Module: Segment

Extended by:
Entity
Included in:
NLP::GdepChunk, NLP::GdepToken, NamedEntity, Token
Defined in:
lib/rbbt/segment.rb,
lib/rbbt/segment/encoding.rb,
lib/rbbt/segment/overlaps.rb

Defined Under Namespace

Modules: RangeIndex

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.align(text, parts) ⇒ Object



156
157
158
159
160
161
162
163
164
165
166
# File 'lib/rbbt/segment.rb', line 156

def self.align(text, parts)
  pre_offset = 0
  docid = text.respond_to?(:docid) ? text.docid : nil
  parts.each do |part|
    offset = text.index part
    next if offset.nil?
    Segment.setup(part, pre_offset + offset, docid)
    pre_offset += offset + part.segment_length - 1
    text = text[(offset + part.segment_length - 1)..-1]
  end
end

.ascii(text, replace = nil, &block) ⇒ Object



13
14
15
16
17
# File 'lib/rbbt/segment/encoding.rb', line 13

def self.ascii(text, replace = nil, &block)
  bad = bad_chars(text)
  replace = "?" if replace.nil?
  Transformed.with_transform(text, bad, replace, &block)
end

.bad_chars(text) ⇒ Object



3
4
5
6
7
8
9
10
11
# File 'lib/rbbt/segment/encoding.rb', line 3

def self.bad_chars(text)
  segments = []
  text.chars.each_with_index do |c,i|
    if ! c.ascii_only?
      segments << Segment.setup(c, :offset => i)
    end
  end
  segments
end

.clean_sort(segments) ⇒ Object



103
104
105
106
107
108
109
110
111
# File 'lib/rbbt/segment.rb', line 103

def self.clean_sort(segments)
  sorted = sort(segments).reject{|s| s.offset.nil?}
  overlaps = overlaps(sorted)
  overlaps.each do |s|
    sorted.delete s
  end

  sorted
end

.collisions(main, secondary) ⇒ Object



58
59
60
61
62
# File 'lib/rbbt/segment/overlaps.rb', line 58

def self.collisions(main, secondary)
  secondary.select do |ss|
    main.select{|ms| ms.overlaps? ss }.any?
  end
end

.index(*args) ⇒ Object



168
169
170
# File 'lib/rbbt/segment.rb', line 168

def self.index(*args)
  Segment::RangeIndex.index(*args)
end

.overlaps(sorted_segments) ⇒ Object



91
92
93
94
95
96
97
98
99
100
101
# File 'lib/rbbt/segment.rb', line 91

def self.overlaps(sorted_segments)
  last = nil
  overlaped = []

  sorted_segments.reverse.each do |segment| 
    overlaped << segment if (not last.nil?) and segment.range.end > last 
    last = segment.range.begin
  end

  overlaped
end

.sort(segments, inline = true) ⇒ Object



70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
# File 'lib/rbbt/segment.rb', line 70

def self.sort(segments, inline = true)
  if inline
    segments.sort do |a,b| 
      case
      when ((a.nil? and b.nil?) or (a.offset.nil? and b.offset.nil?))
        0
      when (a.nil? or a.offset.nil?)
        -1
      when (b.nil? or b.offset.nil?)
        +1
      when (not a.range.include? b.offset.to_i and not b.range.include? a.offset.to_i)
        a.offset.to_i <=> b.offset.to_i
      else
        a.segment_length <=> b.segment_length
      end
    end
  else
    segments.sort_by do |segment| segment.offset.to_i || 0 end.reverse
  end
end

.split(text, segments, skip_segments = false) ⇒ Object



113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
# File 'lib/rbbt/segment.rb', line 113

def self.split(text, segments, skip_segments = false)
  sorted_segments = clean_sort segments

  chunks      = []
  segment_end = 0
  text_offset = 0
  sorted_segments.each do |segment|
    return chunks if text.nil? or text.empty?
    next if segment.offset.nil?
    offset = segment.offset - text_offset

    # Consider segment offset. Save pre, or skip if overlap
    case
    when offset < 0 # Overlap, skip
      next
    when offset > 0 # Save pre
      chunk = text[0..offset - 1]
      Segment.setup(chunk, text_offset)
      chunks << chunk
    end

    segment_end = offset + segment.segment_length - 1

    if not skip_segments
      chunk = text[offset..segment_end]
      Segment.setup(chunk, text_offset + offset)
      chunks << chunk
    end

    text_offset += segment_end + 1
    text = text[segment_end + 1..-1]

  end

  if not text.nil? and not text.empty?
    chunk = text.dup
    Segment.setup(chunk, text_offset)
    chunks << chunk
  end

  chunks
end

Instance Method Details

#eendObject



52
53
54
# File 'lib/rbbt/segment.rb', line 52

def eend
  offset.to_i + length - 1
end

#includes?(segment) ⇒ Boolean

Returns:

  • (Boolean)


44
45
46
47
# File 'lib/rbbt/segment/overlaps.rb', line 44

def includes?(segment)
  (segment.offset.to_i >= self.offset.to_i) and
  (segment.offset.to_i + segment.segment_length.to_i <= self.offset.to_i + self.segment_length.to_i)
end

#make_relative(segments, &block) ⇒ Object



22
23
24
25
26
27
28
29
30
# File 'lib/rbbt/segment/overlaps.rb', line 22

def make_relative(segments, &block)
  if block_given?
    segments.each{|s| s.push offset}
    yield(segments)
    segments.each{|s| s.pull offset}
  else
    segments.each{|s| s.push offset}
  end
end

#overlaps(segments) ⇒ Object



54
55
56
# File 'lib/rbbt/segment/overlaps.rb', line 54

def overlaps(segments)
  segments.select{|s| self.overlaps?(s) }
end

#overlaps?(segment) ⇒ Boolean

Returns:

  • (Boolean)


49
50
51
52
# File 'lib/rbbt/segment/overlaps.rb', line 49

def overlaps?(segment)
  segment.offset.to_i >= self.offset.to_i && segment.offset.to_i <= self.eend || 
  self.offset.to_i >= segment.offset.to_i && self.offset.to_i <= segment.eend
end

#pull(offset) ⇒ Object



2
3
4
5
6
7
8
9
10
# File 'lib/rbbt/segment/overlaps.rb', line 2

def pull(offset)
  if self.offset.nil? or offset.nil?
    self.offset = nil
  else
    self.offset += offset 
  end

  self
end

#push(offset) ⇒ Object



12
13
14
15
16
17
18
19
20
# File 'lib/rbbt/segment/overlaps.rb', line 12

def push(offset)
  if self.offset.nil? or offset.nil?
    self.offset = nil
  else
    self.offset -= offset 
  end

  self
end

#rangeObject



56
57
58
# File 'lib/rbbt/segment.rb', line 56

def range
  (offset.to_i..eend)
end

#range_in(container = nil) ⇒ Object



32
33
34
35
36
37
38
39
40
41
42
# File 'lib/rbbt/segment/overlaps.rb', line 32

def range_in(container = nil)
  raise "No offset specified" if offset.nil?
  case
  when (Segment === container and not container.offset.nil?)
    ((offset - container.offset)..(self.eend - container.offset))
  when Integer === container
    ((offset - container)..(self.eend - container))
  else
    range
  end
end

#segment_lengthObject



48
49
50
# File 'lib/rbbt/segment.rb', line 48

def segment_length
  length
end