Module: Transformed
- Defined in:
- lib/rbbt/segment/transformed.rb
Instance Attribute Summary collapse
-
#transformation_stack ⇒ Object
Returns the value of attribute transformation_stack.
-
#transformed_segments ⇒ Object
Returns the value of attribute transformed_segments.
Class Method Summary collapse
- .transform(text, segments, replacement = nil, &block) ⇒ Object
- .with_transform(text, replace_segments, replacement = nil) ⇒ Object
Instance Method Summary collapse
- #fix_segment(segment, range, diff) ⇒ Object
- #replace_segments(segments, replacement = nil, strict = false, &block) ⇒ Object
-
#restore(segments = [], first_only = false) ⇒ Object
Restore the sentence from all transformation.
- #shift(segment_o) ⇒ Object
Instance Attribute Details
#transformation_stack ⇒ Object
Returns the value of attribute transformation_stack.
25 26 27 |
# File 'lib/rbbt/segment/transformed.rb', line 25 def transformation_stack @transformation_stack end |
#transformed_segments ⇒ Object
Returns the value of attribute transformed_segments.
25 26 27 |
# File 'lib/rbbt/segment/transformed.rb', line 25 def transformed_segments @transformed_segments end |
Class Method Details
.transform(text, segments, replacement = nil, &block) ⇒ Object
3 4 5 6 7 8 9 10 11 |
# File 'lib/rbbt/segment/transformed.rb', line 3 def self.transform(text, segments, replacement = nil, &block) block = replacement if Proc === replacement text.extend Transformed text.replace_segments(segments, replacement, &block) text end |
.with_transform(text, replace_segments, replacement = nil) ⇒ Object
13 14 15 16 17 18 19 20 21 22 23 |
# File 'lib/rbbt/segment/transformed.rb', line 13 def self.with_transform(text, replace_segments, replacement = nil) text.extend Transformed text.replace_segments(replace_segments, replacement) segments = yield text segments = [] unless Array === segments && Segment === segments.first text.restore(segments, true) end |
Instance Method Details
#fix_segment(segment, range, diff) ⇒ Object
139 140 141 142 143 144 145 146 147 148 149 150 151 152 |
# File 'lib/rbbt/segment/transformed.rb', line 139 def fix_segment(segment, range, diff) case # Before when segment.eend < range.begin # After when segment.offset.to_i > range.end + diff segment.offset = segment.offset.to_i - diff # Includes when (segment.offset.to_i <= range.begin and segment.eend >= range.end + diff) segment.replace self[segment.offset.to_i..segment.eend - diff] else raise "Segment overlaps with transformation: #{Misc.fingerprint(segment)} (#{segment.range} & #{range.begin}..#{range.end + diff})" end end |
#replace_segments(segments, replacement = nil, strict = false, &block) ⇒ Object
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 |
# File 'lib/rbbt/segment/transformed.rb', line 64 def replace_segments(segments, replacement = nil, strict = false, &block) @transformed_segments ||= {} @transformation_stack ||= [] stack = [] segments = [segments] unless Array === segments orig_length = self.length offset = self.respond_to?(:offset) ? self.offset.to_i : 0 segments = segments.collect do |s| if Segment === s s elsif String === s matches = self.scan(s) Segment.align(self, matches) end end.flatten segments = segments.select do |s| shift = shift s.range s_offset = s.offset.to_i s_offset += shift.first if shift s_offset >= offset && s_offset <= offset + self.length - 1 end Segment.clean_sort(segments).each do |segment| next if segment.offset.nil? shift = shift segment.range next if shift.nil? shift_begin, shift_end = shift text_offset = self.respond_to?(:offset)? self.offset.to_i : 0 updated_begin = segment.offset.to_i + shift_begin - text_offset updated_end = segment.range.last + shift_end - text_offset updated_range = (updated_begin..updated_end) updated_text = self[updated_begin..updated_end] if updated_text.nil? Log.warn "Range outside of segment: #{self.length} #{segment.range} (#{updated_range})" next end #raise "error '#{segment}' => '#{updated_text}'" if updated_text != segment original_text = segment.dup segment.replace updated_text case when block_given? new = block.call(segment) when String === replacement new = replacement when Proc === replacement new = replacement.call(segment) end diff = new.length - segment.segment_length self[updated_begin..updated_end] = new @transformed_segments[segment.object_id] = [segment.range, diff, updated_text, updated_range, @transformed_segments.size] segment.replace original_text stack << segment.object_id end @transformation_stack << stack end |
#restore(segments = [], first_only = false) ⇒ Object
Restore the sentence from all transformation. Segments that are passed as parameters are restored from transformed space to original space
156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 |
# File 'lib/rbbt/segment/transformed.rb', line 156 def restore(segments = [], first_only = false) return segments if @transformation_stack.empty? if first_only @transformation_stack.pop.reverse.each do |id| segment_info = @transformed_segments.delete id orig_range, diff, text, range = segment_info new_range = (range.begin..range.last + diff) self[new_range] = text segments = segments.collect do |segment| next segment unless Segment === segment begin fix_segment(segment, range, diff) segment rescue Log.low "Skipped: " + $!. next end end.compact if Array === segments end segments else while @transformation_stack.any? restore(segments, true) end segments end end |
#shift(segment_o) ⇒ Object
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
# File 'lib/rbbt/segment/transformed.rb', line 27 def shift(segment_o) begin_shift = 0 end_shift = 0 text_offset = self.respond_to?(:offset)? self.offset.to_i : 0 @transformed_segments.sort_by{|id, info| info.last}.each{|id,info| pseg_o, diff, utext, pseg_u, index = info pseg_u = ((pseg_u.begin + text_offset)..(pseg_u.last + text_offset)) case # Before when segment_o.last + end_shift < pseg_u.begin # After when (segment_o.begin + begin_shift > pseg_u.last) begin_shift += diff end_shift += diff # Includes when (segment_o.begin + begin_shift <= pseg_u.begin and segment_o.last + end_shift >= pseg_u.last) end_shift += diff # Inside when (segment_o.begin + begin_shift >= pseg_u.begin and segment_o.last + end_shift <= pseg_u.last) return nil # Overlaps start when (segment_o.begin + begin_shift <= pseg_u.begin and segment_o.last + end_shift <= pseg_u.last) return nil # Overlaps end when (segment_o.begin + begin_shift >= pseg_u.begin and segment_o.last + end_shift >= pseg_u.last) return nil else raise "Unknown overlaps: #{segment_o.inspect} - #{pseg_u.inspect}" end } [begin_shift, end_shift] end |