bk 0.0.1 → 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- data/Gemfile +1 -7
- data/README.md +58 -25
- data/lib/bk/version.rb +1 -1
- data/test/common.rb +4 -0
- data/test/test_building_tree.rb +1 -2
- data/test/test_import_and_export.rb +1 -2
- data/test/test_querying_tree.rb +2 -2
- metadata +9 -9
- data/test/test_all.rb +0 -3
data/Gemfile
CHANGED
data/README.md
CHANGED
@@ -1,57 +1,78 @@
|
|
1
1
|
# BK-Tree implementation in Ruby
|
2
2
|
|
3
|
-
|
3
|
+
BK-trees can be used to efficiently locate strings' best matches from within a
|
4
|
+
large set. If you don’t know what a BK-tree is, these links should provide a
|
5
|
+
good explanation and introduction.
|
4
6
|
|
5
7
|
* [Damn Cool Algorithms, Part 1: BK-Trees](http://blog.notdot.net/2007/4/Damn-Cool-Algorithms-Part-1-BK-Trees)
|
6
8
|
* [Fast Approximate String Matching in a Dictionary](http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.21.3317)
|
7
9
|
|
10
|
+
|
11
|
+
## Installation
|
12
|
+
|
13
|
+
BK is available as a Ruby gem:
|
14
|
+
|
15
|
+
gem install bk
|
16
|
+
|
8
17
|
## Usage
|
9
18
|
|
10
|
-
|
11
|
-
|
19
|
+
```ruby
|
20
|
+
require "bk"
|
21
|
+
tree = BK::Tree.new # Use the default Levenshtein distance algorithm
|
22
|
+
```
|
12
23
|
|
13
24
|
Add items to the tree:
|
14
25
|
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
26
|
+
```ruby
|
27
|
+
tree.add "cat"
|
28
|
+
tree.add "dog"
|
29
|
+
tree.add "monkey"
|
30
|
+
tree.add "donkey"
|
31
|
+
```
|
19
32
|
|
20
33
|
Find all items within distance 1 of ‘munkey’:
|
21
34
|
|
22
|
-
|
23
|
-
|
35
|
+
```ruby
|
36
|
+
tree.query("munkey", 1)
|
37
|
+
# => {"monkey"=>1}
|
38
|
+
```
|
24
39
|
|
25
40
|
Find all items within distance 2 of ‘munkey’:
|
26
41
|
|
27
|
-
|
28
|
-
|
42
|
+
```ruby
|
43
|
+
tree.query("munkey", 2)
|
44
|
+
# => {"donkey"=>2, "monkey"=>1}
|
45
|
+
```
|
29
46
|
|
30
47
|
You can specify a custom distance algorithm by passing an object that responds
|
31
48
|
to `call(a, b)` with a number:
|
32
49
|
|
33
|
-
|
34
|
-
|
35
|
-
|
50
|
+
```ruby
|
51
|
+
custom_algorithm = lambda{ |a, b|
|
52
|
+
Text::Levenshtein.distance(a, b)
|
53
|
+
}
|
36
54
|
|
37
|
-
|
55
|
+
# or, more tersely:
|
56
|
+
|
57
|
+
custom_algorithm = Text::Levenshtein.public_method(:distance)
|
58
|
+
|
59
|
+
tree = BK::Tree.new(custom_algorithm)
|
60
|
+
```
|
38
61
|
|
39
62
|
Note that the result *must* satisfy the
|
40
63
|
_triangle inequality_, i.e. _d(x,z) ≤ d(x,y) + d(y,z)_.
|
41
64
|
|
42
65
|
The precomputed tree can be exported to and reimported later from an IO-like object:
|
43
66
|
|
44
|
-
|
45
|
-
|
46
|
-
|
47
|
-
|
48
|
-
File.open("tree", "rb") do |f|
|
49
|
-
tree = BK::Tree.import(f)
|
50
|
-
end
|
67
|
+
```ruby
|
68
|
+
File.open("tree", "wb") do |f|
|
69
|
+
tree.export(f)
|
70
|
+
end
|
51
71
|
|
52
|
-
|
53
|
-
|
54
|
-
|
72
|
+
File.open("tree", "rb") do |f|
|
73
|
+
tree = BK::Tree.import(f)
|
74
|
+
end
|
75
|
+
```
|
55
76
|
|
56
77
|
## Performance
|
57
78
|
|
@@ -75,3 +96,15 @@ As the threshold increases, the benefit is reduced. At threshold 3:
|
|
75
96
|
|
76
97
|
* Memory usage: around 6 MB for a 20,000-word tree.
|
77
98
|
* Maximum tree depth is limited by the stack.
|
99
|
+
|
100
|
+
## Testing
|
101
|
+
|
102
|
+
rake test
|
103
|
+
|
104
|
+
...or, for specific tests:
|
105
|
+
|
106
|
+
ruby -Itest test/test_building_tree.rb
|
107
|
+
|
108
|
+
## Licensing
|
109
|
+
|
110
|
+
MIT (see COPYING.txt)
|
data/lib/bk/version.rb
CHANGED
data/test/common.rb
ADDED
data/test/test_building_tree.rb
CHANGED
data/test/test_querying_tree.rb
CHANGED
@@ -1,5 +1,4 @@
|
|
1
|
-
|
2
|
-
require 'test/unit'
|
1
|
+
require 'common'
|
3
2
|
require 'bk'
|
4
3
|
|
5
4
|
class BKTreeQueryAccuracyTest < Test::Unit::TestCase
|
@@ -33,6 +32,7 @@ class BKTreeSearchSpaceTest < Test::Unit::TestCase
|
|
33
32
|
def initialize
|
34
33
|
@history = []
|
35
34
|
@counting = false
|
35
|
+
@recording = false
|
36
36
|
end
|
37
37
|
|
38
38
|
def call(a, b)
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: bk
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0
|
4
|
+
version: 1.0.0
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -9,11 +9,11 @@ authors:
|
|
9
9
|
autorequire:
|
10
10
|
bindir: bin
|
11
11
|
cert_chain: []
|
12
|
-
date: 2012-
|
12
|
+
date: 2012-07-04 00:00:00.000000000Z
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: text
|
16
|
-
requirement: &
|
16
|
+
requirement: &2152922040 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,10 +21,10 @@ dependencies:
|
|
21
21
|
version: '0'
|
22
22
|
type: :runtime
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *2152922040
|
25
25
|
- !ruby/object:Gem::Dependency
|
26
|
-
name:
|
27
|
-
requirement: &
|
26
|
+
name: rake
|
27
|
+
requirement: &2152921620 !ruby/object:Gem::Requirement
|
28
28
|
none: false
|
29
29
|
requirements:
|
30
30
|
- - ! '>='
|
@@ -32,7 +32,7 @@ dependencies:
|
|
32
32
|
version: '0'
|
33
33
|
type: :development
|
34
34
|
prerelease: false
|
35
|
-
version_requirements: *
|
35
|
+
version_requirements: *2152921620
|
36
36
|
description: Burkhard Keller Tree implementation in Ruby
|
37
37
|
email: pbattley@gmail.com
|
38
38
|
executables: []
|
@@ -47,7 +47,7 @@ files:
|
|
47
47
|
- lib/bk.rb
|
48
48
|
- samples/graph.rb
|
49
49
|
- samples/performance.rb
|
50
|
-
- test/
|
50
|
+
- test/common.rb
|
51
51
|
- test/test_building_tree.rb
|
52
52
|
- test/test_import_and_export.rb
|
53
53
|
- test/test_querying_tree.rb
|
@@ -76,7 +76,7 @@ signing_key:
|
|
76
76
|
specification_version: 3
|
77
77
|
summary: Burkhard Keller Tree implementation in Ruby
|
78
78
|
test_files:
|
79
|
-
- test/
|
79
|
+
- test/common.rb
|
80
80
|
- test/test_building_tree.rb
|
81
81
|
- test/test_import_and_export.rb
|
82
82
|
- test/test_querying_tree.rb
|
data/test/test_all.rb
DELETED