picky 4.27.1 → 4.28.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/picky/backends/google_hash.rb +4 -1
- data/lib/picky/backends/memory/basic.rb +13 -7
- data/lib/picky/extensions/string.rb +11 -5
- data/lib/picky/extensions/symbol.rb +8 -2
- data/lib/picky/index.rb +8 -0
- data/lib/picky/indexes.rb +12 -0
- data/lib/picky/loader.rb +3 -1
- data/lib/picky/optimizers/memory/array_deduplicator.rb +26 -0
- data/lib/picky/optimizers.rb +7 -0
- data/lib/picky/query/tokens.rb +1 -1
- data/lib/picky.rb +2 -4
- data/spec/functional/backends/file_spec.rb +11 -11
- data/spec/functional/no_dump_hint_spec.rb +3 -3
- data/spec/functional/optimize_memory_spec.rb +44 -0
- data/spec/lib/extensions/string_spec.rb +3 -0
- data/spec/lib/indexes_indexing_spec.rb +1 -1
- data/spec/lib/picky_spec.rb +6 -6
- data/spec/lib/query/tokens_spec.rb +15 -1
- metadata +7 -17
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 1070a597a2893ecafe13461959d34eb72b774fb1
|
4
|
+
data.tar.gz: 187eba56c158b4a3301360c5321423a37e3410af
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 399f943b00ce22fffd54c7cb961155fa6386bd898a3d5566f9778c5e883cdb380da9c873eda6644bf1a5d7629e7fe3916e6db64f7049df75f196880e381d0d95
|
7
|
+
data.tar.gz: 7eaa2a8156eb638476fa08edf92617e67ff687096dc631470402f63b7ce91cc0813ee936badd31d4caf7f41bc16e6431e09eeded924ef3b112189224d37df464
|
@@ -15,9 +15,9 @@ module Picky
|
|
15
15
|
|
16
16
|
include Helpers::File
|
17
17
|
|
18
|
-
# This file's
|
18
|
+
# This file's cache file without extensions.
|
19
19
|
#
|
20
|
-
attr_reader :
|
20
|
+
attr_reader :cache_file_path
|
21
21
|
|
22
22
|
# What hash type to use. Default: ::Hash
|
23
23
|
#
|
@@ -26,11 +26,11 @@ module Picky
|
|
26
26
|
# An index cache takes a path, without file extension,
|
27
27
|
# which will be provided by the subclasses.
|
28
28
|
#
|
29
|
-
def initialize
|
30
|
-
@
|
31
|
-
@hash_type
|
32
|
-
@empty
|
33
|
-
@initial
|
29
|
+
def initialize cache_file_path, hash_type = Hash, options = {}
|
30
|
+
@cache_file_path = cache_file_path
|
31
|
+
@hash_type = hash_type
|
32
|
+
@empty = options[:empty]
|
33
|
+
@initial = options[:initial]
|
34
34
|
end
|
35
35
|
|
36
36
|
# The default extension for index files is "index".
|
@@ -38,6 +38,12 @@ module Picky
|
|
38
38
|
def extension
|
39
39
|
:index
|
40
40
|
end
|
41
|
+
def type
|
42
|
+
:memory
|
43
|
+
end
|
44
|
+
def cache_path
|
45
|
+
[cache_file_path, type, extension].join(?.)
|
46
|
+
end
|
41
47
|
|
42
48
|
# The empty index that is used for putting the index
|
43
49
|
# together before it is dumped into the files.
|
@@ -27,17 +27,23 @@ class String
|
|
27
27
|
#
|
28
28
|
def each_subtoken from_length = 1, range = nil
|
29
29
|
sub = self
|
30
|
-
|
31
|
-
|
32
|
-
|
30
|
+
|
31
|
+
if range
|
32
|
+
unless (range.first.zero? && range.last == -1)
|
33
|
+
sub = sub[range].freeze
|
34
|
+
end
|
35
|
+
end
|
36
|
+
|
33
37
|
yield sub
|
34
|
-
|
38
|
+
|
35
39
|
size = sub.size
|
36
40
|
from_length = size + from_length + 1 if from_length < 0
|
37
41
|
from_length = size if size < from_length
|
38
42
|
from_length = 1 if from_length < 1
|
39
43
|
|
40
|
-
size.downto(from_length + 1) { yield sub = sub.chop }
|
44
|
+
size.downto(from_length + 1) { yield sub = sub.chop.freeze }
|
45
|
+
|
46
|
+
sub = nil
|
41
47
|
end
|
42
48
|
|
43
49
|
# 'keys'.each_intoken # => yields each of ['keys', 'key', 'eys', 'ke', 'ey', 'ys', 'k', 'e', 'y', 's']
|
@@ -30,7 +30,11 @@ class Symbol
|
|
30
30
|
def each_subtoken from_length = 1, range = nil
|
31
31
|
sub = self.id2name
|
32
32
|
|
33
|
-
|
33
|
+
if range
|
34
|
+
unless (range.first.zero? && range.last == -1)
|
35
|
+
sub = sub[range]
|
36
|
+
end
|
37
|
+
end
|
34
38
|
|
35
39
|
yield sub.intern
|
36
40
|
|
@@ -39,7 +43,9 @@ class Symbol
|
|
39
43
|
from_length = size if size < from_length
|
40
44
|
from_length = 1 if from_length < 1
|
41
45
|
|
42
|
-
size.downto(from_length + 1)
|
46
|
+
size.downto(from_length + 1) do
|
47
|
+
yield sub.chop!.intern
|
48
|
+
end
|
43
49
|
end
|
44
50
|
|
45
51
|
# :keys.each_intoken # => yields each of [:keys, :key, :eys, :ke, :ey, :ys, :k, :e, :y, :s]
|
data/lib/picky/index.rb
CHANGED
@@ -134,6 +134,14 @@ module Picky
|
|
134
134
|
require_relative 'index/hints'
|
135
135
|
@hints = Hints.new hints
|
136
136
|
end
|
137
|
+
|
138
|
+
# Explicitly trigger memory optimization.
|
139
|
+
#
|
140
|
+
def optimize_memory array_references = Hash.new
|
141
|
+
dedup = Picky::Optimizers::Memory::ArrayDeduplicator.new
|
142
|
+
dedup.deduplicate categories.map(&:exact).map(&:inverted), array_references
|
143
|
+
dedup.deduplicate categories.map(&:partial).map(&:inverted), array_references
|
144
|
+
end
|
137
145
|
|
138
146
|
# TODO Doc.
|
139
147
|
#
|
data/lib/picky/indexes.rb
CHANGED
@@ -43,6 +43,18 @@ module Picky
|
|
43
43
|
@indexes = []
|
44
44
|
@index_mapping = Hash.new
|
45
45
|
end
|
46
|
+
|
47
|
+
# Tries to optimize the memory usage of the indexes.
|
48
|
+
#
|
49
|
+
def optimize_memory array_references = Hash.new
|
50
|
+
dedup = Picky::Optimizers::Memory::ArrayDeduplicator.new
|
51
|
+
@indexes.each do |index|
|
52
|
+
index.optimize_memory array_references
|
53
|
+
end
|
54
|
+
end
|
55
|
+
def self.optimize_memory array_references = Hash.new
|
56
|
+
self.instance.optimize_memory array_references
|
57
|
+
end
|
46
58
|
|
47
59
|
# Registers an index with the indexes.
|
48
60
|
#
|
data/lib/picky/loader.rb
CHANGED
@@ -69,7 +69,9 @@ module Picky
|
|
69
69
|
load_relative 'helpers/measuring',
|
70
70
|
'helpers/indexing',
|
71
71
|
'helpers/identification',
|
72
|
-
'splitter'
|
72
|
+
'splitter',
|
73
|
+
'optimizers',
|
74
|
+
'optimizers/memory/array_deduplicator'
|
73
75
|
end
|
74
76
|
def load_index_generation_strategies
|
75
77
|
load_relative 'indexers/base',
|
@@ -0,0 +1,26 @@
|
|
1
|
+
module Picky::Optimizers::Memory
|
2
|
+
|
3
|
+
# Straightforward implementation of an array deduplicator.
|
4
|
+
# Tries to find duplicate instances of Array values in a hash
|
5
|
+
# and points references that point to a duplicate to one of the
|
6
|
+
# Array instances.
|
7
|
+
#
|
8
|
+
# TODO Could we have C-Ruby point to parts of another Array?
|
9
|
+
#
|
10
|
+
class ArrayDeduplicator
|
11
|
+
|
12
|
+
def deduplicate hashes, array_references = Hash.new
|
13
|
+
hashes.inject(array_references) do |array_references, hash|
|
14
|
+
deduplicate_hash hash, array_references
|
15
|
+
array_references
|
16
|
+
end
|
17
|
+
end
|
18
|
+
|
19
|
+
def deduplicate_hash hash, array_references
|
20
|
+
hash.each do |k, ary|
|
21
|
+
hash[k] = (array_references[ary] ||= ary)
|
22
|
+
end
|
23
|
+
end
|
24
|
+
|
25
|
+
end
|
26
|
+
end
|
data/lib/picky/query/tokens.rb
CHANGED
data/lib/picky.rb
CHANGED
@@ -11,10 +11,8 @@ module Picky
|
|
11
11
|
# Set default encoding.
|
12
12
|
# (Note: Rails does this already as well)
|
13
13
|
#
|
14
|
-
|
15
|
-
|
16
|
-
# Encoding.default_external = Encoding::UTF_8
|
17
|
-
# Encoding.default_internal = Encoding::UTF_8
|
14
|
+
Encoding.default_external = Encoding::UTF_8
|
15
|
+
Encoding.default_internal = Encoding::UTF_8
|
18
16
|
|
19
17
|
# External libraries.
|
20
18
|
#
|
@@ -7,7 +7,7 @@ require 'spec_helper'
|
|
7
7
|
#
|
8
8
|
describe Picky::Backends::File do
|
9
9
|
|
10
|
-
class
|
10
|
+
class PoetryBook
|
11
11
|
attr_reader :id, :title, :author
|
12
12
|
def initialize id, title, author
|
13
13
|
@id, @title, @author = id, title, author
|
@@ -44,7 +44,7 @@ describe Picky::Backends::File do
|
|
44
44
|
books.search('title').ids.should == []
|
45
45
|
end
|
46
46
|
it 'handles removing with more than one entry' do
|
47
|
-
data.add
|
47
|
+
data.add PoetryBook.new(2, 'title', 'author')
|
48
48
|
|
49
49
|
books.search('title').ids.should == ['2', '1']
|
50
50
|
|
@@ -53,8 +53,8 @@ describe Picky::Backends::File do
|
|
53
53
|
books.search('title').ids.should == ['2']
|
54
54
|
end
|
55
55
|
it 'handles removing with three entries' do
|
56
|
-
data.add
|
57
|
-
data.add
|
56
|
+
data.add PoetryBook.new(2, 'title', 'author')
|
57
|
+
data.add PoetryBook.new(3, 'title', 'author')
|
58
58
|
|
59
59
|
books.search('title').ids.should == ['3', '2', '1']
|
60
60
|
|
@@ -63,7 +63,7 @@ describe Picky::Backends::File do
|
|
63
63
|
books.search('title').ids.should == ['3', '2']
|
64
64
|
end
|
65
65
|
it 'handles replacing' do
|
66
|
-
data.replace
|
66
|
+
data.replace PoetryBook.new(1, 'toitle', 'oithor')
|
67
67
|
|
68
68
|
books.search('title').ids.should == []
|
69
69
|
books.search('toitle').ids.should == ['1']
|
@@ -100,7 +100,7 @@ describe Picky::Backends::File do
|
|
100
100
|
books.search('title').ids.should == []
|
101
101
|
end
|
102
102
|
it 'handles removing with more than one entry' do
|
103
|
-
data.add
|
103
|
+
data.add PoetryBook.new(2, 'title', 'author')
|
104
104
|
|
105
105
|
books.search('title').ids.should == [2, 1]
|
106
106
|
|
@@ -109,8 +109,8 @@ describe Picky::Backends::File do
|
|
109
109
|
books.search('title').ids.should == [2]
|
110
110
|
end
|
111
111
|
it 'handles removing with three entries' do
|
112
|
-
data.add
|
113
|
-
data.add
|
112
|
+
data.add PoetryBook.new(2, 'title', 'author')
|
113
|
+
data.add PoetryBook.new(3, 'title', 'author')
|
114
114
|
|
115
115
|
books.search('title').ids.should == [3, 2, 1]
|
116
116
|
|
@@ -119,7 +119,7 @@ describe Picky::Backends::File do
|
|
119
119
|
books.search('title').ids.should == [3, 2]
|
120
120
|
end
|
121
121
|
it 'handles replacing' do
|
122
|
-
data.replace
|
122
|
+
data.replace PoetryBook.new(1, 'toitle', 'oithor')
|
123
123
|
|
124
124
|
books.search('title').ids.should == []
|
125
125
|
books.search('toitle').ids.should == [1]
|
@@ -144,7 +144,7 @@ describe Picky::Backends::File do
|
|
144
144
|
data.backend described_class.new
|
145
145
|
data.clear
|
146
146
|
|
147
|
-
data.add
|
147
|
+
data.add PoetryBook.new(1, 'title', 'author')
|
148
148
|
end
|
149
149
|
|
150
150
|
instance_eval &its_to_s
|
@@ -157,7 +157,7 @@ describe Picky::Backends::File do
|
|
157
157
|
data.backend described_class.new
|
158
158
|
data.clear
|
159
159
|
|
160
|
-
data.add
|
160
|
+
data.add PoetryBook.new(1, 'title', 'author')
|
161
161
|
end
|
162
162
|
|
163
163
|
instance_eval &its_to_i
|
@@ -5,7 +5,7 @@ require 'ostruct'
|
|
5
5
|
|
6
6
|
describe "Hint: no_dump" do
|
7
7
|
|
8
|
-
|
8
|
+
ComicBook = Struct.new(:id, :title, :author)
|
9
9
|
|
10
10
|
let(:index) do
|
11
11
|
Picky::Index.new :no_dump do
|
@@ -18,14 +18,14 @@ describe "Hint: no_dump" do
|
|
18
18
|
let(:try) { Picky::Search.new index }
|
19
19
|
|
20
20
|
it 'can index and search' do
|
21
|
-
index.replace
|
21
|
+
index.replace ComicBook.new(2, "Title", "Author")
|
22
22
|
|
23
23
|
try.search("title:title").ids.should == [2]
|
24
24
|
end
|
25
25
|
|
26
26
|
context 'dumping and loading' do
|
27
27
|
it "raises" do
|
28
|
-
index.replace
|
28
|
+
index.replace ComicBook.new(2, "Title New", "Author New")
|
29
29
|
|
30
30
|
expect {
|
31
31
|
index.dump
|
@@ -0,0 +1,44 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
#
|
3
|
+
require 'spec_helper'
|
4
|
+
|
5
|
+
describe "Memory optimization" do
|
6
|
+
|
7
|
+
it 'saves memory' do
|
8
|
+
# Remove all indexes.
|
9
|
+
Picky::Indexes.clear_indexes
|
10
|
+
|
11
|
+
index = Picky::Index.new :memory_optimization do
|
12
|
+
category :text1
|
13
|
+
category :text2
|
14
|
+
category :text3
|
15
|
+
category :text4
|
16
|
+
end
|
17
|
+
try = Picky::Search.new index
|
18
|
+
|
19
|
+
thing = Struct.new(:id, :text1, :text2, :text3, :text4)
|
20
|
+
|
21
|
+
require 'objspace'
|
22
|
+
|
23
|
+
GC.start
|
24
|
+
memsize_without_added_thing = ObjectSpace.memsize_of_all(Array)
|
25
|
+
GC.start
|
26
|
+
|
27
|
+
index.add thing.new(1, 'one', 'two', 'three', 'four')
|
28
|
+
|
29
|
+
GC.start
|
30
|
+
memsize_with_added_thing = ObjectSpace.memsize_of_all(Array)
|
31
|
+
GC.start
|
32
|
+
|
33
|
+
memsize_with_added_thing.should > memsize_without_added_thing
|
34
|
+
|
35
|
+
Picky::Indexes.optimize_memory
|
36
|
+
|
37
|
+
GC.start
|
38
|
+
memsize_with_optimized_memory = ObjectSpace.memsize_of_all(Array)
|
39
|
+
GC.start
|
40
|
+
|
41
|
+
memsize_with_optimized_memory.should < memsize_with_added_thing
|
42
|
+
end
|
43
|
+
|
44
|
+
end
|
@@ -45,7 +45,7 @@ describe Picky::Indexes do
|
|
45
45
|
end
|
46
46
|
end
|
47
47
|
def self.it_forwards_each name
|
48
|
-
describe name do
|
48
|
+
describe "forwarding ##{name}" do
|
49
49
|
it "calls #{name} on each in order" do
|
50
50
|
@index1.should_receive(name).once.with(no_args).ordered
|
51
51
|
@index2.should_receive(name).once.with(no_args).ordered
|
data/spec/lib/picky_spec.rb
CHANGED
@@ -2,17 +2,17 @@ require 'spec_helper'
|
|
2
2
|
|
3
3
|
describe Picky do
|
4
4
|
|
5
|
-
it 'sets the right
|
5
|
+
it 'sets the right external encoding' do
|
6
6
|
Encoding.default_external.should == Encoding::UTF_8
|
7
7
|
end
|
8
|
-
# THINK What to set default
|
8
|
+
# THINK What to set default internal encoding to?
|
9
9
|
#
|
10
|
-
|
11
|
-
|
12
|
-
|
10
|
+
it 'sets the right internal encoding' do
|
11
|
+
Encoding.default_internal.should == Encoding::UTF_8
|
12
|
+
end
|
13
13
|
|
14
14
|
it 'loads in a simple ruby environment with the defined requirements' do
|
15
|
-
#TODO Picky.root is set to /spec/temp in spec_helper, so is this the "best" way?
|
15
|
+
# TODO Picky.root is set to /spec/temp in spec_helper, so is this the "best" way?
|
16
16
|
load_path = File.expand_path('../../../lib', __FILE__)
|
17
17
|
ruby = File.join(RbConfig::CONFIG['bindir'], RbConfig::CONFIG['ruby_install_name']).sub(/.*\s.*/m, '"\&"')
|
18
18
|
|
@@ -224,10 +224,24 @@ describe Picky::Query::Tokens do
|
|
224
224
|
it 'should work correctly' do
|
225
225
|
(@tokens + @other).to_s.should == 'Hello~ I~ Am A* Token~'
|
226
226
|
end
|
227
|
+
it 'should work correctly' do
|
228
|
+
(@tokens + @other).texts.should == ['hello', 'i', 'am', 'a', 'token']
|
229
|
+
end
|
230
|
+
it 'should work correctly' do
|
231
|
+
(@tokens + @other).originals.should == ['Hello~', 'I~', 'Am', 'A*', 'Token~']
|
232
|
+
end
|
233
|
+
it 'correctly handles ignore_unassigned' do
|
234
|
+
(@tokens + @other).ignore_unassigned.should == false
|
235
|
+
end
|
236
|
+
it 'correctly handles ignore_unassigned' do
|
237
|
+
yet_another = described_class.new [], true
|
238
|
+
(@tokens + yet_another).ignore_unassigned.should == true
|
239
|
+
(yet_another + @tokens).ignore_unassigned.should == true
|
240
|
+
end
|
227
241
|
end
|
228
242
|
|
229
243
|
def self.it_should_forward name
|
230
|
-
describe name do
|
244
|
+
describe "forwarding ##{name}" do
|
231
245
|
before(:each) do
|
232
246
|
@internal_tokens = double :internal_tokens
|
233
247
|
@tokens = described_class.new @internal_tokens
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: picky
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 4.
|
4
|
+
version: 4.28.1
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Florian Hanke
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2015-
|
11
|
+
date: 2015-06-28 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: activesupport
|
@@ -52,20 +52,6 @@ dependencies:
|
|
52
52
|
- - "~>"
|
53
53
|
- !ruby/object:Gem::Version
|
54
54
|
version: '2009.0'
|
55
|
-
- !ruby/object:Gem::Dependency
|
56
|
-
name: google_hash
|
57
|
-
requirement: !ruby/object:Gem::Requirement
|
58
|
-
requirements:
|
59
|
-
- - "~>"
|
60
|
-
- !ruby/object:Gem::Version
|
61
|
-
version: '0.8'
|
62
|
-
type: :runtime
|
63
|
-
prerelease: false
|
64
|
-
version_requirements: !ruby/object:Gem::Requirement
|
65
|
-
requirements:
|
66
|
-
- - "~>"
|
67
|
-
- !ruby/object:Gem::Version
|
68
|
-
version: '0.8'
|
69
55
|
description: Fast Ruby semantic text search engine with comfortable single field interface.
|
70
56
|
email: florian.hanke+picky@gmail.com
|
71
57
|
executables:
|
@@ -182,6 +168,8 @@ files:
|
|
182
168
|
- lib/picky/loggers/default.rb
|
183
169
|
- lib/picky/loggers/silent.rb
|
184
170
|
- lib/picky/loggers/verbose.rb
|
171
|
+
- lib/picky/optimizers.rb
|
172
|
+
- lib/picky/optimizers/memory/array_deduplicator.rb
|
185
173
|
- lib/picky/performant.rb
|
186
174
|
- lib/picky/platforms/macruby.rb
|
187
175
|
- lib/picky/pool.rb
|
@@ -255,6 +243,7 @@ files:
|
|
255
243
|
- spec/functional/non_specific_ids_larger_than_20_spec.rb
|
256
244
|
- spec/functional/object_use_spec.rb
|
257
245
|
- spec/functional/only_spec.rb
|
246
|
+
- spec/functional/optimize_memory_spec.rb
|
258
247
|
- spec/functional/or_spec.rb
|
259
248
|
- spec/functional/pool_spec.rb
|
260
249
|
- spec/functional/range_queries_spec.rb
|
@@ -405,7 +394,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
405
394
|
version: '0'
|
406
395
|
requirements: []
|
407
396
|
rubyforge_project: http://rubyforge.org/projects/picky
|
408
|
-
rubygems_version: 2.4.
|
397
|
+
rubygems_version: 2.4.6
|
409
398
|
signing_key:
|
410
399
|
specification_version: 4
|
411
400
|
summary: 'Picky: Semantic Search Engine. Clever Interface. Good Tools.'
|
@@ -439,6 +428,7 @@ test_files:
|
|
439
428
|
- spec/functional/non_specific_ids_larger_than_20_spec.rb
|
440
429
|
- spec/functional/object_use_spec.rb
|
441
430
|
- spec/functional/only_spec.rb
|
431
|
+
- spec/functional/optimize_memory_spec.rb
|
442
432
|
- spec/functional/or_spec.rb
|
443
433
|
- spec/functional/pool_spec.rb
|
444
434
|
- spec/functional/range_queries_spec.rb
|