Method: JsonDiff.array_pairing

Defined in:
lib/json-diff/diff.rb

.array_pairing(before, after, options) ⇒ Object

[[before index, after index, similarity]],

removed: [before index],
added: [after index]

  • options: procedure taking (before, after) objects. Returns a probability between 0 and 1 of how likely after is a modification of before, or nil if it cannot determine it.



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
155
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/json-diff/diff.rb', line 118

def self.array_pairing(before, after, options)
  # Array containing the array of similarities from before to after.
  similarities = before.map do |before_item|
    after.map do |after_item|
      similarity(before_item, after_item, options)
    end
  end

  # Array containing the array of couples of indices, sorted by similarity.
  indices = before.map.with_index do |before_item, before_index|
    after.map.with_index do |after_item, after_index|
      [before_index, after_index]
    end
  end

  # Sort them in O(n^2 log(n)).
  indices.map! do |couples|
    couples.sort! do |a, b|
      a_before_index = a[0]
      b_before_index = b[0]
      a_after_index = a[1]
      b_after_index = b[1]

      similarities[b_before_index][b_after_index] <=> similarities[a_before_index][a_after_index]
    end
  end
  # Sort the toplevel.
  indices.sort! do |a, b|
    a_top_before_index = a[0][0]
    a_top_after_index = a[0][1]
    b_top_before_index = b[0][0]
    b_top_after_index = b[0][1]

    similarities[b_top_before_index][b_top_after_index] <=> similarities[a_top_before_index][a_top_after_index]
  end

  # Map from indices to boolean (true if paired).
  before_paired = {}
  after_paired = {}

  num_pairs = [before.size, after.size].min

  pairs = (0...num_pairs).map do |_|
    unpaired_before_index = indices.index { |a| !before_paired[a[0][0]] }
    unpaired_after_index = indices[unpaired_before_index].index { |a| !after_paired[a[1]] }
    unpaired_couple = indices[unpaired_before_index][unpaired_after_index]
    before_paired[unpaired_couple[0]] = true
    after_paired[unpaired_couple[1]] = true

    [unpaired_couple[0], unpaired_couple[1],
      similarities[unpaired_couple[0]][unpaired_couple[1]]]
  end

  if before.size < after.size
    added = after.map.with_index { |_, i| i} - after_paired.keys
    removed = []
  else
    removed = before.map.with_index { |_, i| i } - before_paired.keys
    added = []
  end

  {
    pairs: pairs,
    removed: removed,
    added: added,
  }
end