polysearch 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/Gemfile +8 -0
- data/Gemfile.lock +180 -0
- data/LICENSE.txt +21 -0
- data/README.md +67 -0
- data/Rakefile +4 -0
- data/app/models/concerns/searchable.rb +137 -0
- data/app/models/record.rb +99 -0
- data/bin/console +15 -0
- data/bin/setup +8 -0
- data/bin/standardize +4 -0
- data/lib/generators/polysearch/migration_generator.rb +29 -0
- data/lib/generators/polysearch/templates/migration.rb +37 -0
- data/lib/polysearch.rb +10 -0
- data/lib/polysearch/version.rb +5 -0
- data/polysearch.gemspec +28 -0
- metadata +130 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA256:
|
3
|
+
metadata.gz: 21bf6d369ed9516908e5ac52bb3b5688335e05952e920741cccd40d73b2279d8
|
4
|
+
data.tar.gz: 7257fa7063ad0b250c7d3b602a7acc5bc3b42984b6fefa47202629eedbb03018
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 8995e6d99585aadc2add2e47b9d18829313dd29853468c706e0297a4361b633acbe593db5fcd34e0a2890cc1832f4758d95f74d3a10ad990fde0dc41e91f348b
|
7
|
+
data.tar.gz: c6e75132f957e688b23ef5499fc8af50189edeb470ed664ad3dab4ca5096a3a343e7e33902d98fb0db2113b7993610b8126ee20653bac026787c7c3397c9d380
|
data/Gemfile
ADDED
data/Gemfile.lock
ADDED
@@ -0,0 +1,180 @@
|
|
1
|
+
PATH
|
2
|
+
remote: .
|
3
|
+
specs:
|
4
|
+
polysearch (0.1.0)
|
5
|
+
rails (>= 6.0)
|
6
|
+
|
7
|
+
GEM
|
8
|
+
remote: https://rubygems.org/
|
9
|
+
specs:
|
10
|
+
actioncable (6.1.3.2)
|
11
|
+
actionpack (= 6.1.3.2)
|
12
|
+
activesupport (= 6.1.3.2)
|
13
|
+
nio4r (~> 2.0)
|
14
|
+
websocket-driver (>= 0.6.1)
|
15
|
+
actionmailbox (6.1.3.2)
|
16
|
+
actionpack (= 6.1.3.2)
|
17
|
+
activejob (= 6.1.3.2)
|
18
|
+
activerecord (= 6.1.3.2)
|
19
|
+
activestorage (= 6.1.3.2)
|
20
|
+
activesupport (= 6.1.3.2)
|
21
|
+
mail (>= 2.7.1)
|
22
|
+
actionmailer (6.1.3.2)
|
23
|
+
actionpack (= 6.1.3.2)
|
24
|
+
actionview (= 6.1.3.2)
|
25
|
+
activejob (= 6.1.3.2)
|
26
|
+
activesupport (= 6.1.3.2)
|
27
|
+
mail (~> 2.5, >= 2.5.4)
|
28
|
+
rails-dom-testing (~> 2.0)
|
29
|
+
actionpack (6.1.3.2)
|
30
|
+
actionview (= 6.1.3.2)
|
31
|
+
activesupport (= 6.1.3.2)
|
32
|
+
rack (~> 2.0, >= 2.0.9)
|
33
|
+
rack-test (>= 0.6.3)
|
34
|
+
rails-dom-testing (~> 2.0)
|
35
|
+
rails-html-sanitizer (~> 1.0, >= 1.2.0)
|
36
|
+
actiontext (6.1.3.2)
|
37
|
+
actionpack (= 6.1.3.2)
|
38
|
+
activerecord (= 6.1.3.2)
|
39
|
+
activestorage (= 6.1.3.2)
|
40
|
+
activesupport (= 6.1.3.2)
|
41
|
+
nokogiri (>= 1.8.5)
|
42
|
+
actionview (6.1.3.2)
|
43
|
+
activesupport (= 6.1.3.2)
|
44
|
+
builder (~> 3.1)
|
45
|
+
erubi (~> 1.4)
|
46
|
+
rails-dom-testing (~> 2.0)
|
47
|
+
rails-html-sanitizer (~> 1.1, >= 1.2.0)
|
48
|
+
activejob (6.1.3.2)
|
49
|
+
activesupport (= 6.1.3.2)
|
50
|
+
globalid (>= 0.3.6)
|
51
|
+
activemodel (6.1.3.2)
|
52
|
+
activesupport (= 6.1.3.2)
|
53
|
+
activerecord (6.1.3.2)
|
54
|
+
activemodel (= 6.1.3.2)
|
55
|
+
activesupport (= 6.1.3.2)
|
56
|
+
activestorage (6.1.3.2)
|
57
|
+
actionpack (= 6.1.3.2)
|
58
|
+
activejob (= 6.1.3.2)
|
59
|
+
activerecord (= 6.1.3.2)
|
60
|
+
activesupport (= 6.1.3.2)
|
61
|
+
marcel (~> 1.0.0)
|
62
|
+
mini_mime (~> 1.0.2)
|
63
|
+
activesupport (6.1.3.2)
|
64
|
+
concurrent-ruby (~> 1.0, >= 1.0.2)
|
65
|
+
i18n (>= 1.6, < 2)
|
66
|
+
minitest (>= 5.1)
|
67
|
+
tzinfo (~> 2.0)
|
68
|
+
zeitwerk (~> 2.3)
|
69
|
+
ast (2.4.2)
|
70
|
+
builder (3.2.4)
|
71
|
+
concurrent-ruby (1.1.8)
|
72
|
+
crass (1.0.6)
|
73
|
+
erubi (1.10.0)
|
74
|
+
globalid (0.4.2)
|
75
|
+
activesupport (>= 4.2.0)
|
76
|
+
i18n (1.8.10)
|
77
|
+
concurrent-ruby (~> 1.0)
|
78
|
+
loofah (2.9.1)
|
79
|
+
crass (~> 1.0.2)
|
80
|
+
nokogiri (>= 1.5.9)
|
81
|
+
magic_frozen_string_literal (1.2.0)
|
82
|
+
mail (2.7.1)
|
83
|
+
mini_mime (>= 0.1.1)
|
84
|
+
marcel (1.0.1)
|
85
|
+
method_source (1.0.0)
|
86
|
+
mini_mime (1.0.3)
|
87
|
+
mini_portile2 (2.5.1)
|
88
|
+
minitest (5.14.4)
|
89
|
+
nio4r (2.5.7)
|
90
|
+
nokogiri (1.11.4)
|
91
|
+
mini_portile2 (~> 2.5.0)
|
92
|
+
racc (~> 1.4)
|
93
|
+
nokogiri (1.11.4-arm64-darwin)
|
94
|
+
racc (~> 1.4)
|
95
|
+
parallel (1.20.1)
|
96
|
+
parser (3.0.1.1)
|
97
|
+
ast (~> 2.4.1)
|
98
|
+
racc (1.5.2)
|
99
|
+
rack (2.2.3)
|
100
|
+
rack-test (1.1.0)
|
101
|
+
rack (>= 1.0, < 3)
|
102
|
+
rails (6.1.3.2)
|
103
|
+
actioncable (= 6.1.3.2)
|
104
|
+
actionmailbox (= 6.1.3.2)
|
105
|
+
actionmailer (= 6.1.3.2)
|
106
|
+
actionpack (= 6.1.3.2)
|
107
|
+
actiontext (= 6.1.3.2)
|
108
|
+
actionview (= 6.1.3.2)
|
109
|
+
activejob (= 6.1.3.2)
|
110
|
+
activemodel (= 6.1.3.2)
|
111
|
+
activerecord (= 6.1.3.2)
|
112
|
+
activestorage (= 6.1.3.2)
|
113
|
+
activesupport (= 6.1.3.2)
|
114
|
+
bundler (>= 1.15.0)
|
115
|
+
railties (= 6.1.3.2)
|
116
|
+
sprockets-rails (>= 2.0.0)
|
117
|
+
rails-dom-testing (2.0.3)
|
118
|
+
activesupport (>= 4.2.0)
|
119
|
+
nokogiri (>= 1.6)
|
120
|
+
rails-html-sanitizer (1.3.0)
|
121
|
+
loofah (~> 2.3)
|
122
|
+
railties (6.1.3.2)
|
123
|
+
actionpack (= 6.1.3.2)
|
124
|
+
activesupport (= 6.1.3.2)
|
125
|
+
method_source
|
126
|
+
rake (>= 0.8.7)
|
127
|
+
thor (~> 1.0)
|
128
|
+
rainbow (3.0.0)
|
129
|
+
rake (13.0.3)
|
130
|
+
regexp_parser (2.1.1)
|
131
|
+
rexml (3.2.5)
|
132
|
+
rubocop (1.14.0)
|
133
|
+
parallel (~> 1.10)
|
134
|
+
parser (>= 3.0.0.0)
|
135
|
+
rainbow (>= 2.2.2, < 4.0)
|
136
|
+
regexp_parser (>= 1.8, < 3.0)
|
137
|
+
rexml
|
138
|
+
rubocop-ast (>= 1.5.0, < 2.0)
|
139
|
+
ruby-progressbar (~> 1.7)
|
140
|
+
unicode-display_width (>= 1.4.0, < 3.0)
|
141
|
+
rubocop-ast (1.5.0)
|
142
|
+
parser (>= 3.0.1.1)
|
143
|
+
rubocop-performance (1.11.2)
|
144
|
+
rubocop (>= 1.7.0, < 2.0)
|
145
|
+
rubocop-ast (>= 0.4.0)
|
146
|
+
ruby-progressbar (1.11.0)
|
147
|
+
sprockets (4.0.2)
|
148
|
+
concurrent-ruby (~> 1.0)
|
149
|
+
rack (> 1, < 3)
|
150
|
+
sprockets-rails (3.2.2)
|
151
|
+
actionpack (>= 4.0)
|
152
|
+
activesupport (>= 4.0)
|
153
|
+
sprockets (>= 3.0.0)
|
154
|
+
standard (1.1.1)
|
155
|
+
rubocop (= 1.14.0)
|
156
|
+
rubocop-performance (= 1.11.2)
|
157
|
+
standardrb (1.0.0)
|
158
|
+
standard
|
159
|
+
thor (1.1.0)
|
160
|
+
tzinfo (2.0.4)
|
161
|
+
concurrent-ruby (~> 1.0)
|
162
|
+
unicode-display_width (2.0.0)
|
163
|
+
websocket-driver (0.7.3)
|
164
|
+
websocket-extensions (>= 0.1.0)
|
165
|
+
websocket-extensions (0.1.5)
|
166
|
+
zeitwerk (2.4.2)
|
167
|
+
|
168
|
+
PLATFORMS
|
169
|
+
aarch64-linux-musl
|
170
|
+
arm64-darwin-20
|
171
|
+
|
172
|
+
DEPENDENCIES
|
173
|
+
bundler (~> 2.0)
|
174
|
+
magic_frozen_string_literal
|
175
|
+
polysearch!
|
176
|
+
rake (~> 13.0)
|
177
|
+
standardrb
|
178
|
+
|
179
|
+
BUNDLED WITH
|
180
|
+
2.2.15
|
data/LICENSE.txt
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
The MIT License (MIT)
|
2
|
+
|
3
|
+
Copyright (c) 2021 Hopsoft
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
7
|
+
in the Software without restriction, including without limitation the rights
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
10
|
+
furnished to do so, subject to the following conditions:
|
11
|
+
|
12
|
+
The above copyright notice and this permission notice shall be included in
|
13
|
+
all copies or substantial portions of the Software.
|
14
|
+
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
21
|
+
THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,67 @@
|
|
1
|
+
# Polysearch
|
2
|
+
|
3
|
+
#### Simplified polymorphic full text + similarity search based on postgres
|
4
|
+
|
5
|
+
## Requirements
|
6
|
+
|
7
|
+
- Postgresql >= 11
|
8
|
+
- Rails >= 6.0
|
9
|
+
|
10
|
+
## Usage
|
11
|
+
|
12
|
+
1. Add the gem to your project
|
13
|
+
|
14
|
+
```sh
|
15
|
+
bundle add polysearch
|
16
|
+
```
|
17
|
+
|
18
|
+
1. Run the generator
|
19
|
+
|
20
|
+
```sh
|
21
|
+
bundle exec rails g polysearch:migration
|
22
|
+
```
|
23
|
+
|
24
|
+
You can also specify a datatype that your app uses for primary keys (default is `bigint`).
|
25
|
+
For example, if your application uses `uuid` primary keys, you install the migration like this.
|
26
|
+
|
27
|
+
```sh
|
28
|
+
bundle exec rails g polysearch:migration uuid
|
29
|
+
```
|
30
|
+
|
31
|
+
1. Migrate the database
|
32
|
+
|
33
|
+
```sh
|
34
|
+
bundle exec rails db:migrate
|
35
|
+
```
|
36
|
+
|
37
|
+
1. Update the model(s) you'd like to search
|
38
|
+
|
39
|
+
```ruby
|
40
|
+
class User < ApplicationRecord
|
41
|
+
include Polysearch::Searchable
|
42
|
+
|
43
|
+
after_save_commit :update_polysearch
|
44
|
+
|
45
|
+
def to_tsvectors
|
46
|
+
[]
|
47
|
+
.then { |list| list << make_tsvector(first_name, weight: "A") }
|
48
|
+
.then { |list| list << make_tsvector(last_name, weight: "A") }
|
49
|
+
.then { |list| list << make_tsvector(email, weight: "B") }
|
50
|
+
end
|
51
|
+
end
|
52
|
+
```
|
53
|
+
|
54
|
+
1. Start searching
|
55
|
+
|
56
|
+
```
|
57
|
+
User.create first_name: "Nate", last_name: "Hopkins", email: "nhopkins@mailinator.com"
|
58
|
+
|
59
|
+
User.polysearch("nate")
|
60
|
+
User.polysearch("ntae") # misspellings also return results
|
61
|
+
User.polysearch("nate").where(created_at: 1.day.ago..Current.time) # active record chaining
|
62
|
+
User.polysearch("nate").order(created_at: :desc) # chain additional ordering after the polysearch scope
|
63
|
+
```
|
64
|
+
|
65
|
+
## License
|
66
|
+
|
67
|
+
The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
|
data/Rakefile
ADDED
@@ -0,0 +1,137 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module Polysearch
|
4
|
+
module Searchable
|
5
|
+
extend ActiveSupport::Concern
|
6
|
+
|
7
|
+
module ClassMethods
|
8
|
+
def ngrams(value, min: 1, max: 24)
|
9
|
+
value = value.to_s
|
10
|
+
Set.new.tap do |set|
|
11
|
+
(min..max).each do |num|
|
12
|
+
value.scan(/\w{#{num}}/).each { |item| set << item }
|
13
|
+
end
|
14
|
+
end.to_a
|
15
|
+
end
|
16
|
+
|
17
|
+
def fts_words(value)
|
18
|
+
Loofah.fragment(value.to_s).scrub!(:whitewash).to_text.gsub(/\W/, " ").squeeze(" ").downcase.split
|
19
|
+
end
|
20
|
+
|
21
|
+
def fts_string(value)
|
22
|
+
value = fts_words(value).join(" ")
|
23
|
+
value = value.chop while value.bytes.size > 2046
|
24
|
+
value
|
25
|
+
end
|
26
|
+
|
27
|
+
def sanitize_sql_value(value)
|
28
|
+
sanitize_sql_array ["?", value]
|
29
|
+
end
|
30
|
+
end
|
31
|
+
|
32
|
+
delegate :ngrams, :fts_words, :fts_string, :sanitize_sql_value, to: "self.class"
|
33
|
+
delegate :quote_column_name, to: "self.class.connection"
|
34
|
+
|
35
|
+
included do
|
36
|
+
has_one :polysearch, as: :searchable, class_name: "Polysearch::Record", inverse_of: "searchable"
|
37
|
+
after_destroy :destroy_polysearch
|
38
|
+
|
39
|
+
scope :polysearch, ->(value) {
|
40
|
+
if value.blank?
|
41
|
+
all
|
42
|
+
else
|
43
|
+
fts_rank_alias = "#{table_name.singularize}_fts_rank"
|
44
|
+
similarity_rank_alias = "#{table_name.singularize}_similarity_rank"
|
45
|
+
|
46
|
+
fts = Polysearch::Record
|
47
|
+
.select_fts_rank(value, :searchable_id, rank_alias: fts_rank_alias)
|
48
|
+
.select_similarity_rank(value, :searchable_id, rank_alias: similarity_rank_alias)
|
49
|
+
.where(searchable_type: name)
|
50
|
+
.fts(value).or(Polysearch::Record.similar(value))
|
51
|
+
|
52
|
+
query = <<~SQL
|
53
|
+
SELECT searchables.*, fts.#{fts_rank_alias}, fts.#{similarity_rank_alias} from (#{fts.to_sql}) fts
|
54
|
+
LEFT JOIN LATERAL (select * from #{table_name} WHERE id = fts.searchable_id) searchables ON TRUE
|
55
|
+
SQL
|
56
|
+
|
57
|
+
select(Arel.star).from(Arel::Nodes::SqlLiteral.new("(#{query})").as(table_name))
|
58
|
+
.reorder(fts_rank_alias => :desc, similarity_rank_alias => :desc)
|
59
|
+
end
|
60
|
+
}
|
61
|
+
end
|
62
|
+
|
63
|
+
def update_polysearch
|
64
|
+
tsvectors = to_tsvectors.compact.uniq
|
65
|
+
return if tsvectors.blank?
|
66
|
+
tsvectors.pop while tsvectors.size > 500
|
67
|
+
tsvectors.concat similarity_words_tsvectors
|
68
|
+
tsvector = tsvectors.join(" || ")
|
69
|
+
fts = Polysearch::Record.where(searchable: self).first_or_create
|
70
|
+
fts.update_value tsvector
|
71
|
+
fts.update_columns words: similarity_words.join(" ")
|
72
|
+
end
|
73
|
+
|
74
|
+
# Polysearch::Searchable#to_tsvectors is abstract... a noop by default
|
75
|
+
# it must be implemented in including ActiveRecord models if this behavior is desired
|
76
|
+
#
|
77
|
+
# Example:
|
78
|
+
#
|
79
|
+
# def to_tsvectors
|
80
|
+
# []
|
81
|
+
# .then { |result| result << make_tsvector(EXAMPLE_COLUMN_OR_PROPERTY, weight: "A") }
|
82
|
+
# .then { |result| EXAMPLE_TAGS_COLUMN.each_with_object(result) { |tag, memo| memo << make_tsvector(tag, weight: "B") } }
|
83
|
+
# end
|
84
|
+
#
|
85
|
+
def to_tsvectors
|
86
|
+
[]
|
87
|
+
end
|
88
|
+
|
89
|
+
def similarity_words
|
90
|
+
tsvectors = to_tsvectors.compact.uniq
|
91
|
+
return [] if tsvectors.blank?
|
92
|
+
tsvector = tsvectors.join(" || ")
|
93
|
+
|
94
|
+
ts_stat = Arel::Nodes::NamedFunction.new("ts_stat", [
|
95
|
+
Arel::Nodes::SqlLiteral.new(sanitize_sql_value("SELECT #{tsvector}"))
|
96
|
+
])
|
97
|
+
length = Arel::Nodes::NamedFunction.new("length", [Arel::Nodes::SqlLiteral.new(quote_column_name(:word))])
|
98
|
+
query = self.class.select(:word).from(ts_stat.to_sql).where(length.gteq(3)).to_sql
|
99
|
+
result = self.class.connection.execute(query)
|
100
|
+
result.values.flatten
|
101
|
+
end
|
102
|
+
|
103
|
+
def similarity_ngrams
|
104
|
+
similarity_words.each_with_object(Set.new) do |word, memo|
|
105
|
+
ngrams(word).each { |ngram| memo << ngram }
|
106
|
+
end.to_a.sort_by(&:length)
|
107
|
+
end
|
108
|
+
|
109
|
+
def similarity_words_tsvectors(weight: "D")
|
110
|
+
similarity_ngrams.each_with_object([]) do |ngram, memo|
|
111
|
+
memo << make_tsvector(ngram, weight: weight)
|
112
|
+
end
|
113
|
+
end
|
114
|
+
|
115
|
+
protected
|
116
|
+
|
117
|
+
def make_tsvector(value, weight: "D")
|
118
|
+
value = fts_string(value).gsub(/\W/, " ").squeeze.downcase
|
119
|
+
return nil if value.blank?
|
120
|
+
to_tsv = Arel::Nodes::NamedFunction.new("to_tsvector", [
|
121
|
+
Arel::Nodes::SqlLiteral.new("'simple'"),
|
122
|
+
Arel::Nodes::SqlLiteral.new(sanitize_sql_value(value))
|
123
|
+
])
|
124
|
+
setweight = Arel::Nodes::NamedFunction.new("setweight", [
|
125
|
+
to_tsv,
|
126
|
+
Arel::Nodes::SqlLiteral.new(sanitize_sql_value(weight))
|
127
|
+
])
|
128
|
+
setweight.to_sql
|
129
|
+
end
|
130
|
+
|
131
|
+
private
|
132
|
+
|
133
|
+
def destroy_polysearch
|
134
|
+
polysearch&.destroy
|
135
|
+
end
|
136
|
+
end
|
137
|
+
end
|
@@ -0,0 +1,99 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
# == Schema Information
|
4
|
+
#
|
5
|
+
# Table name: polysearches
|
6
|
+
#
|
7
|
+
# id :uuid not null, primary key
|
8
|
+
# searchable_type :string not null
|
9
|
+
# value :tsvector
|
10
|
+
# words :text
|
11
|
+
# created_at :datetime not null
|
12
|
+
# updated_at :datetime not null
|
13
|
+
# searchable_id :uuid not null
|
14
|
+
#
|
15
|
+
# Indexes
|
16
|
+
#
|
17
|
+
# index_polysearches_on_created_at (created_at)
|
18
|
+
# index_polysearches_on_searchable_type_and_searchable_id (searchable_type,searchable_id) UNIQUE
|
19
|
+
# index_polysearches_on_updated_at (updated_at)
|
20
|
+
# index_polysearches_on_value (value) USING gin
|
21
|
+
# index_polysearches_on_words (words) USING gin
|
22
|
+
#
|
23
|
+
module Polysearch
|
24
|
+
class Record < ActiveRecord::Base
|
25
|
+
# extends ...................................................................
|
26
|
+
# includes ..................................................................
|
27
|
+
|
28
|
+
# relationships .............................................................
|
29
|
+
belongs_to :searchable, polymorphic: true, inverse_of: "polysearch"
|
30
|
+
|
31
|
+
# validations ...............................................................
|
32
|
+
# callbacks .................................................................
|
33
|
+
|
34
|
+
# scopes ....................................................................
|
35
|
+
|
36
|
+
scope :select_fts_rank, ->(value, *selects, rank_alias: nil) {
|
37
|
+
value = value.to_s.gsub(/\W/, " ").squeeze(" ").downcase.strip
|
38
|
+
value = Arel::Nodes::SqlLiteral.new(sanitize_sql_array(["?", value.to_s]))
|
39
|
+
plainto_tsquery = Arel::Nodes::NamedFunction.new("plainto_tsquery", [Arel::Nodes::SqlLiteral.new("'simple'"), value])
|
40
|
+
ts_rank = Arel::Nodes::NamedFunction.new("ts_rank", [arel_table[:value], plainto_tsquery])
|
41
|
+
|
42
|
+
rank_alias ||= "fts_rank"
|
43
|
+
selects << Arel.star if selects.blank?
|
44
|
+
selects << ts_rank.as(rank_alias)
|
45
|
+
select(*selects).order("#{rank_alias} desc")
|
46
|
+
}
|
47
|
+
|
48
|
+
scope :fts, ->(value) {
|
49
|
+
value = value.to_s.gsub(/\W/, " ").squeeze(" ").downcase.strip
|
50
|
+
value = Arel::Nodes::SqlLiteral.new(sanitize_sql_array(["?", value.to_s]))
|
51
|
+
plainto_tsquery = Arel::Nodes::NamedFunction.new("plainto_tsquery", [Arel::Nodes::SqlLiteral.new("'simple'"), value])
|
52
|
+
where(Arel::Nodes::InfixOperation.new("@@", arel_table[:value], plainto_tsquery))
|
53
|
+
}
|
54
|
+
|
55
|
+
scope :select_similarity_rank, ->(value, *selects, rank_alias: nil) {
|
56
|
+
value = value.to_s.gsub(/\W/, " ").squeeze(" ").downcase.strip
|
57
|
+
value = Arel::Nodes::SqlLiteral.new(sanitize_sql_array(["?", value.to_s]))
|
58
|
+
|
59
|
+
rank_alias ||= "similarity_rank"
|
60
|
+
selects << Arel.star if selects.blank?
|
61
|
+
selects << Arel::Nodes::NamedFunction.new("similarity", [arel_table[:words], value]).as(rank_alias)
|
62
|
+
select(*selects).order("#{rank_alias} desc")
|
63
|
+
}
|
64
|
+
|
65
|
+
scope :similar, ->(value, range: 0.01) {
|
66
|
+
value = value.to_s.gsub(/\W/, " ").squeeze(" ").downcase.strip
|
67
|
+
value = Arel::Nodes::SqlLiteral.new(sanitize_sql_array(["?", value.to_s]))
|
68
|
+
where Arel::Nodes::NamedFunction.new("similarity", [arel_table[:words], value]).gteq(range)
|
69
|
+
}
|
70
|
+
|
71
|
+
# additional config (i.e. accepts_nested_attribute_for etc...) ..............
|
72
|
+
self.table_name = "polysearches"
|
73
|
+
self.primary_key = :id
|
74
|
+
|
75
|
+
# class methods .............................................................
|
76
|
+
class << self
|
77
|
+
end
|
78
|
+
|
79
|
+
# public instance methods ...................................................
|
80
|
+
|
81
|
+
def update_value(tsvector_sql)
|
82
|
+
sql = <<~SQL
|
83
|
+
UPDATE polysearches
|
84
|
+
SET value = (#{tsvector_sql})
|
85
|
+
WHERE searchable_type = :searchable_type
|
86
|
+
AND searchable_id = :searchable_id;
|
87
|
+
SQL
|
88
|
+
self.class.connection.execute self.class.sanitize_sql_array([
|
89
|
+
sql,
|
90
|
+
searchable_type: searchable_type,
|
91
|
+
searchable_id: searchable_id
|
92
|
+
])
|
93
|
+
end
|
94
|
+
|
95
|
+
# protected instance methods ................................................
|
96
|
+
|
97
|
+
# private instance methods ..................................................
|
98
|
+
end
|
99
|
+
end
|
data/bin/console
ADDED
@@ -0,0 +1,15 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
# frozen_string_literal: true
|
3
|
+
|
4
|
+
require "bundler/setup"
|
5
|
+
require "polysearch"
|
6
|
+
|
7
|
+
# You can add fixtures and/or initialization code here to make experimenting
|
8
|
+
# with your gem easier. You can also use a different console, if you like.
|
9
|
+
|
10
|
+
# (If you use this, don't forget to add pry to your Gemfile!)
|
11
|
+
# require "pry"
|
12
|
+
# Pry.start
|
13
|
+
|
14
|
+
require "irb"
|
15
|
+
IRB.start(__FILE__)
|
data/bin/setup
ADDED
data/bin/standardize
ADDED
@@ -0,0 +1,29 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
# require "rails/generators/base"
|
4
|
+
|
5
|
+
module Polysearch
|
6
|
+
module Generators
|
7
|
+
class MigrationGenerator < Rails::Generators::Base
|
8
|
+
desc "Copy polysearch database migrations into your project"
|
9
|
+
argument :searchable_id_datatype, type: :string, default: "bigint"
|
10
|
+
source_root File.expand_path("../templates", __FILE__)
|
11
|
+
|
12
|
+
def copy_polysearch_migration
|
13
|
+
template "migration.rb", "db/migrate/#{timestamp}_add_polysearch.rb",
|
14
|
+
searchable_id_datatype: searchable_id_datatype,
|
15
|
+
migration_version: migration_version
|
16
|
+
end
|
17
|
+
|
18
|
+
private
|
19
|
+
|
20
|
+
def timestamp
|
21
|
+
DateTime.current.strftime "%Y%m%d%H%M%S"
|
22
|
+
end
|
23
|
+
|
24
|
+
def migration_version
|
25
|
+
"[#{Rails::VERSION::MAJOR}.#{Rails::VERSION::MINOR}]"
|
26
|
+
end
|
27
|
+
end
|
28
|
+
end
|
29
|
+
end
|
@@ -0,0 +1,37 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class AddPolysearch < ActiveRecord::Migration<%= migration_version %>
|
4
|
+
def change
|
5
|
+
enable_extension :pg_trgm
|
6
|
+
|
7
|
+
create_table :polysearches, primary_key: [:id, :searchable_type], options: "PARTITION BY HASH (searchable_type)" do |t|
|
8
|
+
t.uuid :id, null: false, default: "gen_random_uuid()"
|
9
|
+
t.string :searchable_type, null: false
|
10
|
+
t.<%= searchable_id_datatype %> :searchable_id, null: false
|
11
|
+
t.text :words
|
12
|
+
t.tsvector :value
|
13
|
+
t.timestamps
|
14
|
+
|
15
|
+
t.index [:searchable_type, :searchable_id], unique: true
|
16
|
+
t.index :words, using: :gin, opclass: :gin_trgm_ops
|
17
|
+
t.index :value, using: :gin
|
18
|
+
t.index :created_at
|
19
|
+
t.index :updated_at
|
20
|
+
end
|
21
|
+
|
22
|
+
reversible do |dir|
|
23
|
+
dir.up do
|
24
|
+
execute "CREATE TABLE polysearches_01 PARTITION OF polysearches FOR VALUES WITH (MODULUS 4, REMAINDER 0);"
|
25
|
+
execute "CREATE TABLE polysearches_02 PARTITION OF polysearches FOR VALUES WITH (MODULUS 4, REMAINDER 1);"
|
26
|
+
execute "CREATE TABLE polysearches_03 PARTITION OF polysearches FOR VALUES WITH (MODULUS 4, REMAINDER 2);"
|
27
|
+
execute "CREATE TABLE polysearches_04 PARTITION OF polysearches FOR VALUES WITH (MODULUS 4, REMAINDER 3);"
|
28
|
+
end
|
29
|
+
dir.down do
|
30
|
+
drop_table :polysearches_01
|
31
|
+
drop_table :polysearches_02
|
32
|
+
drop_table :polysearches_03
|
33
|
+
drop_table :polysearches_04
|
34
|
+
end
|
35
|
+
end
|
36
|
+
end
|
37
|
+
end
|
data/lib/polysearch.rb
ADDED
data/polysearch.gemspec
ADDED
@@ -0,0 +1,28 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require File.expand_path("../lib/polysearch/version", __FILE__)
|
4
|
+
|
5
|
+
Gem::Specification.new do |gem|
|
6
|
+
gem.name = "polysearch"
|
7
|
+
gem.license = "MIT"
|
8
|
+
gem.version = Polysearch::VERSION
|
9
|
+
gem.authors = ["Nathan Hopkins"]
|
10
|
+
gem.email = ["natehop@gmail.com"]
|
11
|
+
gem.homepage = "https://github.com/hopsoft/polysearch"
|
12
|
+
gem.summary = "Simplified polymorphic full text + similarity search based on postgres"
|
13
|
+
|
14
|
+
gem.metadata = {
|
15
|
+
"homepage_uri" => gem.homepage,
|
16
|
+
"source_code_uri" => gem.homepage
|
17
|
+
}
|
18
|
+
|
19
|
+
gem.files = Dir["app/**/*", "lib/**/*", "bin/*", "[A-Z]*"]
|
20
|
+
gem.test_files = Dir["test/**/*.rb"]
|
21
|
+
|
22
|
+
gem.add_dependency "rails", ">= 6.0"
|
23
|
+
|
24
|
+
gem.add_development_dependency "bundler", "~> 2.0"
|
25
|
+
gem.add_development_dependency "rake"
|
26
|
+
gem.add_development_dependency "standardrb"
|
27
|
+
gem.add_development_dependency "magic_frozen_string_literal"
|
28
|
+
end
|
metadata
ADDED
@@ -0,0 +1,130 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: polysearch
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.1.0
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Nathan Hopkins
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2021-05-18 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: rails
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - ">="
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '6.0'
|
20
|
+
type: :runtime
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - ">="
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '6.0'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: bundler
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '2.0'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '2.0'
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: rake
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - ">="
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - ">="
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '0'
|
55
|
+
- !ruby/object:Gem::Dependency
|
56
|
+
name: standardrb
|
57
|
+
requirement: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - ">="
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
62
|
+
type: :development
|
63
|
+
prerelease: false
|
64
|
+
version_requirements: !ruby/object:Gem::Requirement
|
65
|
+
requirements:
|
66
|
+
- - ">="
|
67
|
+
- !ruby/object:Gem::Version
|
68
|
+
version: '0'
|
69
|
+
- !ruby/object:Gem::Dependency
|
70
|
+
name: magic_frozen_string_literal
|
71
|
+
requirement: !ruby/object:Gem::Requirement
|
72
|
+
requirements:
|
73
|
+
- - ">="
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: '0'
|
76
|
+
type: :development
|
77
|
+
prerelease: false
|
78
|
+
version_requirements: !ruby/object:Gem::Requirement
|
79
|
+
requirements:
|
80
|
+
- - ">="
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: '0'
|
83
|
+
description:
|
84
|
+
email:
|
85
|
+
- natehop@gmail.com
|
86
|
+
executables: []
|
87
|
+
extensions: []
|
88
|
+
extra_rdoc_files: []
|
89
|
+
files:
|
90
|
+
- Gemfile
|
91
|
+
- Gemfile.lock
|
92
|
+
- LICENSE.txt
|
93
|
+
- README.md
|
94
|
+
- Rakefile
|
95
|
+
- app/models/concerns/searchable.rb
|
96
|
+
- app/models/record.rb
|
97
|
+
- bin/console
|
98
|
+
- bin/setup
|
99
|
+
- bin/standardize
|
100
|
+
- lib/generators/polysearch/migration_generator.rb
|
101
|
+
- lib/generators/polysearch/templates/migration.rb
|
102
|
+
- lib/polysearch.rb
|
103
|
+
- lib/polysearch/version.rb
|
104
|
+
- polysearch.gemspec
|
105
|
+
homepage: https://github.com/hopsoft/polysearch
|
106
|
+
licenses:
|
107
|
+
- MIT
|
108
|
+
metadata:
|
109
|
+
homepage_uri: https://github.com/hopsoft/polysearch
|
110
|
+
source_code_uri: https://github.com/hopsoft/polysearch
|
111
|
+
post_install_message:
|
112
|
+
rdoc_options: []
|
113
|
+
require_paths:
|
114
|
+
- lib
|
115
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
116
|
+
requirements:
|
117
|
+
- - ">="
|
118
|
+
- !ruby/object:Gem::Version
|
119
|
+
version: '0'
|
120
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
121
|
+
requirements:
|
122
|
+
- - ">="
|
123
|
+
- !ruby/object:Gem::Version
|
124
|
+
version: '0'
|
125
|
+
requirements: []
|
126
|
+
rubygems_version: 3.2.15
|
127
|
+
signing_key:
|
128
|
+
specification_version: 4
|
129
|
+
summary: Simplified polymorphic full text + similarity search based on postgres
|
130
|
+
test_files: []
|