dwc_agent 0.1.15 → 0.1.16
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/dwc_agent/cleaner.rb +2 -2
- data/lib/dwc_agent/constants.rb +10 -2
- data/lib/dwc_agent/parser.rb +1 -0
- data/lib/dwc_agent/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 42cdbedccdf7273ea32dba89ec741e732d743b2c
|
4
|
+
data.tar.gz: 3b547d5836201c1f0ac2e63ef48b07abdda15205
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 59de6c725b3df8667b4f856da723ae92f228a2e37f238f98a2199a0b846fcfcd2ceced0be1f52f868eee294053370ddb0e19592666d10858e7d80e9f37d2bf77
|
7
|
+
data.tar.gz: 789f96f530c38b47928cf1935c23d058aae3a2214da296dbe0d5cd385f8648109eb6d5c319839af99ef6c440db39807b88580a39e2a8bb70b517d2dae2dfb03f
|
data/lib/dwc_agent/cleaner.rb
CHANGED
@@ -48,12 +48,12 @@ module DwcAgent
|
|
48
48
|
(parsed_namae.given == parsed_namae.given.upcase ||
|
49
49
|
parsed_namae.given == parsed_namae.given.downcase) &&
|
50
50
|
!parsed_namae.given.include?(".") &&
|
51
|
-
parsed_namae.given.length >= 4
|
51
|
+
parsed_namae.given.gsub(".","").length >= 4
|
52
52
|
parsed_namae.given = CapitalizeNames.capitalize(parsed_namae.given)
|
53
53
|
end
|
54
54
|
|
55
55
|
if parsed_namae.given && /[A-Za-z]\./.match(parsed_namae.given)
|
56
|
-
parsed_namae.given = CapitalizeNames.capitalize(parsed_namae.given).
|
56
|
+
parsed_namae.given = CapitalizeNames.capitalize(parsed_namae.given).gsub(/[a-z]\./, &:upcase)
|
57
57
|
end
|
58
58
|
|
59
59
|
parsed_namae.normalize_initials
|
data/lib/dwc_agent/constants.rb
CHANGED
@@ -2,7 +2,7 @@ module DwcAgent
|
|
2
2
|
STRIP_OUT = %r{
|
3
3
|
\s*?\d+\.\d+|
|
4
4
|
\b\d+\(?(?i:[[:alpha:]])\)?\b|
|
5
|
-
\b[,;]?\s*(?i:et
|
5
|
+
\b[,;]?\s*(?i:et\.?\s+al)\.?|
|
6
6
|
\bu\.\s*a\.|
|
7
7
|
\b[,;]?\s*(?i:and|&)?\s*(?i:others)\s*\b|
|
8
8
|
\b[,;]?\s*(?i:etc)\.?|
|
@@ -127,7 +127,15 @@ module DwcAgent
|
|
127
127
|
'>' => '',
|
128
128
|
'<' => '',
|
129
129
|
'{' => '',
|
130
|
-
'}' => ''
|
130
|
+
'}' => '',
|
131
|
+
'@' => '',
|
132
|
+
'%' => ''
|
133
|
+
}
|
134
|
+
|
135
|
+
PHRASE_SUBS = {
|
136
|
+
'Dr\.' => 'Dr. ',
|
137
|
+
'Mr\.' => 'Mr. ',
|
138
|
+
'Mrs\.' => 'Mrs. '
|
131
139
|
}
|
132
140
|
|
133
141
|
COMPLEX_SEPARATORS = %r{
|
data/lib/dwc_agent/parser.rb
CHANGED
@@ -16,6 +16,7 @@ module DwcAgent
|
|
16
16
|
residual_terminators_regex = Regexp.new SPLIT_BY.to_s + %r{\s*\z}.to_s
|
17
17
|
cleaned = name.gsub(STRIP_OUT, ' ')
|
18
18
|
.gsub(/[#{CHAR_SUBS.keys.join('\\')}]/, CHAR_SUBS)
|
19
|
+
.gsub(/(#{PHRASE_SUBS.keys.join('|')})/, PHRASE_SUBS)
|
19
20
|
.gsub(/([A-Z]{1}\.)([[:alpha:]]{2,})/, '\1 \2')
|
20
21
|
.gsub(COMPLEX_SEPARATORS, '\1 | \2')
|
21
22
|
.gsub(residual_terminators_regex, '')
|
data/lib/dwc_agent/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: dwc_agent
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.16
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- David P. Shorthouse
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2019-01-
|
11
|
+
date: 2019-01-17 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: namae
|