 
# File 'lib/twitter_cldr/segmentation/possible_word.rb', line 20

def candidates(cursor, dictionary, end_pos)
  start = cursor.position

  if start != @offset
    @offset = start
    @count, _, @lengths, @prefix = dictionary.matches(
      cursor, end_pos - start, POSSIBLE_WORD_LIST_MAX
    )

    # dictionary leaves text after longest prefix, not longest word, so back up.
    if @count <= 0
      cursor.position = start
    end
  end

  if @count > 0
    cursor.position = start + @lengths[@count - 1]
  end

  @current = @count - 1
  @mark = @current

  return @count
end

#longest_prefix ⇒ Object

return the longest prefix this candidate location shares with a dictionary word



64
65
66
 
# File 'lib/twitter_cldr/segmentation/possible_word.rb', line 64

def longest_prefix
  @prefix
end

#mark_current ⇒ Object

mark the current candidate as the one we like



69
70
71
 
# File 'lib/twitter_cldr/segmentation/possible_word.rb', line 69

def mark_current
  @mark = @current
end