Class: Autosuggest

Inherits:
Object
  • Object
show all
Defined in:
lib/autosuggest.rb,
lib/autosuggest/version.rb

Constant Summary collapse

VERSION =
"0.1.0"

Instance Method Summary collapse

Constructor Details

#initialize(top_queries) ⇒ Autosuggest

Returns a new instance of Autosuggest.



7
8
9
10
11
12
13
14
15
# File 'lib/autosuggest.rb', line 7

def initialize(top_queries)
  @top_queries = top_queries
  @concepts = {}
  @words = Set.new
  @non_duplicates = Set.new
  @blacklisted_words = Set.new
  @preferred_queries = {}
  @profane_words = Set.new(Obscenity::Base.blacklist)
end

Instance Method Details

#add_concept(name, values) ⇒ Object



17
18
19
# File 'lib/autosuggest.rb', line 17

def add_concept(name, values)
  @concepts[name] = Set.new(values.compact.uniq.map(&:downcase))
end

#blacklist_words(words) ⇒ Object



45
46
47
48
49
# File 'lib/autosuggest.rb', line 45

def blacklist_words(words)
  words.each do |word|
    @blacklisted_words << word.downcase
  end
end

#not_duplicates(pairs) ⇒ Object



39
40
41
42
43
# File 'lib/autosuggest.rb', line 39

def not_duplicates(pairs)
  pairs.each do |pair|
    @non_duplicates << pair.map(&:downcase).sort
  end
end

#parse_words(phrases, options = {}) ⇒ Object



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
# File 'lib/autosuggest.rb', line 21

def parse_words(phrases, options = {})
  min = options[:min] || 1

  word_counts = Hash.new(0)
  phrases.each do |phrase|
    words = tokenize(phrase)
    words.each do |word|
      word_counts[word] += 1
    end
  end

  word_counts.select { |_, c| c >= min }.each do |word, _|
    @words << word
  end

  word_counts
end

#prefer(queries) ⇒ Object



51
52
53
54
55
# File 'lib/autosuggest.rb', line 51

def prefer(queries)
  queries.each do |query|
    @preferred_queries[normalize_query(query)] ||= query
  end
end

#pretty_suggestionsObject



128
129
130
131
132
133
134
# File 'lib/autosuggest.rb', line 128

def pretty_suggestions
  str = "%-30s   %5s   %s\n" % %w(Query Score Notes)
  suggestions.each do |suggestion|
    str << "%-30s   %5d   %s\n" % [suggestion[:query], suggestion[:score], suggestion[:notes].join(", ")]
  end
  str
end

#suggestionsObject



57
58
59
60
61
62
63
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
# File 'lib/autosuggest.rb', line 57

def suggestions
  stemmed_queries = {}
  added_queries = Set.new
  @top_queries.sort_by { |_query, count| -count }.map do |query, count|
    query = query.to_s

    # TODO do not ignore silently
    next if query.length < 2

    stemmed_query = normalize_query(query)

    # get preferred term
    preferred_query = @preferred_queries[stemmed_query]
    if preferred_query && preferred_query != query
      original_query, query = query, preferred_query
    end

    # exclude duplicates
    duplicate = stemmed_queries[stemmed_query]
    stemmed_queries[stemmed_query] ||= query

    # also detect possibly misspelled duplicates
    # TODO use top query as duplicate
    if !duplicate && query.length > 4
      edits(query).each do |edited_query|
        if added_queries.include?(edited_query)
          duplicate = edited_query
          break
        end
      end
    end
    if duplicate && @non_duplicates.include?([duplicate, query].sort)
      duplicate = nil
    end
    added_queries << query unless duplicate

    # find concepts
    concepts = []
    @concepts.each do |name, values|
      concepts << name if values.include?(query)
    end

    # exclude misspellings that are not brands
    misspelling = @words.any? && misspellings?(query)

    profane = blacklisted?(query, @profane_words)

    blacklisted = blacklisted?(query, @blacklisted_words)

    notes = []
    notes << "duplicate of #{duplicate}" if duplicate
    notes.concat(concepts)
    notes << "misspelling" if misspelling
    notes << "profane" if profane
    notes << "blacklisted" if blacklisted
    notes << "originally #{original_query}" if original_query

    {
      query: query,
      original_query: original_query,
      score: count,
      duplicate: duplicate,
      concepts: concepts,
      misspelling: misspelling,
      profane: profane,
      blacklisted: blacklisted,
      notes: notes
    }
  end
end