Class: Autosuggest

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

Defined Under Namespace

Modules: Generators

Constant Summary collapse

VERSION =
"0.1.3"

Instance Method Summary collapse

Constructor Details

#initialize(top_queries) ⇒ Autosuggest

Returns a new instance of Autosuggest.



13
14
15
16
17
18
19
20
21
22
23
24
# File 'lib/autosuggest.rb', line 13

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

Instance Method Details

#add_concept(name, values) ⇒ Object



26
27
28
29
30
# File 'lib/autosuggest.rb', line 26

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

#blacklist_words(words) ⇒ Object



61
62
63
64
65
# File 'lib/autosuggest.rb', line 61

def blacklist_words(words)
  warn "[autosuggest] blacklist_words is deprecated. Use block_words instead."
  add_nodes(@blacklisted_words, words)
  words
end

#block_words(words) ⇒ Object



56
57
58
59
# File 'lib/autosuggest.rb', line 56

def block_words(words)
  add_nodes(@blocked_words, words)
  words
end

#not_duplicates(pairs) ⇒ Object



50
51
52
53
54
# File 'lib/autosuggest.rb', line 50

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

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



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

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



67
68
69
70
71
# File 'lib/autosuggest.rb', line 67

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

#pretty_suggestionsObject



154
155
156
157
158
159
160
# File 'lib/autosuggest.rb', line 154

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

#suggestions(filter: false) ⇒ Object

TODO add queries method for filter: false and make suggestions use filter: true in 0.2.0



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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
# File 'lib/autosuggest.rb', line 74

def suggestions(filter: false)
  stemmed_queries = {}
  added_queries = Set.new
  results = @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

    tokens = tokenize(query)

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

    profane = blocked?(tokens, @profane_words)
    blocked = blocked?(tokens, @blocked_words)
    blacklisted = blocked?(tokens, @blacklisted_words)

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

    result = {
      query: query,
      original_query: original_query,
      score: count,
      duplicate: duplicate,
      concepts: concepts,
      misspelling: misspelling,
      profane: profane,
      blocked: blocked
    }
    result[:blacklisted] = blacklisted if @blacklisted_words.any?
    result[:notes] = notes
    result
  end
  if filter
    results.reject! { |s| s[:duplicate] || s[:misspelling] || s[:profane] || s[:blocked] }
  end
  results
end