Module: Philiprehberger::FuzzyMatch

Defined in:
lib/philiprehberger/fuzzy_match.rb,
lib/philiprehberger/fuzzy_match/dice.rb,
lib/philiprehberger/fuzzy_match/version.rb,
lib/philiprehberger/fuzzy_match/levenshtein.rb,
lib/philiprehberger/fuzzy_match/jaro_winkler.rb

Defined Under Namespace

Modules: Dice, JaroWinkler, Levenshtein

Constant Summary collapse

VERSION = 
'0.1.6'

Class Method Summary collapse

Class Method Details

.best(query, candidates, threshold: 0.0) ⇒ Object 



32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 32

def self.best(query, candidates, threshold: 0.0)
  return nil if candidates.empty?

  best_result = nil
  best_score = -1.0

  candidates.each do |candidate|
    score = ratio(query, candidate.to_s)
    if score > best_score
      best_result = candidate
      best_score = score
    end
  end

  return nil if best_score < threshold

  { match: best_result, score: best_score.round(4) }
end

.dice_coefficient(str_a, str_b) ⇒ Object 



18
19
20
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 18

def self.dice_coefficient(str_a, str_b)
  Dice.coefficient(str_a, str_b)
end

.jaro_winkler(str_a, str_b) ⇒ Object 



14
15
16
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 14

def self.jaro_winkler(str_a, str_b)
  JaroWinkler.similarity(str_a, str_b)
end

.levenshtein(str_a, str_b) ⇒ Object 



10
11
12
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 10

def self.levenshtein(str_a, str_b)
  Levenshtein.distance(str_a, str_b)
end

.ratio(str_a, str_b) ⇒ Object 



22
23
24
25
26
27
28
29
30
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 22

def self.ratio(str_a, str_b)
  a = str_a.to_s.downcase
  b = str_b.to_s.downcase
  max_len = [a.length, b.length].max
  return 1.0 if max_len.zero?

  distance = Levenshtein.distance(a, b)
  1.0 - (distance.to_f / max_len)
end

.search(query, candidates, threshold: 0.3) ⇒ Object 



51
52
53
54
55
56
57
58
59
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 51

def self.search(query, candidates, threshold: 0.3)
  scored = candidates.map do |candidate|
    score = ratio(query, candidate.to_s)
    { match: candidate, score: score.round(4) }
  end

  results = scored.select { |r| r[:score] >= threshold }
  results.sort_by { |r| -r[:score] }
end

.suggest(query, candidates, threshold: 0.6, max: 5) ⇒ Object 



61
62
63
64
 
# File 'lib/philiprehberger/fuzzy_match.rb', line 61

def self.suggest(query, candidates, threshold: 0.6, max: 5)
  results = search(query, candidates, threshold: threshold)
  results.first(max).map { |r| r[:match] }
end