text_alignment 0.2.6 → 0.2.7
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile +2 -1
- data/Gemfile.lock +3 -1
- data/bin/align_annotations +4 -0
- data/lib/text_alignment/approximate_fit.rb +31 -17
- data/lib/text_alignment/version.rb +1 -1
- data/text_alignment.gemspec +1 -0
- metadata +17 -3
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 45bf8e55a66daff4d79765aec0e4d7482151aa7b86749e054e5d93030ffffabe
|
4
|
+
data.tar.gz: 6251835e35077e2ae828666026a663ae63be7f8d7b7d785b4898763021f546e2
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 4750b2507b4eb172123ca9a82320c2456e66d3f06ca3e3bdd52cf31055e9cb46af53013a484dbb87b1bb3236701cc8a69e1c4c64b576ee41a3a4957c3431be65
|
7
|
+
data.tar.gz: 1819b94e47bdf064d53712bb667f15d74f8e37bd49df7976075c3cbdc0a992d25bb318f76bfc0522cea0b594d002bcab0ebcef6112bf9e26128c245a7fb67bcf
|
data/Gemfile
CHANGED
data/Gemfile.lock
CHANGED
@@ -15,6 +15,7 @@ GEM
|
|
15
15
|
rspec-support (~> 3.0.0)
|
16
16
|
rspec-support (3.0.4)
|
17
17
|
ruby-dictionary (1.1.1)
|
18
|
+
string-similarity (2.1.0)
|
18
19
|
|
19
20
|
PLATFORMS
|
20
21
|
ruby
|
@@ -23,9 +24,10 @@ DEPENDENCIES
|
|
23
24
|
diff-lcs (~> 1.3)
|
24
25
|
rspec (~> 3.0)
|
25
26
|
ruby-dictionary (~> 1.1, >= 1.1.1)
|
27
|
+
string-similarity (~> 2.1)
|
26
28
|
|
27
29
|
RUBY VERSION
|
28
|
-
ruby 2.
|
30
|
+
ruby 2.5.5p157
|
29
31
|
|
30
32
|
BUNDLED WITH
|
31
33
|
1.17.3
|
data/bin/align_annotations
CHANGED
@@ -42,6 +42,10 @@ puts
|
|
42
42
|
puts '[Denotations transformed]'
|
43
43
|
new_denotations = align.transform_hdenotations(denotations)
|
44
44
|
pp new_denotations
|
45
|
+
puts
|
46
|
+
puts '[Annotations transformed]'
|
47
|
+
anns2[:denotations] = new_denotations
|
48
|
+
puts anns2.to_json
|
45
49
|
|
46
50
|
# p align.common_elements
|
47
51
|
# puts "---------------"
|
@@ -1,11 +1,14 @@
|
|
1
1
|
#!/usr/bin/env ruby
|
2
|
+
require 'string-similarity'
|
3
|
+
|
2
4
|
module TextAlignment; end unless defined? TextAlignment
|
3
5
|
|
4
6
|
# approximate the location of str1 in str2
|
5
7
|
module TextAlignment
|
6
8
|
SIGNATURE_NGRAM = 5
|
7
9
|
MIN_LENGTH_FOR_APPROXIMATION = 50
|
8
|
-
BUFFER_RATE = 0.
|
10
|
+
BUFFER_RATE = 0.1
|
11
|
+
TEXT_SIMILARITY_TRESHOLD = 0.8
|
9
12
|
end
|
10
13
|
|
11
14
|
class << TextAlignment
|
@@ -22,29 +25,40 @@ class << TextAlignment
|
|
22
25
|
# If there is no shared n-gram found, it may mean there is no serious overlap between the two strings
|
23
26
|
return nil, nil if ngram_shared.empty?
|
24
27
|
|
25
|
-
|
26
|
-
|
28
|
+
signature_ngrams = ngram_shared.select{|g| ngram2.count(g) == 1}
|
29
|
+
return nil, nil if signature_ngrams.empty? #raise "no signature ngram"
|
30
|
+
|
31
|
+
fit_begin, fit_end = nil, nil
|
32
|
+
signature_ngrams.each do |signature_ngram|
|
33
|
+
loc_signature_ngram_in_str1 = str1.index(signature_ngram)
|
34
|
+
loc_signature_ngram_in_str2 = str2.index(signature_ngram)
|
27
35
|
|
28
|
-
|
29
|
-
|
30
|
-
|
31
|
-
fit_begin = 0 if fit_begin < 0
|
36
|
+
# approximate the beginning of the fit
|
37
|
+
fit_begin = loc_signature_ngram_in_str2 - loc_signature_ngram_in_str1 - (loc_signature_ngram_in_str1 * TextAlignment::BUFFER_RATE).to_i
|
38
|
+
fit_begin = 0 if fit_begin < 0
|
32
39
|
|
33
|
-
|
34
|
-
|
40
|
+
# approximate the end of the fit
|
41
|
+
offset_end = str1.length - loc_signature_ngram_in_str1
|
42
|
+
fit_end = loc_signature_ngram_in_str2 + offset_end + (offset_end * TextAlignment::BUFFER_RATE).to_i
|
43
|
+
fit_end = str2.length if fit_end > str2.length
|
35
44
|
|
36
|
-
|
37
|
-
|
38
|
-
|
39
|
-
|
40
|
-
return nil, nil if signature_ngram.nil? # raise "no signature ngram"
|
41
|
-
offset = str1.length - str1.rindex(signature_ngram)
|
42
|
-
fit_end = str2.rindex(signature_ngram) + offset + (offset * TextAlignment::BUFFER_RATE).to_i
|
43
|
-
fit_end = str2.length if fit_end > str2.length
|
45
|
+
text_similarity = text_similarity(str1, str2[fit_begin ... fit_end])
|
46
|
+
break if text_similarity > TextAlignment::TEXT_SIMILARITY_TRESHOLD
|
47
|
+
fit_begin, fit_end = nil, nil
|
48
|
+
end
|
44
49
|
|
45
50
|
return nil, nil if fit_begin >= fit_end
|
46
51
|
return fit_begin, fit_end
|
47
52
|
end
|
53
|
+
|
54
|
+
private
|
55
|
+
|
56
|
+
def text_similarity(str1, str2, ngram_order = 3)
|
57
|
+
_str1 = str1.delete(" \t\r\n")
|
58
|
+
_str2 = str2.delete(" \t\r\n")
|
59
|
+
String::Similarity.cosine(_str1, _str2, ngram:2)
|
60
|
+
end
|
61
|
+
|
48
62
|
end
|
49
63
|
|
50
64
|
if __FILE__ == $0
|
data/text_alignment.gemspec
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: text_alignment
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.2.
|
4
|
+
version: 0.2.7
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Jin-Dong Kim
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date:
|
11
|
+
date: 2020-07-02 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: ruby-dictionary
|
@@ -30,6 +30,20 @@ dependencies:
|
|
30
30
|
- - ">="
|
31
31
|
- !ruby/object:Gem::Version
|
32
32
|
version: 1.1.1
|
33
|
+
- !ruby/object:Gem::Dependency
|
34
|
+
name: string-similarity
|
35
|
+
requirement: !ruby/object:Gem::Requirement
|
36
|
+
requirements:
|
37
|
+
- - "~>"
|
38
|
+
- !ruby/object:Gem::Version
|
39
|
+
version: '2.1'
|
40
|
+
type: :development
|
41
|
+
prerelease: false
|
42
|
+
version_requirements: !ruby/object:Gem::Requirement
|
43
|
+
requirements:
|
44
|
+
- - "~>"
|
45
|
+
- !ruby/object:Gem::Version
|
46
|
+
version: '2.1'
|
33
47
|
- !ruby/object:Gem::Dependency
|
34
48
|
name: rspec
|
35
49
|
requirement: !ruby/object:Gem::Requirement
|
@@ -98,7 +112,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
98
112
|
- !ruby/object:Gem::Version
|
99
113
|
version: '0'
|
100
114
|
requirements: []
|
101
|
-
rubygems_version: 3.0.
|
115
|
+
rubygems_version: 3.0.8
|
102
116
|
signing_key:
|
103
117
|
specification_version: 4
|
104
118
|
summary: Ruby class for aligning two character strings
|