fuzzy_match 1.3.1 → 1.3.2
Sign up to get free protection for your applications and to get access to all the features.
- data/Gemfile +12 -2
- data/History.txt +13 -0
- data/README.markdown +10 -6
- data/benchmark/before-with-free.txt +21 -21
- data/benchmark/before.txt +21 -21
- data/benchmark/memory.rb +6 -6
- data/examples/bts_aircraft/{blockings.csv → groupings.csv} +0 -0
- data/examples/bts_aircraft/test_bts_aircraft.rb +6 -6
- data/fuzzy_match.gemspec +1 -10
- data/lib/fuzzy_match.rb +41 -33
- data/lib/fuzzy_match/result.rb +1 -1
- data/lib/fuzzy_match/rule.rb +14 -0
- data/lib/fuzzy_match/rule/grouping.rb +32 -0
- data/lib/fuzzy_match/rule/identity.rb +19 -0
- data/lib/fuzzy_match/rule/normalizer.rb +20 -0
- data/lib/fuzzy_match/rule/stop_word.rb +11 -0
- data/lib/fuzzy_match/version.rb +1 -1
- data/test/helper.rb +3 -1
- data/test/test_fuzzy_match.rb +188 -124
- data/test/test_fuzzy_match_convoluted.rb.disabled +12 -12
- data/test/{test_blocking.rb → test_grouping.rb} +6 -6
- data/test/test_identity.rb +8 -8
- data/test/test_normalizer.rb +2 -2
- data/test/test_wrapper.rb +1 -1
- metadata +15 -101
- data/lib/fuzzy_match/blocking.rb +0 -36
- data/lib/fuzzy_match/identity.rb +0 -23
- data/lib/fuzzy_match/normalizer.rb +0 -28
- data/lib/fuzzy_match/stop_word.rb +0 -19
@@ -42,7 +42,7 @@ class TestFuzzyMatchConvoluted < MiniTest::Spec
|
|
42
42
|
]
|
43
43
|
@tightenings = []
|
44
44
|
@identities = []
|
45
|
-
@
|
45
|
+
@groupings = []
|
46
46
|
@positives = []
|
47
47
|
@negatives = []
|
48
48
|
end
|
@@ -55,23 +55,23 @@ class TestFuzzyMatchConvoluted < MiniTest::Spec
|
|
55
55
|
@_ltd ||= FuzzyMatch.new @haystack,
|
56
56
|
:tightenings => @tightenings,
|
57
57
|
:identities => @identities,
|
58
|
-
:
|
58
|
+
:groupings => @groupings,
|
59
59
|
:positives => @positives,
|
60
60
|
:negatives => @negatives,
|
61
|
-
:
|
61
|
+
:grouping_only => @grouping_only,
|
62
62
|
:log => $log
|
63
63
|
end
|
64
64
|
|
65
|
-
should "optionally only pay attention to things that match
|
65
|
+
should "optionally only pay attention to things that match groupings" do
|
66
66
|
assert_equal @a_haystack, ltd.improver.match(@a_needle)
|
67
67
|
|
68
68
|
clear_ltd
|
69
|
-
@
|
69
|
+
@grouping_only = true
|
70
70
|
assert_equal nil, ltd.improver.match(@a_needle)
|
71
71
|
|
72
72
|
clear_ltd
|
73
|
-
@
|
74
|
-
@
|
73
|
+
@grouping_only = true
|
74
|
+
@groupings.push ['/dash/i']
|
75
75
|
assert_equal @a_haystack, ltd.improver.match(@a_needle)
|
76
76
|
end
|
77
77
|
|
@@ -111,7 +111,7 @@ class TestFuzzyMatchConvoluted < MiniTest::Spec
|
|
111
111
|
end
|
112
112
|
end
|
113
113
|
|
114
|
-
should "have a false match without
|
114
|
+
should "have a false match without grouping" do
|
115
115
|
# @d_needle will be our victim
|
116
116
|
@haystack.push @d_lookalike
|
117
117
|
@tightenings.push @t_1
|
@@ -119,19 +119,19 @@ class TestFuzzyMatchConvoluted < MiniTest::Spec
|
|
119
119
|
assert_equal @d_lookalike, ltd.improver.match(@d_needle)
|
120
120
|
end
|
121
121
|
|
122
|
-
should "do
|
122
|
+
should "do grouping if the needle matches a group" do
|
123
123
|
# @d_needle will be our victim
|
124
124
|
@haystack.push @d_lookalike
|
125
125
|
@tightenings.push @t_1
|
126
|
-
@
|
126
|
+
@groupings.push ['/(bombardier|de ?havilland)/i']
|
127
127
|
|
128
128
|
assert_equal @d_haystack, ltd.improver.match(@d_needle)
|
129
129
|
end
|
130
130
|
|
131
|
-
should "treat
|
131
|
+
should "treat groups as exclusive" do
|
132
132
|
@haystack = [ @d_needle ]
|
133
133
|
@tightenings.push @t_1
|
134
|
-
@
|
134
|
+
@groupings.push ['/(bombardier|de ?havilland)/i']
|
135
135
|
|
136
136
|
assert_equal nil, ltd.improver.match(@d_lookalike)
|
137
137
|
end
|
@@ -1,28 +1,28 @@
|
|
1
1
|
require 'helper'
|
2
2
|
|
3
|
-
|
3
|
+
describe FuzzyMatch::Rule::Grouping do
|
4
4
|
it %{matches a single string argument} do
|
5
|
-
b = FuzzyMatch::
|
5
|
+
b = FuzzyMatch::Rule::Grouping.new %r{apple}
|
6
6
|
b.match?('2 apples').must_equal true
|
7
7
|
end
|
8
8
|
|
9
9
|
it %{embraces case insensitivity} do
|
10
|
-
b = FuzzyMatch::
|
10
|
+
b = FuzzyMatch::Rule::Grouping.new %r{apple}i
|
11
11
|
b.match?('2 Apples').must_equal true
|
12
12
|
end
|
13
13
|
|
14
14
|
it %{joins two string arguments} do
|
15
|
-
b = FuzzyMatch::
|
15
|
+
b = FuzzyMatch::Rule::Grouping.new %r{apple}
|
16
16
|
b.join?('apple', '2 apples').must_equal true
|
17
17
|
end
|
18
18
|
|
19
19
|
it %{fails to join two string arguments} do
|
20
|
-
b = FuzzyMatch::
|
20
|
+
b = FuzzyMatch::Rule::Grouping.new %r{apple}
|
21
21
|
b.join?('orange', '2 apples').must_equal false
|
22
22
|
end
|
23
23
|
|
24
24
|
it %{returns nil instead of false when it has no information} do
|
25
|
-
b = FuzzyMatch::
|
25
|
+
b = FuzzyMatch::Rule::Grouping.new %r{apple}
|
26
26
|
b.join?('orange', 'orange').must_be_nil
|
27
27
|
end
|
28
28
|
end
|
data/test/test_identity.rb
CHANGED
@@ -1,36 +1,36 @@
|
|
1
1
|
require 'helper'
|
2
2
|
|
3
|
-
|
3
|
+
describe FuzzyMatch::Rule::Identity do
|
4
4
|
it %{determines whether two records COULD be identical} do
|
5
|
-
i = FuzzyMatch::Identity.new %r{(A)[ ]*(\d)}
|
5
|
+
i = FuzzyMatch::Rule::Identity.new %r{(A)[ ]*(\d)}
|
6
6
|
i.identical?('A1', 'A 1foobar').must_equal true
|
7
7
|
end
|
8
8
|
|
9
9
|
it %{determines that two records MUST NOT be identical} do
|
10
|
-
i = FuzzyMatch::Identity.new %r{(A)[ ]*(\d)}
|
10
|
+
i = FuzzyMatch::Rule::Identity.new %r{(A)[ ]*(\d)}
|
11
11
|
i.identical?('A1', 'A 2foobar').must_equal false
|
12
12
|
end
|
13
13
|
|
14
14
|
it %{returns nil indicating no information} do
|
15
|
-
i = FuzzyMatch::Identity.new %r{(A)[ ]*(\d)}
|
15
|
+
i = FuzzyMatch::Rule::Identity.new %r{(A)[ ]*(\d)}
|
16
16
|
i.identical?('B1', 'A 2foobar').must_equal nil
|
17
17
|
end
|
18
18
|
|
19
19
|
it %{can be initialized with a regexp} do
|
20
|
-
i = FuzzyMatch::Identity.new %r{\A\\?/(.*)etc/mysql\$$}
|
20
|
+
i = FuzzyMatch::Rule::Identity.new %r{\A\\?/(.*)etc/mysql\$$}
|
21
21
|
i.regexp.must_equal %r{\A\\?/(.*)etc/mysql\$$}
|
22
22
|
end
|
23
23
|
|
24
24
|
it %{can be initialized from a string (via to_regexp gem)} do
|
25
|
-
i = FuzzyMatch::Identity.new '%r{\A\\\?/(.*)etc/mysql\$$}'
|
25
|
+
i = FuzzyMatch::Rule::Identity.new '%r{\A\\\?/(.*)etc/mysql\$$}'
|
26
26
|
i.regexp.must_equal %r{\A\\?/(.*)etc/mysql\$$}
|
27
27
|
|
28
|
-
i = FuzzyMatch::Identity.new '/\A\\\?\/(.*)etc\/mysql\$$/'
|
28
|
+
i = FuzzyMatch::Rule::Identity.new '/\A\\\?\/(.*)etc\/mysql\$$/'
|
29
29
|
i.regexp.must_equal %r{\A\\?/(.*)etc/mysql\$$}
|
30
30
|
end
|
31
31
|
|
32
32
|
it %{embraces case insensitivity} do
|
33
|
-
i = FuzzyMatch::Identity.new %r{(A)[ ]*(\d)}i
|
33
|
+
i = FuzzyMatch::Rule::Identity.new %r{(A)[ ]*(\d)}i
|
34
34
|
i.identical?('A1', 'a 1foobar').must_equal true
|
35
35
|
end
|
36
36
|
end
|
data/test/test_normalizer.rb
CHANGED
@@ -1,8 +1,8 @@
|
|
1
1
|
require 'helper'
|
2
2
|
|
3
|
-
|
3
|
+
describe FuzzyMatch::Rule::Normalizer do
|
4
4
|
it %{applies itself to a string argument} do
|
5
|
-
t = FuzzyMatch::Normalizer.new %r{(Ford )[ ]*(F)[\- ]*(\d\d\d)}i
|
5
|
+
t = FuzzyMatch::Rule::Normalizer.new %r{(Ford )[ ]*(F)[\- ]*(\d\d\d)}i
|
6
6
|
t.apply('Ford F-350').must_equal 'Ford F350'
|
7
7
|
t.apply('Ford F150').must_equal 'Ford F150'
|
8
8
|
t.apply('Ford F 350').must_equal 'Ford F350'
|
data/test/test_wrapper.rb
CHANGED
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: fuzzy_match
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.3.
|
4
|
+
version: 1.3.2
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -9,99 +9,11 @@ authors:
|
|
9
9
|
autorequire:
|
10
10
|
bindir: bin
|
11
11
|
cert_chain: []
|
12
|
-
date: 2012-02-
|
12
|
+
date: 2012-02-24 00:00:00.000000000 Z
|
13
13
|
dependencies:
|
14
|
-
- !ruby/object:Gem::Dependency
|
15
|
-
name: minitest
|
16
|
-
requirement: &2165315840 !ruby/object:Gem::Requirement
|
17
|
-
none: false
|
18
|
-
requirements:
|
19
|
-
- - ! '>='
|
20
|
-
- !ruby/object:Gem::Version
|
21
|
-
version: '0'
|
22
|
-
type: :development
|
23
|
-
prerelease: false
|
24
|
-
version_requirements: *2165315840
|
25
|
-
- !ruby/object:Gem::Dependency
|
26
|
-
name: activerecord
|
27
|
-
requirement: &2165314740 !ruby/object:Gem::Requirement
|
28
|
-
none: false
|
29
|
-
requirements:
|
30
|
-
- - ! '>='
|
31
|
-
- !ruby/object:Gem::Version
|
32
|
-
version: '3'
|
33
|
-
type: :development
|
34
|
-
prerelease: false
|
35
|
-
version_requirements: *2165314740
|
36
|
-
- !ruby/object:Gem::Dependency
|
37
|
-
name: mysql2
|
38
|
-
requirement: &2165314140 !ruby/object:Gem::Requirement
|
39
|
-
none: false
|
40
|
-
requirements:
|
41
|
-
- - ! '>='
|
42
|
-
- !ruby/object:Gem::Version
|
43
|
-
version: '0'
|
44
|
-
type: :development
|
45
|
-
prerelease: false
|
46
|
-
version_requirements: *2165314140
|
47
|
-
- !ruby/object:Gem::Dependency
|
48
|
-
name: cohort_scope
|
49
|
-
requirement: &2165313620 !ruby/object:Gem::Requirement
|
50
|
-
none: false
|
51
|
-
requirements:
|
52
|
-
- - ! '>='
|
53
|
-
- !ruby/object:Gem::Version
|
54
|
-
version: '0'
|
55
|
-
type: :development
|
56
|
-
prerelease: false
|
57
|
-
version_requirements: *2165313620
|
58
|
-
- !ruby/object:Gem::Dependency
|
59
|
-
name: weighted_average
|
60
|
-
requirement: &2165312980 !ruby/object:Gem::Requirement
|
61
|
-
none: false
|
62
|
-
requirements:
|
63
|
-
- - ! '>='
|
64
|
-
- !ruby/object:Gem::Version
|
65
|
-
version: '0'
|
66
|
-
type: :development
|
67
|
-
prerelease: false
|
68
|
-
version_requirements: *2165312980
|
69
|
-
- !ruby/object:Gem::Dependency
|
70
|
-
name: rake
|
71
|
-
requirement: &2165312300 !ruby/object:Gem::Requirement
|
72
|
-
none: false
|
73
|
-
requirements:
|
74
|
-
- - ! '>='
|
75
|
-
- !ruby/object:Gem::Version
|
76
|
-
version: '0'
|
77
|
-
type: :development
|
78
|
-
prerelease: false
|
79
|
-
version_requirements: *2165312300
|
80
|
-
- !ruby/object:Gem::Dependency
|
81
|
-
name: yard
|
82
|
-
requirement: &2165311320 !ruby/object:Gem::Requirement
|
83
|
-
none: false
|
84
|
-
requirements:
|
85
|
-
- - ! '>='
|
86
|
-
- !ruby/object:Gem::Version
|
87
|
-
version: '0'
|
88
|
-
type: :development
|
89
|
-
prerelease: false
|
90
|
-
version_requirements: *2165311320
|
91
|
-
- !ruby/object:Gem::Dependency
|
92
|
-
name: amatch
|
93
|
-
requirement: &2165310720 !ruby/object:Gem::Requirement
|
94
|
-
none: false
|
95
|
-
requirements:
|
96
|
-
- - ! '>='
|
97
|
-
- !ruby/object:Gem::Version
|
98
|
-
version: '0'
|
99
|
-
type: :development
|
100
|
-
prerelease: false
|
101
|
-
version_requirements: *2165310720
|
102
14
|
- !ruby/object:Gem::Dependency
|
103
15
|
name: activesupport
|
104
|
-
requirement: &
|
16
|
+
requirement: &2151876580 !ruby/object:Gem::Requirement
|
105
17
|
none: false
|
106
18
|
requirements:
|
107
19
|
- - ! '>='
|
@@ -109,10 +21,10 @@ dependencies:
|
|
109
21
|
version: '3'
|
110
22
|
type: :runtime
|
111
23
|
prerelease: false
|
112
|
-
version_requirements: *
|
24
|
+
version_requirements: *2151876580
|
113
25
|
- !ruby/object:Gem::Dependency
|
114
26
|
name: to_regexp
|
115
|
-
requirement: &
|
27
|
+
requirement: &2151871760 !ruby/object:Gem::Requirement
|
116
28
|
none: false
|
117
29
|
requirements:
|
118
30
|
- - ! '>='
|
@@ -120,7 +32,7 @@ dependencies:
|
|
120
32
|
version: 0.0.3
|
121
33
|
type: :runtime
|
122
34
|
prerelease: false
|
123
|
-
version_requirements: *
|
35
|
+
version_requirements: *2151871760
|
124
36
|
description: Find a needle in a haystack using string similarity and (optionally)
|
125
37
|
regexp rules. Replaces loose_tight_dictionary.
|
126
38
|
email:
|
@@ -132,6 +44,7 @@ files:
|
|
132
44
|
- .document
|
133
45
|
- .gitignore
|
134
46
|
- Gemfile
|
47
|
+
- History.txt
|
135
48
|
- LICENSE
|
136
49
|
- README.markdown
|
137
50
|
- Rakefile
|
@@ -145,7 +58,7 @@ files:
|
|
145
58
|
- examples/bts_aircraft/5-2-D.htm
|
146
59
|
- examples/bts_aircraft/5-2-E.htm
|
147
60
|
- examples/bts_aircraft/5-2-G.htm
|
148
|
-
- examples/bts_aircraft/
|
61
|
+
- examples/bts_aircraft/groupings.csv
|
149
62
|
- examples/bts_aircraft/identities.csv
|
150
63
|
- examples/bts_aircraft/negatives.csv
|
151
64
|
- examples/bts_aircraft/normalizers.csv
|
@@ -156,24 +69,25 @@ files:
|
|
156
69
|
- examples/icao-bts.xls
|
157
70
|
- fuzzy_match.gemspec
|
158
71
|
- lib/fuzzy_match.rb
|
159
|
-
- lib/fuzzy_match/blocking.rb
|
160
72
|
- lib/fuzzy_match/cached_result.rb
|
161
|
-
- lib/fuzzy_match/identity.rb
|
162
|
-
- lib/fuzzy_match/normalizer.rb
|
163
73
|
- lib/fuzzy_match/result.rb
|
74
|
+
- lib/fuzzy_match/rule.rb
|
75
|
+
- lib/fuzzy_match/rule/grouping.rb
|
76
|
+
- lib/fuzzy_match/rule/identity.rb
|
77
|
+
- lib/fuzzy_match/rule/normalizer.rb
|
78
|
+
- lib/fuzzy_match/rule/stop_word.rb
|
164
79
|
- lib/fuzzy_match/score.rb
|
165
80
|
- lib/fuzzy_match/score/amatch.rb
|
166
81
|
- lib/fuzzy_match/score/pure_ruby.rb
|
167
82
|
- lib/fuzzy_match/similarity.rb
|
168
|
-
- lib/fuzzy_match/stop_word.rb
|
169
83
|
- lib/fuzzy_match/version.rb
|
170
84
|
- lib/fuzzy_match/wrapper.rb
|
171
85
|
- test/helper.rb
|
172
86
|
- test/test_amatch.rb
|
173
|
-
- test/test_blocking.rb
|
174
87
|
- test/test_cache.rb
|
175
88
|
- test/test_fuzzy_match.rb
|
176
89
|
- test/test_fuzzy_match_convoluted.rb.disabled
|
90
|
+
- test/test_grouping.rb
|
177
91
|
- test/test_identity.rb
|
178
92
|
- test/test_normalizer.rb
|
179
93
|
- test/test_wrapper.rb
|
@@ -205,10 +119,10 @@ summary: Find a needle in a haystack using string similarity and (optionally) re
|
|
205
119
|
test_files:
|
206
120
|
- test/helper.rb
|
207
121
|
- test/test_amatch.rb
|
208
|
-
- test/test_blocking.rb
|
209
122
|
- test/test_cache.rb
|
210
123
|
- test/test_fuzzy_match.rb
|
211
124
|
- test/test_fuzzy_match_convoluted.rb.disabled
|
125
|
+
- test/test_grouping.rb
|
212
126
|
- test/test_identity.rb
|
213
127
|
- test/test_normalizer.rb
|
214
128
|
- test/test_wrapper.rb
|
data/lib/fuzzy_match/blocking.rb
DELETED
@@ -1,36 +0,0 @@
|
|
1
|
-
class FuzzyMatch
|
2
|
-
# "Record linkage typically involves two main steps: blocking and scoring..."
|
3
|
-
# http://en.wikipedia.org/wiki/Record_linkage
|
4
|
-
#
|
5
|
-
# Blockings effectively divide up the haystack into groups that match a pattern
|
6
|
-
#
|
7
|
-
# A blocking (as in a grouping) comes into effect when a str matches.
|
8
|
-
# Then the needle must also match the blocking's regexp.
|
9
|
-
class Blocking
|
10
|
-
attr_reader :regexp
|
11
|
-
|
12
|
-
def initialize(regexp_or_str)
|
13
|
-
@regexp = regexp_or_str.to_regexp
|
14
|
-
end
|
15
|
-
|
16
|
-
def match?(str)
|
17
|
-
!!(regexp.match(str))
|
18
|
-
end
|
19
|
-
|
20
|
-
# If a blocking "joins" two strings, that means they both fit into it.
|
21
|
-
#
|
22
|
-
# Returns false if they certainly don't fit this blocking.
|
23
|
-
# Returns nil if the blocking doesn't apply, i.e. str2 doesn't fit the blocking.
|
24
|
-
def join?(str1, str2)
|
25
|
-
if str2_match_data = regexp.match(str2)
|
26
|
-
if str1_match_data = regexp.match(str1)
|
27
|
-
str2_match_data.captures.join.downcase == str1_match_data.captures.join.downcase
|
28
|
-
else
|
29
|
-
false
|
30
|
-
end
|
31
|
-
else
|
32
|
-
nil
|
33
|
-
end
|
34
|
-
end
|
35
|
-
end
|
36
|
-
end
|
data/lib/fuzzy_match/identity.rb
DELETED
@@ -1,23 +0,0 @@
|
|
1
|
-
class FuzzyMatch
|
2
|
-
# Identities take effect when needle and haystack both match a regexp
|
3
|
-
# Then the captured part of the regexp has to match exactly
|
4
|
-
class Identity
|
5
|
-
attr_reader :regexp
|
6
|
-
|
7
|
-
def initialize(regexp_or_str)
|
8
|
-
@regexp = regexp_or_str.to_regexp
|
9
|
-
end
|
10
|
-
|
11
|
-
# Two strings are "identical" if they both match this identity and the captures are equal.
|
12
|
-
#
|
13
|
-
# Only returns true/false if both strings match the regexp.
|
14
|
-
# Otherwise returns nil.
|
15
|
-
def identical?(str1, str2)
|
16
|
-
if str1_match_data = regexp.match(str1) and match_data = regexp.match(str2)
|
17
|
-
str1_match_data.captures.join.downcase == match_data.captures.join.downcase
|
18
|
-
else
|
19
|
-
nil
|
20
|
-
end
|
21
|
-
end
|
22
|
-
end
|
23
|
-
end
|
@@ -1,28 +0,0 @@
|
|
1
|
-
class FuzzyMatch
|
2
|
-
# A normalizer just strips a string down to its core
|
3
|
-
class Normalizer
|
4
|
-
attr_reader :regexp
|
5
|
-
|
6
|
-
def initialize(regexp_or_str)
|
7
|
-
@regexp = regexp_or_str.to_regexp
|
8
|
-
end
|
9
|
-
|
10
|
-
# A normalizer applies when its regexp matches and captures a new (shorter) string
|
11
|
-
def apply?(str)
|
12
|
-
!!(regexp.match(str))
|
13
|
-
end
|
14
|
-
|
15
|
-
# The result of applying a normalizer is just all the captures put together.
|
16
|
-
def apply(str)
|
17
|
-
if match_data = regexp.match(str)
|
18
|
-
match_data.captures.join
|
19
|
-
else
|
20
|
-
str
|
21
|
-
end
|
22
|
-
end
|
23
|
-
|
24
|
-
def inspect
|
25
|
-
"#<FuzzyMatch::Normalizer regexp=#{regexp.inspect}>"
|
26
|
-
end
|
27
|
-
end
|
28
|
-
end
|